Home

HPSPICE MOSFET Models

1. model nch nmos Level 49 Tnom 27 0 nch 1 024685E 17 tox 1 00000E 08 xj 1 00000E 07 lint 3 75860E 08 wint 2 02101528644562E 07 vthO 6094574 k1 5341038 k2 1 703463E 03 k3 17 24589 dvt0 1767506 dvt1 5109418 dvt2 2 0 05 nlx 9 979638E 08 w0 le 6 k3b 4 139039 vsat 97662 05 ua 1 748481E 09 ub 3 178541E 18 uc 1 3623e 10 rdsw 298 873 u0 307 2991 prwb 2 24e 4 a0 4976366 keta 2 195445E 02 al 0332883 a2 9 voff 9 623903E 02 nFactor 8408191 cit 3 994609E 04 cdsc 1 130797E 04 cdscb 2 4e 5 eta0 0145072 etab 3 870303E 03 dsub 4116711 pclm 1 813153 pdiblcl 2 003703E 02 pdiblc2 00129051 pdiblcb 1 034e 3 drout 4380235 pscbel 5 752058E 08 pscbe2 7 510319E 05 pvag 6370527 prt 68 7 ngate 1 e20 alpha0 1 e 7 beta0 28 4 prwg 0 001 ags 1 2 dvt0w20 58 dvtlw 5 3e6 dvt2w 0 0032 kt1 3 kt2 03 at 33000 ute 1 5 ual 4 31E 09 ubl 7 61E 18 ucl 2 2 378e 10 ktll 1e 8 wr b0 1e 7 bl le 7 dwg 5e 8 dwb 2e 8 delta 0 015 cgdl le 10 cgsl le 10 cgbo le 10 xpart 0 0 cgdo 0 4e 9 cgso 0 4e 9 432 HSPICE MOSFET Models Manual X 2005 09 cle 0 6 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models clc20 1e 6 ckappa 0 6 PMOS Model This is an example of a PMOS model for the Level 49 MOSFET VTHO is negative model pch PMOS Level 49 ktl 3 t 33000 te 1 5 Tn
2. pslot Igbmod MOS101model pmodel MOS101igbMod pslot Igcmod MOS101model pmodel MOS101igcMog pslot Igso ptran MOS101Igs pslot Igdo ptran gt MOS101Igd pslot Igb ptran gt MOS101Igb pslot Igcs ptran MOS101Igcs pslot Igcd ptran MOS101IgcG9 pslot gigsos ptran MOS101gIgss pslot gigsog ptran MOS101gIgsg pslot gigcsg ptran MOS101gIgcsg pslot gigcsd ptran MOS101gIgcsd pslot gigcsb ptran MOS101gIgcsb pslot gigcss ptran MOS101gIgcss pslot gigdod ptran MOS101gIgdd pslot gigdog ptran MOS101gIgdg pslot gigcdg ptran MOS101gIgcdg pslot gigcdd ptran gt MOS101lgIgcdd pslot gigcdb ptran MOS101gIgcdb pslot gigcds ptran MOS101gIgcds pslot gigbg ptran MOS101gIgbg pslot gigbd ptran MOS101gIgbd pslot gigbb ptran MOS101gIgbb pslot gigbs ptran MOS101gIigbs TOPO101 of topo101 h is accessible through a pointer to topvar in CMI_VAR of CMldef h for example in file extcmi mos101 CMlmos101 eval c struct TOPO101 topol01 struct TOPO101 pslot topovar By doing this additional eletrical biases can be accessed using TOPO101 member variables vgms vdbs vsbs vdes vses qdef and so forth All quantities of TOPO101 are calculated and assigned within the evaluation function For a detailed application see CMII101evaluate and MOS101EvalNoise in example file
3. o T o N o w o A o Volts Lin CAPOP 2 Parameterized Modified Meyer Capacitance The CAPOP 2 Meyer capacitance model is the more general form of Meyer capacitance The CAPOP 1 Meyer capacitance model is the special case of CAPOP 2 if CF1 0 CF2 0 1 and CF3 1 In the following equations G G D and D are smooth factors You cannot change the values of these parameters Definition cap COXscaled Weff Leff Gate Bulk Capacitance cgb Accumulation vgs lt vfb vsb cgb cap Depletion vgs x vth cgs NERONE NER NES vgs vsb vfb 7 1442 GAMMA HSPICE MOSFET Models Manual 83 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models 84 Inversion vgs gt vth Gt cap 4 GAMMA PHI vsb 2 4 PHI SE GAMMA cgb These equations replace GAMMA with effective Y for model levels higher than 4 Gate Source Capacitance cgs Low vds vds lt 0 1 Accumulation vgs lt vth CF1 cgs CF5 cap G D Depletion vgs lt vth CF2 CF1 aa vgs vth CFl Ei vds Ng B cgs CFS cap SEH CE L CF2 2 CF2 vds D 2 Strong Inversion vgs gt vth max CF2 CF1 CF3 vds UPDATE 0 Strong Inversion vgs gt vth CF2 CF1 UPDATE 1 cgs CFS cap 1 T vgs vth CFl vds Y 2 vges vth CF1 vds High vds Vas 2 0 1 Accumulation vgs vth CF1 cgs CF3 cap G Dt CF1 0 cgs CF
4. Refer to extcmi enhancements STARTKS0078251 for testcode and testcases An Extension to Support BSIM4 Topology HSPICE provides device modeling capability for MOSFET models with BSIM4 topology through a Generalized Customer CMI To use this capability add the cmiflag option to the netlist to load the dynamically linked CMI library The topology ID number 101 is assigned internally for MOSFET models developed through this Generalized Customer CMI This differs from the existing Customer CMI which uses topology ID number 0 Therefore you must assign the topology ID number topoid to 101 in the CMI_MOSDEF structure located in your CMIZZZ c file where is the model name To better understand BSIM4 topology Figure 45 illustrates that with settings of RGATEMOD 3 RBODYMOD 1 and TRNQSMOD 1 virtually every circuit can be modeled 556 HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Extended Topology Figure 45 High level Equivalent Circuits for BSIM4 Model xg o S zo CC mgo g lg zo ig 6 F Rs fs V Rd fd V xs OAM 9 WO xd O ngs In this figure RBPB RBPD RBPS RBDB and RBSB constitute the substrate resistance network R_Gvyettg is gate electrode resistance R_Gycrg is channel reflected gate resistance Rd Rs consists of both electrode diffusion drain source resistance and drain Source bias dependent resistance LDD
5. VSH s y T pra leff COX Leff FDS UFDS VFDS UFDS vds The preceding equations describe piecewise linear variations of vbi as a function of vds If you do not specify VFDS this model uses the first equation for vbi Note The Level 6 MOSFET device model calculates model parameters such as VTO PHI and GAMMA if you did not specify them see Common Threshold Voltage Equations on page 58 Multi Level Gamma VBO gt 0 Use Multi Level Gamma to model MOS transistors with lon Implanted channels The doping concentration under the gate is approximated as step functions GAMMA represents the corresponding body effects coefficient for the implant layer LGAMMA represents the corresponding body effects coefficient for the substrate Figure 27 shows the variation of vth as a function of vsb for Multi Level Gamma HSPICE MOSFET Models Manual 161 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model 162 Figure 27 Threshold Voltage Variation Vsb VBO p PHI Vsb 2 PHI The following equations calculate the threshold voltage for different regions Channel Depletion Region is in the Implant Layer vsb VBO y yi vth vbi yi vsb PHI vbi Channel Depletion Region Expands into the Bulk vsb gt VBO VTO yi PHI y yb vth vbi b vsb PHI 2 vbi vtb yb PHI For the threshold voltage to be continuous at
6. cece eee eee 378 Synopsys Parameters lille eae 379 References oss pex awe Ded DRESDEN ahs dedu DER es 379 7 BSIM MOSFET Models Levels 47 to 65 LLuuuuuulsusss 381 Level 47 BSIM3 Version 2 MOS Model 000 00 eeee eee 382 Using the BSIM3 Version 2 MOS Model 05 386 NOTES aaa NA NAKA ANG AA a etre a IE LUN ME 387 Leff and Weff Equations for BSIM3 Version 2 0 liuius 389 Level 47 Model Equations iilii 390 Threshold Voltage 0 cece ete 390 Mobility of Carrier 391 Drain Saturation Voltage liliis ee 391 Liriear Region oos peed ieee pew ete aided Hala 393 Saturation Region 000 c cee eee 393 Drain Gutrent suc uestre meridemreeus venie Mada ws wale 394 Subthreshold Region 00 00 ccc eee eee 394 Transition Region for subthMod 2 only 55 395 Temperature Compensation cece e eee 396 xii HSPICE MOSFET Models Manual X 2005 09 PMOS Model 0 0000 c eee ee eee Level 49 and 53 BSIM3v3 MOS Models Selecting Model Versions cee eee eee Recommended BSIM3v3 Version 00005 Version 3 2 Features cc eens Nonquasi Static NQS Model 0000 eee eee HSPICE Junction Diode Model and Area Calculation Method TSMC Diode Model 0 000 cc cee eee eee eee BSIM3v3 STILOD 0 0 0000 ccc Parameter Differe
7. 000 cece eee 242 Intrinsic Noise Model Equations 0 0000 c ee eee eee 243 Thermal Noise 00 00 cece eee 243 Flicker Noise rr 243 Operating Point Information 243 Numerical values of model internal variables 243 Transconductance efficiency factor aaa 244 HSPICE MOSFET Models Manual ix X 2005 09 Contents Early voltage beer eae ee Oh T e eyes 244 Overdrive voltage 6 cc eens 244 SPICE like threshold voltage a 244 Saturation voltage 0 c cee eee 244 Saturation non saturation flag 00 0 cece eee eee 244 Estimation and Limits of Static Intrinsic Model Parameters 245 Model Updates Description liliis 246 Revision September 1997 00 0 0 cee eae 247 Revision Il July 1998 0 0000 eee 247 Corrections from EPFL R11 March 1999 248 Corrections from EPFL R12 July 30 1999 248 Level 58 University of Florida SOI 000 eee eee 249 NOLES c orient red Bah ied eee eee Eee E 250 Level 58 FD SOI MOSFET Model Parameters 250 Level 58 NFD SOI MOSFET Model Parameters 254 Notess ost tee bete ole have A ate angat ata tette afta 259 Level 58 Template Output 0 0 00 cee 259 Level 61 RPI a Si TFT Model 0 0 0 0 0 0 ccc cc es 260 ModeliFeat res 42 tes scien Dep er Dre Sev Ee gas 260 Using Level 61
8. In the preceding equations vt is the thermal voltage The following equation calculates the lgs current for v von Vos Von I Lj von vde v e Jes VOS gt VON Li Lj v vde vs gs The following equation calculates the vde value used in the preceding equation vde min v Vi Note The modified threshold voltage von due to NFS specification is also used in strong inversion instead of vi mostly in the mobility equations If WIC 3 the Level 2 model calculates the subthreshold current differently In this case the ly current is Tas lasg vde Vep isub NO jp ND oop Voss Vas NO and ND are functions of effective device width and length HSPICE MOSFET Models Manual 135 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 3 IDS Empirical Model LEVEL 3 IDS Empirical Model 136 This section describes the LEVEL 3 IDS Empirical model parameters and equations LEVEL 3 Model Parameters MOSFET Level 3 uses only the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters LEVEL 3 Model Equations The LEVEL 3 model equations follow IDS Equations The following equations describe how the LEVEL 3 MOSFET model calculates the ly drain current Cutoff Region vgs lt vth Ij 0 See subthreshold current On Region Vgs gt Vth Ti p ursi l E Pvde vde The following equations calculate values used in the preceding equation
9. 00 0 cece eee ee 183 IDS EqQuationS sss pasted Bak ni ue ee rri e dei d 183 Effective Channel Length and Width 0005 183 HSPICE MOSFET Models Manual vii X 2005 09 Contents Effective Substrate Doping nsub 0000 0c eee ee 183 Threshold Voltage vith 0 00 0 eee 184 Saturation Voltage vdsat 0 0 00 eee eee 184 Effective Mobility ueff llle 184 Channel Length Modulation 00 0c eee eee 186 Subthreshold Current ldS 00 00 eee ee 187 LEVEL 27 SOSFET Model 000000 e cece eee eee 188 LEVEL 27 Model Parameters 00 e cece eee eee 190 Non Fully Depleted SOI Model cece eee eee 192 Model Components 00000 cece eee 192 Obtaining Model Parameters 0000 cee eee eee eee 193 Fully Depleted SOI Model Considerations 00 194 LEVEL 38 IDS Cypress Depletion Model 00000 eee eee 195 LEVEL 38 Model Parameters 0 0 c eee eee eee 197 LEVEL 38 Model Equations elles 197 IDS Equations sne 0 000 197 Threshold Voltage vth 0 0 0 eee 199 Saturation Voltage vdsat eee 201 Mobility Reduction UBeff 00 e eee ee 201 Channel Length Modulation 0000 e eee eee 202 Subthreshold Current idS 00 0c c eee eee 202 Example Model File 0 00 c cee ete 203 Mobility Model 0 0 ae 203 Body Effect s barbe L
10. 58 81 Pcav uy T v The following equations calculate values used in the preceding equations Wer W Bi Un A B UBeff cox 7 eff eff COX CS 2 TIE Cay CS Za cox cs DP 1e 4 DNB n NI 1le4 E d eee a OTD a DP vde min v Vi The following sections describe the saturation voltage threshold voltage and effective y Threshold Voltage vj The VTO model parameter is an extrapolated zero bias threshold voltage for a large device The following equations calculate the effective threshold voltage including the device size effects and the terminal voltages Vg Vg Bd vch Y O4 Baysa The following equations calculate values used in the preceding equation 1 2 Vip VTO pd vch yg PD Bas D sce vch LMI UB cox cav HSPICE MOSFET Models Manual 151 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model 152 B 25b sg nal Yo cav AE nd DNB ie NI nd DNB DP le 4 The following equation computes the effective y including the small device size effects Y Yo 1 scf 19 ncf The following equations calculate values used in the preceding equation If SCM lt 0 then scf 0 Otherwise scf E 41 2 eff XJ If NWM sO then ncf 0 Otherwise NWM X 1 2 ncf w eff The following equation calculates the xg value used in the preceding equation 25 d Mq DNB Note If Vgs lt Vin the surface is invert
11. Base the derivatives conductances and capacitances on the polarity conventions for the bias and current 564 HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Conventions The following code demonstrates the required polarity reversal for currents Von and Vdsat for PMOS devices if model gt type lt 0 pslot ids pslot gt ids pslot ibs pslot gt ibs pslot ibd pslot gt ibd pslot von pslot von pslot vdsat pslot vdsat Source Drain Reversal Conventions If you operate the MOSFET in the reverse mode when Vds 0 for N channel or Vds 0 for P channel then HSPICE or HSPICE RF performs the appropriate computations This includes a variable transformation Vds Vds Vgs gt Vgd Vbs gt Vbd and an interchange of the source and drain terminals You do not see this transformation but it simplifies the model coding task Thread Safe Model Code HSPICE or HSPICE RF uses shared memory multithreading algorithms to evaluate the model To ensure thread safe model code adhere to the following rules Do not use static variables in CMI Evaluate CMIDiodeEval CMIWriteError Or MI Noise orin functions that these routines call Never write to a global variable when you execute CMI Evaluate CMIDiodeEval CMIWriteError Of CMI Noise HSPICE MOSFET Models Manual 565 X 2005 09 8
12. lt LEVEL vab lt keyname 1 val1 gt lt keyname2 val2 gt lt VERSION version_number gt or MODEL mname NMOS lt ENCMODE 0 15 LEVEL vab lt keyname1 val1 gt lt keyname2 val2 gt VERSION version numbers Parameter Description mname Model name Elements refer to the model by this name PMOS Identifies a p channel MOSFET model NMOS Identifies an n channel MOSFET model ENCMODE Applicable to BSIM4 level 54 and BSIM3v3 level 49 and 53 use to suppress warning messages originating in CMI code while inside encrypted code Default is O off Set to off if this parameter is not present This parameter cannot be overwritten through the instance line LEVEL Use the LEVEL parameter to select from several MOSFET model types Default 1 0 VERSION Specifies the version number of the model for LEVEL 13 BSIM and LEVEL 39 BSIM2 models only See the MODEL statement description for information about the effects of this parameter HSPICE MOSFET Models Manual 13 X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates Example MODEL MODP PMOS LEVEL 7 VTO 3 25 GAMMA 1 0 MODEL MODN NMOS LEVEL 2 VTO 1 85 TOX 735e 10 MODEL MODN NMOS LEVEL 39 TOX 2 0e 02 TEMP 2 5e 01 VERSION 95 1 MOSFET Output Templates Several MOSFET models produce an output template consisting of a set of parameters that specify the output of state variables stored charges capacitances pa
13. nt CMImos3Evaluate CMI VAR char char nt CMImos3DiodeEval CMI VAR char char nt CMImos3Noise CMI VAR char char nt CMImos3PrintModel char nt CMImos3FreeModel char nt CMImos3FreeInstance char char nt CMImos3WriteError int char nt CMImos3Start void nt CMImos3Conclude void extended model topology O normal mos 1 berkeley SOI local tatic MOS3model Mos3Model tatic MOS3instance Mos3Instance tatic CMI MOSDEF CMI mos3def S char amp Mos3Model char amp Mos3I Qa aaaaaaaaaa S x NGC mo s31 ResetMode mos3 mos3AssignMP mos3AssignIP mos3SetupMode Imos3SetupIn mos 31 ResetIn Senet mo 3 mos3Noise Evaluate DiodeEva mo s31 mos3FreeModel Imos3FreeInsta PrintMode l l 1 nce nstance stance stance HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Interface Variables CMImos3 CMImos3Start CMImos3Conclude li export CMI_MOSDEF pCMI_mos3def amp CMI_mos3def Note The last 8 functions are optional If you do not define a function replace it with NULL CMI ResetModel This routine initializes all parameters of a model After initialization all model parameters become undefined in a netlist model card The pmos flag sets the transistor type
14. uble rgeomod flag to control the type of contacts uble tnoimod flag to select thermal noise model tnoimod 0 1 uble trnqsmod flag to select NQS model for transient analysis uble acnqsmod flag to select NQS model for AC analysis uble qgmid mid gate charge for rgatemod 3 uble qcheq channel charge under quasi static uble qcdump channel charge deficit surplus uble cqdb transcapacitance dqcheq dVd uble cqgb transcapacitance dqcheq dVg uble cqsb transcapacitance dqcheq dVs uble cqbb transcapacitance dqcheq dVb cqdb cqgb 559 8 Customer Common Model Interface Extended Topology cqsb double nois_irg thermal noise contributed by gate electrode resistance current 2 double nois irbps thermal noise contributed by resistance between ib and sb current 2 double nois irbpd thermal noise contributed by resistance between ib and db current 2 double nois irbpb thermal noise contributed by resistance between ib and xb current 2 double nois irbsb thermal noise contributed by resistance between xb and sb current 2 double nois irbdb thermal noise contributed by resistance between xb and db current 2 double nois igs shot noise associated with Igso Igcs current 2 double nois igd sh current 2 double nois igb shot noise ass
15. HSPICE style MODEL n1 nmos Level 47 XL 0 le6 LD 0 15e 6 SatMod 2 SubthMod 2 BulkMod 1 CGSO 0 3e 9 CGDO 0 3e 9 CGBO 0 SPICE3 style MODEL n2 nmos Level 47 LD 20 1e 6 SatMod 2 SubthMod 2 BulkMod 1 CGSO 0 3e 9 CGDO 0 3e 9 CGBO 0 HSPICE MOSFET Models Manual 389 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model 390 Level 47 Model Equations The following model equations are based on the BSIM3 source code Threshold Voltage Model Parameters Ving Kl K2 6 Nie K3 W T ox Vip Di Dy Dy N N pear Nsub Y Yo Vp Vow VorXp TREF Vin Vino KU Rcs J ien T K3B apt A K3 K3B V Wr a op Vin TEMP DTEMP 273 15 raeg TREF 273 15 AV in Olep Vy s Dy Leg Dn Leg 0 Leg Dy Bal 2l 2exp L L W 3 Tox Ais me Dy Vy X dep I si Es 0 E Vos q N peak T If you do not specify Phi as a model parameter then N 0 2 V4 In pest Npeak and ni in cm 3 l Vim K T q 1 5 n 1 45e10 exp 21 5565981 Eg 2 V 30013 300 15 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model Eg 1 16 7 02e 4 T T 1108 0 If you do not specify K1 and K2 as model parameters simulation calculates them as follows Ki 35 2 Ko 49 Vom A D Vs m JO K
16. Units Default Description CTP EG F1EX GAP1 GAP2 LAMEX MJ MJSW PTA 1 K eV eV K K 1 K V K 0 0 7 02e 4 1108 1 0 0 5 0 33 0 0 HSPICE MOSFET Models Manual X 2005 09 Junction sidewall capacitance CJSW temperature coefficient If TLEVC 1 CTP overrides the default temperature compensation Energy gap for pn junction diode for TLEV 0 or 1 default 1 11 for TLEV 2 default 1 16 1 17 silicon 0 69 Schottky barrier diode 0 67 germanium 1 52 gallium arsenide Bulk junction bottom grading coefficient First bandgap correction factor from Sze alpha term 7 02e 4 silicon 4 73e 4 silicon 4 56e 4 germanium 5 41e 4 gallium arsenide Second bandgap correction factor from Sze beta term 1108 silicon 636 silicon 210 germanium 204 gallium arsenide LAMBDA temperature coefficient Emission coefficient Bulk junction bottom grading coefficient Bulk junction sidewall grading coefficient Junction potential PB temperature coefficient If you set TLEVC to 1 or 2 PTA overrides the default temperature compensation 99 2 Technical Summary of MOSFET Models Temperature Parameters and Equations Table 28 Temperature Effects Parameters Continued Name Units Default Description Alias PTC VPK 0 0 Fermi potential PHI temperature coefficient If you set TLEVC to 1 or 2 PTC overrides the d
17. rthO Normalized thermal resistance moC W 0 xbjt Power dependence of jbjt on the none 2 temperature xdif Power dependence of jdif on the none 2 temperature xrec Power dependence of jrec on the none 20 temperature xtun Power dependence of jtun on the none 0 temperature HSPICE MOSFET Models Manual X 2005 09 503 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model Table 144 MOSFET Level 60 Model Parameter Notes Note Explanation nl 1 nl 2 nC 1 nC 2 nC 3 nC 4 504 Capmod 0 and 1 do not calculate the dynamic depletion Therefore ddMod does not work with capmod BSIMSOI refers to a substrate of the silicon below the buried oxide not the well region in BSIMG It calculates the backgate flatband voltage Vipp and the parameters related to the source drain diffusion bottom capacitance Vegth Vsdtb Csdmin e Positive Ngyp is the same type of doping as the body e Negative Nub is the opposite type of doping If you do not specify cgso simulation calculates it if you specify dic greater than O then cgso pl dlc cox cgs1 if the previously calculated cgso lt 0 then cgso 0 else cgso 0 6 Tsi cox Calculates Cgdo similar to Csdo If nsub is positive then 102 n Var iof 2 0 3 n n 20 else Vu PA J 03 N sub If nsub is positive then 12 n 5 753x10 n Osa 2 Tog sb Ys sub q nj Chox Vedi Vsafb Vsa
18. scf 1 i 1 2 LEM rad LAMBDA PHI vsp ul L LGAMMA The XJ model parameter modifies the GAMMA model parameter by the short channel factor gl 2 LAMBDA gis XJscaled DIE i 1 2 PHI vsb SCM vds He Er leff XJscaled LING vds The gl factor generally replaces the scf factor for the multi level GAMMA model The gw factor modifies GAMMA to compute the narrow width effect _ 14NWM xd weff The following equation calculates the xd value used in the preceding equation ee xd q DNB gW Finally the effective y including short channel and narrow width effects is y GAMMA gw gl scf Effective Built in Voltage vbi The Level 6 model includes the narrow width effect This effect is the increase in threshold voltage due to the extra bulk charge at the edge of the channel To use this effect with the NWE model parameter modify vbi Modify vbi to use the short channel effect which decreases threshold voltage due to the induced potential barrier lowering effect To include this effect you must specify either the FDS parameter or the UFDS and VFDS model parameters HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model The following equations calculate vbi which sums up the preceding features vds lt VFDS or VFDS 0 Leff VSH 83 FDS vds COX P Leff vds VFDS i 1 2 LDscaled esi VTO y PHI 1 PHI b
19. 0 1vu WVFB 0 08v u Leff 1 10 m lu Weff 2 10 m 2u LREFeff 2 10 m 2u WREFeff 1 10 m 10u 1 1 1 1 zvfb VFBO LVFB ea HI WVFB 4 LI Leff LREFeff Weff WREFeff 1 1 1 1 b 0 35v 0 1v LI 0 08v LIL zvfi v vel TET veiw ju To zvfb 0 35v 0 05v 0 032v zvfb 0 368v LEVEL 28 Model Equations The LEVEL 28 model equations follow Effective Channel Length and Width The effective channel length and width for LEVEL 28 is consistent with the LEVEL 3 model L W and the M multiplier are from the MODEL statement in the netlist SCALE and SCALM are options If you do not specify any scaling options or multipliers then Leff L XL 2 LD Weff W XW 2 WD Note If you specify LDAC and WDAC in the MODEL statement Leff L XL 2 LDAC Weff W XW 2 WDAC Lscaled L SCALE Wscaled W SCALE XLscaled XL SCALM LDscaled LD SCALM HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modified BSIM Model XWscaled XW SCALM WDscaled WD SCALM Leff Lscaled LMLT XLscaled 2 LDscaled LREFeff LREFscaled LMLT XLREFscaled 2 LDscaled Weff M Wscaled WMLT XWREFscaled 2 WDscaled WREFeff M WREFscaled WMLT XWscaled 2 WDscaled Threshold Voltage Effective model parameter values for the threshold voltage after you adjust the device size are zphi zvfb zk1 zk2 zeta zx2e zx3e zgammn and zetamn Simulation calculates these values
20. 1 1 Parasitic bipolar flag 0 off 1 on 254 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 58 University of Florida SOI Table 54 MOSFET Level 58 Flag Parameters Continued Parameter Unit Default Typical Value Description SELFT 0 TPG 2 TPS 0 Self heating flag 0 no self heating 1 approximate model 2 full self heating Type of gate polysilicon 1 opposite to body 1 same as body Type of substrate 1 opposite to body 1 same as body Table 55 MOSFET Level 58 Structural Parameters Parameter Unit Default Typical Value Description TOXF m 1 063 3 8 x10 9 Front gate oxide thickness TOXB m 0 5e 6 80 400 x109 Back gate oxide thickness NSUB cm3 1 0 15 4915 4917 Substrate doping density NGATE cm3 0 0 1019 1920 Poly gate doping density 0 for no poly gate depletion NDS cm3 5 0e19 4919 4920 Source drain doping density TF m 0 2e 6 3 8 x109 Silicon film thickness TB m 0 1e 6 30 100 x109 Film body thickness THALO m 0 0 Halo thickness 0 for no halo NBL cm3 5 0e16 1017 1018 Low body doping density NBH cm3 5 0e17 4919 49209 Halo doping density HSPICE MOSFET Models Manual X 2005 09 255 5 Standard MOSFET Models Levels 50 to 64 Level 58 University of Florida SOI 256 Table 55 MOSFET Level 58 Structural Parameters Continued Parameter Unit Default Typical Value Description NHAL
21. 600 example capacitance 71 cascoding 63 gate capacitance 71 MODEL CARDS NMOS model 432 MOSFETs BSIM LEVEL 13 339 data fitting 580 Empirical 142 gate capacitance 71 IDS LEVEL 7 164 166 PMOS model 433 EXD model parameter 41 EXJ model parameter 41 EXP model parameter 41 EXS model parameter 41 F FC model parameter 41 field effect transistor 28 See also MOSFETs JFETs flicker noise 96 Fluke Mosaid model 5 Frohman Bentchkowski equations 173 G gate capacitance 92 AMI 93 charge sharing coefficient 77 example 71 length width 95 LEVEL 39 367 372 model CAPOP 39 369 modeling 578 parameters 64 74 plotting 72 SPICE 78 gate direct tunneling current 553 GE CRD Franz model 5 GE Intersil model 5 generalized customer CMI 553 activating enhancements 563 BSIM4 556 gate direct tunneling current 553 geometry MOSFETs model parameters 43 HSPICE MOSFET Models Manual X 2005 09 scaling 12 transistor field effect 29 global scaling 12 GMIN option 11 GMINDC option 11 39 Grove Frohman model 4 H HDIF model parameter 43 HP a Si TFT model 204 equations 206 topology 211 HSPICE junction diode model 402 model enhancements 571 VERSION parameter 332 HSPICE AL equation 179 IDS Cypress depletion model 6 equations LEVEL 1 128 LEVEL 13 333 LEVEL 2 130 LEVEL 3 136 LEVEL 38 197 LEVEL 5 145 LEVEL 6 159 168 LEVEL 8 183 LEVEL 38 Cypress model 195 LEVEL 5 equations 150 model 144 LEVEL 6 equations 159 168 exampl
22. MOSFET Output Templates 2 Technical Summary of MOSFET Models Nonplanar and Planar Technologies llle Nonplanar techonology HSPICE MOSFET Models Manual X 2005 09 xiii xiv XV Xvi Xvi xvii 27 27 27 28 28 32 Contents Equation Variables 0c cee eee 32 Using the MOSFET Current Convention 0000 34 Using MOSFET Equivalent Circuits a 35 MOSFET Diode Models 0 000 e cnet EES 39 Selecting MOSFET Diode Models 00 00 eee nee ee 39 Enhancing Convergence 02 22 cece eee eres 39 MOSFET Diode Model Parameters a 40 Using an ACM 0 MOS Diode 000 eee ees 43 Calculating Effective Areas and Peripheries 45 Calculating Effective Saturation Current 000005 45 Calculating Effective Drain and Source Resistances 45 Using an ACM 1 MOS Diode 0000 e eee eee 46 Calculating Effective Areas and Peripheries 47 Calculating Effective Saturation Current 000005 48 Calculating Effective Drain and Source Resistances 48 Using an ACM 2 MOS Diode 0000 0c eee eee 49 Calculating Effective Areas and Peripheries 51 Calculating Effective Saturation Currents 51 Calculating Effective Drain and Source Resistances 52 Using an ACM 3 MOS Diode 2 0c
23. Set Level 61 to identify the model as the AIM SPCIE MOS15 a Si TFT model The default value for L is 100um and the default value for W is 100um Level 61 is a 3 terminal model This model does not include a bulk node therefore simulation does not append parasitic drain bulk or source build diodes are appended to the model You can specify a fourth node but it does not affect simulation results HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 61 RPI a Si TFT Model 4 The default room temperature is 25 C in Synopsys circuit simulators but is 27 C in most other simulators When comparing to other simulators use TEMP 27 or OPTION TNOM 27 to set the simulation temperature to 27 in the netlist Example The following is an example of how Level 61 modifies the Synopsys MOSFET model and element statement mckt drain gate source nch L 10e 6 W 10e 6 MODEL nch nmos Level 61 alphasat 0 6 cgdo 0 0 cgso 0 0 def0 0 6 delta 5 0 el 0 35 emu 0 06 eps 11 epsi 7 4 gamma 0 4 gmin 1e23 iol 3e 14 kasat 0 006 kvt 0 036 lambda 0 0008 m 2 5 muband 0 001 rd 0 0 rs 0 0 sima0 1e 14 tnom 27 tox 1 0e 7 v0 0 12 vaa 7 5e3 vdsl 7 vfb 3 vgsl 7 vmin 0 3 vto 0 0 Name Unit Default Description ALPHASAT 0 6 Saturation modulation parameter CGDO F m 0 0 Gate drain overlap capacitance per meter channel width CGSO
24. Table 111 MOSFET Level 54 Instance Parameters Parameter Unit Default Description RDC ohm 0 0 Drain contact resistance for the per finger device RSC ohm 0 0 Source contact resistance for the per finger device HSPICE MOSFET Models Manual 443 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 111 MOSFET Level 54 Instance Parameters Continued Parameter Unit Default Description DELVTO V 0 0 Shift in the zero bias threshold voltage VTHO DELVTO MULUO 1 0 Low field mobility UO multiplier DELK1 y1 2 Shift in the body bias coefficient K1 DELNFCT 0 0 Shift in subthreshold swing factor NFACTOR Table 112 MOSFET Level 54 Warning Messages Parameter Effective Value Warning Message RDC RD_eff RD_diffusion RDC NF Warning if RDC lt 0 0 and reset RDC to 0 0 RSC RS_eff RS_diffusion RSC NF Warning if RSC lt 0 0 and reset RSC to 0 0 DELVTO VTHO_eff VTHO DELVTO N A DELVTO MULUO UO eff UO MULUO Warning if MULUO 0 0 and reset MULUO to 1 0 DELK1 K1 eff K1 DELK1 Warning if DELK1 lt K1 and reset DELK1 to 0 0 DELNFCT NFACTOR_eff NFACTOR DELNFCT N A Table 113 Model Selectors Controllers MOSFET Level 54 Parameter Default Binnable Description VERSION 4 10 NA Model version number BINUNIT 1 NA Binning unit selector PARAMCHK 1 NA Switch for the parameter value check 444 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Le
25. W W Bs KP A ug COX 29 eff eff vde min v Vdsat GAMMA f b f I sen 4 PHI v Note In the above equation the factor 4 should be 2 but because SPICE uses a factor of 4 this model uses a factor of 4 as well HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 3 IDS Empirical Model The f parameter specifies the narrow width effect _ DELTA 1 2T Eg o Wa 4 COX The f term expresses the effect of the short channel XJ caled p LD scaled W C W ae _ LD scaled i W p f Lg XI scaled XJ scaled XJ scaled W X PHI v ka PIAN d Ng NSUB W W 2 0 0631353 0 8013292 E 0 011107772 P l scale W XJ c scaled scale Effective Channel Length and Width The following equations determine the effective channel length and width in the LEVEL 3 model Lg a L scaled LMLT XL cated 2 P LD cated DEL ated Wert M Wscajgg WMLT XW calea 7 WP scaled LREFF LREF scaled LMLT AL calea 7 2 P LD scaled DEL caled WREF g M WREF au WMLT KW caled 7 2 WD scaled Threshold Voltage vj The following equation calculates the effective threshold voltage including the device size and terminal voltage effects S 6 14e 22 ETA Wa 112 Yih vhi ee ae Vds GAMMA fs PHI Vig tf PHI Ven eff HSPICE MOSFET Models Manual 137 X 2005 09 4 Standard MOSFET
26. compatibility WREF gt co use WREF 0 WREF caleg WREF SCALM Metallurgical junction depth XJ scaled XJ SCALM Junction depth Length bias accounts for the masking and etching effects length XLecaled XL SCALM Difference between the physical on the wafer and the drawn reference channel length Use this parameter to calculate Lepp only if DL 0 XLscaled XL SCALM Difference between the physical on the wafer and the drawn reference channel length XLREFscaled XLREF SCALM Width bias accounts for the masking and etching effects width XW scaled XW SCALM Difference between the physical on the wafer and the drawn S D active width Use this parameter to calculate Wa only if DW 0 XWecaled XW SCALM 39 28 39 117 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 26 Effective Width and Length Parameters Continued Name Alias Units Default Description Level XWREF m 0 0 Difference between the physical on the wafer 28 39 and the drawn reference channel width XWREFscaled XWREF SCALM Table 27 Threshold Voltage Parameters Name Alias Units Default Description Level BetaGam 1 0 Body effect transition ratio 38 CAV 0 0 Thermal voltage multiplier for the weak 8 inversion equation DELTA 0 0 Narrow width factor for adjusting the threshold 2 3 8 DNB cm 0 0 Surface doping density 5 38 NSUB 1 cm3 1 0015 Substrate doping 6 7
27. instance multiplier output only double vges eletrical external vges v xg v is double vgms eletrical internal vgms v mg v is double vdbs eletrical body nody bias vdbs v db v is double vsbs eletrical body nody bias vsbs v sb v is double vdes electrical external vdes v xd v is double vses electrical external double double gstoto double gstotod double gstotog double gstotob double gstotos double gdtoto double 558 vses v xs v is qdef internal NQS gdtotod dgdtoto dvd AY bias M4 version number M factor up to 4 50 compat to handle tible template gate to internal source bias mid gate to source bias at at drain to source bias source to source bias qdei f v nqs pt electrical source conductance dgstoto dvd vs dgstoto dvg vs dgstoto dvb vs electrical es es es gstotod gstotog gstotob drain conductance vdes vds Ey 4 the drain side to source the source side to source HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Extended Topology
28. 1 2e 3 1 251551 gt 0 570545e 4 Vn 0 8 15e 3 75 0 6875 egarg 9 399920 vtherm 3 215466e 2 phi t 0 64 1 251551 0 3238962 0 4770964 2 vt Aphi t Ij beta 0 5 5 724507e 4 Simulation results T 25 1d 6 91200e 04 T 100 id 25 72451e 04 These results agree with the hand calculations LEVEL 4 IDS MOS Model The LEVEL 4 MOS model is the same as the LEVEL 2 model with the following exceptions No narrow width effects h 1 No short channel effects y GAMMA For lateral diffusion LDscalea LD XJ SCALM If you specify XJ the LD default 0 75 If you do not specify XJ the default is 0 TPG the model parameter for type of gate materials defaults to zero AL gate The default is 1 for other levels If you do not specify VTO this parameter computes VTO see Common Threshold Voltage Equations on page 58 Starting in 2001 4 2 MOSFET Level 4 and Level 9 support both M and AREA scaling HSPICE MOSFET Models Manual 143 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model LEVEL 5 IDS Model This section describes the LEVEL 5 IDS model parameters and equations Note This model uses micrometer units rather than the typical meter units Units and defaults are often unique in LEVEL 5 Level 5 does not use the SCALM option LEVEL 5 Model Parameters MOSFET Level 5 uses the generic MOSFET model parameters described in Chapter 3
29. 536 HSPICE MOSFET Models Manual X 2005 09 CMI AssignModelParm 8 Customer Common Model Interface Interface Variables This routine sets the value of a model parameter Syntax int CMI AssignModelParm char pmodel char pname double value Parameter Description pmodel Pointer to the model instance pname String of the parameter name You cannot use character strings as parameter values in HSPICE RF value Parameter value Example int ifdef STDC CMImos3AssignMP char pmodel char pname double value else CMImos3AssignMP pmodel pname value char pmodel char pname double value endif int param CMImos3GetMpar pname CMImos3SetMpar param return 0 int CMImos3AssignMP HSPICE MOSFET Models Manual X 2005 09 amp param value MOS3model pmodel tf 537 8 Customer Common Model Interface Interface Variables CMI AssignlnstanceParm This routine sets the value of an instance parameter Syntax int CMI AssignInstanceParm char pinst char pname double value Parameter Description pinst Pointer to the instance pname String of the parameter name You cannot use character strings as parameter values in HSPICE RF value Parameter value Example int ifdef STDC CMImos3AssignIP char ptran char pname double value else CMImos3AssignIP
30. 7 61e 18 Temperature coefficient for Up uci 1 V 0 056 Temperature coefficient for Ug ute 1 5 Mobility temperature exponent xbjt 1 Power dependence of jp on the temperature xdif XBJT Power dependence of jq on the temperature xrec 1 Power dependence of je on the temperature xtun 0 Power dependence of ji on the temperature HSPICE MOSFET Models Manual X 2005 09 475 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model 476 Notes BSIMPD2 01 supports capmod 2 and 3 only It does not support capmod 0 and 1 Modern SOI technology commonly uses source drain extension or LDD The source drain junction depth Xj can be different from the silicon film thickness Tsi By default if you do not specify Xj simulation sets it to Tsi Xj cannot be greater than Tsi BSIMPD refers the substrate to the silicon below the buried oxide not to the well region in BSIM3 to calculate the backgate flatband voltage Vfbb and the parameters related to the source drain diffusion bottom capacitance Vsdth Vsdfb Csdmin e Positive nsub means the same type of doping as the body e Negative nsub means the opposite type of doping New WOFLK Parameter The following equation models the SPICE flicker noise current density used in both UCB SOI code and the Synopsys Level 57 MOSFET model for noiMod 1 and 4 Sid f I2 Hz KF Ids AF Cox Leff 2 f EF 1 However if AF is not equal to
31. DC sweeps have been checked against SPICE3e2 The default setting for CAPOP is CAPOP 13 which is the BSIM1 charge conserving capacitance model This model does not use the BSIM3 capacitance model The Level 47 model supports the TNOM model parameter name as an alias for TREF The conventional terminology is TREF which is supported as a model parameter in all Synopsys MOS levels Level 47 supports the TNOM alternative name for compatibility with SPICE3 The default room temperature is 25 C in Synopsys simulators but is 27 C in SPICES If you specify the BSIM3 model parameters at 27 C add TREF 27 to the model so that simulation correctly interprets the model parameters To set the nominal simulation temperature to 27 add OPTION TNOM 27 to the netlist when you test the Synopsys model versus SPICES The default of DERIV is zero the analytical method You can set DERIV to 1 for the finite difference method Analytic derivatives in the SPICE3e2 code are not exact in some regions Setting DERIV 1 returns more accurate derivatives GM GDS and GMBS but consumes more CPU time You can select one of three ways to calculate Vp e Using K1 and K2 values that you specify e Using GAMMA1 GAMMA2 VBM and VBX values that you enter in the MODEL statement e Using NPEAK NSUB XT and VBM values that you specify You can enter the NPEAK and U0 model parameters in meters or centimeters Simulation converts NPEAK to cm3 as follows if
32. Dynamic Macromodeling Dynamic Model Switcher ECL Compiler ECO Compiler EDAnavigator Encore Encore PQ Evaccess ExpressModel Floorplan Manager Formal Model Checker FoundryModel FPGA Compiler Il FPGA Express Frame Compiler Galaxy Gatran HANEX HDL Advisor HDL Compiler Hercules Hercules Explorer Hercules ll Hierarchical Optimization Technology High Performance Option HotPlace HSIM 4S HSPICE Link iN Tandem Integrator Interactive Waveform Viewer i Virtual Stepper Jupiter Jupiter DP JupiterXT JupiterXT ASIC JVXtreme Liberty Libra Passport Library Compiler Libra Visa Magellan Mars Mars Rail Mars Xtalk Medici Metacapture Metacircuit Metamanager Metamixsim Milkyway ModelSource Module Compiler MS 3200 MS 3400 Nova Product Family Nova ExploreRTL Nova Trans Nova VeriLint Nova VHDLlint Optimum Silicon Orion ec Parasitic View Passport Planet Planet PL Planet RTL Polaris Polaris CBS Polaris MT Power Compiler PowerCODE PowerGate ProFPGA ProGen Prospector Protocol Compiler PSMGen Raphael Raphael NES RoadRunner RTL Analyzer Saturn ScanBand Schematic Compiler Scirocco Scirocco i Shadow Debugger Silicon Blueprint Silicon Early Access SinglePass SoC Smart Extraction SmartLicense SmartModel Library Softwire Source Level Design Star Star DC Star MS Star MTB Star Power Star Rail Star RC Star RCXT Star Sim Star SimXT Star Time Star XP SWIFT Taurus TimeSlice TimeTracker Timi
33. LDIF 0 5 um HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models Table 11 ACM 0 MOS Diode Parameters Parameter Description CJ 1e 10 F m of gate width Note the change from F m in ACM 0 to F m CJSW 2e 10 F m of gate width JS 1e 14 A m of gate width Note the change from A m in ACM 0 to A m JSW 1e 13 A m of gate width NRD number of squares for drain resistance NRS number of squares for source resistance Calculating Effective Areas and Peripheries For ACM 1 simulation calculates the effective areas and peripheries as follows ADeff Weff WMLT ASeff Weff WMLT PDff Weff PSeff 2 Weff The following equation calculates the Weff value used in the preceding equations Weff M Wscaled WMLT XWscaled Note The Weff value is not the same as the weff value in the LEVEL 1 2 3 6 and 13 models The 2 WDscaled term is not subtracted HSPICE MOSFET Models Manual 47 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models 48 Calculating Effective Saturation Current For ACM 1 the MOS diode effective saturation currents are calculated as follows Source Diode Saturation Current Define val JSscaled ASeff JSWscaled PSeff If val gt O then isbs val Otherwise isbs M IS Drain Diode Saturation Current Define val JSscaled ADeff J5Wscaled PDeff If val gt O then isbd val Otherwise isbd M IS Calcul
34. LEVEL 13 Equations eee eae Effective Channel Length and Width IDS Equations 0 0000 e eee eee Threshold Voltage llle Saturation Voltage vdsat ids Subthreshold Current aa Contents Resistors and Capacitors Generated with Interconnects Temperature Effect 22200005 Charge Based Capacitance Model Regions Charge Expressions Preventing Negative Output Conductance Calculations Using LEVEL 13 Equations Compatibility Notes lille Model Parameter Naming SPICE Synopsys Model Parameter Differences ParasitiCS im sre nh AGANE LA MAA eed te Temperature Compensation UPDATE Parameter aaan IDS and VGS Curves for PMOS and NMOS LEVEL 28 Modified BSIM Model LEVEL 28 Features omoi fuk lx LEVEL 28 Model Equations 4 Effective Channel Length and Width Threshold Voltage llle Effective Mobility llle Saturation Voltage vdsat Transition Points llle Strong Inversion Current 4 Weak Inversion Current 00 5 LEVEL 39 BSIM2 Model 00 000 LEVEL 39 Model Parameters 00 Other Device Model Parameters that Affect BSIM2 HSPICE MOSFET Models Manual X
35. LKIO LKIO LUO LUB LVFRC LFSB MOB THETA 104 s cm V 10 s cm um um cm um V s 10 4A s cm V 104V12 s cm HSPICE MOSFET Models Manual X 2005 09 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Residue current coefficient BFRC sensitivity to the effective channel length FRC sensitivity to the effective channel length Length dependent implant channel mobility modifier Length dependent residue current coefficient UO sensitivity to the effective channel length VFRC sensitivity to the effective channel length FSB sensitivity to the effective channel length Selects a mobility equation You can set this parameter to MOB 0 or MOB 7 MOB 7 changes both the model and the channel length calculation The MOB 7 flag invokes the channel length modulation and mobility equations in MOSFET LEVEL 3 In MOSFET Level 8 you can set MOB to 2 3 6 or 7 Mobility modulation MOSFET models use THETA only if MOB 7 A typical value is THETA 5e 2 38 38 38 38 38 38 38 2 40 123 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 28 Mobility Parameters Continued Name Alias Units Default Description Level v UB UO cm V s UCRIT V cm V cm UEFF UEXP F2 UH cm V s UHSAT um V UO UB UBO cm V s 124 0 0 0 0 1 0e4 1e4 0 0 900 N 300 P 0 0 600
36. Maal Sod kT n 5 753x10 En else 2 tog xu oe oM SP sub n box Vain Vsafb acto Y uos HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Table 144 MOSFET Level 60 Model Parameter Notes Continued Note Explanation nC 5 X z 28964 C E Esi C Cs ddep box sddep 7 6 sddep T X gt sdmin T n u 10 sddep q Csddep Chox nC 6 If you do not specify cf then simulation calculates it eps Zeek aeo T T Ox nT 1 For mobmod 1 and 2 the unit is m V2 Default is 5 6E 11 For mobmod 3 the unit is 1 V and the default is 0 056 Level 65 SSIMSOI Model Level 65 is a surface potential charge based and partially depleted SOI MOSFET model developed by Motorola semiconductor Model Feature The SSIMSOI model includes the following features A simple linear body resistance and body tie parasitic for body contacted SOI devices for several capacitances and conductances are added A lateral bipolar model used to account for both the BUT current and the associated diffusion capacitance A Additional components of the channel edge diode generation and recombination current Side wall and areal components of diode currents are removed Shot noise sources associated with the parasitic currents used to model the observed Lorentzian noise spectra in SOI The body bias dependence for DIBL is removed to prevent convergence diff
37. PHP tnom GAP2 dphpdt thom Surface Potential Temperature Equations TLEVC 0 ok m MN D P t egnom__ eg t PHI t PHI KUIBA in SONO vd TLEVC 1 PHI t PHI PTC PAt If you do not specify the PHI parameter simulation calculates it as PHI t 2 v i n SSE The intrinsic carrier concentration ni must be temperature updated and it is calculated from the silicon bandgap at room temperature ni 145e16 1 exp 6G 1 tno tnom 2 vt t TLEVC 2 PHI t PHI PTC PAt TLEVC 3 PHI t PHI dphidt At If TLEVC 3 and TLEV 0 or 1 then egnom 3 vt tnom 1 16 egnom 2 Eee LAP PHI a tnom 110 dphidt thom TLEV 2 egnom 3 vt tnom EG egnom 2 Luci BO Lt PHI tnom GAP dphidt tnom HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models Temperature Parameters and Equations Threshold Voltage Temperature Equations The threshold temperature equations are TLEV 0 vbi t vbi tnom Ree PHI BOM EBM eg t 2 VTO t vbi t GAMMA PHI t 2 TLEV 1 VTO t VTO TCV PAt vbi t VTO t GAMMA b PHI t TLEV 2 GAMMA VTO t vro 1 2 PHI dphidt At vbi t VTO t GAMMA P PHI t 2 Mobility Temperature Equations The MOS mobility temperature equations are t BEX UO t
38. Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level CAP_BS LX28 Extrinsic drain to substrate 57 58 Capacitances Meyer and Charge Conservation CAP_BS csbox csesw csbox is the substrate to source bottom capacitance csesw is the substrate to source sidewall capacitance CAP BD LX29 Bias dependent bulk drain capacitance All except 57 58 CAP BD LX29 Extrinsic source to substrate 57 58 Capacitances Meyer and Charge Conservation CAP_BD cdbox cdesw cdbox is the substrate to drain bottom capacitance cdesw is the substrate to drain sidewall capacitance CQS LX31 Channel charge current CQS All CDGBO LX32 CDGBO dQd dVg Meyer and Charge All except 54 Conservation 57 59 60 CDGBO LX32 Intrinsic drain to gate capacitance 54 57 59 60 CDDBO LX33 CDDBO dQd dVd Meyer and Charge All except 54 Conservation 57 59 60 CDDBO LX33 Intrinsic drain capacitance 54 57 59 60 CDSBO LX34 CDSBO dQd dVs All HSPICE MOSFET Models Manual X 2005 09 Drain to source capacitance Meyer and Charge Conservation 19 1 Overview of MOSFET Models MOSFET Output Templates 20 Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level QE LX35 Substrate charge QE Meyer and 57 58 59 Charge Conservation CQE LX36 Substrate charge current CQE Meyer 57 58 59 and Charge Conservation CDEBO LX37 CDEBO dQd dVe intrinsic
39. The automatic model selection program searches a data file for a MOSFET model where the width and length are within the range specified in the MOSFET element statement Simulation then uses this model statement To search a data file for MOSFET models within a specified range of width and length 1 Provide a root extension for the model reference name in the MODEL statement 2 Usethe model geometric range parameters LMIN LMAX WMIN and WMAX HSPICE MOSFET Models Manual 9 X 2005 09 1 Overview of MOSFET Models Selecting Models These model parameters define the range of physical length and width dimensions to which the MOSFET model applies Example 1 If the model reference name in the element statement is NCH the model selection program examines the models with the same root model reference name NCH such as NCH 1 NCH 2 or NCH A The model selection program selects the first MOSFET model statement whose geometric range parameters include the width and length specified in the associated MOSFET element statement Example 2 The following example shows how to call the MOSFET model selection program from a data file The model selector program examines the MODEL statements where the model reference names have the root extensions NCHAN 2 NCHAN 3 NCHY 20 and NCHY 50 The netlist for this example is located in the following directory Sinstalldir demo hspice mos selector sp Setting MOSFET Control Options
40. The following equations calculate values used in the preceding equation xul vgs vth body xul zul zx2ul vsb zx3ul vds VDDM UPDATE 2 ve zul zx2ul Pvsb zx3ul vds VDDM UPDATE 0 1 Leff xul Threshold Voltage You can express the threshold voltage as vth zvfb zphi gamma zphi vsb xeta Pvds The following equations calculate values used in the preceding equation gamma zkl zKk2 b zphi vsb xeta zeta zx2e Pvsb zx3e vds VDDM UPDATE 0 2 xeta zeta zx2e zphi vsb zx3e vds VDDM UPDATE 1 334 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model Saturation Voltage vdsat The following equation calculates the saturation voltage in the BSIM Level 13 model vgs vth 1 2 vdsat body arg ids Subthreshold Current Simulation calculates the isub subthreshold current if znO is less than 200 Ilim Iexp Ilim Iexp isub The following equations calculate values used in the preceding equation vgs vth vd Iexp Boc vt el8 ng nv 1 e Weff 2 XxX JI Ilim 4 5 B vt B uo CO Leff xn zn0 znb bvsb znd vds Simulation also adds the isub current to the ids current in the strong inversion Resistors and Capacitors Generated with Interconnects Refer to the wire model table resistor element for the model parameters that you used For an exampl
41. Vps Vpse exp Vegans 1 LKINK MK Akinkt VKINK VpsE TRAT 1 3 Vin AED Vosat sar VGTE Threshold Voltage If you do not specify VTO then V7 VON DVT Otherwise V7 VTO Temperature Dependence Vry Vr DVTO TEMP TNOM u MUI DMUI TEMP TNOM LASAT eff Osat ASAT DASAT TEMP TNOM Capacitance CAPMOD 0 2 Vpsar VpsEx V C Car C poe gs 3 Cees 2V psar ad 2 VpbsAT 2 Ci CC geet gd 1 3 ged L ee HSPICE MOSFET Models Manual 275 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model C Ceca 7 n Verx ng en EE s Ned Vin C C OX V 1 ETACO exp sx ETACO Vx Wore Lere j Cox cam ETACO ETAC00 Vpspx Vps Vosex 1 V NMC4MC a l DSA Verx Vas Vrx CAPMOD 1 If ZEROC equals 1 Vos Ves Vrx Phi 0 6 phi mg VS Cos Coq 0 phi x8 Vast Cox 2 Cox _ 276 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model 2Cpy Vps2 Vpsar C us 3 gd 0 Vgst 2 0 Vos lt Vosar Pa io alan tees 8 3 L M Vpsar Vpsgx J NE 2 Cus EE 1 DSAT DSE s 3 Z Vpsar VpsEx j W L E Vos Cox ra s Vpsex EA Te Qx E Vis MCMC 1 L Vpsa Geometry Effect Weg W XW Leg L XL Self Heating Self heating is turned on if self heating parameters SHMOD 1 and RTHO gt 0
42. char char CMI PrintModel char CMI FreeModel char CMI FreeInstance char char CMI WriteError int char CMI Start void CMI_Conclude void extended model topology 0 is normal mos 1 is berkeley SOI int CMI etc topoid _MOSDEF All routines return O if they succeed or a non zero integer an error code that you define for a warning or error The following sections describe each entry HSPICE MOSFET Models Manual 533 X 2005 09 8 Customer Common Model Interface Interface Variables 534 HSPICE or HSPICE RF extracts examples for the first seven functions from the MOS3 implementation Examples for the remaining eight functions are not part of the actual MOS3 code but the code includes them for demonstration The MOS3 implementation example contains one header file and eight C files All routines are based on SPICE 3 code pModel plnstance When you compile HSPICE or HSPICE HF initializes structures for an interface variable For example i i i i i i i i i i i i i i i S S function declaration int CMImos3ResetModel char int int nt CMImos3ResetInstance char nt CMImos3AssignMP char char double nt CMImos3AssignIP char char double nt CMImos3SetupModel char nt CMImos3SetupInstance char char
43. gate nodes are xg external mg middle and ig internal drain nodes are xd external and id internal source nodes are xs external and is internal bulk nodes are xb external ib internal db internal at the drain side and sb internal at the source side nqs is a non quasi static model node for modeling all intrinsic and extrinsic effects found in ultra small MOSFETs HSPICE MOSFET Models Manual 557 X 2005 09 8 Customer Common Model Interface Extended Topology In addition to the gate current components in CMI _VAR of CMldef h a pre determined header file named topo101 h is provided located in the extcmi mos101 directory in which a data structure TOPO101 has been defined This structure is accessible through a topovar pointer in CMI VAR of the CMldef h file In addition to those existing in Customer CMI data structure CMI VAR the other necessary calculated quantities such as current charge conductance or transconductance capacitance of transcapacitance noise biases derivatives of bias dependent source and drain conductance are passed to the HSPICE engine using TOPO0101 Remember that header file topo101 h is shared internally in HSPICE so you must not modify it Additional member variables introduced in TOPO101 of topo101 h see Figure 44 for node reference are listed as follows double bsim4 ver BE double mult
44. vdsatiiO V 0 9 Nominal drain saturation voltage at threshold for the impact ionization current VECB V 0 026v Electron tunneling from the conduction band VEVB V 0 075v Electron tunneling from the valence band voff V 0 08 Offset voltage in the subthreshold region for large W and L values VrecO V 0 0 Voltage dependent parameter for the recombination current vsat m sec 8e4 Saturation velocity at Temp Tnom vthO V NMOS 0 7 Threshold voltage Vbs 0 for a long wide device PMOS 0 7 VtunO V 0 0 Voltage dependent parameter for the tunneling current w0 m 0 Narrow width parameter wint m 0 0 Width offset fitting parameter from I V without bias wr 1 Width offset from Weff for the Rds calculation 472 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model Table 131 MOSFET Level 57 AC amp Capacitance Parameters Parameter Unit Default Description acde m V 1 0 Exponential coefficient for the charge thickness in the CapMod 3 for the accumulation and depletion regions asd V 0 3 Smoothing parameter for the source drain bottom diffusion cf F m cal Fringing field capacitance of the gate to source drain cgdl F m 0 0 Overlap capacitance for the lightly doped drain gate region cgdo F m 0 Non LDD region drain gate overlap capacitance per channel length CGEO F m 0 Gate substrate overlap capacitance per unit channel length cgsl F m 0 0 Overlap capacitance for the lightly
45. 0 000000E 00 VFB 5 760000E 01 LVFB 0 000000E 00 WVFB 0 000000E 00 PHI 6 500000E 01 LPHI 0 000000E 00 WPHI 0 000000E 00 K1 9 900000E 01 LK1 0 000000E 00 WK1 0 000000E 00 K2 1 290000E 01 LK2 0 000000E 00 WK2 0 000000E 00 ETAO 4 840000E 03 LETAO 0 000000E 00 WETAO 0 000000E 00 ETAB 5 560000E 03 LETAB 0 000000E 00 WETAB 0 000000E 00 MUO 3 000000E 02 MUOB 0 000000E 00 LMUOB 0 000000E 00 HSPICE MOSFET Models Manual 377 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model 378 WMUOB 0 000000E 00 MUSO 7 050000E 02 LMUSO 0 000000E 00 WMUSO 0 000000E 00 MUSB 0 000000E 00 LMUSB 0 000000E 00 WMUSB 0 000000E 00 MU20 1 170000E 00 LMU20 0 000000E 00 WMU20 0 000000E 00 MU2B 0 000000E 00 LMU2B 0 000000E 00 WMU2B 0 000000E 00 MU2G 0 000000E 00 LMU2G 0 000000E 00 WMU2G 0 000000E 00 MU30 3 000000E 01 LMU30 0 0000
46. DNS NI 1 cm3 0 0 Surface substrate doping 6 7 DVIN V 0 0 Adjusts the empirical surface inversion voltage 38 DVSBC V 0 0 Adjusts the empirical body effect transition 38 voltage E1 3 9 Dielectric constant of first film 40 E2 0 0 Dielectric constant of second film 40 ETA y 0 0 Static feedback factor for adjusting the 3 40 Level 40 threshold voltage difficulty of band bending 0 0 Drain induced barrier lowering DIBL effect 8 coefficient for the threshold voltage 0 0 Channel length independent drain induced 38 118 barrier lowering HSPICE MOSFET Models Manual X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 27 Threshold Voltage Parameters Continued Name Alias Units Default Description Level FDS 0 0 Field drain to source Controls the threshold 6 7 reduction due to the source drain electric field FSS NFS cm2 v 0 0 Number of fast surface states 5 38 GAMMA y1 2 0 5276 Body effect factor If you do not specify 1 2 3 GAMMA simulation calculates it from NSUB y1 2 Body effect factor 6 7 f you do not specify GAMMA simulation calculates it from DNB GAMMA is the body effect if vsb lt VBO If vsb gt VBO simulation uses LGAMMA GAMMA LGAMMA and VBO perform a two step approximation of a non homogeneous substrate LBetaGam um 0 0 BetaGam dependence on the channel length 38 LDVSBC V um 0 0 Adjusts the L dependent body effect transition 38 voltage LETA DIB
47. HSPICE MOSFET Models Manual X 2005 09 uble gdtotog dgdtoto dvg vdes vds uble gdtotob dgdtoto dvb vdes vds uble gdtotos gdtotod gdtotog gdtotob uble Igidlo electrical GIDL current uble ggidlob dIgidlo dvb uble ggidlog dIgidlo dvg uble ggidlod dIgidlo dvd uble Igislo electrical GISL current uble ggislob dIgislo dvb uble ggislog dIgislo dvg uble ggislos dIgislo dvs uble gcrgo channel reflected gate conductance uble gcrgod dgcrgo dVd uble gcrgog dgcrgo dVg uble gcrgob dgcrgo dVb uble gcrgos dgcrgo dVs dgcrgo dVd dgcrgo dVg dgcrgo dVb uble xparto charge partition flag used for NOS modeling uble gtau NQS term to model finite channel time constant uble grgeltd gate electrode conductance uble CoxWL gate oxide capacitance for NQS uble grbpd conductance between ib and db uble grbdb conductance between xb and db uble grbpb conductance between ib and xb uble grbps conductance between ib and sb uble grbsb conductance between xb and sb uble rgatemod flag to control gate resistance model rgatemod 0 1 2 3 uble rbodymod flag to control substrate resistance model rbodymod 0 1 2 uble rdsmod flag to control the use of source drain resistance model rdsmod 0 internal 1 external
48. LD SCALM LDIF m 0 Length of lightly doped diffusion adjacent to the gate ACM 1 2 LDIFscaled LDIF SCALM WMLT 1 Width diffusion layer shrink reduction factor XJ m 0 Metallurgical junction depth XJscaled XJ SCALM XW WDEL DW m 0 Accounts for masking and etching effects XWscaled XW SCALM Using an ACM 0 MOS Diode Figure 11 shows the parameter value settings fora MOSFET diode designed with a MOSFET that has a channel length of 3 um and a channel width of 10 um HSPICE MOSFET Models Manual 43 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models Figure 11 ACM 0 MOS Diode Source Gate Drain Contact gt KLD Example A transistor might include LD 5mm W 10mm L 3mm Parameter Description AD area of drain about 80 pm AS area of source about 80 pm CJ 4e 4 F m CJSW 1e 10 F m JS 1e 8 A m JSW 1e 13 A m NRD number of squares for drain resistance NRS number of squares for source resistance PD sidewall of drain about 36 um PS sidewall of source about 36 um 44 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models Calculating Effective Areas and Peripheries For ACM 0 simulation calculates the effective areas and peripheries as follows ADeff M AD WMLT SCALE ASeff M AS WMLT SCALE PDeff M PD WMLT SCALE PSef
49. MOSFET Models Manual 153 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model 154 The following equation calculates the AL value used in the preceding equation 1 3 AL 1e4 271363 XI Vas v4 PHI PHT nal in 129 nal The AL parameter is in microns if XJ is in microns and nat is in cm3 Subthreshold Current Ig If device leakage currents become important for operation near or below the normal threshold voltage then this model considers the subthreshold characteristics The Level 5 MOSFET model uses the subthreshold model only if the number of fast surface states FSS is greater than 1e10 The following equation determines the effective threshold voltage von von Vu fast The following equation calculates the fast value used in the preceding equation fast s 14 45099 rl 5 cox 2 6 v If von lt vinyp then simulation substitutes viny for von Note The Level 5 MOSFET device model uses the following subthreshold model only if vg lt von and if the device is either in partial or full enhancement mode Otherwise it uses the model in enhancement mode ZENH 1 The subthreshold current calculated below includes the residual DC current If Vgs von then Partial Enhancement v4s vg lt vde 2 Lj Bio NI vde cay von vg vde YE i cav y de v e us eer HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1
50. Note If you specify VMAX simulation calculates a different Vasat value Refer to the Vladimirescu document for details Mobility Reduction ue The mobility of carriers in the channel decreases as speeds of the carriers approach their scattering limited velocity The mobility degradation for the LEVEL 2 MOS model uses two different equations depending on the MOB mobility equation selector value If MOB 0 default ma UCRIT E UEXP a ala cox vs vg UTRA Bv 2 Because ue is less than UO the program uses the above equation if the bracket term is less than one otherwise the program uses Ug UO HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 2 IDS Grove Frohman Model If MOB 7 THETA O roe UO ef 14 THETA vg Vin vgs vth ueff UO If MOB 7 THETA 0 r UCRIT E UEXP Hep UO Cox MER If MOB 7 VMAX gt 0 eff eff 1 vde T aic VMAX Lo Channel Length Modulation To include the channel length modulation effect the LEVEL 2 MOS model modifies the lgs current lis L Re ds 1 A Pv i If you do not specify the LAMBDA model parameter the model calculates the X value LAMBDA gt 0 A LAMBDA VMAX gt 0 NSUB gt 0 and LAMBDA 0 X VMAX X 2 1 2 VMAX X d d d eff ds ef eff VMAX 0 NSUBSO and LAMBDA 0 HSPICE MOSFET Models Manual 133 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 2 IDS Grove Frohman Model 13
51. PHI vde vsb 3 2 PHI vsb 32 In the preceding equation vbi2 is the same as vbi for vsb gt VBO vd ids pa oag DV vde 5 yi VBO PHI vsb PHI yi yb VBO PHI v50 vsb For VBO vde vsb VBO the source side gate depletion region is in the implant layer but the drain side gate depletion region expands into the bulk area Alternate DC Model ISPICE model To invoke this model set the KU 1 model parameter Then the model computes vfu and vfa scale factors to scale both the vds voltage and the ids current These scale factors are functions of ECRIT and the vgs voltage The following equations compute the vfa and vfu factors KU Weser E LL LLL o2 KU 4 a KU 1 vfa KA vfu MAD HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model The following equation calculates the a value used in the preceding equations ECRIT Leff vgs vth Note The vfu factor is always less than one The following equation modifies the ids current NU 1 ids vfu MBL ids For NU 0 the vfu MBL factor is set to one The ids current is a function of the effective drain to source voltage vde vde min vds vfa vsat vdsat vfa vsat This alternate model is generally coupled with the mobility normal field equations MOB 3 and the channel length modulation drain field equation CLM 3 The mobility equations use
52. Potential for Good Fit to Data 574 Generally the model with the largest number of parameters has the best potential fit For the purpose of comparing the models simulation counts the number of parameters in two ways Measure Number of Parameters Simulation counts only the drain current parameters not the diode or series resistance nor gate capacitance and impact ionization parameters because these are almost the same for all levels LEVEL 2 VTO PHI GAMMA XJ DELTA UO ECRIT UCRIT UTRA UEXP NSUB LAMBDA NFS total 13 LEVEL 3 VTO PHI GAMMA XJ DELTA ETA UO THETA VMAX NSUB KAPPA NFS total 12 LEVEL 13 VFBO PHIO K1 K2 ETAO X2E X3E MUZ X2M X3MS MUS X2MS UOO X2UO Ul X2U1 X3U1 NO NDO NBO plus L and W variation parameters total 20 3 60 LEVEL 28 similar to LEVEL 13 minus MUS X2MS plus X33M WFAC WFACU total 21 3 63 HSPICE MOSFET Models Manual X 2005 09 B Comparing MOS Models Ease of Fit to Data LEVEL 39 VGHIGH VGLOW VFB K1 K2 ETAO ETAB MUO MUOB MUSO MUSB MU20 MU2B MU2G MU30 MU3B MU40 MU4B MUAG UAO UAB UBO UBB U10 U1B UID NO NB ND plus L and W parameters total 33 3 99 Measure Minimal Number of Parameters The minimal number of parameters is a subset of the above parameters which you use to fit a specific W L device LEVEL 2 and 3 drop
53. Source drain diffusion resistance and contact model selector specifies the end S D contact type point wide or merged and how to compute the S D parasitics resistance RSC Source contact resistance for per finger device TRNQSMOD Transient NQS model selector W BSIM4 MOSFET channel width in meters WNFLAG Turn on to select bin model based on width per NF for multi finger devices Improvements Over BSIM3v3 BSIM4 includes the following major improvements and additions over BSIM3v3 An accurate new model of the intrinsic input resistance Rii for both RF high frequency analog and high speed digital applications A flexible substrate resistance network for RF modeling Anew accurate channel thermal noise model and a noise partition model for the induced gate noise A non quasi static NQS model consistent with the Rii based RF model and a consistent AC model that accounts for the NQS effect in both transconductances and capacitances An accurate gate direct tunneling model HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Acomprehensive and versatile geometry dependent parasitics model for source drain connections and multi finger devices An improved model for steep vertical retrograde doping profiles A better model for pocket implanted devices in Vth the bulk charge effect model and Rout An asymmetrical and bias dependent source drain res
54. Table 141 MOSFET Level 60 DC Parameters SPICE Description Unit Default See Symbol Table 144 vthO Threshold voltage Vp 0 for the long and 0 7 nl 3 wide device k1 First order body effect coefficient y1 2 0 6 k2 Second order body effect coefficient 0 k3 Narrow width coefficient 0 k3b Body effect coefficient of k3 1 V 0 Vbsa Transition body voltage offset V 0 delp Constant for limiting Vbseff to fs V 0 02 Kb1 Coefficient of Vbs0 dependency on Ves 1 Kb3 Coefficient of Vbs0 dependency on Vys at the 1 subthreshold region 496 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model Table 141 MOSFET Level 60 DC Parameters Continued SPICE Description Unit Default See Symbol Table 144 DvbdO First coefficient of VbsO Leff dependency V 0 Dvbdi Second coefficient of VbsO Leff dependency V 0 w0 Narrow width parameter m 0 nlx Lateral non uniform doping parameter m 1 74e 7 dvtO First coefficient of the short channel effect on 2 2 Vth dvt1 Second coefficient of the short channel Vth 0 53 effect dvt2 Body bias coefficient of the short channel Vth 1 V 0 032 effect dvtOw First coefficient of the narrow width effect on Vth 0 z for a small channel length dvtiw Second coefficient of the narrow width effecton 5 3e6 Vth for a small channel length dvt2w Body bias coefficient of the narrow width effect 1 V 0 032 on Vth
55. UO cm V s Surface mobility Default 370 n type 215 p type VTO V Threshold voltage Default 0 83 n type 0 74 p type HSPICE MOSFET Models Manual X 2005 09 191 4 Standard MOSFET Models Level 1 to 40 LEVEL 27 SOSFET Model 192 Example The netlist for this example is located in the following directory Sinstalldir demo hspice mos ml27iv sp Non Fully Depleted SOI Model Several MOSFET models are currently available for SOS SOI applications The 3 terminal SOS model LEVEL 27 is stable for circuit design usage but has some limitations This model does not provide for depleted bulk Use it only with applications that are not fully depleted and that do not consider kink effects The following circuit example is a 4 terminal SOI model for incompletely depleted bulk with the kink effect Its sub circuit allows a parasitic capacitance to the substrate In this example the bulk is the region under the channel This model assumes that the substrate is the conductive layer under the insulator For SOI the insulator is usually silicon dioxide and the substrate is silicon For SOS the insulator is sapphire and the substrate is the metal that contacts the back of the integrated circuit die Model Components This model consists of the following subcomponents Core IDS model any level works because the impact ionization and weak inversion models are common to all DC levels The example uses a LEVEL 3
56. Vdsat 132 Mobility Reduction Ug 1 0 2 2 132 Channel Length Modulation ce eee eee 133 Subthreshold Current lgg eee cece eee 135 LEVEL 3 IDS Empirical Model 0 00 cece eee eee 136 LEVEL 3 Model Parameters 00 00 cece tees 136 LEVEL 3 Model Equations 0 00 0 cee eee 136 IDS Equations nues RR EE ED RE REIS eae Rae 136 Effective Channel Length and Width 0005 137 Threshold Voltage Vin lille 137 Saturation Voltage Vgsat liliis 138 Effective Mobility ug llle 138 Channel Length Modulation llli else 139 Subthreshold Current las GT 140 Compatibility Notes cee nee 141 Synopsys Device Model versus SPICES 141 Temperature Compensation ccc eee eee eee 142 Simulation results 0 0 000 cee 143 LEVEL 4 IDS MOS Model 2 2 2 2 00000 cee 143 vi HSPICE MOSFET Models Manual X 2005 09 Contents LEVEL SIDS Mod ls ues bho be Lab Abe ERE 144 LEVEL 5 Model Parameters llle 144 IDSJEQUAlIOLiS tirer etr dete tu uto Se ate bas e e N 145 Effective Channel Length and Width 0055 146 Threshold Voltage v a 146 Saturation Voltage Vasat luus 147 Mobility Reduction UB 147 Channel Length Modulation llli less 148 Subthreshold Current lyg isis 148 Depletion Mode DC Model ZENH 0 00000 else 149 IDS Equations Depletion Model LEV
57. compatibility with previous versions 246 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Revision I September 1997 Description The narrow channel effect on the substrate factor was revised to improve the transcapacitances behavior The narrow channel effect is no longer a function of the vs source voltage but of the v pinch off voltage Consequence the WETA and DW narrow channel effect parameters require different numerical values to achieve the same effect Revision Il July 1998 Intrinsic time constant Description Simulation calculates the tg intrinsic time constant as a function of the effective B factor including vertical field dependent mobility and short channel effects instead of the maximum mobility using the KP parameter Consequence The NGS time constant has an additional gate voltage dependence resulting in more conservative lower estimation of the NQS time constant at high Vg and additional dependence on short channel effects Thermal noise Description Simulation calculates the Sinermai thermal noise power spectral density as a function of the effective B factor including vertical field dependent mobility and short channel effects instead of the maximum mobility using the KP parameter Consequence Sipermai has an additional gate voltage and short channel effect dependence Optional process parameters to calculate electrical in
58. grading coefficient PBS PBS 1 0V No Bottom junction built in potential PBD PBD PBS No Bottom junction built in potential PBSWS PBSWS 1 0V No Isolation edge sidewall junction built in potential PBSWD PBSWD PBSWS No Isolation edge sidewall junction built in potential PBSWGS PBSWGS PBSWS No Gate edge sidewall junction built in potential PBSWGD PBSWGD PBSWS No Gate edge sidewall junction built in potential HSPICE MOSFET Models Manual X 2005 09 459 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 125 Temperature Dependence Parameters Level 54 Parameter Default Binnable Description TNOM 27 X No Temperature at which simulation extracts parameters UTE 1 5 Yes Mobility temperature exponent KT1 0 11V Yes Temperature coefficient for the threshold voltage KT1L 0 0Vm Yes Channel length dependence of the temperature coefficient for the threshold voltage KT2 0 022 Yes Body bias coefficient of the Vy temperature effect UA1 1 0e 9m V Yes Temperature coefficient for UA UB1 1 0e 18 m Yes Temperature coefficient for UB v UC1 0 056 V for Yes Temperature coefficient for UC MOBMOD 1 0 056e 9m V 2 for MOBMOD 0 and 2 AT 3 3e4m s Yes Temperature coefficient for the saturation velocity PRT 0 00hm m Yes Temperature coefficient for Rdsw NJS NJD NJS 1 0 No Emission coefficients of junction for the source and NJD NJS drain junctions XTIS XTID XTIS 3 0 No Junction current temperature exponents for the XTID XTI
59. s0 epsfm is the effective equivalent dielectric constant of the insulator layers fval 0 84 Sens ng 1 2 2 x FREQ TAU Cgdi f Cfmlw f efm 0 8 exp fval FEFF vgs vth vds Cgsi f Cfmlw f efm 0 8 exp fval FEFF vgs vth vds vdsx vdsat Cgdi Cfmlw Cgsi EM L 54 If vdsx 40 then cdnorm vdsx vesx vini Normalized drain current gm vdsx BAS yg vgsx vth vdsx cdl beta cdnorm Drain current without velocity saturation effect beta beta fgate idrain beta cdnorm gm beta gm dfgdvg cdl Velocity saturation factor if vwAx o then sj fdrain vdsx vdsc fgate fdrain dfddvg dfgdvg HSPICE MOSFET Models Manual 209 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 40 HP a Si TFT Model 210 fdrain dfddvd vdsc Strong inversion current gm fdrain gm t dfddvg idrain BAS gy fdrain gds g dfddvd idrain idrain fdrain idrain beta beta fdrain Ids idrain f GO vgs DEFF vds f gm s GO gds feds fp GO DEFF gm Weak inversion current if vgs vos then vgs paa idrain idrain exp Gra Ids idrain f GO vgs DEFF vds _ idrain _ EM ifi vt xn gm fKgm GO vgs von vt xn BAS gy BAS gp exp gds 8ds 5 f GO DEFF vdsx 0 Ids f GO vgs DEFF vds gm GO BAS gy beta vgsx vth If vFs 0 and vgs lt
60. severity of the warning for five parameters Level 49 issues a warning that the model exceeded the parameter range but continues with the simulation e However the Berkeley release issues a fatal error and aborts the simulation These five parameters are NGATE DVT1W DVT1 DSUB and DROUT See Parameter Range Limits on page 428 for more details e Improvements in numerical stability Provides improvements in numerical stability In most practical situations these improvements do not affect compliance with the Berkeley release but improve the convergence and the simulation time evel 53 maintains full compliance with the Berkeley release including numerically identical model equations identical parameter default values and identical parameter range limits Level 49 and 53 both support the following instance parameters along with the DELVTO instance parameter for local mismatch and NBTI negative bias temperature instability modeling MULUO low field mobility UO multiplier Default 1 0 e MULUA first order mobility degradation coefficient UA multiplier e MULUB second order mobility degradation coefficient UB multiplier HSPICE MOSFET Models Manual 397 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models 398 Both Levels 49 and 53 support a superset of model parameters that include HSPICE specific parameters For Level 53 in all cases HSPICE specific parameters default
61. silicon gate construction with an ion implant used to obtain the depletion characteristics A special model is required for depletion devices because the implant used to create the negative threshold also results in a complicated impurity concentration profile in the substrate The implant profile changes the basis for the traditional calculation of the QB bulk charge The additional charge from the implant QBI must be calculated This implanted layer also forms an additional channel offering a conductive pathway through the bulk silicon as well as through the surface channel This second pathway can cause difficulties when trying to model a depletion device using existing MOS models The surface channel partially shields the bulk channel from the oxide interface and the mobility of the bulk silicon can be substantially higher Yet with all of these differences a depletion model still can share the same theoretical basis as the Ihantola and Moll gradual channel model The depletion model differs from the Ihantola and Moll model Implant charge accounted for Finite implant thickness DP Assumes two channels a surface channel and a bulk channel Bulk channel has a bulk mobility UH Assumes that the bulk gain is different from the surface gain HSPICE MOSFET Models Manual 149 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model In the depletion model the gain is lower at low gate voltages and hig
62. vde vsb 3 2 PHI syl 2 The following equations calculate values used in the preceding equation vde min vds vdsat NWEscaled n 21l weff ueff _ weff B ueff COX Leff Then vbi and y values define the narrow width effect The NWE or NWM model parameters also specify the narrow width effect The vbi and y parameters specify the short channel effect Effective Channel Length and Width The following equations calculate the effective channel length and width from the drawn length and width leff Lscaled LMLT XLscaled 2 P LDscaled DELscaled weff M Wscaled WMLT XWscaled 2 PWDscaled LREFeff LREFscaled LMLT XLscaled 2 P LDscaled DELscaled WREFeff M WREFscaled WMLT XWscaled 2 PWDscaled Threshold Voltage vth The following equation determines the effective threshold voltage vth vbi y PHI vsb The vbi and y built in voltage value depends on the specified model parameters HSPICE MOSFET Models Manual 159 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model 160 Single Gamma VBO 0 If you set the VBO model parameter to zero simulation uses the single gamma model which treats the LGAMMA parameter as a junction depth To modify the GAMMA parameter for the short channel effect this model then uses the scf factor which the Poon and Yau formulation computes In this case simulation multiplies LGAMMA by the SCALM option
63. vo tno KP t KP E tno Fi n a no Channel Length Modulation Temperature Equation If you specify the LAMEX model parameter then the temperature modifies the LAMBDA value LAMBDA ft LAMBDA 1 LAMEX At HSPICE MOSFET Models Manual 105 X 2005 09 2 Technical Summary of MOSFET Models Temperature Parameters and Equations Calculating Diode Resistance Temperature Equations The following equations are examples of the effective drain and source resistance RD t RS 1 TRD At RS t RS 1 TRS At 106 HSPICE MOSFET Models Manual X 2005 09 3 Common MOSFET Model Parameters Lists and describes parameters that are common to several or all MOSFET model levels Parameters that are unique to a specific MOSFET model level are described in later chapters as part of the description of the specific model level that uses the parameter HSPICE MOSFET Models Manual 107 X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Basic MOSFET Model Parameters In this section Table 25 lists the basic MOSFET model parameters Table 26 lists effective width and length parameters Table 27 lists threshold voltage parameters Table 28 lists mobility parameters Table 25 Basic MOSFET Model Parameters Name Alias Units Default Description Level LEVEL 1 0 DC model selector All LEVEL 1 default is the Schichman levels Hodges model LEVEL 2
64. 0 PDW Lin iy tale gw del Sak hb GG NG a del d 1 0 2 0 3 0 4 0 584 HSPICE MOSFET Models Manual X 2005 09 LEVEL 13 28 39 Gds Model versus Data gds vs Vds at Vgs 2 3 4 5 Vos 0 B Comparing MOS Models Examples of Data Fitting These models still have a small change in slope of Gds at Vdsat more visible for the Level 13 model than for Level 28 or Level 39 Param Lin PDW Lin 008 MODEL i 00S_DATA pe Figure 58 LEVEL 28 gds versus Vds Curves Param Lin 140 00 2 0 PDW Lin 00S MODEL Dip ub tra 00S DATA aT HSPICE MOSFET Models Manual X 2005 09 585 B Comparing MOS Models Examples of Data Fitting Figure 59 LEVEL 39 gds versus Vds Curves Param Lin PDW Lin TOO AUT Sh AN AA Be TC Te Pe She sl tes ie bitin Le 0 00S_MODEL amp 00S DATA S5 LEVEL 2 3 28 Ids Model versus Data Ids vs Vgs at Vds 0 1 Vgs 0 1 2 3 4 Figure 60 LEVEL 2 Ids versus Vgs Curves 1 00 Param Lin 0 500 0M 1 0 1 50 2 0 PGD Lin C_MODEL C_DATA G 586 HSPICE MOSFET Models Manual X 2005 09 B Comparing MOS Models Examples of Data Fitting Figure 61 LEVEL 28 Ids versus Vgs Curves 1 00 140 00 10 00 1 00 140 00 10 00 Param Lin 1 00 140 0F 10 0F 1 0F So S d 1000F 2 1 UNT X kao
65. 0 NF gt 1 and SD 0 are given in model cards no evaluation of such effect is performed 438 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 109 Supported HSPICE BSIM4 STI LOD Parameters Parameter Unit Default Bin Description SA 0 0 Distance between S D diffusion edge to poly gate instance edge from one side If not given or if lt 0 the parameter stress effect is turned off SB 0 0 Distance between S D diffusion edge to poly gate instance edge from the other side If not given or if lt 0 parameter the stress effect is turned off SD 0 0 Distance between neighboring fingers For NF instance gt 1 If not given or lt 0 stress effect is turned off parameter STIMOD 0 0 V lt 4 3 STI model selector which gives priority to the Also 1 0 V gt 4 3 instance parameter instance 0 No STI effect parameter 1 UCB s STI model 2 TSMC s STI model SAREF M 1e 06 No Reference distance for SA gt 0 0 SBREF M 1e 06 No Reference distance for SB gt 0 0 WLOD M 0 0 No Width parameter for stress effect KUO M 0 0 No Mobility degradation enhancement coefficient for stress effect KVSAT M 0 0 No Saturation velocity degradation enhancement parameter for stress effect 1 0 lt kvsat lt 1 0 TKUO 0 0 No Temperature coefficient of KUO LKUO 0 0 No Length dependence of KUO WKUO 0 0 No Width dependence of KUO LLODKUO 0 0 No
66. 00 Common SPICE Parameters CGDO 1 000000 E 09 CGSO 1 000000 E 09 HSPICE MOSFET Models Manual X 2005 09 CGBO 2 500000E 11 RSH 3 640000E 01 JS 6 BSIM MOSFET Models Levels 13 to 39 1 380000E 0 00E 06 PB 8 000000E 01 PBSW 8 000000E 01 CJ 4 310000E 04 CJSW 3 960000E 10 MJ 4 560000E 01 MJSW 3 020000E 01 Synopsys Parameters ACM 3 LMLT 8 500000E 01 WMLT 8 500000E 01 XL 5 000000E 08 LD 5 000000E 08 XW 3 000000E 07 WD 5 000000E 07 CJGATE 2 000000E 10 HDIF 2 0000 LDIF 2 000000E 07 RS 2 000000E 03 TRS 2 420000E 03 RD 2 000000E 03 TRD 2 420000E 03 TCV 1 420000E 03 BEX 1 720000E 00 FEX 2 820000E 00 LMUO 0 000000E 00 WMUO 0 000000E 00 JSW 2 400000E 12 References References 1 Duster J S Jeng M C Ko P K and Hu C User s Guide for the BSIM2 Parameter Extraction Program and the SPICES with BSIM Implementation Industrial Liaison Program Software Distribution Office University of California Berkeley May 1990 HSPICE MOSFET Models Manual X 2005 09 379 6 BSIM MOSFET Models Levels 13 to 39 References 380 HSPICE MOSFET Models Manual X 2005 09 BSIM MOSFET Models Levels 47 to 65 Lists and describes seven of the new
67. 0e 9 Width dependence coefficient of THE3R GAM1R 145 0e 3 77 0e 3 Drain induced threshold shift coefficient for high gate drive SLGAM1 160 0e 9 105 0e 9 Length dependence of GAM1R SWGAM1 10 0e 9 11 0e 9 Width dependence of GAM1R ETADSR 600 0e 3 600 0e 3 Exponent of drain dependence of GAM1R ALPR 3 0e 3 44 0e 3 Channel length modulation factor ETAALP 150 0e 3 170 0e 3 Exponent of length dependence of ALPR SLALP 5 65e 3 9 0e 3 Coefficient of length dependence of ALPR SWALP m 1 67e 9 180 0e 12 Coefficient of width dependence of ALPR VPR V 340 0e 3 235 0e 3 Characteristic voltage for channel length modulation GAMOOR 18 0e 3 7 0e 3 Drain induced threshold shift coefficient at zero gate drive and zero back bias SLGAMOO m2 20 0e 15 11 0e 15 Length dependence of HSPICE MOSFET Models Manual X 2005 09 GAMOOR 217 5 Standard MOSFET Models Levels 50 to 64 Level 50 Philips MOS9 Model 218 Table 45 MOSFET Level 50 Model Parameters Continued Name Unit Default N Default P Description ETAGAMR 2 0 1 0 Exponent of back bias dependence of zero gate drive drain induced threshold shift MOR 500 0e 3 375 0e 3 Subthreshold slope factor STMO K 0 0 0 0 Temperature dependence coefficient MOR SLMO m 280 0e 6 47 0e 6 Length dependence coefficient of MOR ETAMR 2 0 1 0 Exponent of back bias dependence of the subthreshold slope ZET1R 420 0e 3 1 3 Weak inversion correction factor
68. 2 June 16 1998 3 2 98 2 Version 3 2 1 April 20 1999 3 21 99 2 Version 3 2 2 April 20 1999 3 22 99 2 3 23 01 4 3 2 4 02 2 Version 3 2 Features In June 1998 Berkeley released BSIM3 Version 3 2 which includes the following new features A new intrinsic capacitance model CAPMOD 3 includes finite charge layer thickness effects CAPMOD now defaults to 3 new parameters CAPMOD 3 ACDE and MOIN Improved modeling of C V characteristics at the weak to strong inversion transition new parameters NOFF and VOFFCV Vth dependence on Tox new parameter TOXM Flatband voltage parameter more accurately models different gate materials new parameter VFB Improved substrate current scalability with channel length new parameter APLHA1 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Restructured nonquasi static NQS model includes pole zero analysis and bug fixes NQSMOD is a BSIM3 element parameter HSPICE supports the model but not the element parameter Junction diode model temperature dependence new parameters TCJ TCJSW TCJSWG TPB TPBSW and TPBSWG Adjustable current limiting in the junction diode current model new parameter IJTH Use C V inversion charge equations CAPMOD 0 1 2 3 to calculate the thermal noise if NOIMOD 2 or 4 Eliminated the small negative capacitance values Cgs and Cgd in the accumulat
69. 2 15 TAU 1 64E 07 FEFF 0 5 MCKT 1 2 3 nch L le 05 W 4e 05 LEVEL 40 Model Equations The following equations show model parameters in all capital letters working variables are in lower case Model parameters and the vgs and vds bias voltages are inputs Ids gm and gds are the DC outputs The Cgs gate to source capacitance and the Cgd gate to drain capacitance are the AC outputs The electron charge is q the Boltzmann s constant is k and the permittivity of a vacuum is e0 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 40 HP a Si TFT Model This model applies the SCALE value before evaluating the equations and scales by M after evaluation gm and gds variables are intermediate not final quantities For a complete description of TFT technology and the device physics underlying these equations see the Hewlett Packard HP IC CAP manual Initially Cgdi 0 Cgsi 0 phi PHI vto VTO ANA wo UO If uo o then uo 1 The following equation computes the Cfm dielectric capacitance per unit area 0 El E2 If 71 o and 720 then cfm 09 52 If 72 o and rio then cfm SUAN kp uo Cfm 10 TEMP is the Synopsys device simulation temperature specified in C but converted to K internally to evaluate these equations t C TEMP q eg 2 10 TEMP 312 14 vto vto DELVTOmodel type DELVTOeleme
70. 2 4e3 pmos NOIC 1 4e 12 nmos No VBX at which the depletion region width equals XT 1 4e 12 pmos EM Vim 44e No Flicker noise parameter AF 1 0 No Flicker noise exponent KF 0 0 No Flicker noise coefficient EF 1 0 No Flicker noise frequency exponent 422 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Note See also Noise Models on page 96 for HSPICE noise model usage the NLEV parameter for HSPICE overrides Berkeley NOIMOD parameter Table 104 Junction Parameters MOSFET Levels 49 53 Name Unit Default Bin Description ACM 10 No Area calculation method selector HSPICE specific e ACM 0 3 uses the HSPICE junction models e ACM 10 13 uses the Berkeley junction models Level 49 ACM defaults to 0 JS Alm 0 0 No Bulk junction saturation current Default deviates from BSIM3v3 1 0e JSW A m 0 0 No Sidewall bulk junction saturation current NJ 1 No Emission coefficient use only with the Berkeley junction model ACM 10 13 N 1 No Emission coefficient HSPICE specific use only with the HSPICE junction model ACMz0 3 CJ F m 5 79e No Zero bias bulk junction capacitance Default deviates from BSIM3v3 5 0e CJSW F m 0 0 No Zero bias sidewall bulk junction capacitance Default deviates from BSIM3v3 5 0e CJSWG F m CJSW No Zero bias gate edge sidewall bulk junction capacitance use only with the Berkeley junction model ACM 10 13 CJGATE
71. 414 426 428 431 432 433 434 434 436 438 438 440 442 462 463 464 466 476 477 477 xiii Contents Using BSIM3 SOI PD escoa tereci eae 478 UCB BSIMSOI3 1 0 0 0000 0c teen ees 479 Ideal Full Depletion FD Modeling 000 479 Gate Resistance Modeling cece e eee 480 Gate Resistance Equivalent Circuit 0 00005 480 Enhanced Binning Capability lille 482 Bug FIX68 45 cid ug exe Dre Ee ette Sip Le taxon e RS 482 Level 59 UC Berkeley BSIM3 SOI FD Model 000000 eae 482 Level 59 Model Parameters naaa 484 Level 59 Template Output esee 492 Level 60 UC Berkeley BSIM3 SOI DD Model 0 00000 ae 492 Model Featutes ir cubi etx Aiea we ae RENE ive 493 AA uuo Last dria educta CR D RE ee a Sere eee Boe 493 Level 60 Model Parameters nannaa 495 Level 65 SSIMSOI Model 2 2 0 ene eas 505 Model Feature 000 c cee eee 505 Using Level 65 with Synopsys Simulators 0 00200 506 General Syntax for SSIMSOI 1 2 0 0 0c ee 506 8 Customer Common Model Interface 000s 517 Overview of Customer CMI 2222 nena 518 Directory Structure ax 204 BSA NAG ad DNA LAURA DEL xd wai ep X aeta 519 Running Simulations Using Customer CMI Models 2 521 Adding Proprietary MOS Models 00 0 cece eee eee ee 522 MOS Models on Unix Platforms 0 0 000 e
72. 490 CSDMIN 490 CTHO 491 DELTA 486 DLC 490 DROUT 486 603 Index DSUB 486 DVTO 487 DVTOW 487 DVT1 487 DVT1W 487 DVT2 487 DVT2W 487 DWB 487 DWC 490 DWG 487 ETAO 487 ETAB 487 gate capacitance basic 74 Meyer 76 overlap 75 impact ionization 61 ISBJT 487 ISDIF 487 ISREC 487 ISTUN 487 K1 487 K2 487 K3 488 K3B 488 KB1 488 KETA 488 KT1 491 KT2 491 KTIL 491 LEVEL 484 Level 60 495 LINT 488 MJSWG 490 MOBMOD 484 NCH 485 NDIO 488 NFACTOR 488 NGATE 485 NGIDL 488 NLX 488 NOIMOD 485 noise 96 NSS 489 NSUB 485 NTUN 488 PBSWG 490 PCLM 488 604 PDIBLC1 488 PDIBLC2 488 PRT 491 PRWB 488 PRWG 488 PVAG 488 RBODY 489 RBSH 489 RDSW 489 RSH 489 RTHO 491 SHMOD 485 SIIO 486 SIN 486 SII2 486 SIID 486 TBOX 485 temperature 98 threshold voltage 58 TNOM 491 TOX 485 TSI 485 TT 490 UO 489 UA 489 UA1 491 UB 489 UB1 491 UC 489 UC1 491 UTE 491 VBSA 489 VERSION 398 VOFF 489 VSAT 489 VSDFB 491 VSDTH 491 VTHO 489 WINT 489 WR 489 XBJT 491 XDIF 492 XPART 491 XREC 492 XTUN 492 See also Chapter 20 models AMD 5 6 AMI ASPEC 4 HSPICE MOSFET Models Manual X 2005 09 ASPEC AMI 4 Berkeley BSIM3 SOI 6 463 BSIM3 SOI DD 492 junction 402 BSIM 5 324 equations 332 LEVEL 13 example 339 BSIM2 6 358 362 BSIM3 6 382 equations 390 Leff Weff 389 BSIM3 SOI FD 482 BSIM3v3 MOS 397 NQS 402 CASMOS 5 GEC 5 mode
73. 5 then vc F3 Leff Note If you use the alternate saturation model vde is different for UPDATE 0 than it is for UPDATE 1 2 MOB 6 7 Modified MOB 3 This mobility equation is the same as MOB 3 except that the equation uses VTO instead of vth If you specify MOB 6 the following equation modifies the ids current ids Le FI vgs vth 242 ids UTRA Leff vde Channel Length Modulation The basic MOSFET current equation for ids describes a parabola where the peak corresponds to the drain to source saturation voltage vdsat Long channel MOSFETs generally demonstrate ideal behavior For vds voltages greater than vdsat ids current does not increase As channel length decreases current in the saturation region continues to increase The simulator models this increase in current as a decrease in the effective channel length Except for CLM 5 and 6 this model calculates the channel length modulation equations only when the device is in the saturation region HSPICE MOSFET Models Manual 175 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model The Level 6 MOSFET model provides several channel length modulation equations all except CLM 5 modify the ids equation ids AL Leff AL is the change in channel length due to MOSFET electric fields ids The CLM model parameter designates the channel length modulation equation for the Level 6 MOSFET device model Param
74. 6B VK 1 8 5E 4 8 5E 4 SLPHIB Coefficient of the length dependence of 9B 0 0 SL2PHIB Second coefficient of length dependence 0 0 of 9B SWPHIB Coefficient of the width dependence of 9B 0 0 BETSQ Gain factor for an infinite square transistor AV 2 3 09E 4 1 15E 4 at the reference temperature ETABETR Exponent of the temperature dependence 1 3 0 5 of the gain factor SLETABET Length dependence coefficient of nBR 0 0 HSPICE MOSFET Models Manual X 2005 09 287 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 63 Level 68 MOS11 Parameters Level 11010 Physical Geometry Scaling Continued Name Description Units NMOS PMOS FBET1 Relative mobility decrease due to first 0 0 lateral profile LP1 Characteristic length of first lateral profile m 8E 7 8E 7 FBET2 Relative mobility decrease due to second 0 0 lateral profile LP2 Characteristic length of second lateral M 8E 7 8E 7 profile THESRR Mobility reduction coefficient due to V 1 0 4 0 73 surface roughness scattering for reference transistor at reference temperature ETASR Exponent of temperature dependence of 0 65 0 5 SR SWTHESR Coefficient of the width dependence of 0 0 SR THEPHR Coefficient of the mobility reduction dueto V 1 1 29E 2 1E 3 phonon scattering for the reference transistor at the reference temperature ETAPH Exponent of the temperature dependence 1 35 3 75 of OSR for the reference temperature SWTHEPH C
75. CDO LX4 Channel current IDS All CBSO LX5 DC source bulk diode current CBSO All CBDO LX6 DC drain bulk diode current CBDO All GMO LX7 DC gate transconductance GMO All GDSO LX8 DC drain source conductance GDSO All 16 HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level GMBSO LX9 DC substrate transconductance All except 57 GMBSO 58 59 GMESO LX9 DC substrate transconductance 57 58 59 GMBSO GMESO GBDO LX10 Conductance of the drain diode GBDO All GBSO LX11 Conductance of the source diode GBSO All QB LX12 Total bulk body charge QB Meyerand All Charge Conservation CQB LX13 Bulk body charge current CQB Meyer All and Charge Conservation QG LX14 Total Gate charge QG Meyer and All Charge Conservation CQG LX15 Gate charge current CQG Meyer and All Charge Conservation QD LX16 Total Drain charge QD 49 53 QD LX16 Channel charge QD Meyer and All except 49 Charge Conservation and 53 CQD LX17 Drain charge current CQD 49 53 CQD LX17 Channel charge current CQD Meyer All except 49 and Charge Conservation and 53 CGGBO LX18 CGGBO dQg dVg CGS CGD 4 CGB All except 54 Meyer and Charge Conservation 57 59 60 CGGBO LX18 Intrinsic gate capacitance 54 57 59 60 CGDBO LX19 CGDBO dQg dVd Meyer and Charge All except 54 Conservation 57 59 60 CG
76. CMI_Noise CMI_VAR pvar char pmodel char pinst Parameter Description pvar Pointer to the CMI VAR variable pmodel Pointer to the model pinst Pointer to the instance Example int ifdef STDC CMImos3Noise CMI VAR pslot char pmodel char ptr else CMImos3Noise pslot pmodel ptr CMI VAR pslot char pmodel char ADULT endif double freq fourkt CMI_ENV penv MOS3instance ptran penv pCMIenv pCMIenv is a global ptran MOS3instance ptr fourkt 4 0 BOLTZMAN ptran gt temp 4kT freq pslot gt freq Drain resistor thermal noise as current 2 source pslot nois ird fourkt ptran gt gdpr HSPICE MOSFET Models Manual 543 X 2005 09 8 Customer Common Model Interface Interface Variables Source resistor thermal noise as current 2 source pslot nois irs fourkt ptran gt gspr Assumes that the thermal noise is the current 2 source referenced to the channel The source code for the thermalnoise is not shown here pslot nois idsth thermalnoise model here fourkt Assumes that the flicker 1 f noise is the current 2 Source referenced to the channel The source code for the flickernoise is not shown here pslot nois idsfl flickernoise model here freq return 0 int CMImos3Noise CMI PrintModel This
77. COX GAMMA PHI VTO KP or UCRIT You can also use a simpler mobility reduction model due to the vertical field Simulation uses the THETA mobility reduction coefficient only if you did not specify EO Name Unit Default Range Description LAMBDA 0 5 gt 0 Depletion length coefficient channel length modulation WETA 0 25 Narrow channel effect coefficient LETA 0 1 Short channel effect coefficient Name Unit Default Range Description QO QO A 20 Reverse short channel effect peak charge s m density LK m 0 29E 6 gt 1 0E 8 Reverse short channel effect characteristic length Name Unit Default Range Description IBA 1 m 0 First impact ionization coefficient IBB V m 3 0E8 gt 1 0E8 Second impact ionization coefficient IBN 1 0 2 0 1 Saturation voltage factor for impact ionization Name Unit Default Description TCV V K 1 0E 3 Threshold voltage temperature coefficient BEX 1 5 Mobility temperature exponent UCEX 0 8 Longitudinal critical field temperature exponent IBBT 1 K 9 0E 4 Temperature coefficient for IBB HSPICE MOSFET Models Manual 229 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model 230 Name Unit Default Description AVTO Vm ga Area related threshold voltage mismatch parameter AKP m 0 Area related gain mismatch parameter AGAMMA Vm 0 Area related body effect mismatch parameter a Only DEV values apply to the st
78. Coefficient for the geometry 1 04 0 86 independent part of nSAT PLTETASAT Coefficient for the length dependent 0 0 part of nSAT PWTETASAT Coefficient for the width dependent 0 0 part of nSAT PLWTETASAT Coefficient for the length times width 0 0 dependent part of nSAT POTA1 Coefficient for the geometry K 1 0 0 independent ST a1 part HSPICE MOSFET Models Manual 305 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS PLTA1 Coefficient for the length dependent K 1 0 0 part of ST a1 PWTA1 Coefficient for the width dependent K 1 0 0 part of ST a1 PLWTA1 Coefficient for the length times width K 1 0 0 dependent part of ST a1 POAGIDL Coefficient for the geometry AV 3 0 0 independent part of AGIDL PLAGIDL Coefficient for the length dependence AV 3 0 0 of AGIDL PWAGIDL Coefficient for the width dependence of AV 3 0 0 AGIDL PLWAGIDL Coefficient for the width over length AV 3 0 0 dependence of AGIDL POBGIDL Coefficient for the geometry V 41 0 41 0 independent part of BGIDL PLBGIDL Coefficient for the length dependence V 0 0 of BGIDL PWBGIDL Coefficient for the width dependence of V 0 0 BGIDL PLWBGIDL Coefficient for the length times width V 0 0 dependence of BGIDL POCGIDL Coefficient for the geometry 0 0 independent part of CGIDL PLCGIDL Co
79. Customer CMI Bias Polarity for N and P channel Devices The vds vgs and vbs input biases in CMI VAR are vds vd vs vgs vg vs vbs vb vs If your model code does not distinguish between the n channel and p channel bias then you must negate these biases for the P channel device The example routines multiply the biases by the type model parameter which is 1 for the N device or 1 for the P device HSPICE MOSFET Models Manual 563 X 2005 09 8 Customer Common Model Interface Conventions For example the following is the MOS3 model code if model MOS3type lt 0 P channel vgs VgsExt vds VdsExt vbs VbsExt else N channel vgs VgsExt vds VdsExt vbs VbsExt Use this code in both the CMI Evaluate andthe CMI DiodeEval functions Figure 46 on page 564 shows the convention to output current components For the channel current drain to source is the positive direction For substrate diodes bulk to source and bulk to drain are the positive directions These conventions are the same for both N channel devices and P channel devices The conventions for von are N channel device is on if vgs gt von P channel device is on if vgs lt von Figure 46 MOSFET node1 node2 node3 node4 N channel node1 drain node 11 M1 node2 gate node node4 substrate node 12 M1 3 14 M1 node3 source node I3 M1
80. Customer Common Model Interface Conventions 566 HSPICE MOSFET Models Manual X 2005 09 A Finding Device Libraries Describes how to use the HSPICE automatic model selector to find the proper model for each transistor size For libraries with multiple models of a specific element you can use an automatic model selector in HSPICE to automatically find the proper model for each transistor size This chapter describes how to use the model selector The model selector uses the following criteria LMIN XLREF lt L XL lt LMAX XLREF WMIN XWREFS W XW lt WMAX XWREF If you do not specify XLREF simulation sets it to XL If you do not specify XWREF simulation sets it to XW The model selector syntax is based on a common model root name with a unique extension for each model The following is an example of HSPICE syntax for MOSFET models M1 drain gate source bulk NJ W 2u L 1u MODEL NJ4 NJF WMIN 1 5u WMAX 3u LMIN 8u LMAX 2u MODEL NJ5 NJF WMIN 1 5u WMAX 3u LMIN 2u LMAX 6u Figure 47 on page 568 illustrates the MOSFET model selection method This example illustrates several pch x models with varying drawn channel lengths and widths in the model library The model root name is pch and the extensions are 1 2 6 The NJ4 instance of the NJ Element W 2 w L 1 u requires a model for which 1 5 u lt channel width lt 3 p and 0 8 u lt channel length lt 2 p The automatic mo
81. DC MOS model Subthreshold model the WIC 3 model parameter allows the older models to use the more advanced models found in the BSIM LEVEL 13 LEVEL 28 models The NO model parameter should have a typical value around 1 0 Impact ionization model set ALPHA and VCR parameters to enable the impact ionization model which is available to all MOS DC equations Typical values are ALPHA 0 1 and VCR 18 Charge conservation gate cap model CAPOP 9 XQC 4 prevents the floating bulk node from obtaining extreme values The automatic periphery diode area calculation method ACM is set to 3 to automatically calculate the source and drain resistances and diode junction leakage and the capacitance ACM 3 CJ 0 CJSW 0 CJGATE 4e 10 JS 0 JSW 1e 9 LD 1u HDIF 1 5u RS 40 RD 40 N 1 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 27 SOSFET Model Note These models assume that the source drain diffusions extend to the buried oxide The area part of the diode has no capacitance to bulk However the subcircuit includes linear capacitors to the substrate Obtaining Model Parameters Use the optimizing capabilities in the Level 27 MOSFET model to obtain the core IDS model parameters Use the optimizer to obtain the core model subthreshold and impact ionization parameters The subthreshold model selected is an improved BSIM type of model that was altered for the older models The charge conservatio
82. DELTA which is a W effect parameter LEVEL 13 and 28 drop the L and W terms and the X2E X3E and NDO second order effects LEVEL 39 drops ETAB MU40 MU4B MU4G and ND The resulting minimal parameter counts for the five models are LEVEL 2 12 LEVEL 3 11 LEVEL 13 17 LEVEL 28 18 LEVEL 39 28 Ease of Fit to Data Generally the larger the minimal number of parameters the more time you spend fitting the data Systematic L and W effect parameters of LEVEL 13 28 and 39 make fitting easier because you can optimize individual W L devices Then the individual models can calculate the final model parameters with L and W terms On the other hand the more physical parameters of LEVEL 2 and 3 are helpful because you can more easily predict the value from a knowledge of the process before fitting to I V data Examples of physical parameters are junction depths and doping concentrations Measure Physical Percentage of Parameters Starting with the minimal set of parameters MOSFET models calculate the percentage that are physical HSPICE MOSFET Models Manual 575 X 2005 09 B Comparing MOS Models Robustness and Convergence Properties For LEVEL 2 e PHI XJ UO ECRIT NSUB and NFS are physical e VTO GAMMA UCRIT UTRA UEXP and LAMBDA are empirical which is 5096 physical parameters For LEVEL 3 PHI XJ UO VMAX NSUB and NFS are physical which is 55 For LEVEL 13 only PHIO and MUZ are phy
83. DSUB 0 56 DIBL coefficient exponent DVBDO V 0 First coefficient of the Vys 0 dependency on Leff DVBD1 V 0 Second coefficient of the Vys 0 dependency on Leff 486 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 59 UC Berkeley BSIM3 SOI FD Model Table 136 MOSFET Level 59 DC Parameters Continued Parameter Unit Default Description DVTO 2 2 First coefficient of the short channel effect on Vi DVTOW 0 First coefficient of the narrow width effect on Vy for a small channel length DVT1 0 53 Second coefficient of the short channel effect on Vip DVT1W 5 3e6 Second coefficient of the narrow width effect on Vj for a small channel length DVT2 1N 0 032 Body bias coefficient of the short channel effect on Vj DVT2W 1 V 0 032 Body bias coefficient of the narrow width effect on Vyp for small channel length DWB m v 1 0 0 Coefficient of the substrate body bias dependence for Weff DWG m V 0 0 Coefficient of the gate dependence for Weff EDL m 2e 6 Electron diffusion length ETAO 0 08 DIBL coefficient in the subthreshold region ETAB 1 V 0 07 Body bias coefficient for the DIBL effect in the subthreshold region ISBJT A m 1 0e 6 BJT injection saturation current ISDIF A m 0 Body to source drain injection saturation current ISREC A m 1 0e 5 Recombination in the depletion saturation current ISTUN A m 0 0 Reverse tunneling saturation current K1 y1 2 0 6 Coefficient for the first order
84. F m 0 0 Gate source overlap capacitance per meter channel width DEFO eV 0 6 Dark Fermi level position DELTA 5 Transition width parameter EL eV 0 35 Activation energy of the hole leakage current EMU eV 0 06 Field effect mobility activation energy EPS 11 Relative dielectric constant of the substrate EPSI 7 4 Relative dielectric constant of the gate insulator GAMMA 0 4 Power law mobility parameter HSPICE MOSFET Models Manual 261 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 61 RPI a Si TFT Model 262 Name Unit Default Description GMIN m3ev 1E23 Minimum density of deep states IOL A 3E 14 Zero bias leakage current parameter KASAT 1 X 0 006 Temperature coefficient of ALPHASAT KVT V PX 0 036 Threshold voltage temperature coefficient LAMBDA 1 V 0 0008 Output conductance parameter M 2 5 Knee shape parameter MUBAND q ys 0 001 Conduction band mobility RD m 0 0 Drain resistance RS m 0 0 Source resistance SIGMAO A 1E 14 Minimum leakage current parameter TNOM 0G 25 Parameter measurement temperature TOX m 1E 7 Thin oxide thickness VO V 0 12 Characteristic voltage for deep states VAA V 7 5E3 Characteristic voltage for field effect mobility VDSL V 7 Hole leakage current drain voltage parameter VFB V 3 Flat band voltage VGSL V 7 Hole leakage current gate voltage parameter VMIN V 0 3 Convergence parameter VTO V 0 0 Zero bias threshold voltage HSPICE MOSFET Models Manual X 2005 09 5 Standar
85. HSPICE RF issues a warning and continues HSPICE calls CMI WriteError after every Customer CMI function call 546 HSPICE MOSFET Models Manual X 2005 09 Syntax int CMI _WriteError int err code 8 Customer Common Model Interface Interface Variables char err str Parameter Description err code Error code err str Points to the error message Example int ifdef STDC__ CM void char else CM int char endif EOF Imos3Writel err code Imos3WriteError err code err str err str int err status 0 switch err code case 1 strcpyn err s case 2 strcpyn err s default re Pe strcpyn err s turn err ser Err err str U us 1 Cat tr err s ser Warn us 1 Cat err s us 1 status CMImos3WriteError tr User Cat Ef int HSPICE MOSFET Models Manual X 2005 09 Eval CMI Error err code err str Eval CMI ERR STR LEN ERR STR LE Err Generic CMI ERR STR LEN 547 8 Customer Common Model Interface Interface Variables 548 CMI Start Before simulation this routine runs startup functions that you define Syntax int CMI_Start void Example int ifdef STDC__ CMImos3Start void else CMImos3Start void endif void CMImos3start ret
86. If UPDATE O and vgs vth F gt VF1 1 F4 F3 vgs vth factor factor If UPDATE 1 2 and vgs vth 2 gt VF1 1 factor F4 F1 F3 VF1 F3 vgs vth Table 34 MOB 4 and MOB 5 Universal Field Mobility Reduction Name Alias Units Default Description ECRIT V cm 0 0 Critical electric drain field for mobility reduction Zero indicates an infinite value F1 V cm 0 0 Source drain mobility reduction field typical values are 1e4 to 5e8 MOB 0 0 Selects a mobility equation Set MOB 4 for the critical field equation Set MOB 5 for the critical field equation with an independent drain field UEXP F2 1 v12 0 0 Bulk mobility reduction factor typical values are 0 to 0 5 UTRA F3 V cm 0 0 Critical electric drain field for mobility reduction HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model The MOB 4 equation is the same as the MSINC UN 3 equation The MOB 5 equation is the same as MOB 4 except that F3 substitutes for ECRIT in the vc expression The MOB 5 equation provides a better fit for CMOS devices in the saturation region Do not specify a VMAX value because the mobility equation calculates the velocity saturation 1 COX ves eth 2224 F2 vsb PHD F1 eox ve factor 1 If MOB 4 then vc ECRIT Leff If MOB
87. Jeng Ko and Hu and was released in SPICE3 in 1991 It is designed to model deep submicron devices It uses a Cubic spline to produce a smooth weak inversion transition and has many additional parameters for improved accuracy The GDS transition at VDSAT is markedly smoother than in BSIM1 Future for Model Developments This sequence of models shows a trend towards empirical rather than physical models and an increasing number of parameters It is unfortunate to lose contact with the physics but it can be unavoidable because the physics have become less universal Short channel devices are much more sensitive to the detail of the process I V curves from different manufacturers show qualitative differences in the shape of the curves Therefore the models need to be very flexible requiring a large number of empirical parameters HSPICE MOSFET Models Manual 573 X 2005 09 B Comparing MOS Models Model Equation Evaluation Criteria Model Equation Evaluation Criteria This section describes the following aspects of the model equations Potential for good fit to data Ease of fitting to data Robustness and convergence properties Behavior follows actual devices in all circuit conditions Ability to simulate process variation Gate capacitance modeling Some of these aspects depend on general features of the Synopsys MOSFET models that are the same for all levels Others result in simple objective measures for comparing the levels
88. MOS DATA ala Param Lin J i 1 0 2 0 3 0 4 0 5 0 PDW Lin 90 0U 200 00 Figure 38 IDS vs Vgs for Vds 0 1V Vbs 0 1 2 8 4V Showing Subthreshold Region Model vs Data 1 00 MODEL 140 0U C DATA d coc a 90 0U 7 E 100 140 0U 7 90 0U Param Lin 1 00 140 0U 90 0U l l i s 1 0F 7 9e m a Reh Talons NAL n a eli Dhan Gin gobaute E OF 9 500 0M 1 0 1 50 2 0 2 50 3 0 Volts Lin 376 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model Figure 39 gr lps VS Vgs for Vas 0 1V Vps O 2V BSIM2 Model vs Data 110 0 2 0J MODEL 100 0 E gt DJ DATA z DID 90 0 80 0 Z 70 0 60 0 Param Lin 50 0 40 0 30 0 20 0 10 0 ae Volts Lin Typical BSIM2 Model Listing In this example geometry sensitivities are set to zero because a fit at only one geometry has been performed This example includes extra HSPICE parameters for LDD temperature and geometry MODEL NCH NMOS LEVEL 39 TOX 2 000000E 02 TEMP 2 500000E 01 VDD 5 000000E 00 VGG 5 000000E 00 VBB 5 000000E 00 DL 0 000000E 00 DW 0 000000E 00 VGHIGH 1 270000E 01 LVGHIGH 0 000000E 00 WVGHIGH 0 000000E 00 VGLOW 7 820000E 02 LVGLOW 0 000000E 00 WVGLOW
89. MOS3von O O O o 9 ptran MOS3cd ptran MOS3gds ptran MOS3gm ptran gt MOS3gmbs tran gt MOS3gbd tran gt MOS3gbs tran gt MOS3capgs tran gt MOS3capgd tran MOS3capgb ptran gt MOS3capbd ptran gt MOS3capbs ptran MOS3cbs ptran gt MOS3chd Assign additional CMI VAR elements here for the substrate model and the overlap capacitances return 0 int CMI mos3 Mm Evaluate CMI DiodeEval Based on the bias conditions and the model instance parameter values this routine evaluates the MOS junction diode model equations It then passes all transistor characteristics via the CMI VAR Variable Syntax int CMI _DiodeEval CMI VAR pvar char pmodel char pinst Parameter Description pvar pmodel pinst Pointer to the CMI VAR variable Pointer to the model Pointer to the instance HSPICE MOSFET Models Manual X 2005 09 541 8 Customer Common Model Interface Interface Variables 542 Example int ifdef _ STDC CMImos3DiodeEval CMI_VAR pslot char pmodel char ptr else CMImos3Diode pslot pmodel ptr CMI_VAR pslot char pmodel char ADE fendif CMI_ENV penv MOS3instance ptran penv pCMIenv pCMIenv is global ptran MOS3instance ptr call model evaluation void CMImos3diode penv MOS3m
90. Models Level 1 to 40 LEVEL 3 IDS Empirical Model The following equation calculates the vp value used in the preceding equation Vp V PHI Or vj VTO GAMMA bPHI VTO is the extrapolated zero bias threshold voltage of a large device If you do not specify VTO GAMMA or PHI simulation computes these values see Common Threshold Voltage Equations on page 58 Saturation Voltage Vgsat The LEVEL 3 model determines the saturation voltage due to the channel pinch off at the drain side The VMAX parameter specifies the reduction of the saturation voltage due to the carrier velocity saturation effect _ Vgs Vth V sat 1 f B 2 3 2 Vasat Vsat Vc sat ve The following equation calculates the v value used in the preceding equations _ VMAX Ly AN us The next section defines the us surface mobility parameter If you do not specify the VMAX model parameter then Vdsat Vsat Effective Mobility uer The Level 3 model defines the carrier mobility reduction due to the normal field as the effective surface mobility us UO Vgg Vi Us 9 1 THETA v vy 138 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 3 IDS Empirical Model The VMAX model parameter model determines the degradation of mobility due to the lateral field and the carrier velocity saturation _ us VMAX 0 Mell 7 E Vo Otherwise ui
91. Models Levels 50 to 64 Level 58 University of Florida SOI Table 50 MOSFET Level 58 Flag Parameters Continued Parameter Unit Default Typical Value Description TPG 1 Type of gate polysilicon 1 opposite to body 1 same as body TPS 1 Type of substrate 1 opposite to body 1 same as body Table 51 MOSFET Level 58 Structural Parameters Parameter Unit Default Typical Value Description TOXF m 1 063 3 8 x10 9 Front gate oxide thickness TOXB m 0 5e 6 80 400 x109 Back gate oxide thickness NSUB cm3 1 0015 1015 1017 Substrate doping density NGATE cm3 0 0 1019 1020 Poly gate doping density O for no poly gate depletion NDS cm 5 0e19 4919 4920 Source drain doping density TB m 0 1e 6 30 100 Film body thickness x10 NBODY cm3 5 0e16 1017 1018 Film body doping density LLDD m 0 0 0 05 0 2 x10 LDD LDS region length 0 for no LDD NLDD cm 5 0019 4x1919 LDD LDS doping density 51619 LDD LDS treated as D S extensions DL m 0 0 0 05 0 15 x10 Channel length reduction DW m 0 0 Channel width reduction 0 1 0 5 x108 HSPICE MOSFET Models Manual X 2005 09 251 5 Standard MOSFET Models Levels 50 to 64 Level 58 University of Florida SOI 252 Table 52 MOSFET Level 58 Electrical Parameters Parameter Unit Default Typical Value Description NQFF cm 0 0 1010 Front oxide fixed charge normalized NQFB cm 0 0 191 Back oxide fixed charge normalized
92. Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS PWSDIBL Coefficient for the width dependent V 1 2 0 0 part of odibl PLWSDIBL Coefficient length times width V 1 2 0 0 dependent of odibl POMO Coefficient for the geometry 0 0 independent part of mO PLMO Coefficient for the length dependent 0 0 part of mO PWMO Coefficient for the width dependent 0 0 part of mO PLWMO Coefficient for the length times width 0 0 mO dependent part POSSF Coefficient for the geometry V 1 2 1 2E 2 1 0E 2 independent part of osf PLSSF Coefficient for the length dependent V 1 2 0 0 part of osf PWSSF Coefficient for the width dependent V 1 2 0 0 part of osf PLWSSF Coefficient for the length times width V 1 2 0 0 osf dependent part POALP Coefficient for the geometry 2 5E 2 2 5E 2 independent part of a PLALP Coefficient for the length dependent 0 0 part of a PWALP Coefficient for the width dependent 0 0 part of a HSPICE MOSFET Models Manual 297 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS PLWALP Coefficient for the length times width 0 0 dependent part of a VP Characteristic voltage of the channel V 5
93. NJ 1 0 Emission coefficient NJSW 1 0 Sidewall emission coefficient XTI 3 0 Junction current temperature exponent coefficient CJ 8 397247e 04Fm2 Bottom junction capacitance per unit area at zero bias CJSW 5 0e 10Fm Source drain sidewall junction capacitance per unit area at zero bias CJSWG 50e 10Fm Source drain gate sidewall junction capacitance per unit area at zero bias MJ 0 5 Bottom junction capacitance grading coefficient MJSW 0 33 Source drain sidewall junction capacitance grading coefficient MJSWG 0 33 Source drain gate sidewall junction capacitance grading coefficient PB 1 0V Bottom junction build in potential PBSW 1 0V Source drain sidewall junction build in potential HSPICE MOSFET Models Manual X 2005 09 319 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model Table 80 MOS DIODE Parameter Default Description PBSWG 1 0V Source drain gate sidewall junction build in potential VDIFFJ 0 5V Diode threshold voltage between source drain and substrate Table 81 Subthreshold Swing Parameter Default Description PTHROU 0 0 Correction for steep subthreshold swing Note Model parameter defaults in the above tables are valid only for versions 100 and 110 For other versions please refer to the following table Table 82 Model Parameter Version Defaults Parameter Version 100 110 Others VMAX 1 00e 7 7 00e 6 BGTMP1 9 03e 5 90 25e 6 BGTMP2 3 05e 7 100 0e 9 TOX
94. NPEAK is greater than 1620 simulation multiplies it by 1e 6 and converts UO to m Vsec as follows if UO is greater than 1 simulation multiplies it by 1e 4 You must enter the NSUB parameter in cm units The specified value of VTHOfor p channel in the MODEL statement should be negative HSPICE MOSFET Models Manual 387 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model 10 16 17 18 19 388 The default value of KT1 is 0 11 The negative sign ensures that the absolute value of the threshold decreases with increasing temperature for NMOS and PMOS You cannot set the LITL model parameter below a minimum value of 1 0e 9 m to avoid a possible divide by zero error After you adjust the temperature VSAT cannot go below a minimum value of 1 0e4 m sec to assure that it is positive after temperature compensation Seven model parameters accommodate the temperature dependencies of six temperature dependent model variables These parameters are KT1 and KT2 for VTH UTE for UO AT for VSAT UA1 for UA UB1 for UB and UC1 for UC Set up the temperature conversion between this model and SPICES as follows SPICE3 OPTION TEMP 125 MODEL NCH NMOS Level 8 TNOM 27 HSPICE TEMP 125 MODEL NCH NMOS Level 47 TREF 27 The SCALM option does not affect parameters that are unique to this model but it does af
95. NQFSW qm 0 0 1012 Effective sidewall fixed charge 0 for no narrow width effect NSF cm ev 0 0 1010 Front surface state density NSB cm2 eV 0 0 101 Back surface state density QM 0 0 0 5 Energy quantization parameter 0 for no quantization UO cm V s 7 0e2 200 700 nMOS Low field mobility 70 400 PMOS THETA cm V 7 0e2 0 1 3 x10 8 Mobility degradation coefficient VSAT cm s 1 0e 6 0 5 1 x10 Carrier saturated drift velocity ALPHA cm 0 0 2 45x108 Impact ionization coefficient 0 for no impact ionization BETA V cm 0 0 1 92x108 Impact ionization exponential factor 0 for no impact ionization BGIDL V cm 0 0 4 8 x109 Exponential factor for gate induced drain leakage 0 for no GIDL GAMMA 0 3 0 3 1 0 BOX fringing field weighting factor HSPICE MOSFET Models Manual X 2005 09 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 58 University of Florida SOI Table 52 MOSFET Level 58 Electrical Parameters Continued Parameter Unit Default Typical Value Description KAPPA 0 5 0 5 1 0 BOX fringing field weighting factor JRO Am 1 0e 10 1011 199 Body source drain junction recombination current coefficient M 2 0 1 0 2 0 Body source drain junction recombination ideality factor LDIFF m 1 0e 7 0 1 0 5 x108 Effective diffusion length in source drain SEFF cm s7 1 0e5 0 5 5 x10 Effective recombination velocity in source drain CGFDO Fm
96. No Reference gate oxide thickness VFB V 11 Yes DC flatband voltage NOFF 1 0 Yes l V parameter weak to strong inversion transition VOFFCOV 0 0 Yes C V parameter weak to strong inversion transition 424 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 106 MOSFET Levels 49 53 Version 3 2 Parameters Continued Name Unit Default Bin Description JTH A 0 1 No Diode limiting current ALPHA1 yi 0 0 Yes Substrate current parameter ACDE m V 1 0 Yes Exponential coefficient for the charge thickness in the accumulation and depletion regions MOIN m V 15 0 Yes Coefficient gate bias dependent surface potential TPB V K 0 0 No Temperature coefficient of PB TPBSW V K 0 0 No Temperature coefficient of PBSW TPBSWG V K 0 0 No Temperature coefficient of PBSWG TCJ V K 0 0 No Temperature coefficient of CJ TCJSW V K 0 0 No Temperature coefficient of CJSW TCJSWG V K 0 0 No Temperature coefficient of CJSWG LLC mln LL No Coefficient of the length dependence for the C V channel length offset LWC mwn LW No Coefficient of the width dependence for the C V channel length offset LWLC m n wn LWL No Coefficient of the length and width for the C V channel length offset WLC mvin WL No Coefficient of the length dependence for the C V channel width offset WWC mwwn WW No Coefficient of the width dependence for the C V channel width offset WWLC mWin WWL No Coefficient of the le
97. SHMOD and RTHO are also instance parameters They override the corresponding model parameters The effective thermal resistance and capacitance equations and equivalent circuit are RTHO hi WTHO s7 Wig Cj CTHO W p WTHO s HSPICE MOSFET Models Manual 277 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Level 63 Philips MOS11 Model 278 The Philips MOS Model 11 Level 1100 and 1101 are available as Level 63 in the Synopsys MOSFET models based on the Unclassified Report NL UR 2001 813 by R Langevelde Philips MOS Model 11 Level 1101 is an updated version of Level 1100 It uses the same basic equations as Level 1100 but uses different geometry scaling rules It includes two types of geometrical scaling rules physical rules and binning rules To select these scaling rules use the Version parameter 1100 11010 or 11011 The parasitic diode model includes the Philips JUNCAP Parasitic Diode Model For more information about the MOS Model 11 and the Philips JUNCAP Parasitic Diode Model see http www semiconductors philips com Philips Models Using the Philips MOS11 Model 1 Set Level 63 to identify the model as Philips MOS Model 11 2 Set the MOS11 version Set Version 1 100 to identify the model as Philips MOS Model 11 Level 1100 e Set Version 11010 to identify the model as Philips MOS Model 11 Level 1101 physical geometry s
98. Scaling Name Description Units NMOS PMOS LEVEL Level of this model 63 VERSION Version of this model 11010 LVAR Difference between the actual and the m 0 0 programmed poly silicon gate length LAP Effective channel length reduction per m 4E 8 4E 8 side due to the lateral diffusion of source drain dopant ions WVAR Difference between the actual and m 0 0 programmed field oxide opening WOT Effective reduction of the channel width m 0 0 per side due to the lateral diffusion of the channel stop dopant ions TR Temperature at which simulation 21 21 determines the reference transistor parameters 286 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 63 Level 63 MOS11 Parameters Level 11010 Physical Geometry Scaling Continued Name Description Units NMOS PMOS VFBR Flat band voltage for the reference V 1 05 1 05 transistor at the reference temperature STVFB Temperature dependence coefficient of V K 5E 4 5E 4 VFB KOR Body effect factor for the reference V1 2 0 5 0 5 transistor SLKO Coefficient of the length dependence of 0 0 kO SL2KO Second coefficient of the length 0 0 dependence of kO SWKO Coefficient of the width dependence ofkO 0 0 KPINV Inverse of the body effect factor poly V 1 2 0 0 silicon gate PHIBR Surface potential at the onset of strong V 0 95 0 95 inversion at the reference temperature STPHIB Temperature dependence coefficient of
99. Specific control options set in the OPTION statement used for MOSFET models include the following For flag options O is unset off and 1 is set on Option Description ASPEC This option uses ASPEC MOSFET model defaults and set units Default 0 BYPASS This option avoids recomputing nonlinear functions that do not change with iterations Default 1 MBYPAS BYPASS tolerance multiplier BYTOL MBYPASSxVNTOL Default 1 if DVDT 0 1 2 or as 3 Default 2 if DVDT 4 DEFAD Default drain diode area Default 0 DEFAS Default source diode area Default 0 DEFL Default channel length Default 1e 4m HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models Selecting Models Option Description DEFNRD DEFNRS DEFPD DEFPS DEFW GMIN GMINDC SCALE SCALM WL Default number of squares for drain resistor Default 0 Default number of squares for source resistor Default 0 Default drain diode perimeter Default 0 Default source diode perimeter Default 0 Default channel width Default 1e 4m Pn junction parallel transient conductance Default 1e 12mho Pn junction parallel DC conductance Default 1e 12mho Element scaling factor Default 1 Model scaling factor Default 1 Reverses order in VSIZE MOS element from the default order length width to width length Default 0 The AD element statement overrides the DEFAD default The AS element statement overrides the DEFAS default
100. Tunneling Current 0 0002 e eee eee Additional Instance Parameter support 00 05 An Extension to Support BSIM4 Topology Activating These Enhancements anaana naaa Conventions Bias Polarity for N and P channel Devices 4 5 Source Drain Reversal Conventions 0 0 00 eee eee Thread Safe Model Code 0 0 ccc eee A Finding Device Libraries 0 0 cece ee B Comparing MOS Models 2 2 0000 c eee History and Motivation III Synopsys LEVEL 2 LEVEL 3 LEVEL 13 LEVEL 28 LEVEL 39 HSPICE MOSFET Models Manual X 2005 09 Device Model Enhancements 000000 c au S MR CNN a 536 537 538 539 539 540 541 542 544 545 545 546 548 548 548 550 552 553 553 555 556 563 563 563 565 565 567 571 571 571 572 572 572 573 573 XV Contents Future for Model Developments aaa aaa aaa 573 Model Equation Evaluation Criteria 2 0 0 0 eee 574 Potential for Good Fit to Data es 574 Measure Number of ParameterS 0 lille 574 Measure Minimal Number of Parameters llle 575 Ease of Fit to DIE Iie EAA ETE E A ee re 575 Measure Physical Percentage of Parameters 575 Robustness and Convergence Properties 000ee eee cence 576 Continuous Derivatives i caus ra ar eek eke es eae es 577 Posi
101. V model equations The preceding equations use the following simple forms Weff W 4 WMLT XW HDIFeff HDIF WMLT HSPICE MOSFET Models Manual 403 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Parameter Description W width specified on the element line HDIF heavy diffusion length specified in the model card WMLT shrink factor specified in the model card XW etch mask effect factor specified in the model card Note These equations ignore the SCALM SCALE and M factor effects See MOSFET Diode Models on page 39 ACM 2 for further details TSMC Diode Model Starting in HSPICE version 2003 09 HSPICE MOSFET Level 49 ACM 12 BSIM3 version 3 2 or later supports a TSMC diode model You can use this TSMC diode model to simulate the breakdown effect the resistance induced non ideality factor and geometry dependent reverse current of a diode Order this model directly from Taiwan Semiconductor Manufacturing Company TSMC not from Synopsys See the TSMC web site http www tsmc com BSIM3v3 STI LOD HSPICE BSIM8v3 Level 49 supports UC Berkeley s STI LOD stress effect model see Table 93 To turn on this stress effect model in BSIM3v3 specify STIMOD 1 in your model cards 404 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 93 Supported HSPICE BSIM3v3 STI LOD Parameters Param
102. VERSION parameter in the MODEL statement to move LEVEL 13 BSIM and LEVEL 39 BSIM2 models between versions Using the HSPICE MOSFET Models Manual 331 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 332 VERSION parameter in a LEVEL 13 MODEL statement results in the following changes to the BSIM model Model Version Effect of VERSION on BSIM model 9007B Introduced the LEVEL 13 BSIM model no changes 9007D Removed the K2 limit 92A Changed the TOX parameter default from 1000 A to 200 A 92B Added the K2LIM parameter which specifies the K2 limit 93A Introduced the gds constraints 93A 02 Introduced the VERSION parameter 95 1 Fixed the nonprinting TREF and incorrect GMBS problems 96 1 Changed the flatband voltage temperature adjustment LEVEL 13 Equations This section lists the LEVEL 13 model equations Effective Channel Length and Width The effective channel length and width for LEVEL 13 depends on the specified model parameters If you specify DLO then Leff Lscaled LMLT DLO Ple 6 LREFeff LREFscaled LMLT DLO Ple 6 Otherwise if you specify XL or LD Leff Lscaled LMLT XLscaled 2 PLDscaled LREFeff LREFscaled LMLT XLscaled 2 PLDscaled If you specify DWO then Weff Wscaled WMLT DWO Ple 6 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model WREFeff WREFscaled WMLT DWO Ple 6 Otherwise if y
103. VK 1 3 638e 4 3 638e 4 dependence of BGIDL Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Name Description Units NMOS PMOS LEVEL Level of this model 63 VERSION Version of this model 11011 LVAR Difference between the actual and the m 0 0 programmed poly silicon gate length LAP Effective channel length reduction per m 4E 8 4E 8 side due to lateral diffusion of source drain dopant ions WVAR Difference between the actual and the m 0 0 programmed field oxide opening WOT Effective reduction channel width per m 0 0 side due to lateral diffusion of channel stop dopant ions TR Temperature at which the parameters oc 21 21 for the reference transistor have been determined VFB Flat band voltage for the reference V 1 05 1 05 transistor at the reference temperature POKO Coefficient geometry independent ko V1 2 0 5 0 5 part HSPICE MOSFET Models Manual 293 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS PLKO Coefficient for the length dependent of V1 2 0 0 ko PWKO Coefficient for the width dependent of V1 2 0 0 ko PLWKO Coefficient length times width ko V1 2 0 0 dependent KPINV Inverse of body effect factor poly V 1 2 0 0 silicon gate POPHIB Coefficient geometry independent PB V 0 950 0 950 part PLPHIB Coefficient for the len
104. WXSU1 um V 2 0 0 Width sensitivity XPART 1 0 Selects a gate capacitance charge sharing coefficient Table 84 Diffusion Layer Parameters MOSFET Level 13 Name Alias Units Default Description CJW CJSW F m 0 0 Zero bias bulk junction sidewall capacitance CJM CJ F m 4 5e 5 Zero bias bulk junction bottom capacitance DS m 0 0 Average size variation due to the side etching or the mask compensation not used IJS JS A m 0 Bulk junction saturation current HSPICE MOSFET Models Manual X 2005 09 329 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 330 Table 84 Diffusion Layer Parameters MOSFET Level 13 Continued Name Alias Units Default Description JSW A m 0 0 Sidewall bulk junction saturation current MJO MJ 0 5 Bulk junction bottom grading coefficient MJW MJSW 0 33 Bulk junction sidewall grading coefficient PJ PB V 0 8 Bulk junction bottom potential PJW PHP V 0 8 Bulk junction sidewall potential RSHM RSH ohm sq 0 0 Sheet resistance square WDF m 0 0 Default width of the layer not used The wire model includes poly and metal layer process parameters Table 85 Temperature Parameters Name Alias Units Default Description BEX 1 5 Temperature exponent for the MUZ and MUS mobility parameters FEX 0 0 Temperature exponent for the U1 mobility reduction factor TCV V K 0 0 Flat band voltage temperature coefficient TREF C 25 Temperature at which s
105. after adjusting the length and width and setting the VBS dependence This feature loses some accuracy in the saturation region particularly at high Vgs You might need to qualify BSIM1 models again if the following occur 1 Devices exhibit self heating during characterization which causes declining lgs at high Vgs This does not occur if the device characterization measurement sweeps Vg 2 Extraction produces parameters that result in negative conductance This model attempts voltage simulation outside the characterized range of the device Calculations Using LEVEL 13 Equations To verify the equations start some simple simulation and analysis tests and check the results with a hand calculator Check the threshold vdsat and ids for a very simple model with many parameters set to zero series resistance RSH 0 Turn off diode current JS JSW IS 0 Turn off the LEVEL 13 subthreshold current n0 200 Set the geometry parameters to zero so Leff L 1u Weff W 1u HSPICE MOSFET Models Manual 339 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 340 This test returns the following TOX value 2 00000e 3F 2 m COX The test is at vbs 0 35 so that phi vbs 1 0 The netlist for this test is located in the following directory Sinstalldir demo hspice mos tl sp Simulation Results ids vth vdsat 1 09907e 02 7 45000e 01 3 69000e 00 Calculations at vgs vds 5 vbs 0 35 phi vb
106. and drain HSPICE MOSFET Models Manual 347 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modified BSIM Model 348 Nonuniform doping profile for ion implanted devices m Channel length modulation JSubthreshold conduction Geometric dependence of the electrical parameters LEVEL 28 Model Parameters MOSFET Level 28 uses the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters It also uses the parameters described in this section which apply only to MOSFET Level 28 Table 87 Transistor Process Parameters Name Alias Units Default Description LEVEL 1 MOSFET model level selector Set this parameter to 28 for this model B1 0 0 Lower vdsat transition point LB1 um 0 0 Length sensitivity WB1 um 0 0 Width sensitivity B2 1 Upper vdsat transition point LB2 um 0 0 Length sensitivity WB2 um 0 0 Width sensitivity CGBO F m 2 0e 10 Gate to bulk parasitic capacitance F m of length CGDO F m 1 5e 9 Gate to drain parasitic capacitance F m of width CGSO F m 1 5e 9 Gate to source parasitic capacitance F m of width ETAO 0 0 Linear vds threshold coefficient LETA um 0 0 Length sensitivity HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modified BSIM Model Table 87 Transistor Process Parameters Continued Name Alias Units Default Description WETA um 0 0 Width sensitivity ETAMN 0 0 Minimum linear vds thresho
107. as TOXE TOXP XJ 1 5e 7m Yes S D junction depth GAMMA1 Y1 calculated Yes Body effect coefficient near the surface in equation V12 GAMMA2 Y2 Calculated Yes Body effect coefficient in the bulk in equation V NDEP 1 7e17cm 3 Yes Channel doping concentration at the depletion edge for the zero body bias NSUB 6 0e16cm 3 Yes Substrate doping concentration NGATE 0 0cm 3 Yes Poly Si gate doping concentration NSD 1 0e20cm 3 Yes Source drain doping concentration VBX calculated v No Vps at which the depletion region width equals XT XT 1 55e 7m Yes Doping depth RSH 0 0ohm square No Source drain sheet resistance RSHG 0 1ohm square No Gate electrode sheet resistance 446 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 115 Basic Model Parameters MOSFET Level 54 Parameter Default Binnable Description VTHO or 0 7V NMOS Yes Long channel threshold voltage at Vps 0 VTHO 0 7V PMOS VFB 1 0V Yes Flat band voltage PHIN PHIN 0 0V Yes Non uniform vertical doping effect on the surface potential K1 0 5V 1 2 Yes First order body bias coefficient K2 0 0 Yes Second order body bias coefficient K3 80 0 Yes Narrow width coefficient K3B 0 0v Yes Body effect coefficient of K3 WO 2 5e 6m Yes Narrow width parameter LPEO 1 74e 7m Yes Lateral non uniform doping parameter LPEB 0 0m Yes Lateral non uniform doping effect on K1 VBM 3 0V Yes Maximum applied body bias in
108. can tie the BSIM2 model like any other HSPICE model into the optimizer in a Synopsys circuit simulator to fit to actual device data For more information see Chapter 12 Statistical Analysis and Optimization in the HSPICE Simulation and Analysis Manual An example fit appears at the end of this section Modeling Guidelines Removing Mathematical Anomalies Because of the somewhat arbitrary geometric and bias adjustments made in the original BSIM2 parameters they can take on non physical values or values that are not mathematically allowed in Berkeley SPICE 3 This can lead to illegal function arguments program crashes and unexpected model behavior for example negative conductance You must satisfy the following guidelines and corrections at all geometries of interest and at biases up to double the supply voltages that is to Vg 2 VDD Vgs 2 VGG and Vp 2 VBB To avoid a drain current discontinuity at Vgs Vasat be sure that BI O if AIO 0 To prevent negative ggs be sure that ETA gt 0 MU3 gt 0 and MU4 lt MU3 4 VDD This should ensure a positive gg value at biases up to double the supply voltages To simplify matters set all MU4 parameters to zero You can obtain reasonably good fits to submicron devices without using MU4 HSPICE MOSFET Models Manual 373 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model 374 In the Level 39 MOSFET model U1S cannot become negative A
109. channel length in meters This parameter overrides DEFL in an OPTIONS statement Default DEFL with a maximum of 0 1m W MOSFET channel width in meters This parameter overrides DEFW in an OPTIONS statement DefaultZDEFW M Multiplier to simulate multiple SOI MOSFETS in parallel The M setting affects all channel widths diode leakages capacitances and resistances Default 1 AD Drain diffusion area Overrides DEFAD in the OPTIONS statement Default DEFAD AS Source diffusion area Overrides DEFAS in the OPTIONS statement Default DEFAS PD Perimeter of the drain junction including the channel edge Overrides DEFPD in the OPTIONS statement PS Perimeter of the source junction including the channel edge Overrides DEFPS in the OPTIONS statement NRD Number of squares of drain diffusion for the drain series resistance Overrides DEFNRD in the OPTIONS statement HSPICE MOSFET Models Manual 483 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 59 UC Berkeley BSIM3 SOI FD Model Parameter Description NRS Number of squares of source diffusion for the source series resistance Overrides DEFNRS in the OPTIONS statement NRB Number of squares for the body series resistance RTHO Thermal resistance per unit width f you do not specify RTHO simulation extracts it from the model card f you specify RTHO it overrides RTHO in the model card CTHO Thermal capacitance per unit width f you do not specify CTHO sim
110. charge models See M Chan K Y Hui C Hu and P K Ko IEEE Trans Electron Devices vol ED 45 pp 834 841 1998 The Level 49 53 MOSFET model supports only the model parameter implementation To invoke the NQS model specify the NQSMOD 1 parameter in the model card You can use NQSMOD with any of the CAPMOD Levels 0 3 but only with Version 3 2 Version 3 0 and 3 1 do not support NQS HSPICE Junction Diode Model and Area Calculation Method You can use two junction diode models with both Levels 49 and 53 the HSPICE junction model and the Berkeley junction model For the HSPICE junction model specify the ACM 0 1 2 or 3 model parameter value For the Berkeley junction model specify ACM 10 11 12 or 13 The default ACM value is 0 for Levels 49 or 10 for Level 53 For the junction current junction capacitance and parasitic resistance equations corresponding to ACM 0 1 2 3 see MOSFET Diode Models on page 39 Set ACM 10 11 12 or 13 to enable the Berkeley junction diodes and to add parasitic resistors to the MOSFET The parasitic resistor equations for ACM 10 13 correspond to the ACM 0 3 parasitic resistor equations ACM 10 13 all use the Berkeley junction capacitance model equations Bulk source capacitance if Ps gt Weff Cbs AS Cjbs PS Weff Cjbssw Weff Cjbsswg else Cbs AS Cjbs PS Cjbsswg The AS and PS area and perimeter factors default to 0 if you do not specify th
111. circuits The components of these circuits form the basis for all element and model equations The equivalent circuit for DC sweep is the same as the one used for transient analysis but excludes capacitances Figure 8 through Figure 10 display the MOSFET equivalent circuits The fundamental component in the equivalent circuit is the DC drain to source current ids Noise and AC analyses do not use the actual ids current Instead the model uses the partial derivatives of ids with respect to the vgs vds and vbs terminal voltages The names for these partial derivatives are as follows Transconductance Aids gm Aves Conductance O ids ds NGI ES mds Bulk Transconductance O ids bs BEE err The ids equation describes the basic DC effects of the MOSFET Simulation considers the effects of gate capacitance and of source and drain diodes separately from the DC ids equations Simulation also evaluates the impact ionization equations separately from the DC ids equation even though the ionization effects are added to ids HSPICE MOSFET Models Manual 35 X 2005 09 36 2 Technical Summary of MOSFET Models MOSFET Equivalent Circuits Figure 8 Equivalent Circuit MOSFET Transient Analysis Source O cgb rs Gate cgs vgs a a Uk vgd cgd rd Drain vds z cbs O Vbs vbd 4 T ANN e as ib cba D idb S
112. conductance double csat diode saturation current HSPICE MOSFET Models Manual X 2005 09 doubl doubl doubl doubl doubl doubl doubl doubl 00000000 8 Customer Common Model Interface Interface Variables capds drain to source capacitance nois irg Gate noise current 2 qgso gate to source old charge qgdo gate to drain old charge qgs gate to source charge qgd gate to drain charge vgsold gate to source old voltage vgdold gate to drain old voltage Interface Variables To assign the model instance parameter values and to evaluate the I V C V response you need fifteen interface routines For each new model an interface variable in the CMI_MOSDEF type defines pointers to these routines and to the model instance variables The include CMldef h file includes this variable typedef struct CMI MosDef char char char char int int int int int int int int int int int int int int int int int ModelName 100 InstanceName 100 pModel pInstance modelSize instSize CMI ResetModel char int int CMI ResetInstance char CMI AssignModelParm char char double CMI_AssignInstanceParm char char double CMI SetupModel char CMI SetupInstance char char CMI Evaluate CMI VAR char char CMI DiodeEval CMI VAR char char CMI Noise CMI VAR
113. default 1 Iwl default 0 wl default 0 win default 1 WW default 0 wwn default 1 wwl default 0 dwg default 0 dwb default 0 llc default 0 lwc default 0 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 94 Parameters Excluded from BSIM3 Lite Continued Parameter Comments Iwlc default 0 wlc default 0 WWC default 0 wwlc default 0 bO default 0 b1 default 0 vbx do not define vom do not define xt do not define nsub do not define nlx default 0 std default 1 74e 7 gammat1 do not define gamma2 do not define ngate Recommended default or set 0 k3 default 0 std default 80 k3b default 0 wO no effect dvtO default 0 std default 2 2 dvt1 default 0 std default 0 53 dvt2 default O std default 0 032 dvtOw default 0 HSPICE MOSFET Models Manual X 2005 09 409 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 94 Parameters Excluded from BSIM3 Lite Continued Parameter Comments dvtiw default 0 std default 5 3e6 dvt2w default O std default 0 032 dsub default 0 prwg default 0 prwb default 0 wr Recommended default or set 1 drout default 0 std default 0 56 pdiblc 1 default 0 std default 0 39 cit Recommended default or set 0 alpha0 Recommended default or set 0 for Version 3 2 kt1l default 0 Parameter Bi
114. developed and widely used models Synopsys has introduced LEVELs that are compatible with models developed by UC Berkeley The University of Florida Rensselaer Polytechnic Institute and others This chapter describes Overview of MOSFET Model Types Selecting Models General MOSFET Model Statement MOSFET Output Templates 2 HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models Overview of MOSFET Model Types Overview of MOSFET Model Types Before you can select the appropriate MOSFET model type to use in analysis you need to know the electrical parameters that are critical to your application LEVEL 1 models are most often used to simulate large digital circuits in situations where detailed analog models are not needed LEVEL 1 models offer low simulation time and a relatively high level of accuracy for timing calculations If you need more precision such as for analog data acquisition circuitry use the more detailed models such as the LEVEL 6 IDS model or one of the BSIM models LEVEL 13 28 39 47 49 53 54 57 59 and 60 For precision modeling of integrated circuits the BSIM models consider the variation of model parameters as a function of sensitivity of the geometric parameters The BSIM models also reference a MOS charge conservation model for precision modeling of MOS capacitor effects Use the SOSFET model LEVEL 27 to model silicon on sapphire MOS devices You can include photocurrent ef
115. different diode sidewall capacitances along the gate edge and field edge To select the MOS parasitic diode use the ACM model parameter ACM 0 default chooses SPICE style The alternatives likely to be of most interest to the BSIM2 user are ACM 2 and 3 ACM 2 calculates the diode area based on W XW and HDIF contact to gate spacing You can override the calculation from the element line You can specify LDIF spacer dimension RS and RD source and drain sheet resistance under the spacer for LDD devices and RSH sheet resistance of HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model the heavily doped diffusion Thus simulation properly calculates the total parasitic resistance of the LDD devices ACM 3 uses all features of ACM 2 Its calculations of diode parasitics take into account the sharing of source drains and different junction sidewall capacitances along the gate and field edges Use the GEO parameter to specify source drain sharing from the element line See MOSFET Diode Models on page 39 for details Skewing of Model Parameters As in any other Synopsys model you can set up the BSIM2 model file for skewing to reflect the process variation You can perform Worst Case or Monte Carlo analysis based on fab statistics For more information see Chapter 12 Statistical Analysis and Optimization in the HSPICE Simulation and Analysis Manual HSPICE Optimizer You
116. drain to 57 59 substrate capacitance igso LX38 Gate to Source Current 54 CBEBO LX38 CBEBO dQb dVe intrinsic floating 57 59 body to substrate capacitance igdo LX39 Gate to Drain Current 54 CEEBO LX39 CEEBO dQe dVe intrinsic substrate 57 59 capacitance CEGBO LX40 CEGBO aQe dVg intrinsic substrate to 57 59 gate capacitance CEDBO LX41 CEDBO dQe dVd intrinsic substrate to 57 59 drain capacitance CESBO LX42 CESBO dQe dVs intrinsic substrate to 57 59 source capacitance VBSI LX43 Body source voltage VBS Meyer and 57 58 59 Charge Conservation ICH LX44 Channel current Meyer and Charge 57 58 59 Conservation IBJT LX45 Parasitic BUT collector current Meyer 57 58 59 and Charge Conservation III LX46 Impact ionization current Meyer and 57 58 59 Charge Conservation HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level IGIDL LX47 GIDL current Meyer and Charge 57 58 59 Conservation ITUN LX48 Tunneling current Meyer and Charge 57 58 59 Conservation Qbacko LX49 Internal body charge 57 59 lbp LX50 Body contact current 57 59 Sft LX51 Value of the temperature node with 57 59 shmod 1 VBFLOAT LX52 Internal body node voltage if you do not 57 59 specify the terminal Rbp LX53 Combination of roody and rhalo 57 59 IGB LX54 Gate tunnelin
117. edge sidewall reverse saturation current density JSWD JSWD JSWS No Isolation edge sidewall reverse saturation current density JSWGS JSWGS 0 0A m No Gate edge sidewall reverse saturation current density JSWGD JSWGD JSWGS No Gate edge sidewall reverse saturation current density CJS CJS 5 0e 4 F m No Bottom junction capacitance per unit area at zero bias CJD CJD CJS No Bottom junction capacitance per unit area at zero bias MJS MJS 0 5 No Bottom junction capacitance grading coefficient MJD MJD MJS No Bottom junction capacitance grading coefficient MJSWS MJSWS 0 33 No Isolation edge sidewall junction capacitance grading coefficient 458 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 124 Asymmetric Source Drain Junction Diode Model Parameters MOSFET Level 54 Continued Parameter Default Binnable Description MJSWD MJSWD MJSWS No Isolation edge sidewall junction capacitance grading coefficient CJSWS CJSWSz5 0e 10 No Isolation edge sidewall junction capacitance F m per unit area CJSWD CJSWD CJSWS No Isolation edge sidewall junction capacitance per unit area CJSWGS CJSWGS CJSWS No Gate edge sidewall junction capacitance per unit length CJSWGD CJSWGDZ2CJSWS No Gate edge sidewall junction capacitance per unit length MJSWGS MJSWGS MJSWS_ No Gate edge sidewall junction capacitance grading coefficient MJSWGD MJSWGD MJSWS No Gate edge sidewall junction capacitance
118. eee eee 52 Calculating Effective Areas and Peripheries 53 Effective Saturation Current Calculations 54 Effective Drain and Source Resistances 54 MOS Diode Equations 0 0 00 ees 55 Bene 55 Using MOS Diode Capacitance Equations 000005 55 Common Threshold Voltage Equations liliis 58 Common Threshold Voltage Parameters aa 58 Calculating PHI GAMMA and VTO 000 cee eee 59 MOSFET Impact lonization lille ee 60 Calculating the Impact lonization Equations 61 Calculating Effective Output Conductance aaa 62 Cascoding Example 0 00 c eee eee 63 Gascode GirCulil 2 2 2 oue Roe pore dece BANA n heise le abies ada esa 64 MOS Gate Capacitance Models 0 0 0 ccc cence eae 64 Selecting Capacitor Models 0 0 0 ccc cee eens 64 TWANSCapacCnanCe siete a sedere eg bale a ete he DAG arate banay 66 Operating Point Capacitance Printout 00 0 eae 68 Element Template Printout ccc eee eee 69 iv HSPICE MOSFET Models Manual X 2005 09 Calculating Gate Capacitance Input File 2 2 2 2 2 0 e eee eee Calculations 00000 ce eee Results vance ee aa LR oleae Plotting Gate Capacitances Capacitance Control Options SCAING staat Pekar ease a epee ae teed MOS Gate Capacitance M
119. effective 38 channel width WFRC 10 4 s cm 0 0 FRC sensitivity to the effective 38 channel width WFSB 104v1 2 s em 0 0 FSB sensitivity to the effective 38 channel width WKBeta1 um 0 0 Width dependent implant 38 channel mobility modifier WKIO um 0 0 Width dependent residue 38 WKIO current coefficient WUO WUB cm um V s 0 0 UO sensitivity to the effective 38 channel width WVFRC 10 4A s cm V 0 0 pins wa a the effective 38 channel width HSPICE MOSFET Models Manual 125 X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters 126 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 Lists and describes standard MOSFET models Levels 1 to 40 This chapter describes the following standard MOSFET models Levels 1 to 40 LEVEL 1 IDS Schichman Hodges Model LEVEL 2 IDS Grove Frohman Model LEVEL 3 IDS Empirical Model LEVEL 4 IDS MOS Model LEVEL 5 IDS Model LEVEL 6 LEVEL 7 IDS MOSFET Model LEVEL 7 IDS Model LEVEL 8 IDS Model LEVEL 27 SOSFET Model LEVEL 38 IDS Cypress Depletion Model LEVEL 40 HP a Si TFT Model For information about standard MOSFET Models Levels 50 to 64 see Chapter 5 Standard MOSFET Models Levels 50 to 64 For information on BSIM MOSFET models based on models developed by the University of California at HSPICE MOSFET Models Manual 127 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 1 IDS Schichman Ho
120. every model parameter listed in the Level 50 Model Parameters table Use the JUNCAP model parameter to select one of two available parasitic junction diode models ACM or JUNCAP JUNCAP 1 selects the Philips JUNCAP model JUNCAP 0 default selects the ACM model Philips added a switch named TH3MOD to MOS Model 9 You can use this switch to re activate effective THES clipping which was removed in an earlier version of this model If THSMOD 1 default effective THES can be slightly negative and clipping does not occur If THSMOD 0 this model clips the effective THES to more than zero Model Statement Example HSPICE MOSFET Models Manual model nch nmos Level 50 ler le 6 wer 10e 6 lvar 0 0 lap 0 05e 6 wvar 0 0 wot 0 0 t tr 27 00 vtor 0 8 stvto 0 slvto 0 t si2vto O swvto 0 kor 0 7 slko 0 swko 0 kr 0 3 slk 0 swk 0 phibr 0 65 vsbxr 0 5 slvsbx 0 swvsbx 0 betsq 120e 6 etabet 1 5 thelr 0 3 stthelr 0 slthelr 0 stithel 0 swthel 0 the2r 0 06 stthe2r 0 slthe2r 0 stlthe2 0 swthe2 0 the3r 0 1 stthe3r 0 slthe3r 0 stlthe3 0 swthe3 0 gamir 0 02 slgaml 0 swgaml 0 etadsr 0 60 alpr 0 01 etaalp 0 slalp 0 223 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model swalp 0 vpr 0 4 gamo
121. extcmi mos101 CMImos 101eval c On the HSPICE engine side these quantities are used for convergence checks Jacobian stamping circuit element summary printout template output and so on They are hard coded and are not accessible by you Generalized Customer CMI only supports the following forty template namings that is temp1t1 or equivalently LX1 temp1t2 or equivalently 1X2 temp1t 40 or equivalently x40 HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Conventions These namings are case insensitive You should specify them in the evaluation function according to your need For example see the extcmi mos101 CMImos101eval c file Activating These Enhancements Customer CMI evolved in three phases Here s how to activate the different enhancements Original Customer CMI a Include OPTION CMIFLAG in the netlist b Assign topoid of CMI_MOSFET structure to 0 in the code Original Customer CMI with gate tunneling modeling and additional instance parameter support a Include OPTION CMIFLAG CUSTCMI 1 in the netlist b Assign topoid of CMI_MOSFET structure to 0 in the code Generalized Customer CMI with BSIM4 like topologies and additional instance parameter support a Include OPTION CMIFLAG in the netlist b Assign topoid of CMI MOSFET structure to 101 in the code Conventions This section describes model conventions in the
122. for a small channel length u0 Mobility at Temp Tnom cm V sec a NMOSFET 670 ua First order mobility degradation coefficient m V 2 256 9 ub Second order mobility degradation coefficient m V 5 9e 19 uc Body effect of the mobility degradation 1 V 0465 coefficient vsat Saturation velocity at Temp Tnom m sec 8e4 HSPICE MOSFET Models Manual X 2005 09 497 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model Table 141 MOSFET Level 60 DC Parameters Continued SPICE Description Unit Default See Symbol Table 144 a0 Bulk charge effect coefficient for the channel 1 0 length ags Gate bias coefficient of Ay ix 1 V 0 0 bO Bulk charge effect coefficient for the channel m 0 0 width b1 Bulk charge effect width offset m 0 0 keta Body bias coefficient of the bulk charge effect m 0 6 Abp Coefficient of the Abeff dependency on Vgst 1 0 mxc Fitting parameter for calculating Abeff 0 9 adice0 DICE bulk charge factor 1 A1 First non saturation effect parameter 1 V 0 0 A2 Second non saturation effect parameter 0 1 0 rdsw Parasitic resistance per unit width W mmWr 100 prwb Body effect coefficient of Rdsw 1 V 0 prwg Gate bias effect coefficient of Rdsw 11 2 0 wr Width offset from Weff for calculating Rds 1 wint Width offset fitting parameter of l V without bias m 0 0 lint Length offset fitting parameter of I V without m 0 0 bias dwg Coefficient of the gate de
123. for the DIBL effect of the first output resistance pdiblc2 0 0086 Correction parameter for the DIBL effect of the second output resistance prwb 1 N 0 Body effect coefficient of Rdsw prwg 1 2 0 Gate bias effect coefficient of Rdsw pvag 0 0 Gate dependence of the Early voltage Rbody ohm m2 0 0 Intrinsic body contact sheet resistance Rbsh ohm m 0 0 Extrinsic body contact sheet resistance rdsw Q um 100 Parasitic resistance per unit width rsh ohm square 0 0 Source drain sheet resistance in ohm per square siio 1 V 0 5 First Vg dependence parameter for the impact ionization current sii 1 V 0 1 Second Vgs dependence parameter for the impact ionization current sii2 1 V 0 Third Vg dependence parameter for the impact ionization current siid 1 V 0 Vas dependence parameter of the drain saturation HSPICE MOSFET Models Manual X 2005 09 voltage for the impact ionization current 471 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model Table 130 MOSFET Level 57 DC Parameters Continued Parameter Unit Default Description tii 0 Temperature dependence parameter for the impact ionization current u0 cm V sec NMOS 670 Mobility at Temp Tnom PMOS 250 ua m V 2 25e 9 First order mobility degradation coefficient ub m V 2 5 87e 19 Second order mobility degradation coefficient uc 1 V 0 0465 Body effect of the mobility degradation coefficient Vabjt V 10 Early voltage for the bipolar current
124. gate length LAP m 100 0e 9 25 0e 9 Lateral diffusion per side WVAR m 25 0e 9 130 0e 9 Variation in active width WOT m 0 0 0 0 Channel stop diffusion per side TR C 21 0 21 0 Reference temperature for model HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 50 Philips MOS9 Model Table 45 MOSFET Level 50 Model Parameters Continued Name Unit Default N Default P Description VTOR V 730 0e 3 1 1 Threshold voltage at zero bias STVTO V K 1 2e 3 1 7e 3 Temperature dependence of VTO SLVTO Vm 135 0e 9 35 0e 9 Length dependence of VTO SL2VTO Vm2 0 0 0 0 Second length dependence of VTO SWVTO Vm 130 0e 9 50 0e 9 Width dependence of VTO KOR yv 650 0e 3 470 0e 3 7 Low back bias body factor SLKO V 1 2m 130 0e 9 200 0e 9 Length dependence of KO SWKO vim 2 0e 9 115 0e 9 Width dependence of KO KR y 1 2 110 0e 3 470 0e 3 High back bias body factor SLK y 1 2m 280 0e 9 200 0e 9 Length dependence of K SWK y 1 2m 275 0e 9 115 0e 9 Width dependence of K PHIBR V 650 0e 3 650 0e 3 Strong inversion surface potential VSBXR V 660 0e 3 0 0 Transition voltage for dual k factor model SLVSBX Vm 0 0 0 0 Length dependence of VSBX SWVSBX Vm 675 0e 9 0 0 Width dependence of VSBX BETSQ AV 83 0e 6 26 1e 6 Gain factor of infinite square transistor ETABET 1 6 1 6 Exponent of temperature HSPICE MOSFET Models Manual X 2005 09 dependence of gain factor 215 5 Standard MOSFE
125. gy Q9 x y G D S B The preceding equation uses the positive sign if x y or the negative sign otherwise This equation produces simple continuous analytical expressions for all transcapacitances in terms of the xf pinch off voltage the xr slope factor and derivatives thereof from weak to strong inversion and from non saturation to saturation Normalized Intrinsic Capacitances Set XQC 1 to select a simplified capacitive dynamic model that uses the five intrinsic capacitances corresponding to the Level 55 Equivalent Circuit on page 226 This model ignores the slight bias dependence of the n slope factor The result is the following simple set of functions a m Xo eqs Wu Eso NR ge BT a UAE mc a T 2 3 xp x 3 xp x n 1 cy Jedes os q Cop Ng 1 ees Cap n 1 4 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Total Intrinsic Capacitances Cos gd gb sb db Cox C os gd gb sb db Intrinsic Noise Model Equations The INps current source models the noise between the intrinsic source and the drain This noise includes a thermal noise component and a flicker noise component and has the following Power Spectral Density PSD SINDS Sthermat Slicker Thermal Noise The following equation calculates the PSD thermal noise component u S 4xT 2 jo 4kT B q NS Lig The preceding thermal noise e
126. input formation t mode device mode t qflag flag for charge cap computing uble vds vds bias HSPICE MOSFET Models Manual X 2005 09 double vgs double vbs double gd double gs double cgso double cgdo double cgbo double von double vdsat double ids double gds double gm double gmbs dIds dVbs y 5 3 MOSFET capacitanc capop can have following val 0 or else vgs bias vbs bias device DC information SO ga ga ga Sai urce con 8 Customer Common Model Interface Model Interface Routines te sourc te drain te bulk turation Sk oul tput con model drain conductance ductance overlap capacitance charge model Meyer s model dIds dVds dIds dVgs selection ues Note HSPICE or HSPICE RF does not support Meyer s model int capop Meyer s capacitances int capacitor selector overlap capacitance overlap capacitance turn on voltage voltage drain de current ductance trans conductance substrate trans conductance i s trinsic capacitance overlap capacitance HSPICE or HSPICE RF ignores these 3 capacitances A charge based model formulation is required double capgs Meyer s gate capa
127. inversion for the reference transistor BINV Probability factor for the intrinsic gate V 48 48 tunneling current in inversion IGACCR Gain factor for the intrinsic gate tunneling AV 2 0 0 current in the accumulation for the reference transistor BACC Probability factor for the intrinsic gate V 48 48 tunneling current in the accumulation VFBOV Flat band voltage for the Source Drain V 0 0 overlap extension KOV Body effect factor for the Source Drain V1 2 2 5 2 5 overlap extensions IGOVR Gain factor for the Source Drain overlap AV 2 0 0 tunneling current for reference transistor TOX Thickness of the gate oxide layer m 3 2E 9 3 2E 9 COL Gate overlap capacitance per unitchannel Fm 1 3 2E 10 3 2E 10 length GATENOISE Flag forinclusion exclusion of the induced 0 0 gate thermal noise HSPICE MOSFET Models Manual X 2005 09 285 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 62 Level 68 MOS11 Parameters Level 1100 Continued Name Description Units NMOS PMOS NTR Coefficient of the thermal noise at the J 1 656E 20 1 656E 20 actual temperature NFAR First coefficient of the flicker noise forthe V 1m 4 1 573bE22 1 573E22 reference transistor NFBR Second coefficient of the flicker noise for V 1m 2 4 752E8 4 752E8 the reference transistor NFCR Third coefficient of the flicker noise forthe V 1 0 0 reference transistor Table 63 Level 68 MOS11 Parameters Level 11010 Physical Geometry
128. is located in the following directory Sinstalldir demo hspice mos gatcap sp 72 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Figure 18 Gate Capacitance 100 0F gt CJPLOT ASO 2 CO 90 0F L5 0T CDB 80 0F CDB abala 70 0F CDB E E nim CDB T 60 0F I ponon v 5 Boe CDB 50 0F SS SD tee i ELLO UE eS CDB Si at Tasa e km 40 0F E S NG dex ARARE ARNEE RIA SR KEI E Sa 30 0F 3 E B Ng 20 0F Dune 10 0F E 0 1 L ao Hor TAU al UP i the 3 0 1 0 2 0 3 0 4 0 5 0 PDN Lin Capacitance Control Options The OPTION SCALM CVTOL DCSTEP and DCCAP control options affect the CAPOP models SCALM scales the model parameters CVTOL controls the error tolerance for convergence for the CAPOP 3 model see CAPOP 3 Gate Capacitances Simpson Integration on page 86 DCSTEP models capacitances with a conductance during DC analysis DCCAP calculates capacitances in DC analysis Scaling OPTION SCALM scales the CGBO CGDO CGSO COX LD and WD parameters according to fixed rules which are a function of the parameter s units If the model parameter s units are in meters simulation multiplies the parameter by SCALM For example HSPICE MOSFET Models Manual X 2005 09 73 2 Technical Summary of MOSFET Models MOS Gate Ca
129. is the Grove Frohman model e LEVEL 3 is an empirical model LEVEL 4 is a modified version of Level 2 LEVEL 5 is the IDS model with enhancement and depletion modes LEVEL 6 is the Lattin Jenkins Grove model us using ASPEC style parasitics LEVEL 7 is the Lattin Jenkins Grove model us using SPICE style parasitics LEVEL 8 is an advanced model using finite differences LEVEL 13 is the University of California UC Berkeley BSIM1 model LEVEL 27 is the SOSFET model LEVEL 28 is a Synopsys proprietary model based on the UC Berkeley BSIM1 model Level 13 LEVEL 38 is the Cypress Depletion model LEVEL 39 is the UC Berkeley BSIM2 model LEVEL 40 is the Hewlett Packard amorphous silicon Then Film Transistor a Si TFT model 108 HSPICE MOSFET Models Manual X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 25 Basic MOSFET Model Parameters Continued Name Alias Units Default Description Level LEVEL continued ACM 0 ALPHA v 0 HSPICE MOSFET Models Manual X 2005 09 LEVEL 47 is the UC Berkeley BSIM3 All version 2 model Levels LEVEL 49 is a Synopsys proprietary model based on the UC Berkeley BSIM3 version 3 model Level 53 LEVEL 50 is the Philips MOS9 model LEVEL 53 is the original UC Berkeley BSIMS version 3 model not modified as Level 49 is LEVEL 54 is the UC Berkeley BSIM4 model LEVEL 55 is the EPFL EKV model LE
130. local impact ionization model is physical so do not arbitrarily vary its parameters The LDD option intensifies the model so set LLDD to 0 for large scale circuit simulation and add the unbiased LDD resistance to RD this simplification stops if you specify NLDD 1e19 Level 58 Template Output For a list of output template parameters in the MOSFET models and which parameters this model supports see Table 4 on page 14 HSPICE MOSFET Models Manual 259 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 61 RPI a Si TFT Model Level 61 RPI a Si TFT Model Level 61 in the Synopsys MOSFET models is an AIM SPICE MOS15 amorphous silicon a Si thin film transistor TFT model developed by Rensselaer Polytechnic Institute 260 Model Features Features of the AIM SPICE MOS15 a Si TFT model include Modified charge control model induced charge trapped in localized states Above threshold includes Field effect mobility becoming a function of gate bias e Band mobility dominated by lattice scattering Below threshold includes Fermi level located in deep localized states Relate position of Fermi level including the deep DOS back to the gate bias Empirical expression for current at large negative gate biases for hole induced leakage current Applies interpolation techniques to equations to unify the model Using Level 61 with Synopsys Simulators To simulate using the AIM SPICE MOS15 a Si TFT model 1
131. more details Impact lonization To select impact ionization modeling instead of BSIM2 keep the Al0 0 value and specify the ALPHA ALPHA Vgs Vasat replaces Al in equation for fin the BSIM2 equations section VCR replaces BI and IIRAT multiplies f model parameters Synopsys impact ionization modeling differs from BSIM2 modeling in two ways 1 Abias term Vas Vasat multiplies the exponential and ALPHA values 2 You can use the IIRAT model parameter to partition the impact ionization component of the drain current between the source and the bulk IIRAT multiplies f in the saturation ly equation Thus the IIRAT fraction of the impact ionization current goes to the source and the 1 IIRAT fraction goes to the bulk adding to DB IIRAT defaults to zero that is 100 of impact ionization current goes to the bulk BSIM2 s impact ionization assumes that all of the impact ionization current is part of lgs In other words it flows to the source This assumption can lead to inaccuracies for example in cascode circuits See Calculating the Impact lonization Equations on page 61 for more details Parasitic Diode for Proper LDD Modeling The Level 39 MOSFET model includes alternative MOS parasitic diodes to replace SPICE style MOS parasitic diodes You can use these alternatives to geometrically scale the parasitics with MOS device dimension properly modeling the LDD parasitic resistances shared sources and drains and select
132. negative U1S is physically meaningless and causes negative arguments in a square root function in one of the BSIM2 equations The U1D value should be less than unity between 0 and 1 For reasonable V behavior make sure that K1 2K2 PHI V 20 For the equations to make sense the following must hold N 0 VGLOW 0 and VGHIGH 0 The BSIM2 gate capacitance model in SPICE 3E tends to display negative Cys in the subthreshold This is due to Cgg O as Vgs Vin by construction of the gate charge equation so that Cg Co Cga Cgo Cga Cgo 7 Cab Therefore use CAPOP 13 default until UC Berkeley releases an improved BSIM2 gate capacitance model Modeling Example The following is the result of fitting data from a submicron channel length NMOS device to BSIM2 To fit this data this example uses the Synopsys ATEM characterization software and the Synopsys simulation optimizer HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model Figure 36 Ipg VS Vas for Vgs 1 2 3 4 5V BSIM2 Model vs Data LEVEL 39 BSIM2 MODEL FET N CH VGS 1 2 3 4 5v 14 MAY 2003 15 33 26 DR39 SWD MODEL E IPU P A R A M Ae ee ee r 3 0 VDS LIN HSPICE MOSFET Models Manual 375 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model Figure 37 gds vs Vds for Vgs 2 3 4 5V BSIM2 Model vs Data LOG scale MOS MODEL A
133. of device figures min Whether to minimize the number of drain or source diffusions for even number fingered device rbdb Resistance connected between the internal drain side body node and the external body node rbsb Resistance connected between the internal source side body node and the external body node HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 110 MOSFET Level 54 Parameters Continued Parameter Description rbpb Resistance connected between the internal reference body node and the external body node rbps Resistance connected between the internal reference body node and the internal drain side body node rbpd Resistance connected between the internal reference body node and the internal source side body node trngsmod Transient NQS model selector acngsmod AC small signal NQS model selector rbodymod Substrate resistance network model selector rgatemod Gate resistance model selector geomod Geometry dependent parasitics model selector rgeomod Source Drain diffusion resistance and contact model selector MOSFET Level 54 uses the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters It also uses the parameters described in this section which apply only to MOSFET Level 54 The simulation calculates Rd and Rs as follows Rd TEMP Rd TNOM 1 TRD TEMP TNOM Rs TEMP Rs TNOM 1 TRS TEMP TNOM
134. overlap region delta tox asymmetric angstrom gundtoxovl gtunwdep Accumulation 2 D fringing parameter micron 0 gtunecbm ECB effective mass 0 4 HSPICE MOSFET Models Manual X 2005 09 511 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Table 147 SSIMSOI Model Parasitic Parameters Continued Name Parameter Units Default gtunecba ECB fitting parameter 0 6 gtunecbb ECB barrier height V 3 1 gtunecbbo ECB barrier height V 3 1 gtunhvbm HWB effective mass 0 3 gtunhvba HVB fitting parameter 1 0 gtunhvbb HVB barrier height V 4 5 gtunhvbbo HVB barrier height V 4 5 gtunevbeg EVB energy bandgap V 1 12 gtunevbm EVB effective mass 0 32 gtunevba EVB fitting parameter 0 4 gtunevbb EVB barrier height V 4 2 gtunevbbo EVB barrier height V 3 1 nbit Effective doping parameter for I bit 1 0 ndif Effective doping parameter for Q diffusion 1 0 cjch S D zero bias junction channel side capacitance F m 0 0 mich S D junction channel side grading coefficient 0 5 pbch S D junction channel side built in pot V 0 8 tcppbch S D temperature coefficient for pbch V K calculated seff S D diode QNR recombination velocity cm s 1e5 seffl I dependence parameter for seff 0 seffle Exponent for I dependence for seff 0 512 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Table 147 8 SSIMSOI Model Parasitic Parameters Continued Name Parameter Units Default seffwl W dependence paramet
135. part of NFC POTVFB Coefficient for the geometry VK 1 5 0E 4 5 0E 4 independent part of ST VFB PLTVFB Coefficient for the length dependent VK 1 0 0 part of ST VFB PWTVFB Coefficient for the width dependent VK 1 0 0 part of ST VFB PLWTVFB Coefficient for the length times width VK 1 0 0 dependent part of ST VFB POTPHIB Coefficient for the geometry VK 1 8 5E 4 8 5E 4 independent part of ST 9B PLTPHIB Coefficient for the length dependent VK 1 0 0 part of ST oB PWTPHIB Coefficient for the width dependent VK 1 0 0 part of ST oB PLWTPHIB Coefficient for the length times width VK 1 0 0 dependent part of ST oB POTETABET Coefficient for the geometry 1 30 0 5 independent part of np PLTETABET Coefficient for the length dependent 0 0 part of np PWTETABET Coefficient for the width dependent 0 0 part of np PLWTETABET Coefficient for the length times width 0 0 nsr dependent part HSPICE MOSFET Models Manual X 2005 09 303 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS POTETASR Coefficient for the geometry 0 65 0 5 independent nsr part PLTETASR Coefficient for the length dependent 0 0 part of nsr PWTETASR Coefficient for the width dependent 0 0 part of nsr PLWTETASR Coefficient for the length times width 0 0 dependent part of nsr POTETAPH Coefficient
136. potential 2 KP AN 50 0E 6 Transconductance parameter EO EO V m 1 0E12 gt 1E5 Mobility reduction coefficient UCRIT V m 2 0E6 gt 1E5 Longitudinal critical field a This model can calculate the default values of VTO GAMMA PHI and KP as a function of TOX NSUB UO and VFB for statistical circuit simulation b As Vg VTO also references the bulk Name Unit Default Range Description 0 Oxide thickness IV TOXY m IV NSUBS cm3 0 Channel doping VFB V s Flat band voltage IV eo UO cm Low field mobility IV VMAX m s O Saturation velocity THETA 1V O 20 Mobility reduction coefficient In this example cm is the basic unit for NSUB and UO TOX and VMAX are in m Optional parameter for calculating COX Optional parameter for the dependence of GAMMA on COX and for calculating PHI Optional parameter for calculating VTO as a function of COX GAMMA or PHI Optional parameter for the dependence of KP on COX Optional parameter for calculating UCRIT Optional parameter for mobility reduction due to the vertical field o79270075992 The preceding parameters accommodate the scaling behavior of the process and basic intrinsic model parameters and statistical circuit simulation Simulation uses the TOX NSUB VFB UO and VMAX parameters only if you HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model did not specify
137. preceding equation vy VTO GAMMA P PHI n 1 PHI amp Vip To include the narrow width effect use vp and n To include the narrow width effect specify the DELTA model parameter The effective y specifies the short channel effect To include short channel effects the XJ model parameter must be greater than zero XJ 2 W nin 2 W N1 2 4 1 seated pl 14 s 1 ad 2 y GAMMA x a l scale scale The following equations determine the W and Wg depletion widths W i PHI v i lq NSUB a 2 E 1 2 W repu PH va ev HSPICE MOSFET Models Manual 131 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 2 IDS Grove Frohman Model 132 If you do not specify parameters such as VTO GAMMA and PHI simulation calculates them automatically The Level 2 model uses these parameters to calculate the threshold voltage See Common Threshold Voltage Equations on page 58 Saturation Voltage Vdsat If you do not specify the VMAX model parameter the program computes the saturation voltage due to channel pinch off at the drain side If you specify the corrections for small size effects then V Vp Es T 2 f E L 4 4 o C eire Vdsat T Vsat If you specify ECRIT the program modifies vs4 to include carrier velocity saturation effect 2 241 2 Vdsat Vsat Vc vi v The following equation calculates the v value used in the preceding equation v ERIT E S
138. prohibited It is the reader s responsibility to determine the applicable regulations and to comply with them Disclaimer SYNOPSYS INC AND ITS LICENSORS MAKE NO WARRANTY OF ANY KIND EXPRESS OR IMPLIED WITH REGARD TO THIS MATERIAL INCLUDING BUT NOT LIMITED TO THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE Registered Trademarks Synopsys AMPS Arcadia C Level Design C2HDL C2V C2VHDL Cadabra Calaveras Algorithm CATS CRITIC CSim Design Compiler DesignPower DesignWare EPIC Formality HSIM HSPICE Hypermodel iN Phase in Sync Leda MAST Meta Meta Software ModelTools NanoSim OpenVera PathMill Photolynx Physical Compiler PowerMill PrimeTime RailMill RapidScript Saber SiVL SNUG SolvNet Superlog System Compiler Testify TetraMAX TimeMill TMA VCS Vera and Virtual Stepper are registered trademarks of Synopsys Inc Trademarks Active Parasitics AFGen Apollo Apollo Il Apollo DPII Apollo GA ApolloGAIl Astro Astro Rail Astro Xtalk Aurora AvanTestchip AvanWaves BCView Behavioral Compiler BOA BRT Cedar ChipPlanner Circuit Analysis Columbia Columbia CE Comet 3D Cosmos CosmosEnterprise CosmosLE CosmosScope CosmosSE Cyclelink Davinci DC Expert DC Expert Plus DC Professional DC Ultra DC Ultra Plus Design Advisor Design Analyzer Design Vision DesignerHDL DesignTime DFM Workbench Direct RTL Direct Silicon Access Discovery DW8051 DWPCI
139. routine prints all model parameter names values and units to the standard output HSPICE or HSPICE RF calls this routine for each model after the CMI SetupModel where pmodel points to the model Syntax int CMI PrintModel char pmodel Example int ifdef STDC CMImos3PrintModel har pmodel o S CMImos3PrintModel pmodel char pmodel endif CMI_ENV penv Note The source for CMImos3printmodel is not shown void CMImos3printmodel MOS3model pmodel return 0 int CMImos3PrintModel 544 HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Interface Variables CMI FreeModel This routine frees memory that HSPICE or HSPICE RF previously allocated for model related data After simulation HSPICE or HSPICE RF calls this routine during a loop over all models Syntax int CMI FreeModel char pmodel In the preceding syntax pmode1 points to the model Example int ifdef STDC CMImos3FreeModel char pmodel else CMImos3FreeModel pmodel char pmodel endif free memory allocated for model data Note CMImos3freemodel source code not shown void CMImos3freemodel MOS3model pmodel return 0 int CMImos3FreeModel CMI Freelnstance This routine frees memory that HSPICE or HSPICE RF previously allocated to store instance related data After s
140. she m EPSI Nose DATE ee Nsac ae FOR N sac N pc q Nope sb Level 62 RPI Poli Si TFT Model Level 62 is an AIM SPICE MOS16 poly silicon Poli Si thin film transistor TFT model developed by Rensselaer Polytechnic Institute Model Features Features of the AIM SPICE MOS16 Poli Si TFT model include Adesign based on the crystalline MOSFET model Field effect mobility that becomes a function of the gate bias Effective mobility that accounts for trap states For low Vgs it is the power law For high Vgs it is the constant Reverse bias drain current function of the electric field near the drain and the temperature A design independent of the channel length HSPICE MOSFET Models Manual 265 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model A unified DC model that includes all four regimes for channel lengths down to 4 um e Leakage thermionic emission e Subthreshold diffusion like model e Above threshold c Si like with mFet e Kink impact ionization with feedback An AC model accurately reproduces the Cg frequency dispersion Automatic scaling of model parameters that accurately model a wide range of device geometries Using Level 62 with Synopsys Simulators To simulate using the AIM SPICE MOS16 Poli Si TFT model 1 Set Leve1 62 to identify the model as the AIM SPICE MOS16 Poli Si TFT model The default value for L is 100um and the default value for
141. specify HSPICE MOSFET Models Manual 113 X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 25 Basic MOSFET Model Parameters Continued Name Alias Units Default Description Level VBO VB V 0 0 Reference voltage for the GAMMA switch 6 7 e fvsb lt VBO the equation uses GAMMA f vsb gt VBO the equation uses LGAMMA VCR V 0 Impact ionization critical voltage This 39 parameter includes geometry sensitivity parameters VMAX VMX m s 0 0 Maximum drift velocity of the carriers Zero 2 3 8 VSAT indicates an infinite value 40 Default VMAX value for Level 40 is 166 VMAX cm s 0 0 Maximum drift velocity of the carriers 6 7 VMX Selects a calculation scheme to use for vdsat Zero indicates an infinite value Typical values electrons8 4e6 cm s holes4 3e6 cm s VTIME S 10m Voltage stress 40 ZENH 1 0 Mode flag enhancement Set ZENH 0 0 5 for the depletion mode Table 26 Effective Width and Length Parameters Name Alias Units Default Description Level DEL m 0 0 Channel length reduction on each side 1 2 3 DELscaleg DEL SCALM 24 8 MOSFET Level 13 does not support DEL DEL um 0 0 Channel length reduction on each side 5 WDEL 114 HSPICE MOSFET Models Manual X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 26 Effective Width and Length Parameters Continued Name Alias Units Default Description Level DELVTO V 0 Thres
142. the geometry V 1 8 12E 2 7 9E 2 independent part of 6R PLTHER Coefficient for the length dependent V 1 0 0 part of OR PWTHER Coefficient for the width dependent V 1 0 0 part of OR HSPICE MOSFET Models Manual 295 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS PLWTHER Coefficient length times width 6R V 1 0 0 dependent THER1 Numerator of the gate voltage V 0 0 dependent part of the series resistance for all transistors in bin THER2 Denominator of the gate voltage V 1 0 1 0 dependent part of the series resistance for all transistors in the bin POTHESAT Coefficient for the geometry V 1 2 513E 1 1 728E 1 independent part of Osat PLTHESAT Coefficient for length dependent part V 1 0 0 of Osat PWTHESAT Coefficient for the width dependent V 1 0 0 part of 0sat PLWTHESAT Coefficient length times width 0sat V 1 0 0 dependent POTHETH Coefficient for the geometry V 3 1 0E 5 0 independent part of 6TH PLTHETH Coefficient for the length dependent V 3 0 0 part of 6TH PWTHETH Coefficient for the width dependent V 3 0 0 part of 0TH PLWTHETH Coefficient for the length times width V 3 0 0 0TH dependent part POSDIBL Coefficient for the geometry V 1 2 8 53E 4 3 551E 5 independent part of odibl PLSDIBL Coefficient for the length dependent V 1 2 0 0 part of odibl 296 HSPICE MOSFET
143. this case the lgs Current is Ij lasg vde Vp isub NO 5 ND ofp Voss Vas The isub subthreshold current for LEVEL 3 is the same as for LEVEL 13 see ids Subthreshold Current on page 335 NO and ND are functions of the effective device width and length 140 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 3 IDS Empirical Model Compatibility Notes Synopsys Device Model versus SPICE3 Differences between the Synopsys Level 3 MOSFET device model and Berkeley SPICES can arise in the following situations Small XJ Level 3 and SPICE3 differ for small XJ values typically 50 05 microns Do not use such small values for XJ they are physically unreasonable XJ calculates the short channel reduction of the GAMMA effect GAMMA gt f GAMMA f is normally less than or equal to 1 For very small values of XJ f can be greater than one The Synopsys Level 3 model imposes the limit f lt 1 0 but SPICES allows f 51 0 ETA In this model 8 14 is the constant in the ETA equation which varies the Vas threshold Berkeley SPICES uses 8 15 Solution To convert a SPICE3 model to the Synopsys Level 3 MOSFET device model multiply ETA by 815 814 NSUB Missing If you do not specify NSUB in SPICE3 the KAPPA equation becomes inactive The Synopsys Level 3 MOSFET model generates a default NSUB from GAMMA and the KAPPA equation is active Solution If you do not specify NSUB in the SPICE3 mo
144. this model can still include diffusion conduction from the drain to bulk rather than from the drain to source von vth fast The following equation calculates the fast value used in the preceding equation fast vt 14 LAES 4 Y COX 2 vsb PHI Cutoff Region vgs lt vth PHI ids 0 Weak Inversion vth PHI lt vgs lt von WEX ids ids von vde vsb 1 MN dL deii d fast PHL HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model Strong Inversion vgs gt von ids ids vgs vde vsb Note Strong inversion conditions do not use the modified threshold voltage von If WIC 3 simulation calculates the subthreshold current differently In this case the ids current is ids ids vgs vde vsb isub NOeff NDeff vgs vds NOeff and NDeff are functions of the effective device width and length Effective Mobility ueff All mobility equations have the following general form ueff UO factor Parameter Description ueff Effective mobility at the specified a specified analysis temperature factor Mobility degradation factor Default 1 0 Use the MOB model parameter to select the mobility modulation equation used in the Level 6 MOSFET model Parameter Description MOB 0 No mobility reduction default MOB 1 Gm equation MOB2 Frohman Bentchkowski equation MOB 3 Normal field equation MOB 4 Universal fie
145. to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model Vos VON n e fett B1 Pea von Vey Full Enhancement Vgg Vjp vde gt 0 Ij B1 NI vde Pcav by vde v vy MAT 2 Vo von B von vg vde e fost Example The netlist for this example is located in the following directory Sinstalldir demo hspice mos ml5iv sp LEVEL 6 LEVEL 7 IDS MOSFET Model These models represent ASPEC MSINC and ISPICE MOSFET model equations The only difference between LEVEL 6 and LEVEL 7 equations is the handling of the parasitic elements and the method of temperature compensation See Table 28 on page 122 and Channel Length Modulation on page 133 for those model parameters LEVEL 6 and LEVEL 7 Model Parameters MOSFET Levels 6 and 7 use the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters These levels also use the parameters described in this section which apply only to MOSFET Levels 6 and 7 HSPICE MOSFET Models Manual 155 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model 156 Table 30 Alternate Saturation Model Parameters Name Units Default Description Alias KA 1 0 Alternate saturation model coefficient for the short channel vds scaling factor KU 0 0 Lateral field mobility parameter MAL 0 5 Alternate saturation model exponent of the short channel vds scaling factor MBL 1 0 Exponent for mobility reduct
146. to Vys MU40 cm2 V3 5 O Empirical parameter for the output resistance MU4B cm V s O Sensitivity of the empirical parameter to Vps MU4G cm2 V4 5 O Sensitivity of the empirical parameter to Vg UAO vi 0 First order vertical field mobility reduction factor UAB V2 0 Sensitivity of the first order factor to Vp UBO v 0 Second order vertical field mobility reduction factor UBB v3 0 Sensitivity of the second order factor to Vps U10 y1 0 High drain field velocity saturation mobility reduction factor HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model Table 89 BSIM2 Model Parameters Continued Name Alias Units Default Description U1B v 0 Sensitivity of the mobility reduction factor to Vps U1D V2 0 Sensitivity of the mobility reduction factor to Vgs NO 0 5 Subthreshold swing coefficient NB y1 2 0 Sensitivity of the subthreshold swing to Vps ND y1 0 Sensitivity of the subthreshold swing to Vas VOFO a 0 Threshold offset normalized to NKT q for the subthreshold VOFB v 0 Sensitivity of the offset to Vp VOFD y1 0 Sensitivity of the offset to Vg AIO 0 Impact ionization coefficient AIB Vi 0 Sensitivity of the impact ionization coefficient to Vbs BIO V 0 Impact ionization exponent BIB 0 Sensitivity of the impact ionization exponent to Vbs DELL m Length reduction of the source drain diffusion not used in the Level 39 MOSFET model WDF m Default width n
147. unity it does not scale properly with Weff because lds is approximately proportional to Weff Also without the HSPICE multiplicity factor M factor in equation 1 this model cannot simulate multiple transistors in parallel To solve these problems HSPICE 2002 2 added a WOFLK width normalizing parameter and corrects equation 1 as M KF Weff WOFLK 1 AF Ids AF Cox Leff 2 f EF 2 The default value of WOFLK is 1 0 to switch off the new width scaling model The unit is in meters The next equation handles the flicker noise model noimod 1 amp 4 depending on whether you specify WOFLK If WOFLK lt 0 0 the default case then the flicker noise model of noiMod 1 and 4 uses this equation for backward compatibility M KF Ids AF Cox LeffA2 f EF 3 ELSE M KF We f WOFLK 1 AF Ids AF Cox Leff 2 f EF HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIMS3 SOI Model Level 57 Template Output For a list of output template parameters in the MOSFET models and which parameters this model supports see Table 4 on page 14 Level 57 Updates to BSIM3 SOI PD versions 2 2 2 21 and 2 22 BSIM PD version 2 2 enhances the model flexibility and accuracy from PD version 2 0 and the following are its major features Gate body tunneling substrate current enhances the model accuracy Body contact res
148. us Channel Length Modulation For Vgs gt Vdsat this model computes the channel length modulation factor The VMAX model parameter value determines the amount of channel length reduction AL VMAX 0 AL X KAPPA vj Vasa 11 VMAX gt 0 41 2 2 E X E X Ak dg a i KAPPA X5 va Vasa In the preceding equation E is the lateral electric field at the pinch off point The following equation approximates its value Vet v Vasat E L gp Vdsat The following equation computes the ly saturation current lis Ta Loy HSPICE MOSFET Models Manual 139 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 3 IDS Empirical Model To prevent a zero denominator the AL value is limited L If AL then AL Leg Ar Subthreshold Current Ig This region of operation is characterized by the model parameter for the fast surface state NFS The following equation determines the modified threshold voltage von NFS50 von v fast The following equation calculates the fast value used in the preceding equation q NFS GAMMA f PHI v4 f PHI vy fast vq e COX 2 PHI v The following equations calculate the ly current VgssVON V von Ij Ig von vde v e fast vgs gt von Iis La on vde Vsp Note Strong inversion does not use the modified threshold voltage If WIC 3 the model calculates subthreshold current differently In
149. use all Synopsys device model capacitance options CAPOP CAPOP 2 is the default setting for LEVEL 38 If you set CAPOP 6 AMI capacitance model LEVEL 38 capacitance calculations become identical to those of LEVEL 5 The ACM default parameter ACM 0 in LEVEL 38 invokes SPICE style parasitics You can set ACM to 1 ASPEC or to 2 Synopsys device model All MOSFET models follow this convention You can use OPTION SCALE with the LEVEL 5 MOSFET device model However you cannot use the SCALM option due to the difference in units You also cannot use the DERIV option You must specify the following parameters for MOS LEVEL 38 VTO VT TOX UO UB FRC ECV and NSUB DNB As with LEVEL 5 this model calculates the Ids current according to three gate voltage regions Depletion Region vgs vfb 0 HSPICE MOSFET Models Manual 195 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 38 IDS Cypress Depletion Model 196 The low gate voltage region which the bulk channel dominates Enhancement Region vgs vib gt O vds lt vgs vfb The region defined by high gate voltage and low drain voltage In the enhancement region both channels are fully turned on Partial enhancement region vgs vfb gt O vds gt vgs vib This region has high gate and drain voltages so the surface region is partially turned on and the bulk region is fully turned on To better model depletion region operations empirical fittin
150. uses only the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters LEVEL 2 Model Equations The LEVEL 2 model equations follow IDS Equations This section describes how the LEVEL 2 MOSFET model calculates the drain current of n channel and p channel MOSFETs Cutoff Region V9Ssvth Ij 0 see subthreshold current On Region vos Vt N ge 2 3 2 3 2 lus B Yge Yo Ngo E PHI Vae Vp PHI Vg 21 The following equations calculate values used in the preceding equation Vde min Vas Viasat W TU SI eff n 1 DELTA B KP 130 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 2 IDS Grove Frohman Model Effective Channel Length and Width The Level 2 model calculates effective channel length and width from the drawn length and width L ff Lscaled LMLT XL 2 P LD scaled DEL ated scaled Wer M CW caida WMLT KW cad 2 PWD icd LREF ez LREF scaled LMLT XL scaled 2 LD scaled DEL caled WREF s7 M WREF scaled WMLT XW caled 7 2 PWD aie Threshold Voltage Vj The VTO model parameter is an extrapolated zero bias threshold voltage for a large device The following equation calculates the effective threshold voltage including the device size effects and the terminal voltages Vin Voi Y PHI v The following equation calculates the vp value used in the
151. vim 7 15e 22 1 53xe 22 1stflicker noise coefficient added in release 98 4 NFBR vim 2 16e 06 4 06e 06 2nd flicker noise coefficient HSPICE MOSFET Models Manual added in release 98 4 219 5 Standard MOSFET Models Levels 50 to 64 Level 50 Philips MOS9 Model Table 45 MOSFET Level 50 Model Parameters Continued Name Unit Default N Default P Description NFCR y1 0 0 2 92e 10 3rd flicker noise coefficient added in release 98 4 SL3VTO V 0 0 Third coefficient of the length dependence of Vio SL2KO V1 2m2 0 0 Second coefficient of the length dependence of Kg SL2K V1 2m2 0 0 Second coefficient of the length dependence of K LP1 M 16 6 16 6 Characteristic length of the first profile FBET1 0 0 Relative mobility decrease due to the first profile LP2 M 1E 8 1E 8 Characteristic length of the second profile FBET2 0 0 Relative mobility decrease due to the second profile GTHE1 0 0 Parameter that selects either the old 0 or the new 21 scaling rule of 0 SL2GAMOO 0 0 Second coefficient of the Yoo length dependence 220 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 50 Philips MOS9 Model JUNCAP Model Parameters The following are JUNCAP model parameters specifically for the Philips MOS 9 Level 50 model Table 46 JUNCAP Model Parameters MOSFET Level 50 Name Unit Default Description JUNCAP 0 JUNCAP flag 0 off 1 on DTA 0 0 Temperatu
152. vsb VBO vtb must be vtb VTO yi yb VBO PHI PHI yi is the effective value of GAMMA bis the effective value of LGAMMA The model computes them as y in single gamma models except the scf factor is 1 0 yi GAMMA gw gl yb LGAMMA gw gl HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model Effective Built in Voltage vbi for VBO gt 0 For vds lt VFDS if vsbsVBO vbi VTO yi PHI 1 2 7 1 PHI vsb Dscalod esi meu dee LLL EDU VSH COX Leff FDS vds if vso gt VBO vbi VTO yb PHI 4 yi yb VBO PHI PHI n 1 LDscaled esi PHI vsb Leff MO Oxon y For vds gt VFDS if vsbsVBO vbi VTO yi PHI n 1 PHI vsb LDscaled si Leff COX Leff FDS UFDS VFDS UFDS vds if vsb VBO vbi VTO yb PHI 4 i o VBO PHD 2 PHI 7 4 1 LDscaled esi Leff VES DX Leff P FDS UFDS VFDS UFDS vds PHI vsb Saturation Voltage vdsat UPDATE 0 2 The following formula determines the saturation voltage due to channel pinch off at the drain side TNT 2 EV hi 1 2 vsat vgs vbi 9 JI ugs vbi PHI vsb n 2m Y n HSPICE MOSFET Models Manual 163 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model 164 P A R A P A R A M z gt The following equati
153. w is 100pm Level 62 is a 3 terminal model This model does not include a bulk node therefore simulation does not append a parasitic drain bulk or source bulk diode to the model You can specify a fourth node but it does not affect the simulation results The default room temperature is 25 9C in Synopsys circuit simulators but is 27 9C in most other simulators When comparing to other simulators use TEMP 27 Or OPTION TNOM 27 to set the simulation temperature to 27 9C in the netlist The following is an example of how Level 62 modifies a MOSFET device model and element statement mckt drain gate source nch L 10e 6 W 10e 6 MODEL nch nmos Level 62 266 t vkink 9 1 von 0 0 vto 0 0 asat 1 at 3e 8 blk 0 001 bt 0 0 cgdo 0 0 cgso 0 0 dasat 0 0 dd 1 4e 7 delta 4 0 dg 2 0e 7 dmul 0 0 dvt 0 0 dvto 0 0 eb 0 68 eta 7 etac0 7 etac00 0 i0 6 0 i00 150 lasat Olkink 19e 6 mc 3 0 mk 1 3 mmu 3 0 t muQ 100 mul 0 0022 mus 1 0 rd 0 0 rdx 0 0 rs 0 0 rsx 0 0 tnom 27 tox 1 0e 7 vfb 0 1 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model Table 58 MOSFET Level 62 Model Parameters Name Unit Default Description ASAT 1 Proportionality constant of Vsat AT m V 3E 8 DIBL parameter 1 BLK 0 001 Leakage barrier lowering constant BT mV 1 9E 6 DIBL
154. waveforms generated during HSPICE circuit design simulation Provides key reference information for using HSPICE including syntax and descriptions for commands options parameters elements and more Provides key reference information for using HSPICE device models including passive devices diodes JFET and MESFET devices and BJT devices Searching Across the HSPICE Documentation Set Synopsys includes an index with your HSPICE documentation that lets you search the entire HSPICE documentation set for a particular topic or keyword In a single operation you can instantly generate a list of hits that are hyperlinked to the occurrences of your search term For information on how to perform searches across multiple PDF documents see the HSPICE release notes available on SolvNet at http solvnet synopsys com or the Adobe Reader online help HSPICE MOSFET Models Manual XV About This Manual Other Related Publications Note To use this feature the HSPICE documentation files the Index directory and the index pdx file must reside in the same directory This is the default installation for Synopsys documentation Also Adobe Acrobat must be invoked as a standalone application rather than as a plug in to your web browser Other Related Publications For additional information about HSPICE see The HSPICE release notes available on SolvNet see Accessing SolvNet on page xvii Documentation on the We
155. with Synopsys Simulators 00000 260 Equivalent Circuit leise 263 Model Equations 0 0 cece Ih 263 Drain Current orae sex etico e rette MEX ORE Fries a hs 263 Temperature Dependence 02000 cee eee ee eeaes 264 Capacitance eee teens 265 Level 62 RPI Poli Si TFT Model 0 00 cence ees 265 Model Features 2 f ahi soo i aea e a deeds pete eels ata be dele 265 Using Level 62 with Synopsys Simulators 0 00000 266 Equivalent Circuit llle 272 Model Equations e A EEOAE Ih 272 Drain Current aman efe dee PLANG e dod hae hoe haat 272 Threshold Voltage 00 0c cee ees 275 Temperature Dependence 2000 0c eee eee eaes 275 Capacitance a baga PUNAN bb Sones Mb Lara CR Ae 275 Geometry Effect llle eee 277 Selt FHealindgs uid xiu rene eR ERE alec ated a dna wae 277 Level 63 Philips MOS11 Model slslelsellslele selle 278 Using the Philips MOS11 Model else 278 Description of Parameters eee 279 Level 64 STARC HiSIM Model 0 000 cece ene eee 310 x HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 0 000 cece eee ee BSIM Model Features 00 0c eee eee LEVEL 13 Model Parameters Sensitivity Factors of Model Parameters MODEL VERSION Changes to BSIM Models
156. wla ite PA BA 0 500 0M 1 0 1 50 2 0 2 50 3 00 PGD Lin Figure 62 LEVEL 3 Ids versus Vgs Curves 1 00 EN C MODEL repe C DATA MEE I N 140 00 10 00 1 00 140 00 10 00 Param Lin 1 00 140 0F 10 0F HORS mec e er QN X iz Doe c SN UE ede d ae at in ld steal dn E dou LE 500 0M 1 0 1 50 2 0 2 50 3 00 100 0F 5 head PGD Lin HSPICE MOSFET Models Manual 587 X 2005 09 B Comparing MOS Models Examples of Data Fitting LEVEL 13 28 39 Ids Model versus Data Ids vs Vgs at Vds 0 1 Vbs 0 1 2 3 4 Figure 638 LEVEL 13 Ids versus Vgs Curves 1 00 ru Cc MODEL 140 00 10 00 1 00 Param Lin PGD Lin Figure 64 LEVEL 28 Ids versus Vgs Curves 1 00 P ass 6 MODEL 140 00 10 00 1 00 140 00 10 00 Param Lin 1 00 140 0F 10 0F naala aab L an a a UT iy ib eRe Rg CAL TOES 500 0M 1 1 50 2 0 2 50 3 00 PGD Lin 588 HSPICE MOSFET Models Manual X 2005 09 Figure 65 LEVEL 39 Ids versus Vgs Curves B Comparing MOS Models Examples of Data Fitting Param Lin PGD Lin i L 3 00 4C MODEL LEVEL 2 3 28 Gm lds Model versus Data m gm lds vs Vgs at Vds 0 1 Vos 0 2 The LEVEL 2 and 3 models have spikes at Vgs Vth The data a
157. 0 0 1x10710 Gate drain overlap capacitance CGFSO F m 0 0 1x107 9 Gate source overlap capacitance CGFBO F m 0 0 0 0 Gate body overlap capacitance RD ohm m 0 0 200 1000 Specific drain parasitic resistance RS ohm m 0 0 200 1000 Specific source parasitic resistance RHOB ohm sq 0 0 30x103 Body sheet resistance FNK F A 0 0 0 10 Flicker noise coefficient FNA 1 0 0 5 2 Flicker noise exponent 253 5 Standard MOSFET Models Levels 50 to 64 Level 58 University of Florida SOI Table 53 MOSFET Level 58 Optional Parameters Parameter Unit Default Typical Value Description VFBF V calc 1 nMOS Front gate flatband voltage 1 pMOS VFBB V calc Back gate flatband voltage WKF V calc VFBF Front gate work function difference WKB V calc Back gate work function difference TAUO S calc 107 105 Carrier lifetime in lightly doped regions BFACT 0 3 0 1 0 5 Vps averaging factor for mobility degradation FVBJT 0 0 0 1 BJT current directional partitioning factor 0 for lateral 1D flow RHOSD ohm sq 0 0 50 Source drain sheet resistance Level 58 NFD SOI MOSFET Model Parameters The following tables describe the Level 58 model parameters for non fully depleted NFD SOI including parameter names descriptions units defaults and typical notes Table 54 MOSFET Level 58 Flag Parameters Parameter Unit Default Typical Value Description Level Level 57 for UFSOI NFDMOD 0 Model selector 1 NFD BJT
158. 0 1 16e 06 ags 25 A1 0 A2 1 b0 01 b1 10 rdsw 0 prwg 0 prwb 2 wr 1 rbody 1E0 rbsh 0 0 a0 1 4 keta 0 1 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIMS3 SOI Model ketas 0 2 vsat 135000 dwg 0 dwb 0 alpha0 1e 8 beta0 0 betal 0 05 beta2 0 07 vdsatii0 8 esatii le7 voff 14 nfactor 7 cdsc 00002 cdscb 0 cdscd 0 cit 0 pclm 2 9 pvag 12 pdiblcl 18 pdiblc2 004 pdiblcb 234 drout 2 delta 01 eta0 05 etab 0 dsub 2 rth0 2 005 clc 0000001 cle 6 cf 1e 20 ckappa 6 cgdl 1e 20 t cgsl 1e 20 kt1 3 kt11 0 kt2 022 ute 1 5 ual 4 31e 09 ubl 7 61e 18 uc1 5 6e 11 prt 760 at 22400 cgso le 10 cgdo 1e 10 t cjswg 1e 12 tt 3e 10 asd 0 3 csdesw le 12 tcjswg le 4 mjswg 5 pbswg 1 UCB BSIMS013 1 In addition to BSIMSO13 0 the MOSFET Level 57 model also supports the UCB BSIMSO13 1 model version which includes the following new features that are not available in BSMISO13 0 Ideal Full Depletion FD Modeling BSIMSOI3 0 supports the modeling of these two families of SOI MOSFETs with a SOIMOD switching model flag SOIMOD 0 for partially depleted devices PD SOIMOD 1 for devices that tend to operate in a mixed mode of PD and FD V3 1 also provides an ideal full depletion FD module SOIMOD 2 not available in V3 0 to model FD SOI devices that literally exhibit no floating body behavior
159. 0 2 0 0 1 0 5 x108 0 5 5 x10 1x10 10 Mobility degradation coefficient Carrier saturated drift velocity Impact ionization coefficient 0 for no impact ionization Impact ionization exponential factor 0 for no impact ionization Exponential factor for gate induced drain leakage 0 for no GIDL Effective trap density for trap assisted junction tunneling 0 for no tunneling Body source drain junction recombination current coefficient Body source drain junction recombination ideality factor Effective diffusion length in source drain Effective recombination velocity in source drain Gate drain overlap capacitance 257 5 Standard MOSFET Models Levels 50 to 64 Level 58 University of Florida SOI Table 56 MOSFET Level 58 Electrical Parameters Continued Parameter Unit Default Typical Value Description CGFSO F m 0 0 1x10710 Gate source overlap capacitance CGFBO F m 0 0 0 0 Gate body overlap capacitance RD ohm m 0 0 200 1000 Specific drain parasitic resistance RS ohm m 0 0 200 1000 Specific source parasitic resistance RHOB ohm sq 0 0 30x108 Body sheet resistance FNK F A 0 0 0 10 Flicker noise coefficient FNA 1 0 0 5 2 Flicker noise exponent Table 57 Optional MOSFET Level 58 Parameters Parameter Unit Default Typical Value Description VFBF V calc 1 nMOS Front gate flatband voltage 1 pMOS VFBB V calc Back gate flatband voltage WKF V calc VFBF Fro
160. 00000E 00 ALPR 1 062000E 02 SLALP 9 957000E 01 ALPEXP 1 039000E 00 SWALP 0 000000E 00 VP 5 000000E 02 THETHR 2 413000E 03 THETHEXP 1 000000E 00 SWTHETH 0 000000E 00 SDIBLO 1 060000E 06 SDIBLEXP 6 756000E 00 MOR 1 050000E 03 MOEXP 3 146000E 00 LLMIN 2 000000E 07 AIR 9 938000E 04 STA1 9 300000E 02 SLA1 2 805000E 03 SWA1 0 000000E 00 A2R 4 047000E 01 SLA2 1 000000E 15 SWA2 0 000000E 00 A3R 7 540000E 01 SLA3 8 705000E 08 SWA3 0 000000E 00 TOX 3 200000E 09 COL 3 200000E 10 NT 1 623700E 20 NFAR 1 000000E 00 NFBR 0 000000E 00 NFCR 0 000000E 00 GATENOISE 0 00000E 00 DTA 0 000000E 00 Example 3 model nch nmos level 63 LEVEL 63 VERSION 11011 LVAR 0 LAP 4 0E 08 WVAR 0 0 WOT 0 0 TR 21 VFB 0 105E 01 POKO 0 5 PLKO 0 0 PWKO 0 0 PLWKO 0 0 KPINV 0 0 POPHIB 0 95 PLPHIB 0 0 PWPHIB 0 0 POBET 1 922E 03 POTHESR 3 562E 01 POTHEPH 1 29E 02 POETAMOB 1 4 POTHER 8 120E 02 THER1 0 0 THER2 0 1E 01 POTHESAT 0 2513 POT
161. 005 09 A Finding Device Libraries Example 2 TEMP 25 M1 d g s b N CHN W 10u L 5u Element statement MODEL N CHN LMIN 1u LMAX 4u WMIN 2u WMAX 100u MODEL statement Because Example 1 does not specify multisweep or temperature analysis simulation does not invoke the model selector feature so simulation uses the N CHN model with no problems In Example 2 however the TEMP statement invokes the model selector feature The model selector tries to find a model named N nnn that fits within the length and width ranges specified in the element statement Because the length in the element statement 5 um is not within the 1 to 4 um range specified in the MODEL statement the model selector cannot find a model that matches the element statement and simulation issues a device N not found error message HSPICE MOSFET Models Manual 569 X 2005 09 A Finding Device Libraries 570 HSPICE MOSFET Models Manual X 2005 09 B Comparing MOS Models Compares and reviews the most commonly used MOSFET models This appendix reviews the history strengths and weaknesses of the following commonly used MOSFET models MOSFET Level Description LEVEL 2 SPICE Level 2 LEVEL 3 SPICE Level 3 LEVEL 13 BSIM1 LEVEL 28 Synopsys proprietary model based on BSIM1 LEVEL 39 SIM2 History and Motivation This section describes the history and the motivation for using the Synopsys MOSFET models Syno
162. 00E 00 WMU30 0 000000E 00 MU3B 0 000000E 00 LMU3B 0 000000E 00 WMU3B 0 000000E 00 MU3G 2 970000E 00 LMU3G 0 000000E 00 WMU3G 0 000000E 00 MU40 0 000000E 00 LMU40 0 000000E 00 WMU40 0 000000E 00 MU4B 0 000000E 00 LMU4B 0 000000E 00 WMU4B 0 000000E 00 MU4G 0 000000E 00 LMU4G 0 000000E 00 WMU4G 0 000000E 00 UA0 0 000000E 00 LUAO 0 000000E 00 WUAO 0 000000E 00 UAB 0 000000E 00 LUAB 0 000000E 00 WUAB 0 000000E 00 UBO 7 450000E 03 LUBO 0 000000E 00 WUBO 0 000000E 00 UBB 0 000000E 00 LUBB 0 000000E 00 WUBB 0 000000E 00 U10 0 000000E 00 LU10 7 900000E 01 WU10 0 000000E 00 U1B 0 000000E 00 LUIB 0 000000E 00 WU1B 0 000000E 00 UID 0 000000E 00 LUID 0 000000E 00 WUID 0 000000E 00 NO 8 370000E 01 INO 0 000000E 00 WNO 0 000000E 00 NB 6 660000E 01 LNB 0 000000E 00 WNB 0 000000E 00 ND 0 000000E 00 LND 0 000000E 00 WND 0 000000E 00 VOFO 4 770000E 01 LVOFO 0 000000E 00 WVOFO 0 000000E 00 VOFB 3 400000E 02 LVOFB 0 000000E 00 WVOFB 0 000000E 00 VOFD 6 900000E 02 LVOFD 0 000000E 00 WVOFD 0 000000E 00 AIO 1 840000E 00 LAIO 0 000000E 00 WAIO 0 000000E 00 AIB 0 000000E 00 LAIB 0 000000E 00 WAIB 0 000000E 00 BIO 2 000000E 01 LBIO 0 000000E 00 WBIO 0 000000E 00 BIB 0 000000E 00 LBIB 0 000000E 00 WBIB 0 000000E 00 DELL 0 000000E 00 WDF 0 000000E
163. 05 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Figure 24 Charge Pump Circuit 5 MCAPS SVO 4 0F CD8 Lo A 7 CD8 3 0F 7 B LIE p _CD8 2 0F a cc CD8 Z pc CD8 1 0F VO QUE AES CD8 E T a 0 hh i Lo Lo Se Sse 4 pian mang Ay Ups na ey SSS al S l l 7 MCAPS SV1 amp 4 0F cos ee z PA _ CD8 3 0F f e Z CD8 B o 2 0F R ong ae y OPEM z 2 m 1 0F a Se x wo PER T um CD8 o I a loa Lo del oto he a a eo AP MP ie be hy aa 2 0 1 0 0 1 0 3 0 4 0 5 0 Volts Lin HSPICE MOSFET Models Manual 91 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models 92 The following example applies a pulse through a constant capacitance to the gate of a MOS transistor Ideally if the model conserves charge then the voltage at node 20 should become zero when the input pulse becomes zero Consequently the model that provides voltage closer to zero for node 20 conserves the charge better The results of the CAPOP 4 model are better than the CAPOP 2 model This example compares charge conservation models in SPICE2G 6 and Synopsys device models The results indicate that the Synopsys device models are more accurate Example The netlist for this example is located in the following directory
164. 1 1 0e7 Phonon scattering MUETMP 0 0 Temperature dependence of phonon scattering MUESRO 1 0 Surface roughness scattering MUESR1 7 0e8 Surface roughness scattering NDEP 1 0 Coefficient of the effective electric field NINV 0 5 Coefficient of the effective electric field NINVD 0 0v Modification of NINV BB 2 0 NMOS High field mobility degradation 1 0 PMOS VMAX 1 0e7cm s Maximum saturation velocity HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model Table 73 Level 64 Mobility Parameters Continued Parameter Default Description VOVER 0 0 Velocity overshoot effect VOVERP 0 0 Late dependence of the velocity overshoot RPOCK1 9 9V2 m1 2 a Resistance coefficient caused by the potential barrier RPOCK2 0 0V Resistance coefficient caused by the potential barrier RPOCP1 0 0 Resistance coefficient caused by the potential barrier if VERSION gt 110 RPOCP2 0 0 Resistance coefficient caused by the potential barrier if VERSION gt 110 Table 74 Level 64 Channel Length Modulation Parameters Parameter Default Description CLM1 0 3 Hardness coefficient of the channel contact junction CLM2 0 0 Coefficient for the Qg contribution CLM3 0 0 Coefficient for the Q contribution Table 75 Level 64 Substrate Current Parameters Parameter Default Description SUB1 0 0v Substrate current coefficient 1 SUB2 70 0 Substrate current coefficient 2 SUB3 1 0 Subst
165. 100 0 TA DI DATA 90 0 u 80 0 gt m B 70 0 eg 60 0 N 50 0 Param Lin 40 0 H xai NN 20 0 i 10 0 gt 0 E ia d aaa r roos o A 500 0M 1 0 1 50 2 0 2 50 3 00 PGD Lin Figure 70 LEVEL 28 gm lds versus Vgs Curves Hoo DI MODEL 1000 ZUM DI DATA 900 CX aa 80 0 70 0 60 0 2 a 50 0 a 40 0 Param Lin Or Eno eae ech d Tt E 500 0M 1 0 1 50 2 0 2 50 3 00 PGD Lin 592 HSPICE MOSFET Models Manual X 2005 09 Figure 71 LEVEL 39 gm lds versus Vgs Curves B Comparing MOS Models Examples of Data Fitting 110 0 100 0 N 90 0 800 70 0 60 0 Param Lin 50 0 40 0 a 30 0 20 0 10 0 EM T DI MODEL 0 55 leet akan 500 0M 1 0 1 50 2 0 PGD Lin Gds versus Vds at Vgs 4 Vbs 0 This plot shows the behavior of gds at the linear to saturation transition The LEVEL 3 model has a gds discontinuity HSPICE MOSFET Models Manual X 2005 09 593 B Comparing MOS Models Examples of Data Fitting Figure 72 LEVELs 2 3 28 gds versus Vds Curves dio n c UR p Nag 00S L3 N 00S AN Goes 1 00 N H DS 5 Na E 1 S 8 Ng i ie 140 00 s 1 0 a ied p gea ds 0 1 0 2 0 3 0 4 0 5 0 PDD Lin Figure 73 LEVELs 13 28
166. 12 200E6 0 1 0 4 TOX 0 2 THETA TOX UCRIT V m 2 0E6 2 3E6 1 0E6 25E6 VMAX UO DL m 0 0 5 Lmin 0 5 Lmin XESSSVBD 0 15 Lmin DW m 0 0 1 Wqi 0 5 W min 0 5Wmin XW 2 WD LAMBDA 0 5 0 8 0 3 HSPICE MOSFET Models Manual 245 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Table 48 Static Intrinsic Model Limits Continued Name Unit Default Example Lower Upper Estimation LETA 0 1 0 3 0 2 WETA 0 25 0 2 0 2 QO As 0 0 230E 6 0 LK m 0 29E 6 0 4E 6 0 05E 6 2E 6 IBA 1 m 0 0 2 0E8 0 0 5 0E8 ALPHA VCR Lo IBB V m 3 0E8 2 0E8 1 8E8 4 0E8 VCR Lc IBN 1 0 0 6 0 4 1 0 a a Also compare with optional process parameters b The minimum value of PHI also determines the minimum value of the pinch off voltage Due to the intrinsic temperature dependence of PHI higher temperatures use a lower value which limits the range of simulation for small currents Eox 0 0345E 9 F m q 1 609E 19C k 1 381E 23 J K L Je X COX g 0 104E 9 F m nj 1 45E16 m3 V kT q 0 0259 V at room temperature Note Parameters in this table use m meter as the length unit Lmin is the minimum drawn length of transistors Wmin is the minimum drawn width of transistors Example values are shown for enhancement N channel devices Model Updates Description Synopsys has made several improvements to the original EKV v2 6 MOSFET model Wherever possible these enhancements maintain backward
167. 129 LEVEL 13 332 LEVEL 2 131 LEVEL 28 354 LEVEL 49 428 431 LEVEL 5 146 LEVEL 6 159 parameters LEVEL 1 114 LEVEL 8 183 effective channel width MOSFETs equations LEVEL 2 131 LEVEL 3 137 LEVEL 39 365 effective mobility equations LEVEL 28 355 LEVEL 3 138 LEVEL 6 171 LEVEL 8 184 parameters 171 EFPS option 11 element statements 3 templates 14 69 element parameters MOSFETSs 53 range limits 428 scaling 11 Empirical model 136 equations 136 example 142 energy gap temperature 101 EPFL EKV MOSFETs model 224 equations BSIM LEVEL 13 332 capacitance MOS diode 102 MOSFETSs 55 overlap 77 diodes 55 599 Index Frohman Bentchkowski 173 HSPICE AL 179 impact ionization 61 MOSFETs channel length modulation 105 diode 55 106 impact ionization 61 LEVEL 61 263 LEVEL 62 272 mobility temperature 105 model parameters 59 models BSIM2 362 BSIM3 389 390 Cypress 197 Empirical 136 EPFL EKV 231 243 HP a Si TFT 206 IDS LEVEL 5 145 150 LEVEL 6 159 168 LEVEL 8 183 LEVEL 49 and 53 431 modified BSIM LEVEL 28 354 quasi static 241 Schichman Hodges 128 130 noise 96 surface potential temperature 104 temperature 101 threshold voltage 105 voltage 58 noise 96 243 Normal Field 173 temperature energy gap 101 MOS diode capacitance 102 MOSFETs 101 saturation current 102 voltage 58 Wang s 178 equivalent circuit 32 35 AC analysis 36 AC noise analysis 37 transient analysis 36 variables and constants 32 EXA model parameter 41
168. 19 CAPOP 0 Capacitances 50 0F CAPOPO SVO 3 C68 VOSPO05 P LUN T 1 C68 VOSPO5 d 2S E E 300F 1 C88 V0SP05 S E 2 oque s o Pe 20 0F Ue p cd l 10 0F th on m 0 ES 50 0F 7 c CAPOPO SVO C68 VOSPO05 ee d i QC68 VOSPO5 a Z moapmpm eee 30 0F 5 a C88 V0SP05 a E LIUIUS oO iac ce A 20 0F 5 paga s 10 0F DL EE does d j n 1 0 0 2 0 3 0 4 0 Volts Lin CAPOP 1 Modified Meyer Gate Capacitances Define cap COXscaled Weff Leff In the following equations G G D and D are smooth factors You cannot change the values of these parameters Gate Bulk Capacitance cgb Accumulation vgs lt vfb vsb cgb cap Depletion vgs lt vth cap L qa VES vsb oe GAMMA2 cgb 80 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Strong Inversion vgs gt vth ee L 44 GAMMA vsb PHI vsb PHIJCO EX GAMMA cgb These equations replace GAMMA with effective Y for model levels higher than 4 Gate Source Capacitance cgs Low vds vds lt 0 1 Accumulation vgs lt vth cgs CF5 cap G D Weak Inversion vgs lt vth 0 1 T 2 cgs CFS cap tart IB od en Strong Inversion vgs vth 0 1 vth vds 7 qos Es VOR VON ere an 2 vgs vth
169. 2 Esi xd q nsub If SNVB is zero then y GAMMA You can adjust the y value for the short channel effect the same way as in the LEVEL 2 model NSUB calculates the o value 2 vt 2829 ni HSPICE MOSFET Models Manual 183 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 8 IDS Model Threshold Voltage vth ETA specifies the threshold voltage reduction due to the potential barrier lowering effect vbi VTO g b fb 512622 ETA byds gp vsb 8 COX d Log vth vbi g 4vsb 4 P Modify y for the short channel effect the same as in the LEVEL 2 model to obtain the effective y Saturation Voltage vdsat Level 8 computes the vsat saturation voltage the same way as in the LEVEL 2 model This model includes the carrier velocity effect only if ECRIT is greater than zero ECRIT 0 vdsat vsat vc Jvsat vc The following equation calculates the vc value used in the preceding equation vc ECRIT Lg ECRIT x0 or MOB 7 vdsat vsat This model computes vsat as in the LEVEL 2 model see Saturation Voltage Vgsat ON page 132 Effective Mobility ueff The MOB mobility equation selector controls the mobility reduction equations In the LEVEL 8 model set MOB to 2 3 6 or 7 Default 6 MOB 2 Mobility Reduction e UCRIT UEXP up DU e BIE eff COX vgs vth UTRA Pvde 184 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 8 IDS Mode
170. 2 Use Microsoft Visual C to open the cmimodel cmimodel dsw file In the cmimodel project directory create a new subdirectory named mos222 4 To copy all files from the MOS LEVEL 3 model to the new mos222 subdirectory D project cmimodel gt xcopy mos3 mos222 e In the source files globally replace mos3 with mos222 In all file names replace mos3 with mos222 For example D project cmimodel rename CMImos3defs h CMImos222defs h After you rename all files the new model subdirectory contains the following files CMImos222defs h CMImos222 c CMImos222Getlpar c CMlImos222Setlpar c CMImos222GetMpar c CMImos222SetMpar c CMImos222eval c CMImos222set c CMImos222temp c 7 Load all of the above files from the mos222 subdirectory into the project directory HSPICE MOSFET Models Manual 527 X 2005 09 8 Customer Common Model Interface Testing Customer CMI Models 8 Add the following declaration to the cmimodel link CMImdIDec h file extern CMI MOSDEF pCMI mos222def 9 Add the following branch to the switch clause in the cmimodel link CMlImalLevel h file case222 pCMIDevice char pCMI mos222def break 10 Rebuild the libCMImodel dll file 11 Put the dynamic link library into the same directory as the hspice exe hspicext exe or hspice mt exe executable Testing Customer CMI Models 528 To test a new model in the shared library run a simulation on the mos3 sp input file in the test
171. 2 Vie Vege ae IDEA V or 1 2 sth 2 sth Subthreshold Leakage Current Subthreshold leakage current is the result of the thermionic field emission of carriers through the grain boundary trap states as described in the following equations aes Yos 7 EXD c a T oak IO Wu LTD f Xrre t Xre tl ete XmrE lo X TEE hi Xrpg X 4X TFE lo 4 TFE hi The following equations calculate values for the preceding equations c E _ 0 55eV HSPICE MOSFET Models Manual 273 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model XTFE o 34r of exp TE W for FS fio Karta exp z 5 Sio CA for fofi Ks 1 529 for f gt Py es jefe We C fni We PE a 1 E F F F fa See ee DELTA e 549 1 2 F us E nin F min le T 4 2 XTFE o f15 BIG ep Z zlo z v 72s Vgs VE posco JS ADD DG SiGe Ty 3 m 3 5 W 1 1 EB 4 Vps Igiode 7 100 Wer ex 25 E E exp k T m 027 Mo HSPICE MOSFET Models Manual X 2005 09 274 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model Impact lonization Effect Vet gt Oery large drain biases include the kink effect Level 62 models this effect as impact ionization in a narrow region near the drain and adds the ly impact ionization current to the drain current The expression is en I A V V ex kink kinkt
172. 200 from Canada e Find other local support center telephone numbers at http www synopsys com support support ctr xviii HSPICE MOSFET Models Manual X 2005 09 Overview of MOSFET Models Provides an overview of MOSFET model types and general information on using and selecting MOSFET models A circuit netlist describes the basic functionality of an electronic circuit that you are designing In HSPICE format a netlist consists of a series of elements that define the individual components of the overall circuit You can use your HSPICE format netlist to help you verify analyze and debug your circuit design before you turn that design into actual electronic circuitry Synopsys provides a series of standard models Each model is like a template that defines various versions of each supported element type used in an HSPICE format netlist Individual elements in your netlist can refer to these standard models for their basic definitions When you use these models you can quickly and efficiently create a netlist and simulate your circuit design Referring to standard models in this way reduces the amount of time required to Create the netlist Simulate and debug your circuit design Turn your circuit design into actual circuit hardware Within your netlist each element that refers to a model is known as an instance of that model When your netlist refers to predefined device models you reduce HSPICE MOSFET Models Manua
173. 2005 09 323 324 324 324 330 331 332 332 333 334 335 335 335 336 336 337 339 339 341 341 341 344 344 346 347 347 347 348 352 353 354 354 355 355 356 356 357 357 358 358 362 xi Contents LEVEL 39 Model Equations a a o naa aa anaana 362 Effective Length and Width 0 0000 e ee eee eee 365 Geometry and Bias of Model Parameters 365 Compatibility Notes III 366 SPICES Flag inn eigo t erepta S t aem PE 366 Temperature 2 necis mwaa haba NN LG AH hang hed wes 366 Parasitl6S tma Magna Gn Pee KGAD DUNG ahay ee ES Rte 367 Selecting Gate Capacitance cee eee 367 Unused Parameters 0 000 cece tae 368 MODEL VERSION Changes to BSIM2 Models 368 Preventing Negative Output Conductance 000000 369 Charge based Gate Capacitance Model CAPOP 39 369 Synopsys Device Model Enhancements 0 0000 ee ae 371 Temperature Effects liiis 371 Alternate Gate Capacitance Model 0 00000 eae 372 Impact lonization 0 0 ee 372 Parasitic Diode for Proper LDD Modeling 372 Skewing of Model Parameters 0 0000 e eee eens 373 HSPICE Optimizer 373 Modeling Guidelines Removing Mathematical Anomalies 373 Modeling Example 00 000 eee eee 374 Typical BSIM2 Model Listing 0 0000 cee eee eee 377 Common SPICE Parameters
174. 3 pp 313 315 1985 Ballay N et al Analytic Modeling of Depletion Mode MOSFET with Short and Narrow Channel Effects IEEE PROC Vol 128 Pt l No 6 1981 Tsividis Y Operations and Modeling of the MOS Transistor McGraw Hill New York 1987 p 145 p 241f BFRC s counterpart in BSIM is x2u0 Jeng M C Design and Modeling of Deep Submicrometer MOSFETs Ph D Dissertation University of California Berkeley 1989 Duster J S Jeng M C Ko P K and Hu C User s Guide for the BSIM2 Parameter Extraction Program and the SPICES with BSIM Implementation Industrial Liaison Program Software Distribution Office University of California Berkeley May 1990 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Lists and describes standard MOSFET models Levels 50 to 64 The MOSFET models described in this chapter are the most currently developed and widely used of the standard MOSFET models Synopsys MOSFET device models have introduced Levels that are compatible with models developed by the University of Florida Rensselaer Polytechnic Institute and others This chapter describes the following standard MOSFET models Levels 50 to 64 Level 50 Philips MOS9 Model Level 55 EPFL EKV MOSFET Model Level 58 University of Florida SOI Level 61 RPI a Si TFT Model Level 62 RPI Poli Si TFT Model Level 63 Philips MOS11 Model Level 64 STARC HiSIM Model For information about stand
175. 3 60e 9 5 0e 9 RS 0 0 80 0e 6 RD 0 0 80 0e 6 VFBC 0 722729 1 0 NSUBC 5 946417 1 0e 17 PARL2 2 20e 8 1 0e 17 LP 0 0 15 0e 9 320 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model Table 82 Model Parameter Version Defaults Continued Parameter Version 100 110 Others NSUBP 5 94e 17 1 0e 17 SC1 13 5 0 0 SC2 1 8 0 0 PGD1 0 0 0 01 PGD2 0 0 1 0 PGD3 0 0 0 8 NINVD 0 0 1 0e 9 MUEPH1 1 00e 7 25 0e43 MUEPHO 0 295 0 300 MUESR1 7 00e 8 2 0e415 MUESRO 1 0 2 0 MUETMP 0 0 1 5 SUB1 0 0 10 0 SUB2 70 0 20 0 SUB3 1 0 0 8 CJ 8 397247e 04 5 0e 04 CLM1 0 3 0 7 CLM2 0 0 2 0 CLM3 0 0 1 0 RPOCK1 0 0 0 01 RPOCK2 0 0 0 1 HSPICE MOSFET Models Manual X 2005 09 321 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model Table 82 Model Parameter Version Defaults Continued Parameter Version 100 110 Others RPOCP1 0 0 1 0 VOVER 0 0 0 01 VOVERP 0 0 0 1 QME1 0 0 40 0e 12 QME2 0 0 300 0e 12 GIDL1 0 0 5 0e 3 for HiSIM101 5 0e 6 for others GIDL2 0 0 1 0e 6 GIDL3 0 0 0 3 GLEAK1 0 0 0 01e 6 for HiSIM101 10 0e 3 for others GLEAK2 0 0 20 0e 6 GLEAK3 0 0 0 3 PZADDO 1 0e 3 5 0e 3 NFTRP 100 0e 9 10 0e 9 NFALP 2 00e 15 1 0e 16 To turn off the model effects use the following settings Short Channel Effect SC1 SC2 SC3 0 Reverse Short Channel Effect LP 0 Quantum Mechanical Effect QME1 QME2 QME3 0 Poly D
176. 39 gds versus Vds Curves NS NS 00S Ix 00S p mr 0S 100 _ 3 S N e E B X ae ON psa LA Ea LH POS Ny A 140 00 7 c 7778 Ni no ss r 1 0 2 0 3 0 4 0 5 0 PDD Lin 594 HSPICE MOSFET Models Manual X 2005 09 Gm lds vs Vgs at Vds 0 1 Vbs 0 2 B Comparing MOS Models Examples of Data Fitting This plot shows a gm discontinuity in the LEVEL 2 model related to the UCRIT and UEXP parameters Figure 74 LEVEL 2 gm lds versus Vgs Curves 110 0 100 0 in 900 E 800 Taa S Z i Ge n A has Sil 60 0 2e yr 500 2 V l Param Lin 400 gt WU z M 300 20 0 10 0 i N W DI MODEL zu DI DATA 2H x 500 0M 1 0 1 50 2 0 2 50 PGD Lin 3 00 Figure 75 LEVEL 28 gm lds versus Vgs Curves 110 0 100 0 90 0 800 N 70 0 M 60 0 Param Lin 500 5 400 30 0 20 0 10 0 0 Na boa Tou DI MODEL E DI DATA 500 0M 1 0 1 50 2 0 2 50 PGD Lin 3 00 HSPICE MOSFET Models Manual X 2005 09 595 B Comparing MOS Models Examples of Data Fitting Gm lds versus Vgs at Vds 0 1 Vbs 0 This plot shows the gm lds ratio in the weak inversion transition region The Level 2 3 and 13 models have kinks near the threshold while Level 28 and Level 39 are monotonic Figure 76 LEVELS 2 3 28 gm lds
177. 3temp MOS3model pmodel MOS3instance ptran return 0 int CMImos3SetupInstance CMI Evaluate Based on the bias conditions and the model instance parameter values this routine evaluates the model equations It then passes all transistor characteristics via the CMI VAR variable Syntax int CMI Evaluate CMI VAR pvar char pmodel char pinst Parameter Description pvar Pointer to the CMI VAR variable pmodel Pointer to the model pinst Pointer to the instance Example int ifdef _ STDC CMImos3Evaluate CMI VAR pslot char pmodel char ptr else CMImos3Evaluate pslot pmodel ptr CMI VAR pslot char pmodel char DUE endif CMI_ENV penv MOS3instance ptran penv pCMIenv pCMIenv is global ptran MOS3instance ptr call model evaluation 540 HSPICE MOSFET Models Manual X 2005 09 void CMI pslot psl psl psl psl psl psl psl psl psl psl psl psl psl psl pslot o o o o o o o pslot o o o o o o o TA AA 8 Customer Common Model Interface Interface Variables mos3evaluate penv MOS3model pmodel ptran vgs pslot vds pslot vbs gd gs von ids gds gm gmbs gbd gbs cgs cgd cgb capdb capsb cbso cbdo ptran MOS3drainConductance ptran MOS3sourceConductance ptran gt
178. 4 If MOB 0 X pa Vdsat amp 1 4 s 9 um 2 Leff Vas 4 L 4 J This equation does not include the effect of the field between the gate and the drain It also tends to overestimate the output conductance in the saturation region If MOB 7 Ec a Vdsat 4 1 8 in ue rs ad L 4 4 eff Vds 1 2 This equation does not include the effect of the field between the gate and the pinch off point It also tends to overestimate the output conductance in the saturation region The following equation calculates the X value used in the two preceding equations z E d Ng NSUB Modifying lgs by a factor of 1 Vgs is equivalent to replacing Leff with Le L 4 A Pv L eff eff To prevent the channel length Le from becoming negative the value of Le is limited If Le xwb then simulation replaces Le with xwb 14 xwb Le xwb The following equation calculates the xwb value used in the preceding equation xwb X PBI HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 2 IDS Grove Frohman Model Subthreshold Current Ig The fast surface states model parameter NFS characterizes this region of operation For NFS gt 0 the model determines the modified threshold voltage von von vy fast The following equation calculates the fast value used in the preceding equation Jub cs ndo he cc cu NES Ov 2 PHI v 2 COX
179. 5 charge conservation model parameters 74 storage modeling 64 conductance 35 current convention 34 diodes DC current equations 55 DC model parameters 40 effective areas 45 drain and source resistance 45 48 52 54 saturation current 45 54 equation 55 equations 55 GEO element parameter 53 geometry model parameters 40 model parameters 40 select 39 model parameters 40 resistance model parameters 40 temperature equations 106 suppressing 49 temperature equations 102 effective length and width 95 output conductance 62 energy gap temperature equations 101 equation variables and constants 32 equivalent circuits 35 AC analysis 36 AC noise analysis 37 transient analysis 36 examples NMOS model 432 HSPICE MOSFET Models Manual X 2005 09 Index PMOS model 433 gate capacitance example 71 capacitance model parameters 64 74 overlap capacitance model parameters 74 impact ionization equations 35 61 isoplanar construction 30 31 silicon gate 30 width cut 32 level parameter 13 LEVELs 6 7 UPDATE selector 156 Meyer capacitance model parameters 74 mobility temperature equations 105 model parameters AO 485 A1 485 A2 485 AGIDL 485 AGS 486 ALPHAO 486 ASD 490 AT 491 BO 486 B1 486 BGIDL 486 BSIM3 SOI FD 484 BSIM4 442 CAPMOD 484 CDSC 486 CDSCB 486 CDSCD 486 CF 490 CGDL 490 CGDO 490 CGSL 490 CGSO 490 change conservation 76 CIT 486 CJSWG 490 CKAPPA 490 CLC 490 CLE 490 CSDESW
180. 5 cap G CF120 Weak Inversion vgs vth CF2 CF1 CF1 0 vgs vth CF1 p cgs CF3 cap max CF2 Saturation Region vgs lt vth CF3 vds cgs CF5 cap HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Linear Region vgs gt vth CF3 vds B vgs vth vds Y cgs CF5 cap f TUA om cereal UPDATE 0 CF1 0 CGS i vgs vth CF3 P vds KEAP 2 vgs vth CF3 P vds Cgs Th UPDATE 1 Gate Drain Capacitance cgd Low vas vds 0 1 Accumulation vgs lt vth CF1 cgd CF5 cap G D Weak Inversion vgs lt vth CF2 CF1 VES vth e CF1 d 5 cgd CFS cap D CF2 max 0 1 2 CF2 vds a Strong Inversion vgs 2 vth CF2 CF1 p 1 vgsavih CFl Y cgd CF5 cap max 2 vgs vth CF1 vds High vds vds 0 1 Accumulation vgs lt vth CF1 cgd CF5 cap G DD Saturation Region vgs lt vth CF3 vds cgd CF5 cap DD DD is a function of CF3 if UPDATE 1 Linear Region vgs gt vth CF3 vds vgs vth 7 ka nM Died due IE Le 2 GE o ea sax A i vgs vth CF3 b ore HSPICE MOSFET Models Manual 85 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models 86 Example The netlist for this example is located in the following directory Sinstalldir demo hspice mos capop2 sp Figure 21 CAPOP 2 Capacitances Param Li
181. 507 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Parameter Description LPE Flag to turn on off Ipe related parasitics DTEMP Increases the temperature Table 145 SSIMSO Model intrinsic Parameters Geometry Modifiers and Threshold Voltage Name Parameter Units Default tox Gate oxide thickness Angstrom 250 tbox Back oxide thickness Angstrom 250 tsi Silicon film thickness cm 0 2e 4 vthO Linear region vth reference large MOSFET Vbs 0 V 0 8 nmos 0 8 pmos tcv Temperature coefficient of threshold voltage 1 K 0 vfb Reference large MOSFET flatband voltage V calc tcvfb Temperature coefficient of flatband voltage 1 K ng Poly gate doping density 1 cm 3 ngf Gate oxide fixed charge density 1 cm 2 0 pbias Length modifier use with odif micron 0 dlivcv Length modified for capacitance model micron 0 odif Outdiffusion of s d under gate micron 0 odifs Outdiffusion of s under gate asymmetric micron oidf abias Width modifier micron 0 lidd Ldd spacer width micron 0 508 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Table 145 SSIMSO Model intrinsic Parameters Geometry Modifiers and Threshold Voltage Continued Name Parameter Units Default Ig2ct Distance from contact to poly edge micron 1 0 dbias Diffusion resistor processing bias micron 0 nfs Fast surface state density 1 V cm 2 0 n1 Surface region doping densi
182. 59 DC Parameters Continued Parameter Unit Default Description RBODY ohm m 0 0 Intrinsic body contact sheet resistance RBSH ohm m 0 0 Extrinsic body contact sheet resistance RDSW Qum 100 Parasitic resistance per unit width RSH ohm square 0 0 Source drain sheet resistance in ohm per square UO cm2 V sec NMOS Mobility at Temp Tnom 670 PMOS 250 UA m V 2 25e 9 First order coefficient for mobility degradation UB m V 2 or Second order coefficient for mobility degradation UC 1 V 0 0465 Body effect coefficient for mobility degradation VBSA V 0 Transition body voltage offset VOFF V 0 08 Offset voltage in the subthreshold region for large W and L values VSAT m sec 8e4 Saturation velocity at Temp Tnom VTHO V NMOS Threshold voltage Vbs 0 for a long wide device PMOS 0 7 WO m 0 Narrow width parameter WINT m 0 0 Width offset fitting parameter from l V without bias WR 1 Width offset from Weff for calculating Rgs HSPICE MOSFET Models Manual X 2005 09 489 7 BSIM MOSFET Models Levels 47 to 65 Level 59 UC Berkeley BSIM3 SOI FD Model Table 137 MOSFET Level 59 AC Capacitance Parameters Parameter Unit Default Description ASD V 0 3 Smoothing parameter for the source drain bottom diffusion CF F m cal Gate to source drain fringing field capacitance CGDL F m 0 0 Lightly doped drain gate region overlap capacitance CGDO F m calculated Non LDD region drain gate overlap capacitance per channel length CGEO F m 0 0 Gate
183. 6 LEVEL 3 138 LEVEL 38 201 LEVEL 47 391 LEVEL 5 153 LEVEL 6 163 167 LEVEL 8 184 SCALE option 11 scaling 12 global SCALM override 12 global vs model 12 MOSFETS capacitance parameters 73 SCALM 11 parameter HSPICE MOSFET Models Manual X 2005 09 global scaling 12 overriding in a model 12 scaling by model 12 Schichman Hodges model 4 128 130 sensitivity factors 353 SGS Thomson MOS model 6 Sharp model 6 short channel effect 131 Siemens model 5 6 Sierra 1 model 5 Sierra 2 model 5 silicon gate transistor 30 Siliconix model 5 silicon on sapphire devices 3 SIM2 358 Simpson Integration 86 simulation 128 SOI model 192 SOSFET model 3 5 188 SPICE compatibility BSIM model 341 diodes 49 models 571 MOSFETs LEVEL 13 341 LEVEL 3 141 LEVEL 39 366 models 4 UTRA model parameter 122 Meyer gate capacitances 78 stacked devices 52 STC ITT model 5 subthreshold current equations LEVEL 13 335 LEVEL 2 135 LEVEL 3 140 LEVEL 38 202 LEVEL 5 154 LEVEL 6 169 LEVEL 8 187 surface potential equations 104 T Taylor Huang model 4 temperature HSPICE MOSFET Models Manual X 2005 09 Index compensation BSIM LEVEL 13 344 effect BSIM LEVEL 13 336 LEVEL 39 371 equations 101 MOSFETs channel length modulation 105 diode 102 106 equations 101 mobility 105 parameters 98 surface potential 104 threshold voltage 105 parameters 98 LEVEL 13 330 LEVEL 28 353 LEVEL 49 428 LEVEL 49 and 53 420 LEVEL 57 474 MOSFETs LEVEL 13 330 LE
184. All RSS LV14 Source resistance squares RSS All XQC LV15 Charge sharing coefficient XQC All GDEFF LV16 Effective drain conductance 1 RDeff All rgeoMod is not 0 GSEFF LV17 Effective source conductance 1 RSeff All rgeoMod is not 0 CDSAT LV18 Drain bulk saturation current at 1 volt All bias CSSAT LV19 Source bulk saturation current at 1 volt All bias VDBEFF LV20 X Effective drain bulk voltage All BETAEFF LV21 BETA effective All GAMMAEFF LV22 GAMMA effective All DELTAL LV23 AL MOS6 amount of channel length 1 2 3 6 modulation UBEFF LV24 UB effective 1 2 3 6 HSPICE MOSFET Models Manual 15 X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level VG LV25 VG drive 1 2 3 6 VFBEFF LV26 VFB effective All LV31 Drain current tolerance not used in All HSPICE releases after 95 3 IDSTOL LV32 Source diode current tolerance All IDDTOL LV33 Drain diode current tolerance All COVLGS LV36 Gate source overlap and fringing All capacitances COVLGD LV37 Gate drain overlap and fringing All capacitances COVLGB LV38 Gate bulk overlap capacitances All except 57 and 59 COVLGE LV38 Gate substrate overlap capacitances 57 59 VBS LX1 Bulk source voltage VBS All except 57 and 59 VES LX1 Substrate source voltage VES 57 59 vas LX2 Gate source voltage VGS All VDS LX3 Drain source voltage VDS All
185. As in BSIMSO13 0 the default SOIMOD value is O for BSIMSOI3 1 The following physical modeling components related to the internal SOI body node are critical for accurately modeling PD SOI devices but are not needed for the ideal FD module Thus for the ideal FD module HSPICE ignores these components which makes SOI MOSFET modeling much easier than for PD devices or non ideal FD devices Source Drain to body diode currents Source Bogy Drain parasitic BUT currents mpact ionization currents Gate body direct currents Body related capacitances HSPICE MOSFET Models Manual 479 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model Gate Resistance Modeling BSIMSOI3 1 uses the same gate resistance models as in the BSIM4 model with four options for various gate resistance modeling topologies Table 133 BSIMSOI3 1 Gate Resistance Modeling Topologies Name Unit Default Bin Description NGCON N Number of gate contacts RGATEMOD 0 N Gate resistance model selector e RGATEMOD 0 No gate resistance RGATEMOD 1 Constant gate resistance RGATEMOD 2 Rii model with variable resistance RGATEMOD 3 Rii model with two nodes RSHG Ohm Sq 0 1 N Gate electrode sheet resistance XGL m 0 N Offset of the gate length due to variations in patterning XGW m 0 N Distance from the gate contact to the channel edge in the W direction XRCRG1 12 Y Parameter for distributed c
186. CMI_AssignInstanceParm 538 CMI_AssignModelParm 537 CMI Conclude 548 CMI DiodeEval 541 CMI Evaluate 540 CMI Freelnstance 545 COMI FreeModel 545 CMI Noise 542 CMI PrintModel 544 CMI Resetlnstance 536 CMI ResetModel 535 CMI_SetupInstance 539 CMI_SetupModel 539 CMI Start 548 CMI WriteError 546 common model interface CMI 517 conductance MOSFETs 35 62 preventing negative output 339 369 control options ASPEC 10 BYPASS 10 capacitance 73 conventions HSPICE MOSFET Models Manual X 2005 09 bias polarity 563 source drain reversal conventions 565 convergence MOSFET diodes 39 CSB model parameter 41 current convention 34 Cypress model 5 195 D Dallas Semiconductor model 5 DC current 55 parameters MOSFETs 40 467 DEFAD option 10 DEFAS option 10 DEFL option 10 DEFNRD option 11 DEFNRS option 11 DEFPD option 11 DEFW option 11 depletion MOS devices 3 diffusion 325 diodes capacitance equations 55 102 CMI DiodeEval 541 MOSFETs 9 capacitance equations 55 equations 55 model selector 9 models 9 39 resistance temperature 106 DLAT model parameter 43 DNB model parameter 41 drain current equation 394 DW model parameter 43 E Early voltage 128 effective channel length equations LEVEL 1 129 LEVEL 13 332 LEVEL 28 354 LEVEL 39 365 LEVEL 49 428 431 LEVEL 5 146 HSPICE MOSFET Models Manual X 2005 09 Index LEVEL 6 159 LEVEL 8 183 parameters LEVEL 1 114 LEVEL 3 137 effective channel width equations LEVEL 1
187. Common MOSFET Model Parameters It also uses the parameters described in this section which apply only to MOSFET Level 5 Table 29 Capacitance Parameters for MOSFET Level 5 Name Alias Units Default Description AFC 1 0 Area factor for MOSFET capacitance CAPOP 6 Gate capacitance selector METO um 0 0 Metal overlap on gate Use the ZENH flag mode parameter to select one of two modes enhancement or depletion Parameter Description ZENH 1 This enhancement model default mode is a portion of the Synopsys MOSS device model and is identical to AMI SPICE MOS LEVEL 4 ZENH 0 This depletion model is revised in the Synopsys MOS5 device model from previous depletion mode and is identical to AMI SPICE MOS LEVEL 5 144 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model The Synopsys enhancement and depletion modes are basically identical to the AMI models However the Synopsys enhancement and depletion modes let you choose either SPICE or ASPEC temperature compensation TLEV 1 default uses ASPEC style temperature compensation TLEV 0 uses SPICE style temperature compensation CAPOP 6 represents AMI Gate Capacitance in the Synopsys device models CAPOP3 6 is the default setting for LEVEL 5 only LEVEL 5 models can also use CAPOP 1 2 3 The ACM parameter defaults to 0 in LEVEL 5 invoking SPICE style parasitics You can also set ACM to 1 ASPEO or to 2 Synopsy
188. Conclude Y HSPICE MOSFET Models Manual X 2005 09 549 8 Customer Common Model Interface Internal Routines Internal Routines In the MOS3 implementation example the interface routines in CMImos3 c also call the internal routines in Table 149 Table 149 Common Model Interface Internal Routines Routine Description CMlImos3Getlpar c Get instance parameter index CMlImos3Setlpar c Set instance parameter CMImos3GetMpar c Get model parameter index CMImos3SetMpar c Set model parameter CMImos3eval c Evaluate model equations CMImos3set c Setup a model CMImos3temp c Setup an instance including the temperature effect Figure 43 shows the hierarchical relationship between the interface routines and the internal routines For the automatic script to work the name of the interface variable and all routine files must follow this naming convention pCMI xxxdef CMIxxx c CMIxxxGetlpar c CMIxxxSetlpar c CMIxxxGetMpar c CMIxxxSetMpar c CMIxxxeval c In the preceding syntax xxx is the model name 550 HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Internal Routines Figure 43 Hierarchy of Interface and Internal Routines Interface Routines Internal Routines CMImos3 c CMI CMI ResetModel CMImos3setupModel CMImos3set c CMI Resetlnstance CMImos3temp CMI_SetupModel E CMImos3temp c CMI_mos3def CMI_SetupIn
189. D parameters 484 template output 492 BSIM3 Version 2 MOS model 382 BSIM3v3 model HSPICE 412 MOS 397 BSIM4 generalized customer CMI 556 BSIM4 model parameters 442 STI LOD 438 bulk charge effect 3 transconductance MOSFETs 35 BYPASS option 10 12 C capacitance 597 Index CAPOP model selector 7 control options 73 equations 55 model parameters 40 41 74 selection 65 MOSFETs AC gate 95 BSIM model 336 diodes 55 102 equations 78 96 gate capacitance 72 example 71 length width 95 models 64 SPICE Meyer 78 Meyer model 64 76 models 65 overlap 77 parameters Meyer 76 MOSFETs Cypress 197 IDS LEVEL 5 144 LEVEL 38 197 LEVEL 49 and 53 419 LEVEL 5 144 LEVEL 57 473 plotting 72 capacitor models gate 64 list 7 capacitor transcapacitance 66 CAPOP model parameter 7 65 74 XPART 77 XQC 77 cascoding example 63 CASMOS GEC model 5 GTE model 5 Rutherford model 5 CBD model parameter 40 CBS model parameter 40 CDB model parameter 41 channel length modulation equations LEVEL 2 133 598 LEVEL 3 139 LEVEL 38 202 LEVEL 5 153 LEVEL 6 175 LEVEL 8 186 parameters LEVEL 6 175 LEVEL 8 182 charge conservation 88 90 circuits nonplanar and planar technologies 27 wave processes 28 CJ model parameter 41 CJA model parameter 41 CJGATE model parameter 41 CJP model parameter 41 CJSW model parameter 41 CMI function calling protocol 548 generalized customer 553 models simulations 521 testing 528 variables 533
190. D 2 VSAT 8 0e4m s Yes Saturation velocity AO 1 0 Yes Coefficient of the channel length dependence of the bulk charge effect AGS 0 0v Yes Coefficient of the Vy dependence of the bulk charge effect 448 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 115 Basic Model Parameters MOSFET Level 54 Continued Parameter Default Binnable Description BO 0 0m Yes Bulk charge effect coefficient for the channel width B1 0 0m Yes Bulk charge effect width offset KETA 0 047 V Yes Body bias coefficient of the bulk charge effect A1 0 0v Yes First non saturation effect parameter A2 1 0 Yes Second non saturation factor WINT 0 0m No Channel width offset parameter LINT 0 0m No Channel length offset parameter DWG 0 0m V Yes Coefficient of gate bias dependence of Wr DWB 0 0m V 1 2 Yes Coefficient of the body bias dependence of the War bias dependence VOFF 0 08V Yes Offset voltage in subthreshold region for large W and L values VOFFL 0 0mV No Channel length dependence of VOFF MINV 0 0 Yes Vosterr fitting parameter for the moderate inversion condition NFACTOR 1 0 Yes Subthreshold swing factor ETAO 0 08 Yes DIBL coefficient in the subthreshold region ETAB 0 07V Yes Body bias coefficient for the DIBL effect for the subthreshold DSUB DROUT Yes DIBL coefficient exponent in the subthreshold region CIT 0 0F m2 Yes Interface trap capacitance HSPICE MOSFET Models Manual
191. DBO LX19 Intrinsic gate to drain capacitance 54 57 59 60 HSPICE9 MOSFET Models Manual 17 X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level CGSBO LX20 CGSBO dQg dVd Meyer and Charge All except 54 Conservation 57 59 60 CGSBO LX20 Intrinsic gate to source capacitance 54 57 59 60 CBGBO LX21 CBGBO dQb dVg Meyer and Charge All except 54 Conservation 57 59 60 CBGBO LX21 Intrinsic bulk to gate capacitance 54 CBGBO LX21 Intrinsic floating body to gate capacitance 57 59 60 CBDBO LX22 CBDBO dQb dVd Meyer and Charge All except 54 Conservation 57 59 60 CBDBO LX22 Intrinsic bulk to drain capacitance 54 CBDBO LX22 Intrinsic floating body to drain 57 59 60 capacitance CBSBO LX23 CBSBO dQb dVs Meyer and Charge All except 54 Conservation 57 59 60 CBSBO LX23 Intrinsic bulk to source capacitance 54 CBSBO LX23 Intrinsic floating body to source 57 59 60 capacitance QBD LX24 Drain bulk charge QBD All LX25 Drain bulk charge current CQBD not All used in HSPICE releases after 95 3 QBS LX26 Source bulk charge QBS All LX27 Source bulk charge current CQBS not All used after HSPICE release 95 3 CAP BS LX28 Bias dependent bulk source capacitance All except 57 58 HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates
192. Default Description AO 1 0 Bulk charge effect coefficient for the channel length Al 1 V 0 0 First non saturation effect parameter A2 1 0 Second non saturation effect parameter ABP 1 0 Coefficient of Apert dependency on Vos ADICEO 1 DICE bulk charge factor AGIDL 1 W 0 0 GIDL constant HSPICE MOSFET Models Manual 485 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 59 UC Berkeley BSIM3 SOI FD Model Table 136 MOSFET Level 59 DC Parameters Continued Parameter Unit Default Description AGS 1 V 0 0 Gate bias coefficient of Apu All 1 V 0 0 First Vdsatii parameter for the Leff dependence ALPHAO m V 0 0 First parameter of the impact ionization current ALPHA1 1 V 1 0 Second parameter of the impact ionization current BO m 0 0 Bulk charge effect coefficient for the channel width B1 m 0 0 Width offset for the bulk charge effect BGIDL V m 0 0 GIDL exponential coefficient BII m V 0 0 Second Vdsatii parameter for the Leff dependence CDSC F m 2 4e 4 Drain source to the channel coupling capacitance CDSCB F m 0 Body bias sensitivity of cdsc CDSCD F m 0 Drain bias sensitivity of cdsc Cll s 0 0 First Vdsatii parameter for the Vy dependence CIT F m 0 0 Interface trap capacitance DELP V 0 02 Constant for limiting Vpserto the surface potential DELTA 0 01 Effective Vy parameter DII V 1 0 Second Vdsatii parameter for the Vys dependence DROUT 0 56 L dependence coefficient of the DIBL correction parameter in Rout
193. Deff lt Weff then COBD SW CJGATEscaled PDeff f ACM 2 and PDeff gt Weff then COBD SW CJSWscaled PDeff Weff CJGATEscaled Weff HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Diode Equations If ACM 3 then COBS SW CJSWscaled PSeff CJGATEscaled Weff COBD SW CJSWscaled PDeff CJGATEscaled Weff Source Diode Capacitance If COBS COBS SW gt O and vbs lt 0 then dibs a Srp cops quotas capbs Spe S PR MJSW COBS_SW 1 vbs PHP If COBS COBS SW gt O and vbs gt 0 then capbs TT COBS 1 MJ Sps COBS 1 wr 5 S COBS SW 14MJ8W PHP Otherwise if COBS COBS SW lt 0 then MJ For vbs lt 0 capbs TT 193 LM CBS 1 m Ovbs PB 2772854 cpg 1e ur 98 For vbs gt 0 capbs TT Sup tM CBS 14MJ PB Drain Diode Capacitance If COBD COBD SW gt 0 then For vbd 0 dibd vb M7 TT BD 1 capbd Jha t CO PR MJSW PDeff COBD_SW 1 5 vod PHP HSPICE MOSFET Models Manual 57 X 2005 09 2 Technical Summary of MOSFET Models Common Threshold Voltage Equations For vbd gt 0 dibd ovbd C0BD 1 MJ 28 capbd TT PB COBD SW 1 MJSW ed PHP Otherwise if ADeff CJscaled PDeff CJSWscaled lt 0 then 1 MJ For vbd 0 capbd TT 224 yM CBD 1 ald Qvbd PB B dibd mba For vbd gt 0
194. Description RDSW 200 0 ohm um P Yes Zero bias LLD resistance per unit width for RDSMOD 0 RDSWMIN 0 0 ohm um P No LDD resistance per unit width at high Vgs and zero Vps for RDSMOD 0 RDW 100 0 ohm um WP Yes Zero bias lightly doped drain resistance Rg v per unit width for RDSMOD 1 RDWMIN 0 0 ohm um P No Lightly doped drain resistance per unit width at high Vgs and zero Vps for RDSMOD 1 RSW 100 0 ohm um WP Yes Zero bias lightly doped source resistance R V per unit width for RDSMOD 1 RSWMIN 0 0 ohm um WP No Lightly doped source resistance per unit width at high Vg and zero Vps for RDSMOD 1 PRWG 1 0V Yes Gate bias dependence of the LDD resistance PRWB 0 0v 9 5 Yes Body bias dependence of the LDD resistance WR 1 0 Yes Channel width dependence of the LDD resistance NRS 1 0 No Number of source diffusion squares NRD 1 0 No Number of drain diffusion squares Table 117 Impact lonization Current Model Parameters MOSFET 54 Parameter Default Binnable Description ALPHAO 0 0Am V_ Yes First parameter of the impact ionization current ALPHA1 0 0A V Yes Isub parameter for length scaling BETAO 30 0V Yes Second parameter for the impact ionization current HSPICE MOSFET Models Manual X 2005 09 451 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 118 Gate Induced Drain Leakage Model Parameters MOSFET Level 54 Parameter Default Binnable Description AGIDL 0 0ohm Yes Pre exponential coefficient
195. Duy ae 2 JO CSOs Vpn NOI Vom Y ang EG N yank Bass 42 4 5 Neus petu oer eL m ox ox Pea V 0 ea in 2 Ei If you do not specify Vp as a model parameter then f 1 0e22 N y ET E 2 ni Mobility of Carrier Model Parameters Mo Up U U Ho Meg Vost Vu Vgs V 1 U E u 44 U Vys OX T Drain Saturation Voltage Model Parameters Ay Vsap Xp Aj Ap Raso Ras Rds and Pfactor Ras Raso Rasw le6 Wese HSPICE MOSFET Models Manual 391 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model 392 Pfactor A V A _ if Pfactor 51 simulation sets it to Pfactor 1 Naat x m Vin If Rds O and Pfactor 1 then sar Pepe Mest Prga ka Apule E sat Lot V est For BULKMOD 1 K1 Aj L Ae 1 oru b KETA V Lyt TV Tis For BULKMOD 2 Kici kh Apn Ra KETA V Lg T1 J0 2 Th YI For Vp lt 0 For Vg 2 0 V Tis Su s Jo EQ 22 7 Vos Herf Pit In general Vdsat solves Tmpa Vdsat Vdsat Tmpb Vdsat Tmpc 0 Vig Tmpb i Tne 4 Tmpa Tmpc 2 Tmpa CoU Ray 14 1 Pfactor sat Tmpa Apuk Apu Wege V Tmpb V 2 Pfactor 1 Agi Esar Lg 3 Abi Vost Wor Var Cox Ras 2 Tmpc Voss BE Lg Vost Ue Wo V sat Cox Ras HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model Linear Re
196. E MOSFET Models Manual 441 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model 442 Lnew and Wnew are evaluated as L LMLT XL W WMLT XW Lnew Wnew Similarly dL dW Ldlc Ldlcig LeffCV Wdwc Wdwcig WeffCV WeffCJ and grgeltd are all evaluated from Lnew and Wnew HSPICE Junction Diode Model and ACM BSIM4 now supports Area Calculation Method ACM similar to BSIM3v3 for the following models and corresponding ACM values For the HSPICE junction model specify ACM 0 1 2 or 3 For the Berkeley BSIM4 junction model specify ACM 10 11 12 or 13 For the junction current junction capacitance and parasitic resistance equations corresponding to ACM 0 1 2 3 see MOSFET Diode Models on page 39 Set ACM 10 11 12 or 13 to enable the Berkeley BSIM4 junction diodes and to add parasitic resistors to the MOSFET The parasitic resistor equations for ACM 10 11 12 or 13 correspond to the ACM 0 1 2 or 3 parasitic resistor equations ACM 10 11 12 or 13 all use the Berkeley junction capacitance model equations The default ACM value is 12 For ACM 10 11 12 or 13 usage see Level 49 and 53 BSIM3v3 MOS Models on page 397 CALCACM still only applies to ACM 12 although it is disabled ACM 12 uses the Berkeley BSIM4 junction diodes and parasitic resistors equations if CALCACM equals 1 or 0 The default CALCACM value is 1 Table 110 MOSFET Level 54 Parameters Parameter Description nf Number
197. E MOSFET Models Manual 463 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model 464 General Syntax for BSIM3 SOI Mxxx nd ng ns ne lt np gt lt nb gt lt nT gt mname lt L val gt lt W val gt lt M val gt lt AD val gt lt AS val gt lt PD val gt lt PS val gt lt NRD val gt lt NRS val gt lt NRB val gt lt RTHO val gt lt CTHO val gt lt NBC val gt lt NSEG val gt lt PDBCP val gt lt PSBCP val gt lt AGBCP val gt lt AEBCP val gt lt VBSUSR val gt DELTOX val TNODEOUT t off lt FRBODY gt BJToff val IC Vds Vgs Vbs Ves Vps gt In this syntax the angle brackets indicate optional parameters Parameter Description Mxxx SOI MOSFET element name Must begin with M followed by up to 1023 alphanumeric characters nd Drain terminal node name or number ng Front gate node name or number ns Source terminal node name or number ne Back gate or substrate node name or number np External body contact node name or number nb Internal body node name or number nT Temperature node name or number mname MOSFET model name reference L SOI MOSFET channel length in meters This parameter overrides DEFL in an OPTIONS statement DefaultZDEFL with a maximum of 0 1m W MOSFET channel width in meters This parameter overrides DEFW in an OPTIONS statement DefaultZDEFW M Multiplier to simulate multiple SOI MOSFETs i
198. E 2 5E 2 length modulation POMEXP Coefficient for the geometry 0 2 0 2 independent 1 m part PLMEXP Coefficient for the length dependent 0 0 part of 1 m PWMEXP Coefficient for the width dependent 0 0 part of 1 m PLWMEXP Coefficient of the length times width 1 0 0 m dependent part POA1 Coefficient of the geometry 6 022 6 858 independent a1 part PLA1 Coefficient for the length dependent 0 0 part of a1 PWA1 Coefficient for the width dependent 0 0 part of a1 PLWA1 Coefficient for the length times width 0 0 a1 dependent part POA2 Coefficient for the geometry V 3 802E 1 5 732E 1 independent a2 part PLA2 Coefficient for the length dependent V 0 0 part of a2 PWA2 Coefficient for the width dependent V 0 0 part of a2 298 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS PLWA2 Coefficient of the length times width a2 V 0 0 dependent part POAS3 Coefficient of the geometry 6 407E 1 4 254E 1 independent a3 part PLA3 Coefficient for the length dependent 0 0 part of a3 PWA3 Coefficient for the width dependent 0 0 part of a3 PLWA3 Coefficient for the length times width 0 0 a3 dependent part POIGINV Coefficient for the geometry AV 2 0 0 independent part of IGINV PLIGINV Coefficient for the length dependent AV 2 0 0 part
199. E noise model See Noise Models on page 96 for more information If you do not specify NLEV simulation invokes the Berkeley noise equations Performance Improvements To improve the performance of Levels 49 and 53 reduce the complexity of model equations replacing some calculations with spline functions and compiler optimization For Level 49 the result is a reduction in simulation time of up to 4096 compared to releases before 97 4 while maintaining accuracy to 5 digits or better To use the spline functions set the SFVTFLAG 1 model parameter in the model card SFVTFLAG 0 the default value disables the spline functions For Level 53 all BSIM3v3 non compliant features default to off Reduced Parameter Set BSIM3v3 Model BSIM3 lite Setting the LITE 1 model parameter in Level 49 to invoke the BSIM3v3 lite reduced parameter set model Use it with model binning Without binning to account for geometry effects the full BSIM3v3 model specifies several model parameters However it is often difficult to extract a global BSIM3v3 model that is accurate over the entire geometry range To improve accuracy over a range of geometries you can bin the model parameters That is this model divides the entire length width geometry range into rectangular regions or bins Simulation extracts a different set of parameters for each bin The built in bilinear parameter interpolation scheme maintains continuity over length width at the boundaries bet
200. EAL DE ei dx 204 SALUKALION nan KA LANG AD IRINGAN AA 204 LEVEL 40 HP a Si TFT Model 2 22 2 204 Using the HP a Si TFT Model 2 cece eee eee 204 Effect of SCALE and SCALM 2 222 205 Noise Model nanana anaana aaa 206 DELVTO Elemen eironi e E cee eens 206 Device Model and Element Statement Example 206 LEVEL 40 Model Equations eee eee eee 206 Cutoff Region NFS 0 vgs von a 208 Noncutoff Region NFS 0 0c e eee eee 208 CO gs oer oes nv eet Re ced a eek ade 211 LEVEL 40 Model Topology llis 211 Heterences sso Ou IAE ey eee EA PULS PU LA E eee 211 5 Standard MOSFET Models Levels 50 to 64 213 Level 50 Philips MOS9 Model 0 0000 cece eee eee 214 viii HSPICE MOSFET Models Manual X 2005 09 Contents JUNCAP Model Parameters 0 illie eee eee 221 Using the Philips MOS9 Model 0 000 c eee eee eee 222 Model Statement Example 0c eee eee eee 223 Level 55 EPFL EKV MOSFET Model 0000 cee eee eee 224 Single Equation Model 1 0 0 0 0 0 cc eens 224 Effects Modeled 2 0 0c cect teens 225 Coherence of Static and Dynamic Models 20005 225 Bulk Reference and Symmetry lille 226 EKV Intrinsic Model Parameters eee ee eeaes 227 Static Intrinsic Model Equations 000 c eee eee eee 231 Basic Definitions liliis e
201. EFW in an OPTION statement DefaultZDEFW with a maximum of 0 1m Multiplier to simulate multiple SOI MOSFETS in parallel The M setting affects all channel widths diode leakages capacitances and resistances Default 1 Drain diffusion area Overrides DEFAD in the OPTION statement DefaultZDEFAD Source diffusion area Overrides DEFAS in the OPTION statement Default DEFAS Perimeter of the drain junction including the channel edge Overrides DEFPD in the OPTION statement Perimeter of the source junction including the channel edge Overrides DEFPS in the OPTION statement Number of squares of the drain diffusion for the drain series resistance Overrides DEFNRD in the OPTION statement Number of squares of the source diffusion for the source series resistance Overrides DEFNRS in the OPTION statement Number of squares for the body series resistance Additional drain resistance due to the contact resistance in units of ohms This value overrides the RDC setting in the model specification Default 0 0 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model Parameter Description RSC Additional source resistance due to the contact resistance in units of ohms This value overrides the RDC setting in the model specification Default 0 0 RTHO Thermal resistance per unit width If you do not specify RTHO simulation extracts it from the mode
202. EL 5 2 2a 150 Threshold Voltage Vi llle 151 Saturation Voltage Vgsat acida rc one ha are e an ee 153 Mobility Reduction UBeff llle II 153 Channel Length Modulation llle 153 Subthreshold Current lgs issus 154 LEVEL 6 LEVEL 7 IDS MOSFET Model 0 0 000 cee ee eaee 155 LEVEL 6 and LEVEL 7 Model Parameters 0 0000 e eee 155 UPDATE Parameter for LEVEL 6 and LEVEL 7 156 LEVEL 6 Model Equations UPDATE 0 2 2020055 159 IDS Equations i rria paia a eee 159 Effective Channel Length and Width aaan anaana 159 Threshold Voltage vth nananana nananana 159 Single Gamma VBO 0 nananana anaana 160 Effective Built in Voltage vbi 0000 eee eee 160 Multi Level Gamma VBO gt 0 0 0000 cece ee ee eee 161 Effective Built in Voltage vbi for VBO gt 0 00005 163 Saturation Voltage vdsat UPDATE 0 2 004 163 Saturation Voltage vsat llle sess 167 LEVEL 6 IDS Equations UPDATE21 sels 168 Alternate DC Model ISPICE model lulllsulssss 168 Subthreshold Current ids llli 169 Effective Mobility ueff llli 171 Channel Length Modulation llli lille 175 ASPEC Compatibility llle 180 LEVEL 7 IDS Model lellselsleee ees 181 LEVEL 8 IDS Model 222 2a 182 LEVEL 8 Model Parameters a 182 LEVEL 8 Model Equations
203. EO 2 another device shares the source GEO 3 another device shares the drain and source Figure 14 Stacked Devices and Corresponding GEO Values GEO 2 GEO 3 GEO 1 LD ka HDIF HDIF pais eal S D PE LDIF Calculating Effective Areas and Peripheries ACM 3 calculates the effective areas and peripheries based on the GEO value If you do not specify AD then For GEO 0 or 2 ADeff 2 HDIFeff Weff For GEO 1 or 3 ADeff HDIFeff Weff Otherwise ADeff M AD WMLT SCALE HSPICE MOSFET Models Manual 53 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models 54 If you do not specify AS then For GEO 0 or 1 ASeff 2 HDIFeff Weff For GEO 2 or 3 ASeff HDIFeff Weff Otherwise ASeff M AS WMLT SCALE If you do not specify PD then For GEO 0 or 2 PDeff 4 HDIFeff Weff For GEO 1 or 3 PDeff 2 HDIFeff Otherwise PDeff M PD WMLT SCALE If you do not specify PS then For GEO 0 or 1 PSeff 4 HDIFeff Weff For GEO 2 or 3 PSeff 2 HDIFeff Otherwise PSeff M PS WMLT SCALE Simulation calculates Weff and HDIFeff as follows Weff M Wscaled WMLT XWscaled HDIFeff HDIFscaled WMLT Note The Weff value is not the same as the Weff value in the LEVEL 1 2 3 and 6 models The 2 WDscaled term is not subtracted Effecti
204. ET Level 6 Name Alias Units Default Description LAMBDA 1NKL O vds coefficient LAM1 1 m 0 Channel length coefficient KL 0 vds exponent VGLAM 1 V 0 Gate drive coefficient HSPICE MOSFET Models Manual 179 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model 180 Unlike the other CLM values this equation calculates the channel length modulation AL in all regions of operations and uses it to modify the ids current Leff LAMBDA vdsK 1 VGLAM vgs vth AL 1 LAMI Leff ids ids 1 AL Leff Note LAMBDA is a function of the temperature ASPEC Compatibility To make MOSFET models compatible with ASPEC specify ASPEC 1 in the OPTION statement and LEVEL 6 in the associated MOSFET model statement If you assign the element parameters without keynames specify the parameters in the same sequence as in the general format The Level 6 MOSFET model assigns parameters in the order that you list them in the element statement If parameter names are also element keynames simulation reports errors If you use the ASPEC option several program variations occur The LEVEL model parameter is set to 6 Note Setting LEVEL 6 in the model does not invoke ASPEC MOSFET Option WL 1 General Options SCALE 1e 6 SCALM 1e 6 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 7 IDS Model ASPEC sets the SCALE and SCALM op
205. ET MESFET Level 3 model Statz model and Spice3 This is example code only for using the Customer CMI interface Adding Proprietary MOS Models 522 You can use the Customer CMI interface to enter proprietary models into HSPICE or HSPICE RF This section describes how to use Customer CMI to both add a new MOS model and simplify integration MOS Models on Unix Platforms In the following examples the percent sign is the UNIX shell prompt and installdir points to the directory where you installed HSPICE or HSPICE RF ARCH is the OS type for the computer The Customer CMI supports Sun4 Solaris and HP platforms To create a Customer CMI shared library and add a new model Create a directory environment Modify the configuration file Prepare and modify the model routines Compile the shared library D pe Gar M Set up the runtime shared library path Creating the Directory Environment To create the Customer CMI directory environment 1 Copy the Customer CMI directory from the HSPICE or HSPICE RF release directory to a new location as shown in the following example cp r S installdir cmi home user1 userx model The new Customer CMI directory home user1 userx model cmi is your working directory You must have read and write access to this directory 2 Copy an existing model subdirectory to a new model directory HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Addi
206. ETAZET 170 0e 3 30 0e 3 Exponent of length dependence of ZET1R SLZET1 390 0e 3 2 8 Length dependence coefficient of ZET1R VSBTR V 2 1 100 0 Limiting voltage for back bias dependence SLVSBT Vm 4 4e 6 0 0 Length dependence of VSBTR AiR 6 0 10 0 Weak avalanche current factor STA1 ki 0 0 0 0 Temperature coefficient of A1R SLA1 m 1 3e 6 15 0e 6 Length dependence of A7R SWA1 m 3 0e 6 30 0e 6 Width dependence of A7R HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 50 Philips MOS9 Model Table 45 MOSFET Level 50 Model Parameters Continued Name Unit Default N Default P Description A2R V 38 0 59 0 Exponent of weak avalanche current SLA2 Vm 1 0e 6 8 0e 6 Length dependence of A2R SWA2 Vm 2 0e 6 15 0e 6 Width dependence of A2H A3R 650 0e 3 520 0e 3 Factor of minimum drain bias above which avalanche sets in SLA3 m 550 0e 9 450 0e 9 Length dependence of A3R SWA3 m 0 0 140 0e 9 Width dependence of A3R TOX m 25 0e 9 25 0e 9 Oxide thickness COL F m 320 0e 12 320 0e 12 Gate overlap capacitance per unit width WDOG m 0 0 Characteristic drawn gate width below which dogboning appears FTHE1 0 0 Coefficient describing the width dependence of THE1 for W lt WDOG NFMOD 0 Flicker noise selector 0 selects the old flicker noise model added in release 98 4 NTR J 24 4e 21 21 1e 21 Thermal noise coefficient NFR y2 70 0e 12 21 4e 12 Flicker noise coefficient NFAR
207. F m CJSW No Zero bias gate edge sidewall bulk junction capacitance HSPICE specific use only if ACM 3 PB PHIB V 1 0 No Bulk junction contact potential PBSW V 1 0 No Sidewall bulk junction contact potential PHP V 1 0 No Sidewall bulk junction contact potential HSPICE use only HSPICE MOSFET Models Manual X 2005 09 with the HSPICE junction model ACM 0 3 423 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 104 Junction Parameters MOSFET Levels 49 53 Continued Name Unit Default Bin Description PBSWG V PBSW No Gate edge sidewall bulk junction contact potential use only with the Berkeley junction model ACM 10 13 HSPICE has no equivalent parameter Gate edge contact potential is always set to PHP for the HSPICE junction model MJ 0 5 No Bulk junction grading coefficient MJSW 0 33 No Sidewall bulk junction grading coefficient MJSWG MJSW No Gate edge sidewall bulk junction grading coefficient use only with the Berkeley junction model ACM 10 13 HSPICE has no equivalent parameter Always set the gate edge grading coefficient to MJSW for the HSPICE junction model Note See MOSFET Diode Models on page 39 for HSPICE junction diode model usage Table 105 NonQuasi Static NOS Parameters MOSFET 49 53 Name Unit Default Bin Description ELM 5 0 Yes Elmore constant Table 106 MOSFET Levels 49 53 Version 3 2 Parameters Name Unit Default Bin Description TOXM m TOX
208. FET Model 188 MOSFET Level 27 is a three terminal silicon on sapphire SOS FET transistor model 4 This SOSFET model is based on a sapphire insulator that isolates the substrate and models the behavior of SOS devices more accurately than standard MOSFET models with physically unreal parameter values The SOSFET model also includes a charge conservation model based on the Ward and Dutton model Because the defaults of the SOSFET model parameters depend on the channel length you must specify the SOSLEV model parameter to select either the 5 um or 3 um processing model SOSLEV 1 selects the 5 um model otherwise this model automatically uses the 3 um value including the second order effects default 3 um Note This model does not include bulk nodes If you specify bulk nodes simulation ignores them This model does not use the ACM model parameter because it does not include any junction diodes Also the only value that the CAPOP model parameter accepts is 7 Seven is its own charge conservation model which you cannot use in other MOSFET models HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 27 SOSFET Model Temperature compensation equations for the VTO and UO SOSFET model parameters are the same as those in the MOSFET model Note This model includes a special option for bulk nodes for silicon on sapphire In the model definition if you specify 1 for the bulk node this model
209. G a 2exp ni Pes T tc LIE a E E REN ee OX 394 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model i D p L SP aq Lio A 3 Tox i Xdepo X depo IT ES If subthMod 0 las 8m 8ds 8mb 9 If subthMod 1 V FG Limit Le 1 exp d I 9 ijs gE sj N peak Were y ds 7 J V limit 9 2 0 L tm limit exp n K eff I u0 q Esi N peak Wn y Vos CE Vin Vogt DIBL Vas exp u E 2 06 D T im exp aa UTS eff If subthMod 2 n Vim Dos on ee 1 exp LN s eff P n Vim Transition Region for subthMod 2 only Model Parameters Vishigio Vastow 2 2 lis 1 f Hose te LAYA EE pes V E Valo TAW gslow 2 V Vsshigh La CV cig 2 Ys t V odiis A us ig Vin a Fion l 2 V NG slow HSPICE MOSFET Models Manual 395 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model 396 Vos E mhigh i Voshigh 7 mlow Viton Cashigh 14s10w P KS Smhigh S8mlow um T ios mlow V T V taw Temperature Compensation Model Parameters Ap U Up U 1 KT1 KT2 UTE al V femp V tref KT1 KT2 V Trario 1 UTE ratio uO temp uO tref T V emp Va Gref A T ratio 1 U temp U tref U 41 Tratio 1 U temp U tref U OL sonam 1 U temp U tref Uci C ots 1 PMOS Model In the following ex
210. Gate Capacitance model default for Levels 13 28 39 CAPOP 39 BSIM2 Charge Conserving Gate Capacitance model LEVEL 39 HSPICE MOSFET Models Manual 65 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models 66 CAPOP 4 selects the recommended charge conserving model from among CAPOP 11 12 or 13 for the specified DC model Table 15 CAPOP 4 Selections MOS Level Default CAPOP CAPOP 4 selects 2 2 11 3 2 12 13 28 39 13 13 Other levels 2 11 The proprietary models Levels 5 17 21 22 25 31 33 the SOS model LEVEL 27 and models higher than 49 have their own built in capacitance routines Transcapacitance If a capacitor has two terminals 1 and 2 with charges named Q1 and Q2 on the two terminals that sum to zero for example Q1 Q2 then the charge is a function of the voltage difference between the terminals V12 V1 V2 One quantity C2dQ1 dV12 completely describes the small signal characteristics of the device If a capacitor has four terminals the sum of the charges on the four terminals must equal zero Q1 Q2 Q3 Q4 0 They can depend only on voltage differences but they are otherwise arbitrary functions Because three independent charges Q1 Q2 Q3 are functions of three independent voltages V14 V24 V34 you must specify nine derivatives to describe the small signal characteristics You can consider the four charges separately as functions of the four terminal vo
211. HETH 0E 5 POSDIBL 8 530E 4 POSSF 1 2E 2 POALP 2 5E 2 VP 5 0E 2 POMEXP 0 2 POA1 6 022 POA2 38 02 POA3 0 6407 POBINV 48 POBACC 48 KOV 2 5 TOX 3 2E 09 POCOX 2 980E 14 POCGDO 6 392E 15 POCGSO 6 392E 15 NT 1 656E 20 PONFA 8 323E 22 PONFB 2 514E 7 POTVFB 5 0E 4 POTPHIB 8 5E 4 POTETABET 1 30 POTETASR 0 65 POTETAPH 1 350 NU 2 0 POTNUEXP 5 25 POTETAR 0 95 POTETASAT 1 040 Level 64 STARC HiSIM Model HiSIM Hiroshima university STARC IGFET Model is a publicly available MOSFET model for circuit simulation It uses drift diffusion approximation and a channel surface potential description You can model all MOSFET characteristics closely based on their physical origins by using fewer model parameters about 90 model parameters each parameter set is sufficient for all gate lengths These model parameters are directly related to the MOSFET physics that a simulator can easily extract according to its physical meanings 310 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model The STARC HiSIM model is Level 64 in the Synopsys MOSFET models To use this model specify M1 drain gate source bulk NCH w 4u I lu MODEL NCH NMOS LEVEL 64 HSPICE HiSIM model code is based on the Spice3f5 version that Hiroshima University STARC released at the following web site http www starc jp kaiha
212. HSPICE MOSFET Models Manual Version X 2005 09 September 2005 SYNOPSYS Copyright Notice and Proprietary Information Copyright 2005 Synopsys Inc All rights reserved This software and documentation contain confidential and proprietary information that is the property of Synopsys Inc The software and documentation are furnished under a license agreement and may be used or copied only in accordance with the terms of the license agreement No part of the software and documentation may be reproduced transmitted or translated in any form or by any means electronic mechanical manual optical or otherwise without prior written permission of Synopsys Inc or as expressly provided by the license agreement Right to Copy Documentation The license agreement with Synopsys permits licensee to make copies of the documentation for its internal use only Each copy shall include all copyrights trademarks service marks and proprietary rights notices if any Licensee must assign sequential numbers to all copies These copies shall contain the following legend on the cover page This document is duplicated with the permission of Synopsys Inc for the exclusive use of and its employees This is copy number Destination Control Statement All technical data contained in this publication is subject to the export control laws of the United States of America Disclosure to nationals of other countries contrary to United States law is
213. I a TGAM2 1 SPO GMBS P M AL R I z AN B M Each plot compares IDS VTH VDSAT GM GDS and GMBS as a function of vsb for UPDATE 1 Saturation Voltage vsat To obtain the right value for vsat the following equations calculate two trial values of vsat corresponding to yi and yb 2 d 2 P g 1 2 magi ves vbil NY Tep 2 Eco e pnr vsb 7 2 n yi 2 eR EE 1 2 vsat2 vas Pe 0 PG vas vbi2 zB PHI vsb n Vbit is the built in potential corresponding to yi vbi2 is the built in potential corresponding to yb If vdsat1 vsb lt VBO then vdsat vdsat1 f vdsat2 vsb gt VBO then vdsat vdsat2 To obtain vdsat vc modifies vsat for the carrier velocity saturation HSPICE MOSFET Models Manual X 2005 09 167 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model 168 LEVEL 6 IDS Equations UPDATE 1 You can use one of three equations for ids depending on the region of operation To derive these equations this model integrates the bulk charge vgs vth v v from the source to the drain For vsb VBO vde the model forms an entire gate depletion region in the implant layer l MR ids p c vbil laag vde 3 Yl PHI vde vsb 2 PHI sip In the preceding equation vbit is the same as vbi for vsbsVBO For vsb VBO the entire gate depletion region expands into the bulk area ids p le vbi2 noeh vde 3 yb
214. ICE or HSPICE RF simulator The Customer Common Model Interface CMI is a Synopsys program interface that you can use to add your own proprietary MOSFET models into the Synopsys HSPICE or HSPICE RF simulator HSPICE RF supports the external CMI including the BTA SOI model This chapter describes the following topics Overview of Customer CMI Directory Structure Running Simulations Using Customer CMI Models Adding Proprietary MOS Models Testing Customer CMI Models Model Interface Routines Interface Variables Internal Routines HSPICE MOSFET Models Manual 517 X 2005 09 8 Customer Common Model Interface Overview of Customer CMI Extended Topology Conventions Overview of Customer CMI 518 HSPICE or HSPICE RF uses a dynamically linked shared library to integrate models with the Customer CMI Add the cmiflag global option to load the dynamically linked Customer CMI library libCMImodel Simulation searches for the libCMImodel shared library in the Shspice lib models path If the simulator does not find the library it searches in the Sinstalldir SARCH lib models directory Dynamic loading shares resources more efficiently than static binding does If you run several simulations concurrently HSPICE or HSPICE RF needs only one copy of the dynamically linked Customer CMI model in memory which speeds up the process Theoretically the static linking version is always slightly faster if you run only one simulatio
215. ICKNESS AND CAPACITANCE TOX 165 CGSO 0 CAPOP 2 CHANNEL IMPLANT t NI 1 5e12 KCS 3 DP 0 25 END LEVEL 40 HP a Si TFT Model 204 The Synopsys Level 40 MOSFET device model represents a Hewlett Packard amorphous silicon thin film transistor model MOSFET Level 40 uses only the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters Using the HP a Si TFT Model 1 SetLEVEL 40 to identify the model as the HP a Si TFT model 2 Default value for L is 10um and the default value for W is 40 um Use the M designation for MOSFET rather than the A designation for a Si TFT in the netlist 4 Use the NMOS or PMOS designation for device type rather than the NAT or PAT designation HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 40 HP a Si TFT Model Note Because of the unavailability of p channel TFTs PMOS model testing has been limited 5 LEVEL 40 is a three terminal model It lacks bulk nodes so simulation does not append any parasitic drain bulk or source bulk diodes to this model You can specify a fourth node but it does not affect the simulation results except for GMIN terms 6 Parasitic resistances and overlap capacitances are constant They are not scaled with width length and temperature Capacitance expressions in this model do not conserve charge The TREF parameter
216. L 0 UNDEF 779 0 0 NOTY GLOB 0 UNDEF mul 1853 0 0 NOTY GLOB 0 UNDEF dtou 810 0 0 NOTY GLOB 0 UNDEF iob 822 0 0 NOTY WEAK O UNDEF _ex_deregister 749 0 O NOTY WEAK O UNDEF _ex_register 857 0 0 NOTY GLOB 0 UNDEF atan 878 0 0 NOTY GLOB 0 UNDEF cos 820 0 0 NOTY GLOB 0 UNDEF exp 777 0 0 NOTY GLOB 0 UNDEF fabs 872 0 0 NOTY GLOB 0 UNDEF fprintf 764 0 0 NOTY GLOB 0 UNDEF log 767 0 0 NOTY GLOB 0 UNDEF malloc 842 0 0 NOTY GLOB 0 UNDEF memset 772 0 0 NOTY GLOB 0 UNDEF pow 869 0 0 NOTY GLOB 0 UNDEF sin 790 0 0 NOTY GLOB 0 UNDEF sqrt 758 0 O NOTY GLOB 0 UNDEF strcasecmp 848 0 0 NOTY GLOB 0 UNDEF strcpy 858 0 0 NOTY GLOB 0 UNDEF strlen 800 0 0 NOTY GLOB 0 UNDEF strncpy HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Adding Proprietary MOS Models To satisfy the above undefined symbols when you run simulation use libraries such as libc or libm However any unsatisfied symbols other than those shown in the example can cause an Unable to load problem MOS Models on PC Platforms To add proprietary MOS models on a PC platform follow the steps below These steps create a Dynamic Link Library 1 Copy the Customer CMI directory from the HSPICE or HSPICE RF release directory to a new location For example D xcopy d synopsys cmimodel project cmimodel e
217. L um 0 0 Channel length dependent drain induced 38 barrier lowering LGAMMA y1 2 0 0 This parameter is the body effect factor if vsb gt 6 7 VBO If you use the Poon Yau GAMMA expression LGAMMA is the junction depth in microns Simulation multiplies LGAMMA by SCALM LND um V 0 0 ND length sensitivity 2 3 6 7 8 LNO um 0 0 NO length sensitivity 2 3 6 7 8 LVT LVTO V um 0 0 VT dependence on the channel length 38 HSPICE MOSFET Models Manual X 2005 09 119 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 27 Threshold Voltage Parameters Continued Name Alias Units Default Description Level ND V 1 0 0 Drain subthreshold factor Typical value 1 2 3 6 1 V 7 8 NO 0 0 Gate subthreshold factor Typical value 1 2 3 6 7 8 NFS DFS om 2 1 0 0 Fast surface state density 1 2 3 NF DNF 6 7 8 NFS cm 0 0 Fast surface state density 40 NSS cm 0 0 Surface state density 40 NSUB cm3 1e15 Bulk surface doping If you do not specify 1 2 3 DNB NB NSUB simulation calculates it from GAMMA NWE m 0 0 Narrow width effect direct compensation of 6 7 VTO NWEscaled NWE SCALM NWM 0 0 Narrow width modifier 5 38 0 0 Narrow width modulation of GAMMA 6 7 PHI V 0 576 X Surface inversion potential If you do not 1 2 3 specify PHI HSPICE calculates it from NSUB 8 V 0 8 Built in potential 5 38 V 0 0 Surface potential 40 SCM 0 0 Short channel drain source voltage multiplier 5 38 0 0 S
218. L statement 3 BSIM models 331 examples NMOS model 432 MOSFETSs 13 VERSION parameter 332 models automatic selection 9 bulk charge effect 3 capacitance 7 depletion MOS devices 3 equations LEVEL 61 263 LEVEL 62 272 ion implanted devices 3 MOS compare 571 MOSFETS UFSOI 249 MOSFETs Berkeley BSIM3 SOI 463 junction 402 BSIM 324 equations 332 LEVEL 13 example 339 BSIM2 358 362 BSIM3 382 equations 390 602 Leff Weff 389 BSIM3 SOI DD 492 BSIM3 SOI FD 482 484 492 BSIM3v3 MOS 397 NQS 402 BSIM4 442 Cypress 195 197 Empirical 136 EPFL EKV 224 231 Frohman Bentchkowski equations 173 HP a Si TFT 204 206 Hspice junction 402 IDS LEVEL 5 144 equations 150 LEVEL 6 equations 159 168 example 164 166 LEVEL 6 and LEVEL 7 155 LEVEL 7 181 LEVEL 8 182 183 LEVEL 55 updates 246 levels 3 4 LEVELs 49 and 53 equations 431 modified BSIM LEVEL 28 347 354 MOS 143 Philips MOS9 214 quasi static equations 241 RPI a Si TFT 260 RPI Poli Si TFT 265 Schichman Hodges 128 130 SOSFETs LEVEL 27 188 UFSOI 249 scaling 12 silicon on sapphire 3 SOSFETs 3 specifying 13 modified BSIM LEVEL 28 equations 354 models 347 MOS diodes 9 model 143 MOS2 model 4 MOSS model 4 MOSFETs HSPICE MOSFET Models Manual X 2005 09 Berkeley 492 BEX factor 345 bulk transconductance 35 capacitance effective length and width 95 equations 78 96 Meyer model 64 models 65 scaling parameters 73 CAPOP 74 78 96 channel length modulation 10
219. LD scale scaled scaled eff LREF y LREF 4 LMLT XLREF 2 LD scaled scaled If DW is nonzero Won W cale4 WMLT DW M WREF s WREF j4 WMLT DW M Otherwise Wu Wscaled WMLT XW 2 WD scaled M WREF s7 WREF 4 WMLT XWREF carea 2 WD cated M Geometry and Bias of Model Parameters Most of the BSIM2 parameters include width and length sensitivity parameters You can also specify Synopsys proprietary WL product sensitivity parameters If P is a parameter then its associated width length and WL product sensitivity parameters are WP LP and PP The value of the P parameter adjusted for width length and WL product is pis pe wp LL Lp Ll Wey WREF Lig LREE 1 1 1 1 Ta fa Wey WREF 4 Loy LREF Berkeley SPICE does not use the WREF and LREF terms They are effectively infinite which is the default in the Level 39 MOSFET model HSPICE MOSFET Models Manual 365 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model 366 The following BSIM2 parameters do not have associated geometry sensitivity parameters TOX TEMP not used VDD VGG VBB DL DW BSIM2 parameters ending in 0 are valid at zero bias and they have associated bias sensitivities listed in the BSIM2 parameter table If PB PD and PG are the geometry adjusted vps Vas and Vgs sensitivity parameters and if they are associated with the PO geometry adjusted
220. Leff variation xul Physically xu1 should decrease as 1 Leff at long channels but when dealing with short channel devices you can turn off this variation Set UPDATE 2 to remove the 1 Leff factor in the xu1 equation UPDATE 2 introduces the present BEX usage as the 1 Leff removal ability UPDATE 3 provides the present BEX using the previous xu1 equation HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modified BSIM Model IDS and VGS Curves for PMOS and NMOS The netlists for the IDS and VGS curves for PMOS and NMOS are located in the following directory Sinstalldir demo hspice mos mll3iv sp The m113iv sp file contains examples of the following model parameter and curve descriptions Two Types of Model Parameter Formats Used VGS Curves GM Test GMB CVN7 5 370 PROCESS PC Filename M57R N channel Devices First Model Parameter Format PMOS Model m Wire Model for Poly and Metal Layers Second Model Parameter Format m N Diffusion Layer PMOS Model Wire Model for Poly and Metal Layers LEVEL 28 Modified BSIM Model This section lists the LEVEL 28 parameters and equations for the modified BSIM model LEVEL 28 Features The following are the significant features of the LEVEL 28 model Vertical field dependence of the carrier mobility Carrier velocity saturation Drain induced barrier lowering Depletion charge sharing by the source
221. Length parameter for UO stress effect gt 0 HSPICE MOSFET Models Manual X 2005 09 439 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 109 Supported HSPICE BSIM4 STI LOD Parameters Continued Parameter Unit Default Bin Description WLODKUO 0 0 No Width parameter for UO stress effect gt 0 KVTHO V m 0 0 No Threshold shift parameter for stress effect WKVTHO 0 0 No Width dependence of KVTHO PKVTHO 0 0 No Cross term dependence of KVTHO LLODVTH 0 0 No Length parameter for Vth stress effect gt 0 WLODVTH 0 0 No Width parameter for Vth stress effect 50 STK2 0 0 No K2 shift factor related to VThO change LODK2 m 1 0 No K2 shift modification factor for stress effect 50 STETAO 0 0 No ETAO shift factor related to VTHO change LODETAO M 1 0 No ETAO shift modification factor for stress effect 50 LMLT and WMLT in BSIMA You can use LMLT and WMLT to shrink the length and width in memory design The LMLT and WMLT parameters are unitless and are used to scale usually scale down MOSFET drawn length and width specified in BSIM4 MOSFET instance lines respectively This makes memory design and netlist creation quite convenient because most if not all memory circuits use the smallest feature sizes as the process capability improves even within the same generation of CMOS technology The shrunken device length and width will then be further offset by XL and XW respectively to the actual device size in li
222. M2 input decks shipped with Berkeley SPICE SE It appears that its use in SPICE 3E was as a printback debug aid Saturation charge sharing appears to be fixed at 60 40 S D in the BSIM2 capacitance model For the charge equations see Charge based Gate Capacitance Model CAPOP 39 on page 369 See also Modeling Guidelines Removing Mathematical Anomalies on page 373 HSPICE MOSFET Models Manual 367 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model 368 You can choose other CAPOP values CAPOP 13 recommended selects the BSIM1 based charge conserving capacitance model for the MOSFET LEVEL 13 BSIM1 or LEVEL 28 modified BSIM1 device models This option is the default selection if SPICE3 0 If you use this capacitance model you can use the XPART or XQC model parameters to adjust charge sharing See LEVEL 13 BSIM Model on page 324 for more information Unused Parameters The Level 39 MOSFET model does not use the DELL S D diode length reduction and WDF default device width SPICE model parameters SPICE 3E does not use the DELL function You can specify a default width in the Level 39 MOSFET model on the OPTION line as DEFW which defaults to 100 MODEL VERSION Changes to BSIM2 Models The Level 39 MOSFET model provides a VERSION parameter to the MODEL statement which lets you move LEVEL 13 BSIM and LEVEL 39 BSIM2 models between device model versions Use the VERSION parameter in a LEVEL 13 MODEL st
223. N 250 P Mobility degradation factor 3 Low field bulk mobility 5 Critical field for mobility 2 degradation UCRIT This parameter is the limit where UO surface mobility begins to decrease as specified in the empirical relation MOB 6 UEXP gt 0 8 Critical field for the mobility degradation UEXP operates as a switch MOB 6 UEXPz0 Critical field for mobility degradation Typical value is 0 01 V Effective mobility at the 6 7 specified analysis temperature Critical field exponent in the 2 8 empirical formula that characterizes the surface mobility degradation Typical value in MOSFET Level 8 with MOB 6 is 0 01 V1 Implant channel mobility 5 38 For depletion model only Implant channel mobility 38 saturation factor Low field bulk 2 mobility Simulation calculates 7 this parameter from the KP value that you specify O CQ Oo oF HSPICE MOSFET Models Manual X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 28 Mobility Parameters Continued Name Alias Units Default Description Level UTRA 0 0 Transverse field coefficient 2 8 mobility Traditional SPICE does not use UTRA HSPICE can use UTRA but simulation issues a warning because UTRA can hinder convergence VFRC A s cm2 V 0 0 Field reduction coefficient 38 variation due to the drain bias VST cm s 0 0 Saturation velocity 5 38 WBFRC 10 s cm V 0 0 BFRC sensitivity to the
224. NG qE T 5 145x1 21 1 n 5 x 10 3001s P 556598 2k T 4 ET 2 116 702x 10 7T T 1108 8 If you do not specify GAMMA simulation calculates it using 24 GAMMA 3 4 sieh C OX 9 If you do not specify GAMMA simulation calculates it using C Ox 10 If you do not specify V simulation calculates it using 2 aX Vis E oet 11 The BSIM3 model can calculate V4 in any of three ways Using K1 and K2 values that you specify e Using GAMMA1 GAMMA2 VBM and VBX values that you enter in the MODEL statement e Using NPEAK NSUB XT and VBM values that you specify You can enter the UO model parameter in meters or centimeters Simulation converts UO to m Vsec as follows if UO is greater than 1 it is multiplied by 1e 4 You must enter the NSUB parameter in cm3 units Specify a negative value of VTHO for the p channel in the MODEL statement The PSCBE1 and PSCBE2 model parameters determine the impact ionization current which contributes to the bulk current HSPICE MOSFET Models Manual 427 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Parameter Range Limits Simulation reports either a warning or a fatal error if BSIM3v3 parameters fall outside predefined ranges These range limitations prevent or at least warn of potential numerical problems Level 53 follows exactly the BSIM3v3 range limit reporting scheme Level 49 deviates from
225. O cm 1018 Halo doping density LRSCE m 0 0 0 1x10 Characteristic length for reverse short channel effect 0 for no RSCE LLDD m 0 0 0 05 0 2 x10 LDD LDS region length 0 for no LDD NLDD cm 5 0619 1x1 19 LDD LDS doping density 51619 LDD LDS treated as D S extensions DL m 0 0 0 05 0 15 x10 Channel length reduction DW m 0 0 0 1 0 5 x10 8 Channel width reduction Table 56 MOSFET Level 58 Electrical Parameters Parameter Unit Default Typical Value Description NQFF cm 0 0 1010 Front oxide fixed charge normalized NQFB cm 0 0 101 Back oxide fixed charge normalized NQFSW cm2 0 0 1912 Effective sidewall fixed charge 0 for no narrow width effect QM 0 0 0 5 Energy quantization parameter 0 for no quantization UO cm Vi s 7 0e2 200 700 nMOS Low field mobility 70 400 pMOS HSPICE MOSFET Models Manual X 2005 09 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 58 University of Florida SOI Table 56 MOSFET Level 58 Electrical Parameters Continued Parameter Unit Default Typical Value Description THETA VSAT ALPHA BETA BGIDL NTR JRO LDIFF SEFF CGFDO cm V cm s cm V cm V cm cm s F m 7 0e2 1 0e 6 0 0 0 0 0 0 0 0 1 0e 10 2 0 1 0e 7 1 0e5 0 0 0 1 3 x10 6 0 5 1 x10 2 45x108 1 92x10 4 8 x109 1014 1015 1011 10 1
226. OSFET Models Selecting Models Table 1 MOSFET Model Descriptions Continued All Platforms All Platforms Level MOSFET Model Description including PC except PC 29 not used 30 VTI X 31 Motorola X 32 AMD X 33 National Semiconductor X 34 EPFL not used X 35 Siemens X 36 Sharp X 37 TI X 38 IDS Cypress depletion model X 39 BSIM2 X 41 TI Analog X 46 SGS Thomson MOS LEVEL 3 X 47 BSIM3 Version 2 0 X 49 BSIM3 Version 3 Enhanced X 50 Philips MOS9 X 53 BSIM3 Version 3 Berkeley X 54 UC Berkeley BSIM4 Model X 55 EPFL EKV Model Ver 2 6 R 11 X 57 UC Berkeley BSIM3 SOI MOSFET Model X Ver 2 0 1 HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models Selecting Models Table 1 MOSFET Model Descriptions Continued All Platforms All Platforms Level MOSFET Model Description including PC except PC 58 University of Florida SOI Model Ver 4 5 X Beta 98 4 59 UC Berkeley BSIM3 501 FD Model X 61 RPI a Si TFT Model X 62 RPI Poli Si TFT Model X 63 Philips MOS11 Model X 64 STARC HiSIM Model X not officially released equations are proprietary documentation not provided requires a license and equations are proprietary documentation not provided Selecting MOSFET Capacitors CAPOP is the MOSFET capacitance model parameter This parameter determines which capacitor models to use when modeling the MO
227. PO PHP vio 3 in 5 mu CO CBD t CBD 14MJ 400u Ar FBO JI CBS t CBS 1 MJ 400w At PO 1 CH CJ 14 MJ 400u Ar PED 1 CJSW t CJSW 1 t MISW 400u At PHPO 1 J PHP TLEVC 1 PB t PB PTA bAt PHP t PHP PTP PAt CBD t CBD 1 CTA At CBS t CBS 1 CTA At CJ CJ 1 CTA At CJSW CJSW 1 CTP At HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models Temperature Parameters and Equations TLEVC 2 PB t PB PTA PAt PHP t PHP PTP PAt PB Jm CBD t CBD sor PB i CBS t CBS CO PB CJ t CJ Cra PHP i CJSW t CJSW CA TLEVC 3 PB t PB dpbdt At PHP t PHP dphpdt At At CBD t CBD 1 0 5 Pdpbdt pat At CBS t CBS 1 0 5 bdpbdt pat PB At CJ t CJ 1 0 5 bdpbdt pat CISW t CISW 1 0 5 Pdphpdt por PHP If TLEVC 3 and TLEV 0 or 1 then egnom 3 vt tnom 1 16 egnom 2 Ens OM PB dpbdt tnom 110 tnom egnom 3 vt tnom 1 16 egnom mom PHP Hides nom og tnom HSPICE MOSFET Models Manual 103 X 2005 09 2 Technical Summary of MOSFET Models Temperature Parameters and Equations 104 TLEV 2 egnom 3 vt tnom EG egnom 2 uu Cc MIO ox PB tnom GAP dpbdt tnom egnom 3 vt tnom EG egnom 2 MOM
228. Parameter for Iys and Iya 0 03 PMOS V DLCIG LINT Yes Source drain overlap length for Igs and Igd NIGC 1 0 Yes Parameter for Ics lcd Igg and Iya POXEDGE 1 0 Yes Factor for the gate oxide thickness in the source drain overlap regions PIGCD 1 0 Yes Vgsdependence of Iggg and lyca NTOX 1 0 Yes Exponent for the gate oxide ratio TOXREF 3 0e 9m No Nominal gate oxide thickness for the gate dielectric tunneling current model only HSPICE MOSFET Models Manual X 2005 09 453 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 120 Charge Capacitance Model Parameters MOS 54 Parameter Default Bin Description nab le XPART 0 0 No Charge partition parameter CGSO calculated No Non LDD region source gate overlap capacitance per unit F m channel width CGDO calculated No Non LDD region drain gate overlap capacitance per unit F m channel width CGBO 0 0 F m No Gate bulk overlap capacitance per unit channel length CGSL 0 0F m Yes Overlap capacitance between the gate and the lightly doped source region CGDL 0 0F m Yes Overlap capacitance between gate and lightly doped source region CKAPPAS 0 6V Yes Coefficient of the bias dependent overlap capacitance for the source side CKAPPAD CKAPPAS Yes Coefficient of bias dependent overlap capacitance for the drain side CF calculated Yes Fringing field capacitance F m CLC 1 0e 7m Yes Constant term for the short channel model CLE 0 6 Yes Exponential term for the short channel
229. QC is specified If you specify XQC but you do not specify XPART then e XQC O 5 0 100 e XQC 0 4 40 60 e XQC 0 5 5 50 50 e XQC 1 0 100 e XQC any other value less than 1 40 60 e XQC gt 1 5 0 100 The only difference is the treatment of the 0 parameter value After you specify XPART XQC the gate capacitance ramps from 50 50 at Vds 0 volt linear region to the value with Vds sweep in the saturation region in XPART XQC Ramping the charge sharing coefficient ensures smooth gate capacitance characteristics Overlap Capacitance Equations The overlap capacitors are common to all models You can input them explicitly or the program can calculate them Either way these overlap capacitors must be added into the respective voltage variable capacitors before integration and before the DC operating point reports the combined parallel capacitance HSPICE MOSFET Models Manual 77 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models 78 Gate to Bulk Overlap Capacitance If you specify CGBO then CGBOeff M Leff CGBOscaled Otherwise CGBOeff 2 WDscaled Leff COXscaled M Gate to Source Overlap Capacitance If you specify CGSO then CGSOeff Weff CGSOscaled Otherwise CGSOeff Weff LDscaled METOscaled COXscaled Gate to Drain Overlap Capacitance If you specify CGDO then CGDOeff Weff CGDOscaled Otherwise CGDOeff Weff LDscaled METOscaled COXscaled Simulatio
230. S gate capacitan capacitan Ce that is the gate to drain capacitance the gate to source ce and the gate to bulk capacitance CAPOP selects a specific version of the Meyer and charge conservation model Some capacitor models are tied to specific DC models they are stated as such Others are for general use by any DC model Parameter Description CAPOP 0 SPICE original Meyer gate capacitance model general CAPOP 1 Modified Meyer gate capacitance model general CAPOP 2 Modified Meyer gate capacitance model with parameters general default HSPICE MOSFET Models Manual 7 X 2005 09 1 Overview of MOSFET Models Selecting Models Parameter Description CAPOP 3 Modified Meyer gate capacitance model with parameters and Simpson integration general CAPOP 4 Charge conservation capacitance model analytic LEVELs 2 3 6 7 13 28 and 39 only CAPOP 5 No capacitor model CAPOP 6 AMI capacitor model LEVEL 5 CAPOP 9 Charge conservation model LEVEL 3 CAPOP 11 Ward Dutton model specialized LEVEL 2 CAPOP 12 Ward Dutton model specialized LEVEL 3 CAPOP 13 Generic BSIM Charge Conserving Gate Capacitance model Default for Levels 13 28 and 39 CAPOP 39 BSIM 2 Charge Conserving Gate Capacitance Model LEVEL 39 CAPOP 4 selects the recommended charge conserving model from among CAPOP 11 12 or 13 for the specified DC model Table 2 CAPOP 4 Selections MOS Level Default CAPOP CAPOP 4 se
231. S source and drain junctions TPB 0 0V K No Temperature coefficient of PB TPBSW 0 0V K No Temperature coefficient of PBSW TPBSWG 0 0V K No Temperature coefficient of PBSWG TCJ 0 0K No Temperature coefficient of CJ 460 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 125 Temperature Dependence Parameters Level 54 Continued Parameter Default Binnable Description TCJSW 0 0K No Temperature coefficient of CJSW TCJSWG 0 0K No Temperature coefficient of CJSWG TRD 0 0 1K 1 Temperature coefficient for the drain diffusion and the Rd contact resistances TRS 0 0 1 K Temperature coefficient for the source diffusion and the Rs contact resistances Table 126 dW and dL Parameters MOSFET Level 54 Parameter Default Binnable Description WL 0 0n24N No Coefficient of the length dependence of the width offset WLN 1 0 No Power of the length dependence of the width offset WW 0 0m WWN No Coefficient of the width dependence of the width offset WWN 1 0 No Power of the width dependence of the width offset WWL 0 0 mWWN WLN No Coefficient of the length and width cross term dependence for the width offset LL 0 0mLLN No Coefficient of the length dependence for the length offset LLN 1 0 No Power of the length dependence for the length offset LW 0 0mEXN No Coefficient of the width dependence for the length offset LWN 1 0 No Power of the width dependence length offset L
232. SW V PB Sidewall junction potential 39 PSI 1e 20 Temperature exponential part 40 RD ohm 1 0K External drain resistance 40 RS ohm 1 0K External source resistance 40 RSH ohm sq 0 Source drain sheet resistance 39 112 HSPICE MOSFET Models Manual X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 25 Basic MOSFET Model Parameters Continued Name Alias X Units Default Description Level SNVB 1 V cm3 0 0 Slope of the doping concentration versus 8 vsb element parameter Multiplied by 1e6 SPICE3 0 Selects SPICE3 model compatibility For 39 accurate SPICE3 BSIM2 set SPICE3 1 TAU S 10n Relaxation time constant 40 TCV V C 0 Zero bias threshold voltage temperature 39 coefficient The sign of TCV adjusts automatically for NMOS and PMOS to decrease the magnitude of the threshold with rising temperature TOX m 1e 7 Gate oxide thickness 1 2 3 A 0 0 Oxide thickness 5 38 m 7 0e 8 Oxide thickness 27 TRD 1 K 0 0 Temperature coefficient for the Rd drain 54 diffusion and contact resistances TREF 1 5 Temperature gradient of UO 40 TRS 1 K 0 0 Temperature coefficient for the Rs source 54 diffusion and contact resistances TUH 1 5 Implant channel mobility temperature 5 exponent depletion model only UO cm2 V s Carrier mobility 1 40 Default for LEVEL 40 is 1 0 UO UB cm2 V s 600 N Low field bulk mobility Simulation calculates 2 UBO 250 P this parameter from the KP value that you
233. Sinstalldir demo hspice mos mcap2_a sp Figure 26 Charge Conservation Test Circuit VDD 2 5v 10 VBB 1V CAPOP 5 No Gate Capacitance If CAPOP 5 for no capacitors then simulation does not calculate gate capacitance HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models CAPOP 6 AMI Gate Capacitance Model Define vos E eee vth vfb 2 ox COM OX 2 10 eU CER The following equations calculate the cgs gate capacitance in the different regions 0 5 vth vfb gt vgs cgs 0 0 5 vth vfb vgs lt vth For vgst vds _ 4 cox vgst 3 vth vfb For vgst gt vds cgs bores genes SOCEM 89 41573 pth vfb vgs vth For vgst vds posui orm EET For vgst vds 3 vgst 2 vds 2 cgs arg cox arg vgst E i 2 2 vgst vds 3 The following equations calculate the cgd gate capacitance in the different regions vgs lt vth cgd 0 HSPICE MOSFET Models Manual X 2005 09 93 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models 94 vgs gt vth and vgst lt vds cgd 0 vgs gt vth and vgst gt vds ced arg COX vgst vds 2 vgst vds The following equation combines the cgb gate capacitance with the calculation of both oxide capacitance and depletion capacitance arg 3 vgst vds cgb m E Simulation calculates the o
234. Synopsys Device Model Enhancements In the following expressions model parameters are in all upper case Roman Simulation assumes that you have already adjusted all model parameters without a trailing 0 for both geometry and bias as appropriate Temperature Effects LEVEL 39 enforces TLEV 1 You cannot currently use any other TLEV value The following equation adjusts the threshold voltage for LEVEL 39 TLEV 1 V4 G Vp T K1 9 T V K2 0 T V ETA V The following equations calculate values used in the preceding equation V T V T K1 Jo T K2 T V 1 V TCV T T o nom 3 In the preceding equations the nominal temperature zero bias threshold voltage is V to VFB PHI K1 JPHI K2 PHI Vy K1 JPHI K2 PHI Simulation calculates o T according to the specified TLEVC value The following equation adjusts the mobility T T T i h p A where UU Na i nnam T oni d Co Wop Lig UIS adjusts the velocity saturation U1S T uis J nom This model also includes all of the usual Synopsys model adjustments to capacitances parasitics diodes and resistors HSPICE MOSFET Models Manual 371 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model 372 Alternate Gate Capacitance Model Select CAPOP 13 for the charge conserving capacitance model widely used with LEVEL 13 BSIM1 and LEVEL 28 improved BSIM1 See LEVEL 13 BSIM Model on page 324 for
235. T model The Synopsys implementation of this model is based on Berkeley SPICE 3E2 To provide input to the Level 39 device model assign the model parameters as for other device models You can use a tabular model entry without model parameter names in BSIM1 but notin BSIM2 LEVEL 39 Model Parameters MOSFET Level 39 uses the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters It also uses the parameters described in this section which apply only to MOSFET Level 39 This section lists the BSIM2 parameters their units their defaults if any in the Level 39 MOSFET model and their descriptions Table 89 lists 47 BSIM2 specific parameters The Synopsys model does not use three of the parameters TEMP DELL and DFW The width and length sensitivity parameters are associated with the remaining parameters except the first six TOX VDD VGG VBB DL and DW So the total parameter count is 120 Unlike Berkeley SPICE the Synopsys Level 39 MOSFET model has L and W 358 HSPICE MOSFET Models Manual X 2005 09 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model sensitivity for MUO This count does not include the generic MOS parameters or the WL product sensitivity parameters which are Synopsys enhancements Table 89 BSIM2 Model Parameters Name Alias Units Default Description TOX m 0 02 Gate oxide thickness assumes that TOX gt 1 is i
236. T Models Levels 50 to 64 Level 50 Philips MOS9 Model 216 Table 45 MOSFET Level 50 Model Parameters Continued Name Unit Default N Default P Description THE1R vy 190 0e 3 190 0e 3 Gate induced mobility reduction coefficient STTHE1R V 1 K 0 0 0 0 Temperature dependence coefficient THE1R SLTHE1R Vim 140 0e 9 70 0e 9 Length dependence coefficient of THE1R STLTHE1 V Im K 0 0 0 0 Temperature dependence of the length dependence for THE1R SWTHE1 Vim 58 0e 9 80 0e 9 Width dependence coefficient of THE1R THE2R y 1 2 12 0e 3 165 0e 3 Back bias induced mobility reduction coefficient STTHE2R y 1 2 K 0 0 0 0 Temperature dependence coefficient THE2R SLTHE2R vim 33 0e 9 75 0e 9 Length dependence coefficient of THE2H STLTHE2 y 1 2 mK 0 0 0 0 Temperature dependence of the length dependence for THE2R SWTHE2 V 1 2m 30 0e 9 20 0e 9 Width dependence coefficient of THE2R THESR vi 145 0e 3 27 0e 3 Lateral field induced mobility reduction coefficient STTHESR V 1 K 660 0e 6 0 0 Temperature dependence coefficient of THE3R SLTHE3R Vim 185 0e 9 27 0e 9 Length dependence coefficient of THE3R HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 50 Philips MOS9 Model Table 45 MOSFET Level 50 Model Parameters Continued Name Unit Default N Default P Description STLTHE3 V Im K 620 0e 12 0 0 Temperature dependence of the length dependence for THE3R SWTHE3 Vim 20 0e 9 11
237. T Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS GATENOISE Inclusion exclusion flag of the induced 0 0 gate thermal noise NT Coefficient for the thermal noise atthe J 1 656E 20 1 656E 20 reference temperature PONFA Coefficient for the geometry V 1m 4 8 3823E 22 1 90E 22 independent NFA part PLNFA Coefficient for the length dependent V im 4 0 0 part of NFA PWNFA Coefficient for the width dependent V im 4 0 0 part of NFA PLWNFA Coefficient for the length times width V im 4 O 0 dependent part of NFA PONFB Coefficient for the geometry V 1m 2 2 514E 7 5 043E 6 independent NFB part PLNFB Coefficient for the length dependent V im 2 0 0 part of NFB PWNFB Coefficient for the width dependent V im 2 O 0 part of NFB PLWNFB Coefficient for the length times width V im 2 0 0 dependent part of NFB PONFC Coefficient for the geometry V 1 0 3 627E 10 independent NFC part PLNFC Coefficient for the length dependent V 1 0 0 part of NFC PWNFC Coefficient for the width dependent V 1 0 0 part of NFC 302 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS PLWNFC Coefficient for the length times width V 1 0 0 dependent
238. The L element statement overrides the DEFL default The NRD element statement overrides the DEFNRD default The NRS element statement overrides the DEFNRS default The PD element statement overrides the DEFPD default The PS element statement overrides the DEFPS default The W element statement overrides the DEFW default Scaling Units The SCALE and SCALM options control the units SCALE scales element statement parameters SCALM scales model statement parameters It also affects the MOSFET gate capacitance and diode model parameters HSPICE MOSFET Models Manual 11 X 2005 09 1 Overview of MOSFET Models Selecting Models In this chapter scaling applies only to parameters that you specify as scaled If you specify SCALM as a parameter in a MODEL statement it overrides the SCALM option In this way you can use models with different SCALM values in the same simulation MOSFET parameter scaling follows the same rules as for other model parameters for example Table 3 Model Parameter Scaling Parameter Units Parameter Value meter multiplied by SCALM meter multiplied by SCALM meter divided by SCALM meter divided by SCALM To override global model size scaling for individual MOSFET diode and BJT models that use the OPTION SCALM lt val gt statement include SCALM lt val gt in the MODEL statement OPTION SCALM lt val gt applies globally for JFETs resistors transmission lines and all mod
239. Tp PB 0 40 Bottom junction grading coefficient PS 0 40 Sidewall junction grading coefficient PG 0 40 Gate edge junction grading coefficient TH3MOD 1 Switch that activates THES clipping If THSMOD 1 default effective THES can be slightly negative and clipping does not occur f TH3MOD 0 this model clips the effective THES to more than zero Using the Philips MOS9 Model 1 SetLevel 50 to identify the model as the Philips MOS Model 9 2 The default room temperature is 25 9C in Synopsys circuit simulators but is 27 C in most other simulators When comparing to other simulators set the simulation temperature to 27 use TEMP 27 or OPTION TNOM 27 3 The model parameter set must include the TR model reference temperature which corresponds to TREF in other model levels The default for TR is 21 0 C to match the Philips simulator 4 This model has its own charge based capacitance model Level 50 ignores the CAPOP parameter which selects different capacitance models 5 This model uses analytical derivatives for the conductances This model ignores the DERIV parameter which selects the finite difference method HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 50 Philips MOS9 Model DTEMP increases the temperature of individual elements relative to the circuit temperature Set DTEMP on the element line Defaults are nonzero so use the MODEL statement to set
240. UO stress effect gt 0 KVTHO V m 0 0 No Threshold shift parameter for stress effect WKVTHO 0 0 No Width dependence of KVTHO PKVTHO 0 0 No Cross term dependence of KVTHO LLODVTH 0 0 No Length parameter for Vth stress effect gt 0 WLODVTH 0 0 No Width parameter for Vth stress effect 50 STK2 0 0 No K2 shift factor related to VThO change LODK2 m 1 0 No K2 shift modification factor for stress effect gt 0 STETAO 0 0 No ETAO shift factor related to VTHO change LODETAO M 1 0 No ETAO shift modification factor for stress effect gt 0 406 Parameter Differences Some parameter names differ between the Synopsys model and the Berkeley junction models The Synopsys models ACM 0 3 do not recognize the following BSIM3v3 parameters NJ ignored use N instead CJSWG ignored use CUGATE instead MJSWG ignored HSPICE has no equivalent parameter and simulation sets the gate sidewall grading coefficient MJSW PBSW ignored use PHP instead PBSWG ignored HSPICE has no equivalent parameter and simulation sets the gate sidewall contact potential PHP The Berkeley model ACM 10 11 12 13 does not recognize the following parameters CJGATE ignored use CUSWG instead PHP ignored use PBSW instead HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Noise Model The HSPICE NLEV parameter overrides the BSIM3v3 NOIMOD parameter Specifying NLEV invokes the HSPIC
241. UZ zmus z3ms zx2mz and zx2ms BEX Usage t NBEX MUZ t MUZ tno t NBEX zmus t zmus tno t NBEX zx3ms t ims LL tno t NBEX zx2mz t zx2mz cx tno t NBEX zx2ms t zx2ms E tno This is equivalent to multiplying the final mobility by the factor EL tno HSPICE MOSFET Models Manual 345 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 346 UPDATE Parameter The UPDATE parameter selects between variations of the BSIM equations UPDATE O default is consistent with UCB SPICE3 UPDATE 3 also is consistent with UCB SPICE3 and BEX usage Parameter Description UPDATE 0 UCB compatible previous BEX usage UPDATE 1 Special X2E equation previous BEX usage UPDATE 2 Remove 1 Leff in U1 equation present BEX usage UPDATE 3 UCB compatible present BEX usage Explanations The normal X2E equation is xeta zeta zx2e vsb zx3e vds VDDM The special X2E equation for UPDATE 1 only is xeta zera zx2e zphi vsb zx3e vds VDDM The special X2E equation was developed to match a parameter extraction program If you use a parameter extraction program check the equations carefully The original U1 equation divides by Leff in microns zul zx2ul vsb zx3ul vds VDDM Leff This is one of the few places where Leff explicitly enters into the BSIM equations usually the L adjustment model parameters such as LU1 handles the
242. Using Customer CMI Models Directory Description link test Release mos1 mos2 mos3 Main Customer CMI routines Model testing example Shared library directory Model directories Running Simulations Using Customer CMI Models To specify Customer CMI models use the eve model parameter Levels used in the example models are the same as in Berkeley Spice 3 LEVEL 1 mos1 LEVEL 1 MOS model LEVEL 2 mos2 LEVEL 2 MOS model LEVEL 3 mos3 LEVEL 3 MOS model LEVEL 4 b1 BSIM model LEVEL 5 b2 BSIM2 model LEVEL 8 b3 BSIM3v3 model LEVEL 9 JFET JFET amp MESFET model To simulate a Customer CMI model add the following line in the input netlist OPTION cmiflag The LEVEL 8 example code located in the b3 directory and the Level 49 MOSFET device model are both based on BSIM3 version 3 However the speed of LEVEL 8 is sometimes 20 percent slower than LEVEL 49 in the Synopsys MOSFET models This occurs because LEVEL 49 is carefully implemented to ensure high accuracy and performance In contrast LEVEL 8 in the example code is only an example of using the Customer CMI interface HSPICE MOSFET Models Manual 521 X 2005 09 8 Customer Common Model Interface Adding Proprietary MOS Models Therefore the slower performance of the example code compared to the MOSFET Level 49 model is expected The Level 9 example code is located in the JFET directory and is based on the JF
243. VEL 28 353 temperature compensation 396 example 142 threshold voltage BSIM LEVEL 13 334 equations 58 LEVEL 1 129 LEVEL 13 334 LEVEL 2 131 LEVEL 28 355 LEVEL 3 137 LEVEL 38 199 LEVEL 47 390 LEVEL 5 151 LEVEL 6 159 LEVEL 8 184 temperature 105 parameters 58 LEVEL 1 108 LEVEL 2 122 LEVEL 5 117 TI model 6 topology 552 transcapacitance 66 transconductance 35 transient analysis 36 607 Index transistors field effect 28 isoplanar silicon gate 30 process parameters LEVEL 13 325 329 LEVEL 28 348 LEVEL 49 414 TT model parameter 41 U Universal Field mobility reduction 174 University of California SOI model 463 University of Florida SOI model 249 UPDATE parameter 346 LEVEL 13 346 LEVEL 6 7 156 V VERSION parameter 332 608 VNDS model parameter 40 voltage 58 VTI model 6 W Wang s equation 178 WDAC parameter 96 WDEL model parameter 43 WL option 11 WMLT model parameter 43 WRD model parameter 42 WRS model parameter 42 X XJ model parameter 43 XPART CAPOP model parameter 77 XQC CAPOP model parameter 77 XW model parameter 43 HSPICE MOSFET Models Manual X 2005 09
244. VEL 57 is the UC Berkeley BSIM3 SOI Partially Depleted PD model LEVEL 58 is the University of Florida SOI model LEVEL 59 is the UC Berkeley BSIM3 SOI Fully Depleted FD model LEVEL 60 is the UC Berkeley BSIM3 SOI Dynamically Depleted DD model LEVEL 61 is the Rensselaer Polytechnic Institute RPI a Si TFT model e LEVEL 62 is the Rensselaer Polytechnic Institute RPI poly silicon Thin Film Transistor Poli Si TFT model LEVEL 63 is the Philips MOS11 model LEVEL 64 is the Hiroshima STARC IGFET HiSIM model Selects MOS S D parasitics ACM 0 is 39 SPICE style Use ACM 2 or 3 for LDD Impact ionization coefficient This parameter 39 includes geometry sensitivity parameters Choose between BSIM2 A10 gt 0 and HSPICE ALPHA gt 0 impact ionization modeling Do not use both 109 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 25 Basic MOSFET Model Parameters Continued Name Alias Units Default Description Level CAPOP MOS gate cap model selector CAPOP 39 4 13 for BSIM2 or CAPOP 13 for BSIM1 39 CAPOP 4 is the same as CAPOP 13 If SPICE3 0 default CAPOP 13 If SPICE3 1 default CAPOP 39 CGBO F m Gate to bulk overlap capacitance 39 If you specify WD and TOX but you do not specify CGBO then simulation calculates CGBO CGDO F 1 0p TFT gate to drain overlap capacitance 40 F m Gate to drain overlap capacitance 39 If you specify TOX and you specify eith
245. Vgs lt Vgs Vth W na Ij KP fl 1 LAMBDA vg vg v vas Lg S g 2 5 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 1 IDS Schichman Hodges Model Saturation Region VOS 2 VIS y _ KP W a A 1 LAMBDA v ag Vin eff Effective Channel Length and Width The Level 1 model calculates the effective channel length and width from the drawn length and width L Lialsd LMLT X edel 2 PULD scaled DEL caled eff T W sr M Ws cated WMLT KW salad ni Threshold Voltage Vip 2 PWD caled vsb20 v GAMMA PHI v j vsbz0 vp Vp GAMMA Pun os s PHI The preceding equations define the built in voltage vpi as Vp vg PHI Or vj VTO GAMMA PPHI Note See Common Threshold Voltage Equations on page 58 for calculation of VTO GAMMA and PHI if you do not specify them Saturation Voltage vs The saturation voltage for the LEVEL 1 model is due to the channel pinch off at the drain side The following equation computes this voltage Vsat Yes Yth The LEVEL 1 model does not include the carrier velocity saturation effect HSPICE MOSFET Models Manual 129 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 2 IDS Grove Frohman Model LEVEL 2 IDS Grove Frohman Model This section describes the parameters and equations for the LEVEL 2 IDS Grove Frohman model LEVEL 2 Model Parameters MOSFET Level 2
246. W and removed IBC and ibc cd Level 58 University of Florida SOI UFSOI includes non fully depleted NFD and fully depleted FD SOI models a dynamic mode must not operate between NFD and FD that separately describe two main types of SOI devices The UFSOI version 4 5F model is Level 58 in the Synopsys MOSFET models This model is described in the UFSOI Model User s Manual at http www soi tec ufl edu Some processes use an external contact to the body of the device The Synopsys MOSFET model supports only a 4 terminal device which includes drain front gate source and back gate or substrate The additional body contact is currently not supported so it floats The effects of parasitic diodes in SOI are different from those in the bulk MOSFET The SOI model does not include the MOSFET junction model ACM developed for bulk MOSFETs The general syntax for MOSFET Level 58 in a netlist is Mxxx nd ngf ns ngb mname lt L val gt lt W val gt M val t lt AD val gt lt AS val gt lt PD val gt PS val lt NRD val gt t lt NRS val gt lt NRB val gt RTH val CTH val off t IC Vas Vgfs VGbs In the preceding syntax angle brackets denote optional parameters The arguments are identical to those for the BSIM3 SOI model but the thermal resistance and capacitance have different names Table 49 Thermal Resistance and Capacitance Names Name Description RTH Thermal resistanc
247. WL 0 0 mLWN LLN No Coefficient of the length and width cross term HSPICE MOSFET Models Manual X 2005 09 dependence for the length offset 461 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 126 dW and dL Parameters MOSFET Level 54 Continued Parameter Default Binnable Description LLC LL No Coefficient of the length dependence for the CV channel length offset LWC LW No Coefficient of the width dependence for the CV channel length offset LWLC LWL No Coefficient of the length and width cross term dependence for the CV channel length offset WLC WL No Coefficient of the length dependence for the CV channel width offset WWC WW No Coefficient of the width dependence for the CV channel width offset WWLC WWL No Coefficient of the length and width cross term dependence for the CV channel width offset Table 127 Range Parameters for Model Application MOSFET Level 54 Description Parameter Default Binnable LMIN 0 0u No LMAX 1 0m No WMIN 0 0m No WMAX 1 0m No Minimum channel length Maximum channel length Minimum channel width Maximum channel width Level 54 BSIM4 Template Output List For a list of output template parameters in the MOSFET models and which parameters this model supports see Table 4 on page 14 462 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model Level 57 UC Berkeley BSIM3 SOI Mod
248. X 2005 09 449 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 115 Basic Model Parameters MOSFET Level 54 Continued Parameter Default Binnable Description CDSC 2 4e AF m Yes Coupling capacitance between the source drain and the channel CDSCB 0 0F Vm Yes Body bias sensitivity of CDSC CDSCD 0 0 F Vm Yes Drain bias sensitivity of DCSC PCLM 1 3 Yes Channel length modulation parameter PDIBLC1 0 39 Yes Parameter for the DIBL effect on Rout PDIBLC2 0 0086 Yes Parameter for the DIBL effect on Rout PDIBLCB 0 0 V Yes Body bias coefficient of the DIBL effect on Rout DROUT 0 56 Yes Channel length dependence of the DIBL effect on Rout PSCBE1 4 24e8V m Yes First substrate current induced body effect parameter PSCBE2 1 0e 5m V Yes Second substrate current induced body effect parameter PVAG 0 0 Yes Gate bias dependence of Early voltage DELTA S in 9 01V Yes Parameter for DC Vyseti equation FPROUT 0 0V m9 5 Yes Effect of the pocket implant on Rout degradation PDITS 0 0v Yes Impact of the drain induced Vj shift on Rout PDITSL 0 0m No Channel length dependence of the drain induced V4 shift for Rout PDITSD 0 0v Yes Vds dependence of the drain induced V shift for Rout 450 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 116 Parameters for Asymmetric and Bias Dependent Rds Model MOSFET Level 54 Parameter Default Binnable
249. X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 108 Element Parameter Range Limit MOSFET Levels 49 53 Continued Leff lt 5 0 x 10 Fatal Weff lt 1 0 x 107 Fatal LeffCV lt 5 0 x 103 Fatal WeffCV lt 1 0 x 10 Fatal For a list of output template parameters in the MOSFET models and which parameters this model supports see Table 4 on page 14 Level 49 53 Equations The effective channel length and width in all model equations are Leg Lows 2dL Wer W kaun 2dW Wer W arawn 2dW Wdrawn W WMULT XW Ldrawn L LMULT XL The unprimed We is bias dependent The primed quantity is bias independent dW dW dW V steft dW J6 Vita J dW W i Ww rem ae Bd in L WLN yo p jp bd ee QN o EE A o in pre we WN po we WN C V calculations replace dW with dW DWC Me wc Wwe WLN WWN WLN WWN L W L W C V also replaces dL with HSPICE MOSFET Models Manual 431 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models L L L VERTO zm i a i RET L W L W Note For details of BSIM3 Version 3 equations see the web site http www device eecs berkeley edu bsim3 get html MODEL CARDS NMOS Model This is an example of a NMOS model for the Level 49 MOSFET VTHO is positive
250. a from real devices is monotonic This kink is annoying to analog designers working with devices in the weak and medium 576 HSPICE MOSFET Models Manual X 2005 09 B Comparing MOS Models Behavior Follows Devices in All Circuit Conditions inversion region LEVEL 28 and 39 have solved this problem but they require additional parameters Three other important measures are Continuous Derivatives Levels 2 and 3 fail Levels 13 28 and 39 pass Positive GDS Levels 13 and 39 fail Levels 2 3 and 28 pass Monotonic GM IDS in weak inversion Levels 2 3 and 13 fail Levels 28 and 39 pass Behavior Follows Devices in All Circuit Conditions A model can be a very good fit to IDS data in the normal operating region but it still might not be useful for simulating some circuits The first criterion is that the model should have good temperature dependence The MOSFET models provide temperature dependence parameters for threshold voltage and mobility for all levels The LEVEL 13 28 and 39 models also include an FEX parameter that controls VDSAT variation with temperature The next most important criterion is that the model should include subthreshold current to provide accurate analog simulation Even for digital circuits this aids in convergence All MOSFET models include a subthreshold current Impact ionization causes a drain to bulk current that has a strong effect on cascode circuits MOSFET models provide ALPHA and VCR paramet
251. acitance Model Parameters Continued Name Alias Units Default Description CJSW CJP F m 0 Zero bias sidewall bulk junction capacitance CJSWscaled CJSW SCALM Default 0 CJ CDB CSB CJA F m2 579 11 Zero bias bulk junction capacitance uF m CJscaled CJ SCALM2 for ACM 1 the unit is F m e CJscaled CJ SCALM Default for the ASPEC 0 option is _ esi q NSUB ehe 2 PB CJGATE F m CSJW Zero bias gate edge sidewall bulk junction capacitance ACM 3 only CJGATEscaled CJGATE SCALM Default CUSW for Hspice releases later than H9007D Default 0 for HSPICE releases H9007D and earlier or if you do not specify CJSW FC 0 5 Forward bias depletion capacitance coefficient not used MJ EXA EXJ EXS 0 5 Bulk junction grading coefficient EXD MJSW EXP 0 33 Bulk sidewall junction grading coefficient NSUB DNB NB i cm 1 0e15 Substrate doping PB PHA PHS V 0 8 Bulk junction contact potential PHD PHP V PB Bulk sidewall junction contact potential TT S 0 Transit time HSPICE MOSFET Models Manual X 2005 09 41 2 Technical Summary of MOSFET Models MOSFET Diode Models Table 9 Drain and Source Resistance Model Parameters Name Alias Units Default Description RD ohm sq 0 0 Drain ohmic resistance This parameter is usually the sheet resistance of a lightly doped region for ACM31 RDC ohm 0 0 Additional drain resistance due to contact resistance LRD ohm m 0 Drain resistance l
252. after initialization Syntax int CMI ResetModel char pmodel int pmos int level Parameter Description pmodel Pointer to the model instance pmos 1 if PMOS or 0 if NMOS level Model level value passed from the parser Example int fifdef STDC CMImos3ResetModel char pmodel int pmos else CMImos3ResetModel pmodel pmos char pmodel int pmos endif reset all flags to undefined void memset pmodel 0 sizeof MOS3model HSPICE MOSFET Models Manual 535 X 2005 09 8 Customer Common Model Interface Interface Variables Note contains a model value passed from the parser if pmos MOS3model pmodel MOS3type PMOS MOS3model pmodel MOS3typeGiven 1 return 0 int CMImos3ResetModel CMI Resetlnstance This routine initializes all parameter s in an instance After initialization all instance parameters become undefined in the netlist MOS instance Syntax int CMI ResetInstance char pinst In the preceding syntax pinst points to the instance Example int ifdef STDC__ CMImos3Reset Instance char ptran else CMImos3Reset Instance ptran char ptran fendif void memset ptran O sizeof MOS3instance MOS3instance ptran MOS3w 1 0e 4 MOS3instance ptran MOS31 1 0e 4 return 0 int CMImos3ResetInstance
253. age UCO V m Temperature coefficient BULKMOD 1 Bulk charge model selector XPART 1 Charge partitioning flag VFB V Flat band voltage PVAG 0 Gate dependence of the output resistance UCO has no effect on the model 386 Using the BSIM3 Version 2 MOS Model The Level 47 model uses the same model parameters for the source drain diode current capacitance and resistance as used in the other supported MOS levels The ACM model parameter controls the choice of source drain equations The Level 47 model also uses the same noise equations as the other MOSFET model levels The NLEV parameter controls the choice of noise equations This model like all Synopsys simulation device models can include parameters You can use these parameters to model the process skew either by worst case corners or by Monte Carlo For information about Worst Case and Monte Carlo analysis see Performing Worst Case Analysis and Performing Monte Carlo Analysis in Chapter 12 Statistical Analysis and Optimization in the HSPICE Simulation and Analysis Manual HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model Notes 1 2 Set Level 47 to identify the model as a BSIM3 model This model is based on BSIMS version 2 0 from UC Berkeley Code was received from UC Berkeley in July 1994 in the form of SPICE3e2 Changes announced in a letter from UCB September 13 1994 have been included
254. al drain resistance due to contact RSC ohm 0 Additional source resistance due to contact RSH ohm sq O Drain and source diffusion sheet resistance Note Source and drain resistances are calculated similarly to other HSPICE MOSFET models which are based on the value of ACM HSPICE MOSFET Models Manual X 2005 09 271 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model Equivalent Circuit Gate RSX RDX Cgs TN Z Cya Source i 2 i hh A Drain RS RD Model Equations Drain Current Total Current I Lang I G Fs a ides 1 Iking Subthreshold Current The following is the expression for the subthreshold current W2 VT Vpn Io MUS C V exp jh 2 exp 23 St 4 W 2 V V L2 MUS OEN exp 72 1 exp 23 sub 0X Ly sth s Vo ETA Vy V sth th C S y OX 7 Tox Ver Ves Vreg 272 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model MN AT Vps BT als Ves VST VIX Leg 1 op Br In the preceding equations et is the dielectric constant of the oxide and kg is the Boltzmanns constant Above the Vez gt 0 threshold the following equation calculates the conduction current 2 Hrer Cox W Vps is L Vere Vps 2 d for Vps S Osat V GTE eff sat 2 Hrer Cox Weg Varg Osa for V ps Osa VarE 2 6 Log Hrer MUO 1 CT bak u NG sth V V
255. ample of a PMOS model for the Level 47 MOSFET VTHO is negative model pch PMOS Level 47 Tnom 27 0 Npeak 1 5E123 Tox 7 0E 09 Xj 1 0E 07 dl 0 2E 06 dw 0 1E 06 SatMod 2 SubthMod 2 BulkMod 1 VthO 8 Phi 7 Kl 5 K2 0 03 K3 0 Dvt0 48 Dvtl 6 Dvt2 2 5e 4 Nlx 0 WO O Vsat 9E6 Ua 1E 09 Ub O Uc 3E 02 Rds0 180 Rdsw O U0 7E 03 A02 87 Voff 07 NFactor 1 5 Cit 3E 05 Cdsc 6E 02 Vglow 12 Vghigh 12 Pclm 77 Pdibli 0 Pdibl2 2E 01 Drout O Pscbel 0 Pscbe2 1E 28 Eta 0 Litl 4 5E 08 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Em 0 Ldd 0 ktl 3 kt2 03 At 33000 t Ual 4E 09 Ubl 7E 18 Ucl 0 Level 49 and 53 BSIM3v3 MOS Models The Synopsys Level 49 and Level 53 models are based on the BSIM3v3 MOS model from UC Berkeley Level 49 is an HSPICE enhanced version of BSIM3v3 Level 49 maintains compliance with the UC Berkeley release of BSIM3v3 with the following three exceptions e Default parameter values To eliminate differences in default parameter values Level 49 explicitly assigns the CAPMOD and XPART parameters and sets ACM 10 e Parameter range limits Provides parameter range limits that are identical to that of the Berkeley release Differences occur only in the
256. andard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Effects Modeled The EPFL EKV MOSFET model version 2 6 models the following physical effects Basic geometrical and process related aspects such as oxide thickness junction depth and effective channel length and width Effects of the doping profile Substrate effect Modeling of weak moderate and strong inversion behavior Modeling of mobility effects due to vertical and lateral fields and velocity saturation Short channel effects such as channel length modulation CLM source and drain charge sharing including for narrow channel widths and the reverse short channel effect RSCE Modeling of the substrate current due to impact ionization Quasi static charge based dynamic model Thermal and flicker noise modeling Short distance geometry dependent and bias dependent device matching Coherence of Static and Dynamic Models Simulation derives all aspects of the static the quasi static and the non quasi static NQS dynamic and noise models from the normalized transconductance to current ratio These expressions use symmetric normalized forward and reverse currents For quasi static dynamic operations you can use a charge based model for the node charges and trans capacitances or a simpler capacitances model The dynamic model including the time constant for the NQS model is described in symmetrical terms of the
257. ant in the inversion vfbsd LX75 Flat band Voltage between the Gate and 54 S D diffusions vgse LX76 Effective Gate to Source Voltage 54 VOX LX77 Voltage Across Oxide 54 rdv LX78 Asymmetric and Bias Dependent Source 54 22 Resistance rdsMod 1 HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level rsv LX79 Asymmetric and Bias Dependent Drain 54 Resistance rdsMod 1 cap bsz LX80 Zero voltage bias bulk source 54 capacitance cap_bdz LX81 Zero voltage bias bulk drain capacitance 54 CGGBM LX82 Total gate capacitance including 54 57 59 60 intrinsic and all overlap and fringing components CGDBM LX83 Total gate to drain capacitance including 54 57 59 60 intrinsic and overlap and fringing components CGSBM LX84 Total gate to source capacitance 54 57 59 60 including intrinsic and overlap and fringing components CDDBM LX85 Total drain capacitance including 54 57 59 60 intrinsic overlap and fringing components and junction capacitance CDSBM LX86 Total drain to source capacitance 54 57 60 CDGBM LX87 Total drain to gate capacitance including 54 57 59 60 intrinsic and overlap and fringing components CBGBM LX88 Total bulk to gate floating body to gate 54 57 59 60 capacitance including intrinsic and overlap components CBDBM LX89 Total bulk to drain
258. ard MOSFET Models Levels 1 to 40 see Chapter 4 Standard MOSFET Models Level 1 to 40 For information on BSIM MOSFET models based on models developed by the University of California at HSPICE MOSFET Models Manual 213 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 50 Philips MOS9 Model Berkeley see Chapter 6 BSIM MOSFET Models Levels 13 to 39 and Chapter 7 BSIM MOSFET Models Levels 47 to 65 Level 50 Philips MOS9 Model The Philips MOS Model 9 Level 902 is available as Level 50 in the Synopsys models based on the Unclassified Report NL UR 003 94 by R M D A Velghe 214 D B M Klaassen and F M Klaassen MOSFET Level 50 incorporates all features of Philips MOS 9 except for the gate noise current You can select one of two versions of MOSFET Level 50 ACM Parasitic Diode Model by using the JS JSW N CJ CUSW CUGATE MJ MJSW PB PHP ACM and HDIF parameters This version does not use the older IS parameter To use this model select the JUNCAP 0 default parameter Philips JUNCAP Parasitic Diode Model To use this version select the JUNCAP 1 model parameter For additional information regarding the MOS Model 9 see http www us semiconductors com Philips Models Table 45 MOSFET Level 50 Model Parameters Name Unit Default N Default P Description LER m 1 1e 6 1 25e 6 Reference Leff WER m 20 0e 6 20 0e 6 Reference Weff LVAR m 220 0e 9 460 0e 9 Variation in
259. ares and reviews the most commonly used MOSFET models The HSPICE Documentation Set This manual is a part of the HSPICE documentation set which includes the Xiv following manuals Manual Description HSPICE Simulation and Analysis User Guide HSPICE Signal Integrity Guide HSPICE Applications Manual Describes how to use HSPICE to simulate and analyze your circuit designs This is the main HSPICE user guide Describes how to use HSPICE to maintain signal integrity in your chip design Provides application examples and additional HSPICE user information HSPICE MOSFET Models Manual X 2005 09 About This Manual Searching Across the HSPICE Documentation Set Manual Description HSPICE Command Reference HPSPICE Elements and Device Models Manual HPSPICE MOSFET Models Manual HSPICE RF Manual AvanWaves User Guide HSPICE Quick Reference Guide HSPICE Device Models Quick Reference Guide Provides reference information for HSPICE commands Describes standard models you can use when simulating your circuit designs in HSPICE including passive devices diodes JFET and MESFET devices and BJT devices Describes standard MOSFET models you can use when simulating your circuit designs in HSPICE Describes a special set of analysis and design capabilities added to HSPICE to support RF and high speed circuit design Describes the AvanWaves tool which you can use to display
260. at http www device eecs berkeley edu bsim3soi For example the following is the topovar structure for the BSIM SOI model in the Customer CMI struct TOPO double vps double ves double delTemp T node double selfheat double qsub double qth double cbodcon double gbps double gbpr double gcde double gcse double gjdg double gjdd double gjdb double gjdT double gjsg double gjsd double gjsb double gjsT double cdeb double cbeb double ceeb HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Extended Topology double cgeo add 4 for T double cgT double cdT double cbT double ceT double rth double cth double gmT double gbT double gbpT double gTtg double gTtd double gTtb e doubl bi Enhancemnets for Customer CMI This section describes several modeling capabilities for Customer CMI which covers gate direct tunneling current modeling additional instance parameter support as well as so called Generalized Customer CMI with BSIM4 like topology Gate Direct Tunneling Current Gate direct tunneling currents as shown in Figure 44 are supported in the Customer CMI The tunneling process takes place in the thin gate oxide MOSFETs The current magnitudes become increasingly more significant as the oxide thickness scales down rapidly The HSPICE Customer CMI MOSFET topology meets such modeling requirement
261. atement Table 90 lists the changes in the BSIM model Table 90 BSIM2 Model Features by Version Number Model Version Effect of VERSION on BSIM2 Model 92A LEVEL 39 BSIM2 model introduced no changes 92B No changes 93A Introduced gds constraints fixed a defect in the WMUSB parameter and introduced a defect in the MU4 parameter 93A 02 Introduced the VERSION parameter and fixed an MU4 parameter defect 95 1 Fixed defects that caused PMUSB LDAC and WDAC parameter problems fixed the GMBS defect if you used gds constraints 96 1 Limited ETA ETAB vb520 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model Preventing Negative Output Conductance The Level 39 MOSFET model internally protects against conditions in the LEVEL 13 model that cause convergence problems due to negative output conductance This model imposes the following constraints MU220 ND20 AI20 Simulation imposes these constraints after adjusting the length and width and setting the VBS dependence This feature loses some accuracy in the saturation region particularly at high Vgs Consequently you might need to requalify the BSIM2 models in the following situations 1 Devices exhibit self heating during characterization which causes declining lgs at high Vgs This does not occur if the device characterization measurement sweeps V 2 The extraction technique produces parameters that result in n
262. ates from BSIM3V3 0 Level 49 XPART defaults to 1 CGSO F m p1 see No Non LDD region source gate overlap 1 capacitance per unit channel length CGDO F m p2 see No Non LDD region source gate overlap 2 capacitance per unit channel length CGBO F m 0 No Gate bulk overlap capacitance per unit channel length CGS1 F m 0 0 Yes Lightly doped source gate overlap region capacitance CGD1 F m 0 0 Yes Lightly doped drain gate overlap region capacitance CKAPPA F m 0 6 Yes Coefficient for the lightly doped region overlap capacitance fringing field capacitance CF F m 3 Yes Fringing field capacitance CLC m 0 1e 6 Yes Constant term for short channel model CLE 0 6 Yes Exponential term short channel model VFBCV V 1 0 Yes Flat band voltage used only in CAPMOD 0 C V calculations Table 99 Length and Width Parameters MOSFET 49 53 Name Unit Default Bin Description WINT m 0 0 No Width offset fitting parameter from I V without bias WIN 1 0 No Power of the length dependence of the width offset WW qQWWN 0 0 No Coefficient of the width dependence for the width offset HSPICE MOSFET Models Manual X 2005 09 419 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 99 Length and Width Parameters MOSFET 49 53 Continued Name Unit Default Bin Description WWN 1 0 No Power of the width depends on the width offset WWL qQWWN 0 0 No Coefficient of the length and width cross term for the widt
263. ating Effective Drain and Source Resistances For ACM 1 simulation calculates the effective drain and source resistances as follows Source Hesistance For UPDATE 0 LDscaled LDIFscaled NRS RSH RSC ems ee AA crier e doeet tul MEER RS Y arbi i carte ama Weff M If UPDATE 2 1 LDIF 0 and you specify the ASPEC option then RSeff RSeff x RS NRS RSH RSC Drain Resistance For UPDATE 0 LDscaled LDIFscaled NRD RSH RDC RD Se a Weff M If UPDATE 2 1 LDIF 0 and you specify the ASPEC option then RDeff RDeff L RD NRD RSH RDC HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models See LEVEL 6 LEVEL 7 IDS MOSFET Model on page 155 and LEVEL 7 IDS Model on page 181 for more possibilities Using an ACM 2 MOS Diode If you set the ACM 2 model parameter simulation uses HSPICE style MOS diodes You can use a fold back calculation scheme similar to the ASPEC method retaining full model parameter compatibility with the SPICE procedure This method also supports both lightly doped and heavily doped diffusions the LD LDIF and HDIF parameters set the diffusion type This model preserves the JS JSW CJ and CJSW units used in SPICE for full compatibility ACM 2 automatically generates more reasonable diode parameter values than those for ACM 1 You can generate the ACM 2 geometry in either of two ways AD AS PD and PS element parameters in
264. ation current sent to the source lae Igs normal RAT I_impact lap Igg diode 1 IIRAT I_impact IIRAT defaults to zero which sends all impact ionization current to bulk Leave IIRAT at its default value unless data is available for both drain and bulk current HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Impact lonization Table 14 Impact lonization Model Parameters Name Alias Units Default Description ALPHA 1 V 0 0 Impact ionization current coefficient LALPHA um V 0 0 ALPHA length sensitivity WALPHA um V 0 0 ALPHA width sensitivity VCR V 0 0 Critical voltage LVCR um V 0 0 VCR length sensitivity WVCR um V 0 0 VCR width sensitivity IIRAT 0 0 Portion of impact ionization current sent to the source Calculating the Impact lonization Equations The following equations calculates the impact current due to the impact ionization effect VCReff I impact Ids ALPHAeff vds vdsat evds vdsat The following equations calculate values used in the preceding equation ALPHAeff ALPHA LALPHA le 6 EMEN Leff LREFeff WALPHA le 6 a Weff WREFeff 1 1 VCReff VCR LVCR 1e 6 LI o CPC of LREFeff 1 1 WVCR 1e 6 4 8 CL Weff WREFeff HSPICE MOSFET Models Manual 61 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Impact lonization 62 The following equ
265. ations calculate the LREFeff and WREFeff values used in the preceding equations LREFeff LREF XLREF 2 LD WREFeff WREF XWREF 2 WD Calculating Effective Output Conductance You can use the element template to directly output gds For example PRINT I M1 gds LX8 M1 If you use impact ionization current gds is the derivative of ly only rather than the total drain current which is lygtlqp The complete drain output conductance is g _ la _ Bas Hay Oas Apa 44 DV QV OV OV 09V 8ds t Ebd For example to print the drain output resistance of the M1 device PRINT rout PAR 1 0 LX8 M1 LX10 M1 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Impact lonization Figure 15 Drain Source and Bulk Currents for vgs 3 with IIRAT 0 5 PLOT OF IMPACT IONIZATION CURRENT 11 0CT90 10 13 07 Pai aa ic E Edu BA fl a pice oo eet Lecce bear daa aaa L VOLTS CLIN Cascoding Example Drain to bulk impact ionization current limits the use of cascoding to increase output impedance The following cascode example shows the effect of changing IIRAT If IIRAT is less than 1 0 the drain to bulk current lowers the output impedance of the cascode stage Figure 16 Low frequency AC Analysis Measuring Output Impedance HSPICE MOSFET Models Manual 63 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Ca
266. atistical matching parameters AVTO AGAMMA AKP for Monte Carlo type simulations Default is 1E 6 for all three parameters in some implementations to allow sensitivity analysis of the matching parameters Do not specify LOT for AVTO AGAMMA or AKP Name Unit Default Description KF a 0 Flicker noise coefficient AF 1 Flicker noise exponent a The unit for KF might depend on the flicker noise model that you select if these options are available Name Unit Default Description NOS 0 Non Quasi Static NQS operation switch SATLIMP exp 4 Ratio defining the i i saturation limit XQC 0 4 Charge capacitance model selector a NQS 1 switches Non Quasi Static operation on default is off the NQS model option might not be available in all implementations b Only used for operating point information the SATLIM option might not be available in all implementations c Selects either the charges transcapacitances default or the capacitances only model XQC 0 4 charges transcapacitances model XQC 1 capacitances only model the XQC model option might not be available in all implementations HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Static Intrinsic Model Equations Basic Definitions SCALE 1 045x 10 F m _ Permittivity of silicon SCALE 345x10 F m Permittivity of silicon dioxide 1 602 x 10 c Magnitude of e
267. aturation voltage and drain current equations compared to the regular Multi Level model To use the improved model set the model parameter to UPDATE 1 HSPICE MOSFET Models Manual 165 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model Example You can see an example of a multi level gamma model with UPDATE 2 using a netlist from a previous example Change UPDATE 0 to UPDATE 2 in the netlist located in the following directory Sinstalldir demo hspice mos tgam2 sp Figure 30 Variation of IDS VTH and VDSAT for UPDATE 2 TGAM2 SP MULT LEVEL GAMMA MODEL UPDATE 2 14 MAY 2003 15 46 38 55 0U 52 50U 50 0U 47 6009U 720 0M 700 0M 680 0M gt 245 0M7 240 0M 1 050 1 10 N 235 146M 1 0 166 TGAM2 SWO0 IDS y TGAM2 SW0 VTH TGAM2 SW0 VDSAT 0 a 1 1 150 1 20 VOLTS LIN HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model Figure 31 Variation of GM GDS and GMBS for UPDATE 2 TGAM2 SP MULT LEVEL GAMMA MODEL UPDATE 2 14 MAY 2003 16 04 09 VOLTS LIN TGAM2_1 SP0 350 0U GM 5 zo P 347 500 E AL R 345 0U l g a M 342 50U gt 3 f ie LU r A i N r 1 H i Lu a I I i a i I 1 LI S TGAM2_1 SP0 32 250U GDS P 320UT 6 AL LN A R 31 750Uz z AN oe M 31 500 r LI r 1 F I 1 LI 1 I r i LI H LI Li I L
268. ature coefficient for UC Temperature coefficient for the saturation velocity Temperature coefficient for RDSW Junction current temperature exponent Table 101 Bin Description Parameters MOSFET Levels 49 53 Name Unit Default Bin Description LMIN m 0 0 No Minimum channel length LMAX m 1 0 No Maximum channel length WMIN m 0 0 No Minimum channel width WMAX m 1 0 No Maximum channel width BINUNIT Assumes that weff leff wref Iref units are in microns if BINUNIT 1 or in meters otherwise Table 102 Process Parameters MOSFET Levels 49 53 Name Unit Default Bin Description DTOXCV capmod 3 only GAMMA1 v 2 8 HSPICE MOSFET Models Manual X 2005 09 Difference between the electrical and physical gate oxide thicknesses due to the effects of the gate poly depletion and finite channel charge layer thickness Yes Body effect coefficient near the surface 421 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 102 Process Parameters MOSFET Levels 49 53 Continued Name Unit Default Bin Description GAMMA2 v12 9 Yes Body effect coefficient in the bulk VBX V 10 Yes VBX at which the depletion region width equals XT XT m 1 55e 7 Yes Doping depth Table 103 Noise Parameters MOSFET Levels 49 53 Name Unit Default Bin Description NIOA 1 0020 nmos No Body effect coefficient near the surface 9 9e18 pmos NOIB 5 0e4 nmos No Body effect coefficient in the bulk
269. b which provides PDF documents and is available through SolvNet at http solvnet synopsys com The Synopsys MediaDocs Shop from which you can order printed copies of Synopsys documents at http mediadocs synopsys com You might also want to refer to the documentation for the following related Synopsys products CosmosScope Aurora Raphael VCS Conventions xvi The following conventions are used in Synopsys documentation Convention Description Courier Indicates command syntax Italic Indicates a user defined value such as object_name Bold Indicates user input text you type verbatim in syntax and examples HSPICE MOSFET Models Manual X 2005 09 About This Manual Customer Support Convention Description Denotes optional parameters such as write file f filename Indicates that a parameter can be repeated as many times as necessary pinl pin2 pinN Indicates a choice among alternatives such as low medium high Indicates a continuation of a command line Indicates levels of directory structure Edit gt Copy Indicates a path to a menu command such as opening the Edit menu and choosing Copy Control c Indicates a keyboard combination such as holding down the Control key and pressing c Customer Support Customer support is available through SolvNet online customer support and through contacting the Synopsys Technical Support Center Ac
270. b dVg dIgb dVd dIgb dVb dIgb dVs d Extended Topology to drain tunneling current through G D dlIgso dVg d Igcs dVg dlIgdo dVg ay dIgcd dVg Igb dVg Refer to extcmi enhancements DE52897 for testcode and testcases Additional Instance Parameter support Customer CMI supports the following instance parameter namings to activate this use OPTION CUSTCMI 1 acngsmod delk1 delnfct muluO nf rbdb rbps rbsb rgatemod sa2 sa3 sa4 sa8 saQ sb sb3 sb4 sb5 HSPICE MOSFET Models Manual X 2005 09 Existing HSPICE BSIM4 like instance parameters deltox geomod rbodymod rbpb sa sal sab sab sb1 sb10 sb6 sb7 min rbpd sa10 sa7 sb2 sb8 555 8 Customer Common Model Interface Extended Topology sb9 sd stimod sw1 sw10 sw2 sw3 sw4 sw5 sw6 sw7 sw8 sw9 trnqsmod Among these instance parameters trnqsmod acnqsmod rbodymod rgatemod geomod and st imod are type integer Six instance model flags of integer type with the following fixed names insflg1 insflg2 insflg3 insflg4 insflg5 and insflg6 Ten instance parameters of double type with the following fixed names insprml insprm2 insprm3 insprm10 The syntax and namings look like this ml dgsb t lt geomod val gt lt nf val gt lt insflgl val gt lt insflg2 val gt lt insflg3 val gt t lt insprml val gt insprm2 val lt insprm3 val gt
271. body effect K2 0 Coefficient for the second order body effect HSPICE MOSFET Models Manual X 2005 09 487 7 BSIM MOSFET Models Levels 47 to 65 Level 59 UC Berkeley BSIM3 SOI FD Model Table 136 MOSFET Level 59 DC Parameters Continued Parameter Unit Default Description K3 0 Narrow coefficient K3B 1 V 0 Body effect coefficient of k3 KB1 1 Coefficient of the Vps 0 dependency on Vos KB3 1 Coefficient of the V5 0 dependency on Vgs at subthreshold region KBJT1 m V 0 Parasitic bipolar Early effect coefficient KETA m 0 6 Body bias coefficient of the bulk charge effect LINT m 0 0 Length offset fitting parameter from I V without bias MXC 0 9 Fitting parameter for calculating Aper NDIODE 1 0 Diode non ideality factor NFACTOR 1 Subthreshold swing factor NGIDL V 1 2 GIDL Vg enhancement coefficient NLX m 1 74e 7 Lateral non uniform doping parameter NTUN 10 0 Reverse tunneling non ideality factor PCLM 1 3 Channel length modulation parameter PDIBL1 0 39 First correction parameter for the DIBL effect of the output resistance PDIBL2 0 0086 Second correction parameter for the DIBL effect of the output resistance PRWB 1 1 0 Body effect coefficient of Rdsw PRWG 1 12 0 Gate bias effect coefficient of Rdsw PVAG 0 0 Gate dependence of the Early voltage 488 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 59 UC Berkeley BSIM3 SOI FD Model Table 136 MOSFET Level
272. caling rules e Set Version 11011 to identify the model as Philips MOS Model 11 Level 1101 binning geometry scaling rules HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model 3 The default room temperature is 25 C in Synopsys circuit simulators but is 27 C in most other simulators When comparing to other simulators use TEMP 27 or OPTION TNOM 27 to set the simulation temperature to 27 in the netlist 4 Theset of model parameters should always include the TR model reference temperature which corresponds to TREF in other levels in the Synopsys MOSFET model levels The default for TR is 21 0 to match the Philips simulator 5 This model has its own charge based capacitance model This model ignores the CAPOP parameter which selects different capacitance models 6 This model uses analytical derivatives for the conductances This model ignores the DERIV parameter which selects the finite difference method 7 You can use DTEMP with this model to increase the temperature of individual elements relative to the circuit temperature Set DTEMP on the element line 8 Because the defaults are non zero you should set every Level 63 model parameter in the Model Parameters table in the MODEL statement 9 The general syntax for the MOSFET element is the same as the other standard MOSFET models other than PS and PD In Level 63 PS and PD are the length of the sid
273. capacitance including 54 intrinsic and junction capacitance CBDBM LX89 Total floating body to drain capacitance 57 59 60 including intrinsic and junction capacitance HSPICE9 MOSFET Models Manual 23 X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level CBSBM LX90 Total bulk to source capacitance 54 including intrinsic and junction capacitance CBSBM LX90 Total floating body to source capacitance 57 59 60 including intrinsi c and junction capacitance CAPFG LX91 Fringing capacitance 54 CDEBM LX92 Total drain to substrate capacitance 57 59 60 including intrinsi c and junction capacitance CSGBM LX93 Total source to gate capacitance 57 59 60 including intrinsic and overlap and fringing components CSSBM LX94 Total source capacitance including 57 59 60 intrinsic overlap and fringing components and junction capacitance CSEBM LX95 Total source to substrate capacitance 57 59 60 including intrinsic and junction capacitance CEEBM LX96 Total substrate capacitance including 57 59 60 intrinsic overlap and fringing components and junction capacitance QGI LX97 Intrinsic Gate charge 49 53 QSI LX98 Intrinsic Source charge 49 53 QDI LX99 Intrinsic Drain charge 49 53 QBI LX100 Intrinsic Bulk charge 49 53 Charge Conservation QBI QGI QSI QDI CDDBI LX101 Intrinsic drain
274. capacitance 49 53 24 HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level CBDBI LX102 Intrinsic bulk to drain capacitance 49 53 CBSBI LX103 Intrinsic bulk to source capacitance 49 53 VBDI LX109 Body drain voltage VBD Meyer and 57 58 59 Charge Conservation IGISLO LX110 Gate induced source leakage current 54 GRII LX118 Intrinsic channel reflected gate 54 conductance GRGELTD LX119 Gate electrode conductance 54 HSPICE9 MOSFET Models Manual 25 X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates 26 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models Describes the technology used in all HSPICE MOSFET models Nonplanar and Planar Technologies Two MOSFET fabrication technologies have dominated integrated circuit design nonplanar and planar technologies Nonplanar techonology Nonplanar technology uses metal gates The simplicity of the process generally provides acceptable yields The primary problem with metal gates is metal breakage across the field oxide steps Field oxide grows when oxidizing the silicon surface When the surface is cut it forms a sharp edge Because metal is affixed to these edges to contact the diffusion or make a gate thicker metal must be applied to compensate for the sharp edges This metal tend
275. capbd TT 75 M CBD 1 MJ 5 Common Threshold Voltage Equations 58 Common Threshold Voltage Parameters The parameters described in this section apply to all MOSFET models except Levels 5 and 13 Table 133 MOSFET Common Threshold Voltage Parameters Name Alias Units Default Description DELVTO V 0 0 Zero bias threshold voltage shift GAMMA y1 2 0 527625 Body effect factor If you do not set GAMMA simulation calculates it from NSUB NGATE 1 cm3 Polysilicon gate doping used for analytical models only Undoped polysilicon is represented by a small value If NGATE lt 0 0 itis set to 1e 18 NSS 1 cm2 1 0 Surface state density NSUB DNB NB 1 cm3 1e15 Substrate doping HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models Common Threshold Voltage Equations Table 133 MOSFET Common Threshold Voltage Parameters Continued Name Alias Units Default Description PHI V 0 576036 Surface potential NSUB default 1e15 TPG TPS 1 0 Type of gate material for analytical models LEVEL 4 TPG default 0 The TPG value can be TPG 0 al gate TPG 1 same as source drain diffusion TPG 1 gate type opposite to source drain diffusion VTO VT V P Zero bias threshold voltage Calculating PHI GAMMA and VTO Use the PHI GAMMA and VTO model parameters to calculate threshold voltages If you do not specify these parameters simulation calculates them as follow
276. ce in series with C RS u 0 Source resistance RSX Q 0 Resistance in series with Cg TNOM C 25 Parameter measurement temperature TOX m 1e 7 Thin oxide thickness VO V 0 12 Characteristic voltage for deep states VFB V 0 1 Flat band voltage 268 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model Table 58 MOSFET Level 62 Model Parameters Continued Name Unit Default Description VKINK V 9 1 Kink effect voltage VON V 0 On voltage VSI V 2 0 vgs dependence parameter VST V 2 0 vgs dependence parameter VTO V 0 Zero bias threshold voltage ZEROC 0 Flag for capacitance calculations in capmod 1 capmod 1 set the O capacitance value capmod 0 calculation capacitance XL m 0 Length bias accounts for the masking and etching effects XW m 0 Width bias accounts for the masking and etching effects LMLT 1 Length shrink factor WMLT 1 Width shrink factor Table 59 Model Parameters Specific to Version 2 Name Unit Default Description VERSION 1 1 Version 1 2 Version 2 ME MS 2 5 Long channel saturation transition parameter META 1 ETA floating body parameter MSS 15 Vdse transition parameter VMAX m s 4 00E 004 Saturation velocity HSPICE MOSFET Models Manual X 2005 09 269 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model Table 59 Model Parameters Specific to Version 2 Continued Name Unit Defau
277. cessing SolvNet SolvNet includes an electronic knowledge base of technical articles and answers to frequently asked questions about Synopsys tools SolvNet also gives you access to a wide range of Synopsys online services which include downloading software viewing Documentation on the Web and entering a call to the Support Center HSPICE MOSFET Models Manual xvii X 2005 09 About This Manual Customer Support To access SolvNet 1 Goto the SolvNet Web page at http solvnet synopsys com 2 f prompted enter your user name and password If you do not have a Synopsys user name and password follow the instructions to register with SolvNet If you need help using SolvNet click SolvNet Help in the Support Resources section Contacting the Synopsys Technical Support Center If you have problems questions or suggestions you can contact the Synopsys Technical Support Center in the following ways Open a call to your local support center from the Web by going to http solvnet synopsys com Synopsys user name and password required then clicking Enter a Call to the Support Center Send an e mail message to your local support center e E mail support center synopsys com from within North America Find other local support center e mail addresses at http www synopsys com support support ctr Telephone your local support center Call 800 245 8005 from within the continental United States Call 650 584 4
278. cient of the threshold temperature effect GAMMA1 y1 2 See Level 47 Body effect coefficient near interface Model Equations on page 390 GAMMA2 y See Level 47 Body effect coefficient in the bulk Model Equations on page 390 WO m 2 5e 6 Narrow width effect coefficient NLX m 1 74e 7 Lateral nonuniform doping along the channel TOX m 150e 10 Gate oxide thickness 382 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model Table 91 MOSFET Level 47 Model Parameters Continued Name Unit Default Description XJ m 0 15e 6 Junction depth DL m 0 0 Channel length reduction on one side multiplied by SCALM DW m 0 0 Channel width reduction on one side multiplied by SCALM NPEAK cm3 pg 1 7e17 Peak doping concentration near the interface NSUB cm3 6 0e16 Substrate doping concentration PHI V See Level 47 Surface potential under strong inversion Model Equations on page 390 XT m 1 55e 7 Doping depth VBM V 5 0 Maximum substrate bias VBX V See Level 47 Vps at which the depletion width equals XT Model Equations on page 390 DVTO 2 2 Short channel effect coefficient 0 DVT1 0 53 Short channel effect coefficient 1 DVT2 1 V 0 032 Short channel effect coefficient 2 UO m2 Vsec 0 067 Low field mobility at T TREF 8 0 067 for n channel 0 025 for p channel UA m V 2 25e 9 First order mobility degradation coefficient UA1 m V 4 31e 9 Temperature coefficient of UA HSPICE MOSFET M
279. citance dQg dVgs cgso double capgd Meyer s gate capacitance dQg dVds cgdo double capgb Meyer s gate capacitance dQg dVbs cgbo substrate junction information double ibs substrate source junction leakage current double ibd substrate drain junction leakage current double gbs substrate source junction conductance double gbd substrate drain junction conductance double capbs substrate source junction capacitance double capbd substrate drain junction capacitance double qbs substrate source junction charge double qbd substrate drain junction charge substrate impact ionization current double isub substrate current double gbgs substrate trans conductance dIsub dVgs double gbds substrate trans conductance dIsub dVds double gbbs substrate trans conductance dIsub dVbs charge based model intrinsic terminal charges NOTE these are intrinsic charges ONLY double qg gate charge double qd drain charge HSPICE MOSFET Models Manual 531 X 2005 09 8 Customer Common Model Interface Model Interface Routines 532 double qs source charge charge based model intrinsic trans capacitances NOTE these are intrinsic capacitances ONLY double cggb double cgdb double cgsb double cbgb double cbdb double cbsb doub
280. clude gate drain gate source and gate bulk overlap capacitance and drain bulk and source bulk diode capacitance Drain and source refer to node 1 and 3 of the MOS element physical instead of electrical For the Meyer models where charges such as QD are not well defined Table 17 shows the printout quantities Table 17 Capacitance Printout for Meyer Models Capacitance Value cdtot cgd cdb cgtot cgs cgd cgb cstot cgs csb cbtot cgb csb cdb cgs cgs cgd cgd Element Template Printout The MOS element template printouts for gate capacitance are LX18 to LX23 and LX32 to LX34 From these nine capacitances you can construct the complete four by four matrix of transcapacitances The nine LX printouts are LX18 m dQG dVGB CGGBO LX19 m dQG dVDB CGDBO LX20 m dQG dVSB CGSBO HSPICE MOSFET Models Manual 69 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models 70 LX21 m dQB dVGB CBGBO LX22 m dQB dVDB CBDBO LX23 m dQB dVSB CBSBO LX32 m dQD dVG CDGBO LX33 m dQD dVD CDDBO LX34 m dQD dVS CDSBO These capacitances include gate drain gate source and gate bulk overlap capacitance and drain bulk and source bulk diode capacitance Drain and source refer to node 1 and 3 of the MOS element physical instead of electrical For an NMOS device with source and bulk grounded LX18 is the input capacitance LX33 is the outp
281. cription LVFB V um 0 0 Length sensitivity WVFB V um 0 0 Width sensitivity X2E 1 V 0 0 Vsb correction to the linear vds threshold coefficient LX2E um V 0 0 Length sensitivity WX2E um V 0 0 Width sensitivity X2M X2MZ cm V2 s 0 0 Vsb correction to the low field first order mobility LX2M LX2MZ um cm V2 s 0 0 Length sensitivity WX2M WX2MZ um cm V2 s 0 0 Width sensitivity X2MS cm V2 s 0 0 Vbs reduction to the high drain field mobility LX2MS umPcm V2 s 0 0 Length sensitivity WX2MS umPcm V2 s 0 0 Width sensitivity X2U0 1 V 0 0 Vsb reduction to the GATE field mobility reduction factor LX2UO um V 0 0 Length sensitivity WX2UO um V2 0 0 Width sensitivity X2U1 um V 0 0 Vsb reduction to the DRAIN field mobility reduction factor LX2U1 um2 v 0 0 Length sensitivity WX2U1 um V 0 0 Width sensitivity HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model Table 83 Transistor Parameters MOSFET Level 13 Continued Name Alias Units Default Description X3E 1 V 0 0 Vds correction to the linear vds threshold coefficient LX3E um V 0 0 Length sensitivity WX3E um V 0 0 Width sensitivity X3MS cm2 5 0 Vds reduction to the high V2b s drain field mobility LX3MS umPcm V2ps 0 0 Length sensitivity WX3MS umPcm V2ps 0 0 Width sensitivity X3U1 um V 0 0 Vds reduction to the drain field mobility reduction factor LX3U1 um V 2 0 0 Length sensitivity
282. d 3 2 tert B vas by de d Partial Enhancement vgs vfb vde ids B1 fq 7K10 NP vde con 24 vgs vfb vde s cav Y vde vsb Phid 3 vsb Phid 2 en Ep BI Pear vgs vfb The following equations calculate values used in the preceding equations BI zKBetal UH Weff vde Leff 1 HSAT UHS m Weff UBeff B UBeff cox Leff _ KCS Si cs DP le 4 ER NI le4 DP Phid vt In vde min vds vdsat The temperature dependence of the mobility terms assume the ordinary exponential form TUH UH t UH inom t TUH zUO t zUO tnom tno 198 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 38 IDS Cypress Depletion Model The following equation calculates the continuity term at the body effect transition point Icrit Ha cav vde vsbc Phid 3 vsbc Phid 3 y 1 3 zBetaGam This model uses the preceding equation if vsbsvsbc Otherwise Icrit 0 The following sections describe saturation voltage threshold voltage body effect transition voltage and the y body effect coefficient Threshold Voltage vth The VTO model parameter often called the pinch off is a zero bias threshold voltage extrapolated from a large device operating in the depletion mode The following equation calculates the effective pinch off threshold voltage inc
283. d MEYER control for transition of cgs from weak to strong inversion region for CAPOP 2 only CF3 1 0 Modified MEYER control for the cgs and cgd transition from the saturation region to the linear region as a function of vds for CAPOP 2 only CF4 50 0 Modified MEYER control for the contour of the cgb and cgs smoothing factors CF5 0 667 Modified MEYER control for the capacitance multiplier for cgs in the saturation region CF6 500 0 Modified MEYER control for contour of cgd smoothing factor CGBEX 0 5 cgb exponent for CAPOP 1 only Table 21 Charge Conservation Parameters CAPOP 4 Name Alias Units Default Description XQC 0 5 Coefficient of channel charge share attributed to drain its range is 0 0 to 0 5 This parameter applies only to CAPOP 4 and some of its level dependent aliases 76 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Specifying XQC and XPART for CAPOP 4 9 11 12 13 Parameter rules for the gate capacitance charge sharing coefficient XQC amp XPART in the saturation region If you do not specify either XPART or XQC then simulation uses the 0 100 model f you specify both XPART and XQC then XPART overrides XQC f you specify XPART but you do not specify XQC then e XPART 0 s 40 60 e XPART 0 4 40 60 e XPART 0 5 5 50 50 e XPART 1 5 0 100 e XPART any other value less than 1 40 60 e XPART 51 5 0 100lf X
284. d MOSFET Models Levels 50 to 64 Level 61 RPI a Si TFT Model Equivalent Circuit Gate Cos n rox Coa Ids O e o Source S Drain Model Equations Drain Current s leakage tI lib 8epV ase l LAMBDA Vip Nag Vase TI 1 Vas Vsate V Q V sate sat gte Echi c DX SERD qn W MUBAND L Sch 1 4 Ecni RS RD Echi qn i Nsa t N sb x EPSI e GAMMA sa a TOX NV aa HSPICE MOSFET Models Manual 263 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 61 RPI a Si TFT Model 2 VO em I V othe EPS Me sb SATOX VO EPS Neo Nanygesp 201 N 3 0 10 m t _ 2 VO Viho E EPS 2 V0O V is 2q GMIN Pak V V 27 Lus EA eee ae DELTA 1 SIE 2 VMIN VMIN Vor E Var Vr V V 2 Paa E DELTA 8 1 gfbe 2 VMIN VMIN V oth s VFB Leakage Dt Lmin r Va J Vig FEL 1 1 I TOL exp lex Cope zJ ka P VDS j VGS Al q Vino Vin Inin SIGMAO Vis Temperature Dependence Vino kg TNOM q Vin kg TEMP q EMU 1 1 Ve VAA KE AN z is T z GAMMAW Vino Vr VTO KVT TEMP TNOM 0 ALPHASAT KASAT TEMP TNOM 264 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model Capacitance 2 Vate Vise z Meg ey E e Ni l sate e 8 2 V cute y Cog 443G 1 TERDUM sate e dn C 0 5 EPS W Ci IV N sac
285. d VTO parameters for the LEVEL 5 model The Synopsys device model uses DNB to compute yo and ignores the GAMMA model parameter The following equation computes the y effective body effect including the device size effects Y Yo 1 scf 1 ncf If SCM 0 then scf 0 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model Otherwise XJ 2xd is fpe TT SCM vgs Vsp 2 1 IF NWM lt then ncf 0 Otherwise NWM X 2 Win The following equation calculates the x value used in the preceding equations 3 E 1 2 i si q DNB Saturation Voltage Vgsat ncf The following equation computes the saturation voltage due to the channel pinch off at the drain side y r 4 1 2 Vdsat 7 Vsat If ECV does not equal 1000 the program modifies v to include the carrier velocity saturation effect 2 2 1 2 Vasat Veat t Ve St Ye where v ECV Leg Mobility Reduction UB gry The following equation computes the mobility degradation effect in the LEVEL 5 MOSFET model OP oie 1 FRC v y d 1 2 Ii A Ec vag FSB v UB TOX VST L HSPICE MOSFET Models Manual 147 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model 148 The following equations calculate the L value used in the preceding equation Le Leg linear region e L Lep AL saturation region e The next section describes the AL chan
286. d as the average capacitance times the change in voltage If the capacitance is nonlinear this approximation can be in error To accurately estimate the charge use Simpson s numerical integration rule This method provides charge conservation control To use this model parameter 1 Setthe CAPOP model parameter to 3 and use the existing CAPOP 2 model parameters 2 Modify the OPTION RELV relative voltage tolerance OPTION RELMOS relative current tolerance for MOSFETs and OPTION CVTOL capacitor voltage tolerance settings HSPICE MOSFET Models Manual 87 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models 88 The default of 0 5 is a good nominal value for CVTOL The CVTOL option uses the following equation to set the number of integration steps IV D Von CVTOL Using a large value for CVTOL decreases the number of integration steps for the n to n 1 time interval this yields slightly less accurate integration results Using a small CVTOL value increases the computational load and sometimes severely CAPOP 4 Charge Conservation Capacitance Model The charge conservation method See Ward Donald E and Robert W Dutton A Charge Oriented Model for MOS Transistor is not implemented correctly into the SPICE2G 6 program There are errors in the derivative of charges especially in LEVEL 3 models Also the channel charge partition is not continuous from the linear region to the sat
287. dard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 63 Level 63 MOS11 Parameters Level 11010 Physical Geometry Scaling Continued Name Description Units NMOS PMOS AiR Weak avalanche current factor for z 6 6 reference transistor at reference temperature STA1 Temperature dependence coefficient of at K 1 0 0 SLA1 Coefficient of the length dependence ofai 0 0 SWA1 Coefficient of the width dependence of a1 0 0 A2R Exponent of the weak avalanche current V 38 38 for the reference transistor SLA2 Coefficient of the length dependence of a2 0 0 SWA2 Coefficient of the width dependence of a2 0 0 A3R Drain source voltage factor above which 1 1 weak avalanche occurs for reference transistor SLA3 Coefficient of the length dependence of a3 0 0 SWA3 Coefficient of the width dependence of a3 0 0 IGINVR Gain factor for intrinsic gate tunneling AV 2 0 0 current in inversion for reference transistor BINV Probability factor for intrinsic gate V 48 87 5 tunneling current in inversion IGACCR Gain factor for intrinsic gate tunneling AV 2 0 0 current in accumulation for reference transistor BACC Probability factor for intrinsic gate V 48 48 tunneling current in accumulation VFBOV Flat band voltage for the Source Drain V 0 0 overlap extensions HSPICE MOSFET Models Manual X 2005 09 291 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 63 Level 63 MOS11 Parameters Level 11010 P
288. del set KAPPA 0 in the Synopsys Level 3 MOSFET model LD Missing If you do not specify LD simulation uses the default 0 75XJ The SPICES default for LD is zero Solution If you do not specify LD in the SPICE3 model set LD 0 in the Level 3 MOSFET model Name Symbol Value Boltzmann constant k 1 3806226e 23J K Electron charge e 1 6021918e 19C HSPICE MOSFET Models Manual 141 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 3 IDS Empirical Model 142 Name Symbol Value Permittivity of silicon dioxide 5 3 45314379969e 11F m Permittivity of silicon Esi 1 035943139907e 10F m Temperature Compensation The input file example located in the following directory verifies temperature dependence for MOSFET Level 3 Sinstalldir demo hspice mos tempdep sp This simple model with XJ 0 and KAPPA 0 has a saturation current 2 B beta 0 5 V9 V m 1 fb GAMMA 4 Jphi t Using the model parameters in the input file and the preceding equations produces these results b 1 2e 3 Kai eta 1 2e 3 ref beta COX 5 UO t fb Vin O 8 TCV t tref T SEE E phi t 0 64 Ton vtherm egarg 3 ioe 7 At room temperature beta 1 2 e 3 Vim 0 8 phi t 0 64 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 4 IDS MOS Model 2 5 2 0 8 I4 0 64 Ij 1 2e 3 0 6912e 4 At T 100 beta
289. del selector chooses the pch 4 model because that model satisfies these requirements Similarly the NJ5 transistor requires a model with 1 5 u lt channel width lt 3 u and 2 u lt channel length lt 6 u The pch 5 model HSPICE MOSFET Models Manual 567 X 2005 09 A Finding Device Libraries satisfies these requirements If a device size is out of range for all models the automatic model selector issues an error message Figure 47 Automatic Model Selector Method Drawn Channel Width wminz1 5 m 1m 2m Iminz0 8m gt lt max 2 m Drawn Channel Length If the automatic model selector cannot find a model within a subcircuit the automatic model selector searches the top level If the automatic model selector fails to find a model simulation terminates The following combination of conditions causes the automatic model selector to fail and terminates the simulation The element statement uses a model name that contains a period The model library was not designed for use with the automatic model selector The simulation input includes either a multisweep specification or a TEMP temperature analysis statement The following example illustrates how a period in a model name can cause problems in automatic model selection Example 1 MI dg s b N CHN W 10u L 5u Element statement MODEL N CHN LMIN 1u LMAX 4u WMIN 2u WMAX 100u MODEL statement 568 HSPICE MOSFET Models Manual X 2
290. dels Levels 13 to 39 LEVEL 39 BSIM2 Model 364 Saturation Vgs gt Vasat drain source current ps p V Vin I so S 2 2aK 1 UA V V 4 UB V V A f In the preceding equation the fimpact ionization term is BI f Al eg Vas Vasar Weak Inversion Vgg lt Vi VGLOW VGLOW 0 Subthreshold drain source current lys 7 V ME B Vis 7 Ips p V2 exp VOFF 1 exp 7 1f tm tm The following equations calculate the V and N values used in the preceding equation v s IT ana N NO q tm NB ND g Vis PHI V Strong inversion to weak inversion transition region Vin VGLOWS Vgs lt th VGHIGH 3 Veh aed gt X ONG j 0 The preceding equation replaces Vas Vgs Vin in the linear or saturation drain currents based on Vasat Vgert At the lower boundary Vos Vi Z VGLOW the saturation equation is valid for all Vas that is Vasat Vgett VGLOW 0 to allow a match to the above subthreshold equation To internally determine the Cj coefficients of the Vges cubic spline the Ipg and Algs AV gs conditions must both be continuous at the Vgs Vth VGLOW and Vg Vin VGHIGH boundaries HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model Effective Length and Width If DL is nonzero L L q LMLT DL scale eff LREF y LREF 4 LMLT DL Otherwise L L q LMLT XL 2
291. dels MOS Gate Capacitance Models Example The netlist for this example is located in the following directory Sinstalldir demo hspice mos mcap3 sp Figure 22 CAPOP 4 9 Capacitances for LEVEL 3 Model Param Lin Param Lin 30 0F 20 0F 10 0F 5 8e 18 30 0F 20 0F 10 0F 4 0e 18 1 0 Tene Cer war Par eae See ie Pen Vu or dod es Fa oP SDan Volts Lin CAPOP2 SVO C68 VOSP05 XA C68 VOSP05 g C La CAPOP2 SVO C68 VOSP05 o C68 VOSP05 Cc G ee HSPICE MOSFET Models Manual X 2005 09 89 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Figure 23 CAPOP 2 Capacitances for LEVEL 3 Model cm ID rr xFILE MEYER2 SP MEYER CAPS LEVEL 3 CAPOP 2 21 SEP92 10 1 41 z i p a i CGB i d PP c sg a poy i 1 0 0 1 0 8 0 3 0 4 0 ve VOLTS LIN 5 0 90 The example below tests the charge conservation capacitance model Yang P B D Epler and P K Chaterjee An Investigation of the Charge Conservation Problem and compares the Meyer and charge conservation models As the graph in Figure 25 shows the charge conservation model returns more accurate results Example The netlist for this example is located in the following directory Sinstalldir demo hspice mos chrgpump sp HSPICE MOSFET Models Manual X 20
292. dependent part of IGOV PLIGOV Coefficient for the length dependent AV 2 0 0 part of IGOV PWIGOV Coefficient for the width dependent AV 2 0 0 part of IGOV 300 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS PLWIGOV Coefficient for the width over length AV 2 0 0 dependent part of IGOV TOX Thickness of the gate oxide layer m 3 2E 9 3 2E 9 POCOX Coefficient for the geometry F 2 98E 14 2 717E 14 independent COX part PLCOX Coefficient for the length dependent F 0 0 part of COX PWCOX Coefficient for the width dependent F 0 0 part of COX PLWCOX Coefficient for the length times width F 0 0 dependent part of COX POCGDO Coefficient for the geometry E 6 392E 15 6 358E 15 independent CGDO part PLCGDO Coefficient for the length dependent F 0 0 part of CGDO PWCGDO Coefficient for the width dependent F 0 0 part of CGDO PLWCGDO Coefficient for the width over length F 0 0 dependent part of CGDO POCGSO Coefficient for the geometry F 6 392E 15 6 358E 15 independent part of CGSO PLCGSO Coefficient for the length dependent F 0 0 part of CGSO PWCGSO Coefficient for the width dependent F 0 0 part of CGSO PLWCGSO Coefficient for the width over length F 0 0 dependent part of CGSO HSPICE MOSFET Models Manual 301 X 2005 09 5 Standard MOSFE
293. devices it also becomes a function of the source and drain voltage due to the charge sharing effect HSPICE MOSFET Models Manual 235 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model 236 Slope Factor GAMMA n 2 Vp PHI AV Note The slope factor or body effect factor is primarily a function of the gate voltage and links to the weak inversion slope Large Signal Interpolation Function F v is the large signal interpolation function relating normalized currents to normalized voltages A simple and accurate expression for the transconductance to current ratio consistently formulates The static large signal interpolation function The dynamic model for the intrinsic charges and capacitances The intrinsic time constant and the thermal noise model for the whole range of current from weak to strong inversion Gaga Ee 1 Ips J0 25 i 0 5 Large signal interpolation function y 40 25 i 0 5 y 2y In y You cannot analytically invert this equation However you can use a Newton Raphson iterative scheme to invert this equation Currently this model uses a simplified algorithm that avoids iteration leading to a continuous expression for the large signal interpolation function The inverted large signal interpolation function has the following asymptotes in strong and weak inversion Fo re for v gt gt 0 exp v for v lt lt 0 Forward Normaliz
294. dges Model Berkeley see Chapter 6 BSIM MOSFET Models Levels 13 to 39 and Chapter 7 BSIM MOSFET Models Levels 47 to 65 LEVEL 1 IDS Schichman Hodges Model 128 Use the LEVEL 1 MOSFET model if accuracy is less important to you than simulation turn around time For digital switching circuits especially if you need only a qualitative simulation of the timing and the function LEVEL 1 run time can be about half that of a simulation using the LEVEL 2 model The agreement in timing is approximately 10 The LEVEL 1 model however results in severe inaccuracies in DC transfer functions of any TTL compatible input buffers in the circuit The LAMBDA channel length modulation parameter is equivalent to the inverse of the Early voltage for the bipolar transistor LAMBDA measures the output conductance in the saturation If you specify this parameter the MOSFET has a finite but constant output conductance in saturation If you do not specify LAMBDA the LEVEL 1 model assumes zero output conductance LEVEL 1 Model Parameters MOSFET Level 1 uses only the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters LEVEL 1 Model Equations The LEVEL 1 model equations follow IDS Equations The LEVEL 1 model does not include the carrier mobility degradation the carrier saturation effect or the weak inversion model This model determines the DC current Cutoff Region VOS S Vh 7 00 Linear Region
295. doped source gate region cgso F m calculate Non LDD region source gate overlap capacitance per d channel length cjswg F m 1 e 10 Source drain gate side sidewall junction capacitance per unit width normalized to 100nm Tsi ckappa F m 0 6 Coefficient for the fringing field capacitance for the overlap capacitance in the lightly doped region clc m 0 1e 7 Constant term for the short channel model cle 0 0 Exponential term for the short channel model csdesw F m 0 0 Fringing capacitance per unit length for the source drain sidewall csdmin V cal Minimum capacitance for the source drain bottom diffusion delvt V 0 0 Threshold voltage adjustment for C V dibg m 0 Length offset fitting parameter for the backgate charge HSPICE MOSFET Models Manual X 2005 09 473 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model Table 131 MOSFET Level 57 AC amp Capacitance Parameters Continued Parameter Unit Default Description dic m lint Length offset fitting parameter for the gate charge dicb m lint Length offset fitting parameter for the body charge dwc m wint Width offset fitting parameter from C V fbody 1 0 Scaling factor for the body charge LdifO 1 Channel length dependency coefficient of the diffusion cap mjswg V 0 5 Grading coefficient of the source drain gate side sidewall junction capacitance moin y1 2 15 0 Coefficient for the gate bias dependent surface potential Ndif 1 Power coeffic
296. drain F m 0 0 gate region ckappa Coefficient of the fringing field capacitance for the F m 0 6 overlap capacitance in the lightly doped region cf Fringing field capacitance for the gate to source F m calculated nC 6 drain clc Constant term for the short channel mode m O1x107 cle Exponential term for the short channel mode none 0 0 dic Length offset fitting parameter from C V m lint dwc Width offset fitting parameter from C V m wint Table 143 MOSFET Level 60 Temperature Parameters SPICE Description Unit Default See Symbol Table 144 tnom Temperature at which simulation expects C 27 parameters ute Mobility temperature exponent none 1 5 kt1 Temperature coefficient for the threshold V 0 11 voltage kt11 Channel length dependence of the V m 0 0 temperature coefficient for the threshold voltage kt2 Body bias coefficient of the Vth temperature none 0 022 effect 502 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model Table 143 MOSFET Level 60 Temperature Parameters Continued SPICE Description Unit Default See Symbol Table 144 ual Temperature coefficient for U4 m V 4 31e 9 ub1 Temperature coefficient for Up m V 2 61e 18 uct Temperature coefficient for Uc 1 V 056 nT 1 at Temperature coefficient for the saturation m sec 3 3e4 velocity cthO Normalized thermal capacity moC W sec O0 prt Temperature coefficient for Rdsw Q um 0
297. duction factor LUO um V 0 0 Length sensitivity WUO um v 0 0 Width sensitivity U1 1 V 0 0 Drain field mobility reduction factor LU1 um V 0 0 Length sensitivity WU1 um v 0 0 Width sensitivity VDDM V 5 0 Critical voltage for the high drain field mobility reduction VFBO VFB V 0 3 Flatband voltage 350 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modified BSIM Model Table 87 Transistor Process Parameters Continued Name Alias Units Default Description LVFB V um 0 0 Length sensitivity WVFB V um 0 0 Width sensitivity WFAC 4 Weak inversion factor LWFAC um 0 0 Length sensitivity WWFAC um 0 0 Width sensitivity WFACU 0 0 Second weak inversion factor LWFACU um 0 0 Length sensitivity WWFACU um 0 0 Width sensitivity X2E 1 V 0 0 Vsb correction to the linear vds threshold coefficient LX2E um V 0 0 Length sensitivity WX2E um v 0 0 Width sensitivity X2M X2MZ cm V s 0 0 Vsb correction to the low field first order mobility LX2M LX2MZ um cm V s 0 0 Length sensitivity WX2M WX2MZ um cm V s 0 0 Width sensitivity X2U0 1 V 0 0 Vsb reduction to the GATE field mobility reduction factor LX2UO um V 0 0 Length sensitivity WX2UO um V 0 0 Width sensitivity X2U1 um V 0 0 Vsb reduction to the DRAIN field mobility reduction factor LX2U1 um V 0 0 Length sensitivity HSPICE MOSFET Models Manual 351 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modif
298. e rs Source resistance vsb Source to bulk voltage vds Drain to source voltage vgs Gate to source voltage A t tnom esi 1 0359e 10F m dielectric constant of silicon HSPICE MOSFET Models Manual 33 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Equivalent Circuits Table 5 Equation Variables and Constants Continued Variable Definition Quantity k 1 38062e 23 Boltzmann s constant q 1 60212e 19 electron charge t New temperature of model or element in K tnom tnom TNOM 273 15 This variable represents the nominal temperature of parameter measurements in K user input in C vt k t q vt tnom k tnom q Using the MOSFET Current Convention Figure 7 on page 34 shows the assumed direction of current flow through a MOS transistor To print the drain current use either I M1 or 11 M1 syntax 2 produces the gate current 3 produces the source current 4 produces the substrate current References to bulk are the same as references to the substrate Figure 7 MOSFET Current Convention N channel nd drain node 11 MI nb substrate node w 14 MI ng gate node Jj 12 MI ns source node 13 MI v 34 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Equivalent Circuits Using MOSFET Equivalent Circuits Simulators use three equivalent circuits to analyze MOSFETs DC Transient AC and noise equivalent
299. e see Wire Model for Poly and Metal Layers on page 347 Hesistances Leff r RSH Weff Capacitances c COX Leff Weff 2 CAPSW Leff Weff HSPICE MOSFET Models Manual 335 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 336 Temperature Effect t BEX MUZ t MUZ UPDATE 0 1 t NBEX zmus t zmus UPDATE 0 1 NBEX uo t uo UPDATE 2 tno 4 NFEX xul t xul tno zvfb t zvfb At TCV The following equation calculates the At value used in the preceding equations At t tnom Charge Based Capacitance Model The LEVEL 13 capacitance model conserves charge and has nonreciprocal attributes Using charge as the state variable guarantees charge conservation To obtain the total stored charge in each of the gate bulk and channel regions integrate the distributed charge densities area of the active region The XPART 40 60 model parameter or 0 100 in the saturation region partitions the channel charge into drain and source components This partitioning smoothly changes to 50 50 in the triode region XPART 0 selects 40 60 drain source charge partitioning in the saturation region That is 40 of the channel charge in the saturation region is at the source and 60 is at the drain XPART 1 selects 0 100 for drain source charge partitioning in the saturation region That is 100 of the channel charge in the saturation region is i at the so
300. e unit in K W default is 0 0 CTH Thermal capacitance unit in W s K default is 0 0 HSPICE MOSFET Models Manual 249 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 58 University of Florida SOI 250 Notes The default value for channel length L and width W is 1 0e 6 The present version supports only 4 nodes only floating body devices AB is typically zero specify it accordingly If you activate the self heating option on the model line RTH and CTH define the thermal impedance of the device Typical values are 5e3 for RTH and 1e 12 for CTH but these can vary widely from one device to another For M gt 1 you must specify W AD AS NRD NRS NRB PDJ PSJ RTH and CTH per gate finger The initial condition IC is in the following order Vds drain voltage Vgfs front gate voltage and Vbgs back gate voltage Level 58 FD SOI MOSFET Model Parameters The following tables describe Level 58 model parameters for the fully depleted FD SOI including parameter names descriptions units defaults and typical notes Table 50 MOSFET Level 58 Flag Parameters Parameter Unit Default Typical Value Description Level Level 57 for UFSOI NFDMOD 0 0 Model selector 0 FD BJT 1 1 Parasitic bipolar flag 0 off 1 on SELFT 0 0 Self heating flag 0 no self heating 1 approximate model 2 full self heating HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET
301. e 164 166 LEVEL 7 model 181 LEVEL 8 equations 183 model 182 impact ionization BSIM2 372 MOSFETs 61 equations 35 61 LEVEL 39 372 inactive devices See latent devices HSPICE MOSFET Models Manual X 2005 09 Index interfaces 533 internal routines 550 intrinsic model parameters 245 ion implanted devices 3 IS model parameter 40 isoplanar MOSFETs construction 30 31 width cut 32 silicon gate transistor 30 ISPICE LEVEL 6 model 168 J JS model parameter 40 JSW model parameter 40 JUNCAP model parameters 221 junction parameters 423 L lambda equations 160 161 LATD model parameter 43 latency option 12 latent devices 12 Lattin Jenkins Grove model 4 LD model parameter 43 LDAC parameter 96 LDIF model parameter 43 levels MOSFETs models 3 4 libraries 567 linear region equations 393 LRD model parameter 42 LRS model parameter 42 MBYPASS option 10 12 Meyer capacitance gate 80 model 64 modified 83 parameters 76 MJ model parameter 41 MJSW model parameter 41 mobility equations 391 parameters 601 Index curve fitting 122 LEVEL 2 122 125 LEVEL 5 125 reduction equations LEVEL 2 132 LEVEL 38 201 LEVEL 5 153 model names periods in 568 model parameters ACM 9 basic 108 121 intrinsic limits 245 MOSFETSs 13 172 174 177 177 LEVEL 59 484 range limit 428 scaling 11 model selection automatic 567 569 failure to find a model 568 program 9 568 See also automatic model selection syntax 13 MODE
302. e parameters according to zX X LX Leff WX Weff For example the following equation calculates the zero field surface mobility HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 38 IDS Cypress Depletion Model LUO WUO leff weff Note This model uses mostly micrometer units rather than meter units Units and defaults are often unique in LEVEL 38 The finite difference method calculates the ly derivatives that define the gm gds and gmbs small signal gains This model does not use the SCALM and DERIV options zUO UO LEVEL 38 Model Parameters MOSFET Level 38 uses the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters This level also uses the parameters described in this section which apply only to MOSFET Level 38 Table 44 Capacitance Parameters Name Alias Units Default Description AFC 1 0 Area factor for MOSFET capacitance CAPOP 6 Gate capacitance selector METO um 0 0 Metal overlap on gate LEVEL 38 Model Equations IDS Equations Depletion vgs vfb lt 0 2 ids bi fa 2K10 NI vde cav vgs fb vde YE cav y vde vsb Phid 3 vsb Phid 2 terit HSPICE MOSFET Models Manual 197 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 38 IDS Cypress Depletion Model Enhancement vgs vfb vde 50 ids Bl zKI0 NI vde 3 cav Y vde vsb Phid 2 2 vsb Phi
303. e temperature and the analysis temperature Analysis at TEMP applies only to thermally activated exponentials in the model equations You cannot adjust the model parameter values if you use TEMP Simulation assumes that you extracted the model parameters at TEMP because TEMP is both the reference and analysis temperature For model levels other than 4 BSIM1 and 5 BSIM2 in UCB SPICES simulation adjusts the key model parameters for the difference between TEMP default 27 C and TNOM To specify TEMP in the netlist use TEMP as in the Level 39 MOSFET model In contrast to UCB SPICE s BSIM models the Synopsys Level 39 MOSFET model does provide for temperature analysis The default analysis temperature is 25 C in the Level 39 model Set TEMP in your netlist to change the analysis temperature you cannot use TEMP as a model parameter The Level 39 MOSFET model adjusts the temperature of the key model parameters as explained in Temperature Effect on page 336 Parasitics ACM gt 0 invokes the MOS source drain parasitics in the Level 39 MOSFET device model ACM 0 default is SPICE style See Synopsys Device Model Enhancements on page 371 Selecting Gate Capacitance CAPOP 39 selects the BSIM2 charge conserving capacitance model as shipped with Berkeley SPICE 3E2 This is the default selection if you set SPICE3 1 XPART charge sharing flag is currently not a BSIM2 model parameter despite its specification in the sample BSI
304. echnical Summary of MOSFET Models Temperature Parameters and Equations Parameter Description RD V2 Hz RS V Hz RX ID V2 Hz FN V2 Hz Output thermal noise due to drain resistor Output thermal noise due to source resistor Transfer function of channel thermal or flicker noise to the output This is not a noise it is a transfer coefficient reflecting the contribution of channel thermal or flicker noise to the output Output channel thermal noise ID RX P channel thermal noise Output flicker noise FN RX P flicker noise TOT V2 Hz Total output noise TOT RD RS ID FN Temperature Parameters and Equations 98 Temperature Parameters The following temperature parameters apply to all MOSFET model levels and the associated bulk to drain and bulk to source MOSFET diode within the MOSFET model The TLEV and TLEVC parameters select the temperature equations used to calculate the temperature effects on the model parameters Table 23 Temperature Effects Parameters Name Units Default Description Alias BEX 1 5 Low field mobility UO temperature exponent CTA 1 K 0 0 Junction capacitance CJ temperature coefficient If TLEVC 1 CTA overrides the default temperature compensation HSPICE MOSFET Models Manual X 2005 09 Table 23 2 Technical Summary of MOSFET Models Temperature Parameters and Equations Temperature Effects Parameters Continued Name Alias
305. ed Current Vp Vs D age L t 4 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Velocity Saturation Voltage Vc UCRIT NS Lig Note This equation accounts for the NS multiple series device number n gp 1 Voss Ve iudei Note The Vpgs variable is half the value of the actual saturation voltage Drain to source Saturation Voltage for Reverse Normalized Current M Pa V 4 3 mtp d bg L 0 6 Channel length Modulation Vpss 1 Vp Vs AV 24 V LAMBDA Jip Vj Vis Vine AV Vig Vpsg AV Los Bk XJ COX Vue NG AL LAMBDA L In 1 TE Lc UCRIT Equivalent Channel Length Including Channel length Modulation and Velocity Saturation V V ALAB PP ef UCRIT 25 10 L NS L Lain NS L Note These equations also account for the NS multiple series device number HSPICE MOSFET Models Manual 237 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model 238 2 min Do UE Lug Note This equation prevents the equivalent channel length from becoming zero or negative Reverse Normalized Current Reverse normalized current i 2 92 2 7 2 2 V t Reverse normalized current for the mobility model intrinsic charges capacitances the thermal noise model and the NQS time constant Ve Vp i NG t Transconductance Factor and Mobil
306. ed and a residual DC current exists If Vsp is large enough to make vih gt Vinn then vip is the inversion threshold voltage To determine the residual current this model inserts viny into the las Vsat and mobility equation in place of vg except for Vgs in the exponential term of the subthreshold current The inversion threshold voltage at a specified vg is vinis which the following equation computes q NI cox vinth Ve V HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model Saturation Voltage Vgsat The following equation computes the saturation voltage Vasa y 4 1 2 Vsat Vgs Vat vch 3 L m Vas Vip vch v p Vasat 7 Vsat IF ECV is not equal to 1000 V um the Synopsys device models modify Vsa to include the carrier velocity saturation effect 2 1 2 Vasat Vsat t Vc Osat ve The following equation calculates the v value used in the preceding equation Mobility Reduction UB gry The surface mobility UB depends on terminal voltages as follows 1 FRC E eee EESBR an UB TOX VST le i UB The following equations calculate values used in the preceding equation L Lep Linear region e L L y AL Saturation region e The next section describes the AL channel length modulation effect Channel Length Modulation Modify the ly current to model the channel length modulation effect I Het Leg HSPICE
307. ee 231 Parameter Preprocessing 00 0 cece e eee eese 231 Bulk Referenced Intrinsic Voltages A ee eae 233 Effective Channel Length and Width liuius 234 Short Distance Matching 000 isle 234 Reverse Short channel Effect RSCE 234 Effective Gate Voltage Including RSCE 234 Effective substrate factor including charge sharing for short and narrow channels oce pie eua Eme I DEN Eo DARET 235 Pinch off Voltage Including Short Channel and Narrow Channel Effects 235 Slope Factors co rELIQeED Sebbene bee pp AA a penes 236 Large Signal Interpolation Function 0000 00a 236 Forward Normalized Current 0 000 cece eee eee 236 Velocity Saturation Voltage 0000 eee 237 Drain to source Saturation Voltage for Reverse Normalized Current 237 Channel length Modulation llli 237 Equivalent Channel Length Including Channel length Modulation and Velocity Saturation 0 00 eee 237 Reverse Normalized Current 0000 cece ee eee 238 Transconductance Factor and Mobility Reduction Due to Vertical Field 238 Specific Current 00 eee 239 Drain to source Current 239 Transconductances 00 2 cee ee teens 240 Impact lonization Current liliis 240 Quasi static Model Equations ccc eee eee 241 Dynamic Model for the Intrinsic Node Charges 241 Intrinsic Capacitances
308. ee 522 Creating the Directory Environment llli eeu 522 Preparing Model Routine Files llle 523 Compiling the Shared Library a 524 Choosing a Compiler 00 c cee eens 524 Runtime Shared Library Path 0 000 eee eee ee 526 Troubleshooting 22000 eee e eee eee eee 526 MOS Models on PC Platforms 0 0000 cece eee een eee 527 Testing Customer CMI Models 0 00 cece eee eee eee 528 Model Interface Routines liliis 529 Interface Variables llli 533 pModel plnstance ees 534 COMI ResetModel 0 a 535 xiv HSPICE MOSFET Models Manual X 2005 09 Contents CMI_ResetInstance eee eens COMI AssignModelParm cece eee CMI_AssignInstanceParm 000 00 e eee CMI_SetupModel 0 000 c eee eee COMI Setuplnstance 00 0 0 ee COMI Eval ate e da dp CMI BiodeEval 4 3 erae uerb ror bet ded x nex es CMIZNOISO S mo Ivo de gest eate ec d ade UNIES GMIPrintMode l kaaga aa Eon ES s e DLE IAE t RP COMI FreeModel 0 eee eens CMI Freelnstance 0 a CMI Writ Efron 3 2 katad ey eph isi leesdrer thd malala ca sete A Lalala CMI Start GMI Conclude esi ROLE EVEN Eee ee a Customer CMI Function Calling Protocol aa Internal Routines llle RR Extended Topology eres Enhancemnets for Customer CMI 0000 a Gate Direct
309. ee the range column in the parameter tables then simulation sets its value to the closest acceptable value Intrinsic Parameters Temperature Dependence VTO T VTO TCV T T nom T BEX T NUCEX KP T KP UCRIT T UCRIT Trom d som T T T PHI T PHI 3 V n 4 ET sE T i d on i i Taom gl nom Taom j gl IBB T IBB 10 IBBT T T sano Bulk Referenced Intrinsic Voltages Simulation refers all voltages to the local substrate see Bulk Reference and Symmetry on page 226 Vg Vgg Vgs Vps Intrinsic gate to bulk voltage V Vsp Vgs Intrinsic source to bulk voltage Vp Vpg Vps Ves Intrinsic drain to bulk voltage For P channel devices simulation inverts all signs before processing HSPICE MOSFET Models Manual 233 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model 234 Effective Channel Length and Width Wy W DW Leg L DL Note Unlike the convention in other MOSFET models DL and DW usually permit a negative value due to the above definition Short Distance Matching The inverse of the square root of the transistor area usually suitably describes a random mismatch between two transistors with an identical layout and close to each other The following relationships have been adopted VIO BT INP Were NS Loge KP KP po D ANP Wo NS Log AGAMMA These model equations apply only in Monte Carlo and sens
310. efault temperature compensation PTP VPK 0 0 Junction potential PHP temperature coefficient If TLEVC 1 or 2 PTP overrides the default temperature compensation TCV VPK 0 0 Threshold voltage temperature coefficient Typical values are 1mV for n channel and 1mV for p channel TLEV 0 0 Temperature equation level selector Set TLEV 1 for ASPEC style Default is SPICE style If you invoke the ASPEC option the program sets TLEV for ASPEC TLEVC 0 0 Temperature equation level selector for junction capacitances and potentials Interacts with TLEV Set TLEVC 1 for ASPEC style Default is SPICE style If you invoke the ASPEC option the program sets TLEVC for ASPEC TRD 1 K 00 Temperature coefficient for drain resistor TRS 1 K 0 0 Temperature coefficient for source resistor XTI 0 0 Saturation current temperature exponent Use XTI 3 for silicon diffused junction Set XTI 2 for Schottky barrier diode MOS Temperature Coefficient Sensitivity Parameters Model levels 13 BSIM1 39 BSIM2 and 28 METAMOS include length and width sensitivity parameters as shown in Table 24 Use these parameters with the Automatic Model Selector to enable more accurate modeling for various 100 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models Temperature Parameters and Equations device sizes The default value of each sensitivity parameter is zero to ensure backward compatibility Table 24 MOS Temperature Co
311. effect parameter HSPICE MOSFET Models Manual X 2005 09 467 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model Table 130 MOSFET Level 57 DC Parameters Continued Parameter Unit Default Description Aely V m 0 Channel length dependency of the Early voltage for the bipolar current Agidl 1 W 0 0 GIDL constant ags 1 V 0 0 Gate bias coefficient of Apu Ahli 0 High level injection parameter for the bipolar current alpha0 m V 0 0 First parameter of the impact ionization current bO m 0 0 Bulk charge effect coefficient for the channel width b1 m 0 0 Bulk charge effect width offset beta0 1 V 0 0 First Vgs dependence parameter of the impact ionization current beta1 0 0 Second Vgs dependence parameter of the impact ionization current beta2 V 0 1 Third Vy dependence parameter of the impact ionization current Bgidl V m 0 0 GIDL exponential coefficient cdsc F m 2 4e 4 Drain source to the channel coupling capacitance cdscb F m 0 Body bias sensitivity of cdsc cdscd F m 0 Drain bias sensitivity of cdsc cit F m 0 0 Interface trap capacitance delta 0 01 Effective Vgs parameter drout 0 56 L dependence coefficient of the DIBL correction parameter in Rout dsub 0 56 DIBL coefficient exponent 468 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIMS3 SOI Model Table 130 MOSFET Level 57 DC Parameters Continued Paramete
312. efficient Sensitivity Parameters Sensitivity Parameters Parameter Description Length Width Product BEX Low field mobility UO temperature LBEX WBEX PBEX exponent FEX Velocity saturation temperature exponent LFEX WFEX PFEX TCV Threshold voltage temperature coefficient LTCV WTCV PTCV TRS Temperature coefficient for source resistor LTRS WTRS PTRS TRD Temperature coefficient for drain resistor LTRD WTRD PTRD Temperature Equations This section describes how to use temperature equations Energy Gap Temperature Equations These equations set the energy gap for temperature compensation TLEV Oor 1 tnom egnom 1 16 7 02e 4 inom 1108 0 t2 eg t 1 16 7 02e 4 74 1080 TLEV 2 tnom egnom EG GAP1 mom GAP 2 eg t EG GAPI TI GAP2 HSPICE MOSFET Models Manual 101 X 2005 09 2 Technical Summary of MOSFET Models Temperature Parameters and Equations 102 Saturation Current Temperature Equations isbd t isbd tnom e facln N isbs t isbs tnom e facin N The following equation calculates the facin value used in the preceding equations facln egnom amp 0 x77 in L vt tnom vt t tno MOSFET Diode Models on page 39 defines the isbd and isbs values MOS Diode Capacitance Temperature Equations TLEVC selects the temperature equation level for MOS diode capacitance TLEVC 0 7 Ca egnom _ eg t PB t PB VAP E n Rom Tod PH
313. efficient for the length dependence 0 0 of CGIDL 306 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS PWCGIDL Coefficient for the width dependence of 0 0 CGIDL PLWCGIDL Coefficient for the length times width 0 0 dependence of CGIDL POTBGIDL Coefficient for the geometry VK 1 3 638e 4 3 638e 4 independent part of STBGIDL PLTBGIDL Coefficient for the length dependence VK 1 O 0 of STBGIDL PWTBGIDL Coefficient for the width dependence of VK 1 O 0 STBGIDL PLWTBGIDL Coefficient for the length times width VK 1 0 0 dependence of STBGIDL The following are JUNCAP model parameters specifically for the Philips MOS 11 Level 63 model Table 65 Level 63 JUNCAP Parameters Name Description Units Default DTA Temperature offset of the JUNCAP element with respect toT 0G 0 VR Voltage at which simulation determines parameters V 0 JSGBHR Bottom saturation current density due to generating an electron Am2 1E 03 hole at V Vp JSDBR Bottom saturation current density due to diffusion from back Am2 1E 03 contact JSGSR Sidewall saturation current density due to generating an electron Am 1E 03 hole at V Vp HSPICE MOSFET Models Manual X 2005 09 307 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Mode
314. egative conductance 3 This model simulates the voltage outside the device s characterized range Charge based Gate Capacitance Model CAPOP 39 The BSIM2 gate capacitance model conserves charge and has non reciprocal attributes Using charges as state variables guarantees charge conservation Charge partitioning is fixed at 60 40 S D in saturation and is 50 50 in the linear region Qs Qg Qg Qp in all regions Accumulation region Vgg lt Vos VFB O g Cox Wo L A Vus E V VFB Q zo Q 0 HSPICE MOSFET Models Manual 369 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model Subthreshold region Vhs VFB lt Vgs lt Vth VGLOW O Con W opp L A Yos i YES VFB 2 po Ves Vos VFB H V s Vp VFB Vu Vp VEB VAS 3 Vag gt Vig VEB Vs Q g OF 0 Saturation region Vas gt Vasat 2 Q E 3 CoxWef Log Vost Q bulk The following equations calculate values used in the preceding equation 1 Opux 3CoxWorr LeglVin Vps VEB Q QO pink n DP 2 4 QO 10 3 Cox Weg Leg Yos 4 CoV op Lg Vost Linear region V lt Vasar s 7 a Col 2 dsaf dsa Oj 3 CoxWef Lege Vest V Ori 2 M L Visat sa 1 Q Qpuik oT 3 Cox Megp Leg Vest Visat Le V isa V f Visa u Vds 2 E Vas z L Visit Qpuik 370 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model
315. el The UC Berkeley SOI model BSIM3SOl supports Fully Depleted FD Partially Depleted PD and Dynamically Depleted DD SOI devices of which BSIM3PD2 0 1 for PD SOI devices is Synopsys MOSFET Level 57 For a description of this model see the BSIM3PD2 0 MOSFET MODEL User s Manual at http www device eecs berkeley edu bsim3soi Level 57 uses the UCB Version 2 2 3 model which includes a separate set of the geometry dependence parameters LLC LWC LWLC WLC WWC and WWLO to calculate LeffCV and WeffC V Level 57 also includes a new Full Depletion FD module soiMod 1 This module provides a better fit to FD SOI devices As soiMod 0 default the model is identical to previous BSIMSOI PD models This module also includes gate to channel drain source tunneling currents and overlap components The following enhancements to the BSIMSOI PD version were made starting in the BSIMSOI 3 0 version If the self heating model is on simulation calculates the channel surface potential NDIF includes parameter value checking Simulation calculates the Vgsteff derivatives The default value of CTHO changed from 0 to 1E 5 The DELTOX parameter in the MOS active element M models the relative variation on the transconductance oxide thickness of the MOS in Monte Carlo analysis You can download the BSIMSOI3 0 source code equation parameter list and bug report from http www device eecs berkeley edu bsimsoi HSPIC
316. el 49 and 53 BSIM3v3 MOS Models Table 97 Basic Model Parameters MOSFET Levels 49 53 Continued Name Unit Default Bin Description CDSC F m 2 4e 4 Yes Drain source and channel coupling capacitance CDSCD F Vm2 0 Yes Drain bias sensitivity of CDSC CDSCB F Vm2 0 Yes Body coefficient for CDSC PCLM 1 3 Yes Coefficient of the channel length modulation values x 0 result in an error message and program exit PDIBLC1 0 39 Yes Coefficient 1 for the DIBL drain induced barrier lowering effect PDIBLC2 0 0086 Yes Coefficient 2 for the DIBL effect PDIBLCB 1 V 0 Yes Body effect coefficient of the DIBL effect coefficients DROUT 0 56 Yes Length dependence coefficient of the DIBL correction parameter in Rout PSCBE1 V m 4 24e8 Yes Exponent 1 for the substrate current induced body effect PSCBE2 V m 1 0e 5 Yes Coefficient 2 for the substrate current induced body effect PVAG 0 Yes Gate dependence of Early voltage DELTA V 0 01 Yes Effective Vds parameter ALPHAO m V 0 Yes First parameter of the impact ionization current BETAO V 30 Yes Second parameter of the impact ionization current RSH 0 0 ohm square No Source drain sheet resistance in ohm per square 418 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 98 AC and Capacitance Parameters MOSFET 49 53 Name Unit Default Bin Description XPART 0 No Charge partitioning rate flag default devi
317. els other than MOSFET diode and BJT models You cannot override SCALM in the model Scaling for LEVEL 25 and 33 When using the proprietary LEVEL 25 Rutherford CASMOS or LEVEL 33 National models the SCALE and SCALM options are automatically set to 1e 6 However if you use these models with other scalable models you must explicitly set the SCALE 1e 6 and SCALM 1e 6 options Bypassing Latent Devices Use the BYPASS latency option to decrease simulation time in large designs To speed simulation time this option does not recalculate currents capacitances and conductances if the voltages at the terminal device nodes have not changed The BYPASS option applies to MOSFETs MESFETs JFETs BJTs and diodes Use OPTION BYPASS to set BYPASS BYPASS might reduce simulation accuracy for tightly coupled circuits such as op amps high gain ring oscillators and so on Use OPTION MBYPAS to set MBYPAS to a smaller value for more accurate results HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models General MOSFET Model Statement General MOSFET Model Statement You can use the MODEL statement to include a MOSFET model in your HSPICE netlist For a general description of the MODEL statement see the HSPICE Command Reference The following syntax is for all MOSFET model specifications All related parameter levels are described in their respective sections Syntax MODEL mname PMOS NMOS lt ENCMODE 0 15
318. em on the element line if Vbs lt 0 Cjbs Cj 1 Vbs Pb 3 Cjbssw Cjsw 1 Vbs Pbsw 138v Cjbsswg Cjswg 1 Vbs Pbswg Sg HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models else Cjbs Cj 1 Mj Vbs Pb Cjbssw Cjsw 1 Mjsw Vbs Pbsw Cjbsswg Cjswg 1 Mjswg Vbs Pbswg Bulk drain equations are analogous ACM 10 11 12 13 do not use the HSPICE equations for AS PS AD and PD In accordance with the BSIM3v3 model the default values for these area and perimeter factors are zero To invoke the HSPICE calculations for AS PS AD and PD specify the CALCACM 1 model parameter Note Simulation invokes CALCACM only if ACM 12 The calculations used in ACM 10 11 and 13 are not consistent with the Berkeley diode calculations CALCACM 1 and ACM 12 invoke the following area and perimeter calculations If you do not specify AD on the element line AD 2 HDIFeff Weff else AD AD WMLT 2 If you do not specify AS on the element line AS 2 HDIFeff Weff else AS AS WMLT 2 If you do not specify PS on the element line PS 4 HDIFeff 2 Weff else PS PS WMLT If you do not specify PD on the element line PD 4 HDIFeff 2 Weff else PD PD WMLT Note Weff is not the same Weff used in the BSIM3v3 and Levels 49 and 53 I V C
319. ement summary for the MOSFET Level 54 model HSPICE MOSFET Models Manual 437 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model TSMC Diode Model HSPICE MOSFET Level 54 BSIM4 supports a TSMC junction diode model You can use this TSMC junction diode model to simulate the temperature dependence source body and drain body currents of a junction diode Note For a complete description of this effect model visit the official UCB BSIM web site http www device eecs berkeley edu bsim3 You can order either of these models directly from Taiwan Semiconductor Manufacturing Company TSMC not from Synopsys See the TSMC web site http www tsmc com BSIM4 STI LOD HSPICE BSIM4 supports the full STI Shallow Trench Isolation or LOD Length of Diffusion induced mechanical stress effect model for version 4 3 or later which was first released in the UCB BSIM4 3 0 model version HSPICE BSIM4 turns on the simulation of this effect when the following conditions consistent with those of the UCB BSIM4 model are satisfied if VERSION gt 4 3 if SA gt O and SB gt O and NF 1 or SA gt O and SB gt 0 and NF 1 and SD gt 0 UCB s STI LOD model is turned on If VERSION gt 4 3 the STI model is not dependent on STIMOD 0 or 1 In this case the STI model is applied to the model similar to the UCB STI model see Table 109 When parameter values that satisfy SA gt 0 SB gt
320. emplt6 output template 6 double templt7 output template 7 560 HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Extended Topology double templt8 output template 8 double templt9 output template 9 double templt10 output template 410 double templti11 output template 411 double temp1t12 output template 412 double templt13 output template 413 double temp1t14 output template 414 double templti15 output template 15 double templt16 output template 416 double templt17 output template 17 double templt18 output template 418 double templt19 output template 19 double templt20 output template 420 double templt21 output template 421 double templt22 output template 422 double templt23 output template 423 double templt24 output template 424 double templt25 output template 425 double templt26 output template 426 double templt27 output template 427 double templt28 output template 428 double templt29 output template 429 double templ
321. ength dependence of m 1E 6 1E 6 oSF SWSSF Coefficient of the width dependence of m 0 0 oSF ALPR Factor of the channel length modulation rH 1E 2 1E 2 for the reference transistor SLALP Coefficient of the length dependence ofa 1 1 ALPEXP Exponent of the length dependence ofa 1 1 SWALP Coefficient of the width dependence ofa m 0 0 VP Characteristic voltage channel length V 5E 2 5E 2 modulation LLMIN Minimum effective channel length in a m 1 5E 7 1 5E 7 technology calculates the m smoothing factor AiR Weak avalanche current factor for the 6 6 reference transistor at the reference temperature STA1 Temperature dependence coefficient of a1 K 1 0 0 SLA1 Coefficient of the length dependence of a m 0 0 SWA1 Coefficient of the width dependence of a1 m 0 0 A2R Exponent of the weak avalanche current V 38 38 for the reference transistor SLA2 Coefficient of the length dependence of a2 Vm 0 0 284 HSPICE MOSFET Models Manual X 2005 09 Table 62 Level 68 MOS11 Parameters Level 1100 Continued 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Name Description Units NMOS PMOS SWA2 Coefficient of the width dependence of a2 Vm 0 0 A3R Drain source voltage factor above which 1 1 weak avalanche occurs for the reference transistor SLA3 Coefficient of the length dependence of a3 m 0 0 SWA3 Coefficient of the width dependence of a3 m 0 0 IGINVR Gain factor for the intrinsic gate tunneling AV 2 0 0 current in the
322. ength sensitivity Use this parameter with automatic model selection WRD and PRD to factor a model for the device size WRD ohm m 0 Drain resistance width sensitivity used with LRD PRD ohm m2 0 Drain resistance product area sensitivity used with LRD RS ohm sq 0 0 Source ohmic resistance This parameter is usually the sheet resistance of a lightly doped region for ACM gt 1 LRS ohm m 0 Source resistance length sensitivity Use this parameter with automatic model selection WRS and PRS to factor a model for the device size WRS ohm m 0 Source resistance width sensitivity used with LRS PRS ohm m2 0 Source resistance product area sensitivity used with LRS RSC ohm 0 0 Source resistance due to contact resistance RSH RL ohm sq 0 0 Drain and source diffusion sheet resistance 42 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models Table 10 Using MOS Geometry Model Parameters Name Alias Units Default Description HDIF m 0 Length of heavily doped diffusion from contact to lightly doped region ACM 2 3 only HDIFwscaled HDIF SCALM LD DLALLATD m Lateral diffusion into the channel from the source and drain diffusion f you do not specify LD and XJ LD default 0 0 If you specify LD but you do not specify XJ then simulation calculates LD from Xu Default 0 75 XJ ForLEVEL 4 only lateral diffusion is derived from LD XJ LDscaled
323. ent analysis The GMINDC option creates a parallel conductance across the bulk diodes and drain source for DC analysis These options enhance the convergence properties of the diode model especially when the model has a high off resistance Use the RSH RS and RD parameters to prevent over driving the diode in either a DC or transient forward HSPICE MOSFET Models Manual 39 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models bias condition These parameters also enhance the convergence properties of the diode model MOSFET Diode Model Parameters Table 7 DC Model Parameters Name Alias Units Default Description ACM 0 Area calculation method JS amp m2 0 Bulk junction saturation current JSscaled JS SCALM2 for ACM 1 unit is amp m and JSscaled JS SCALM JSW amp m 0 Sidewall bulk junction saturation current JSWscaled JSW SCALM IS amp 1e 14 Bulk junction saturation current For the ASPEC 1 option default 0 N 1 Emission coefficient NDS 1 Reverse bias slope coefficient VNDS V 1 Reverse diode current transition point Table 8 Capacitance Model Parameters Name Alias Units Default Description CBD F 0 Zero bias bulk drain junction capacitance Used only when CJ and CJSW are 0 CBS F 0 Zero bias bulk source junction 40 capacitance Use only when CJ and CJSW are 0 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models Table 8 Cap
324. ent release that you are using The VERSION model parameter selects among the various Berkeley releases of BSIM3v3 as follows Version 3 0 Berkeley release October 30 1995 default for HSPICE96 1 96 2 96 3 Simulation invokes this version if you specify VERSION 3 0 and HSPVER 98 0 To invoke the Synopsys model version that most accurately represents the Berkeley release of October 1995 specify the VERSION 3 0 and HSPVER 98 0 parameters Version 3 1 Berkeley December 9 1997 default for HSPICE97 1 97 2 97 4 Simulation invokes this version if you specify VERSION 3 1 or 3 11 and HSPVER 98 0 To invoke the Synopsys model version that most accurately represents the Berkeley release of December 1996 specify the VERSION 3 1 or 3 11 and HSPVER 98 0 parameters Berkeley Version 3 0 3 1 bug fixes Berkeley corrected several Version 3 0 and 3 1 bugs in the June 1998 release These fixes are incorporated into HSPICE98 2 which simulation uses if you specify VERSION 3 0 or VERSION 3 1 and HSPVER 98 2 As a result of bug fixes you might notice some differences between Version 3 0 3 1 in HSPICE98 2 and previous Version 3 0 3 1 releases Notably differences occur if you specify PD and PS perimeter factors that are less than Weff PD PS Weff no longer clamp to Weff in Version 3 1 and if DLC and LINT are not identical LeffCV calculation bug in Versions 3 0 and 3 1 For a complete list of bug fixes see the BSIMS web site http www d
325. epletion Effect PGD1 PGD2 PGD3 0 Channel Length Modulation CLM1 CLM2 CLM3 0 Narrow Channel Effect WFC MUEPH2 0 322 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 Lists and describes three of the earliest BSIM type MOSFET models supported by HSPICE This chapter describes three of the earliest Berkeley Short Channel IGFET BSIM type MOSFET device models that HSPICE supports LEVEL 13 BSIM Model LEVEL 28 Modified BSIM Model LEVEL 39 BSIM2 Model These models are all based on models developed by the University of California at Berkeley You can find documentation on BSIM3 and BSIM4 at this website http www eigroup org cmc cmos default htm For descriptions of the newest BSIM models that Synopsys supports see Chapter 7 BSIM MOSFET Models Levels 47 to 65 HSPICE MOSFET Models Manual 323 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model LEVEL 13 BSIM Model 324 LEVEL 13 is based on the SPICE 2G 6 BSIM model which models the device physics of small geometry MOS transistors To invoke the subthreshold region set the NO model parameter low field weak inversion gate drive coefficient to less than 200 Level 13 provides three MOSFET models Wire resistor model compatible with the SPICE BSIM interconnect model for polysilicon and metal layers Simulates resistors and capacitors with interconnects Capacitor model Simulate
326. er DC models with new capacitance equations without having to move to a new DC model You can use the CAPOP model parameter to select the gate capacitance and validate the effects of different capacitance models HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models The CAPOP capacitance model selection parameter selects the capacitor models to use for modeling the MOS gate capacitance gate to drain capacitance gate to source capacitance gate to bulk capacitance You can use CAPOP to select several versions of the Meyer and charge conservation model Some capacitor models are tied to specific DC models DC model level in parentheses below Other models are designated as general any DC model can use them Parameter Description CAPOP 0 SPICE original Meyer gate capacitance model general CAPOP 1 Modified Meyer gate capacitance model general CAPOP 2 Parameterized Modified Meyer gate capacitance model general default CAPOP 3 Parameterized Modified Meyer gate capacitance model with Simpson integration general CAPOP 4 Charge conservation capacitance model analytic Levels 2 3 6 7 13 28 and 39 only CAPOP 5 No capacitor model CAPOP 6 AMI capacitor model LEVEL 5 CAPOP 9 Charge conservation model LEVEL 3 CAPOP 11 Ward Dutton model specialized LEVEL 2 CAPOP 12 Ward Dutton model specialized LEVEL 3 CAPOP 13 Generic BSIM Charge Conserving
327. er Description GEOMOD Geome ry dependent parasitics model selector specifies how the end S D diffusions connect IC Initial guess in the order L BSIM4 MOSFET channel length in meters MIN Whether to minimize the number of drain or source diffusions for even number fingered device mname MOSFET model name reference MULUO Low field mobility UO multiplier nb Bulk terminal node name nd Drain terminal node name NF Number of device fingers ng Gate terminal node name NRD Number of drain diffusion squares NRS Number of source diffusion squares ns Source terminal node name OFF Sets the initial condition to OFF in DC analysis PD Perimeter of the drain junction f PERMOD 0 it excludes the gate edge Otherwise it includes the gate edge PS Perimeter of the source junction f PERMOD 0 it excludes the gate edge Otherwise it includes the gate edge RBDB Resistance connected between dbNode and bNode RBODYMOD Substrate resistance network model selector HSPICE MOSFET Models Manual 435 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model 436 Parameter Description RBPB Resistance connected between bNodePrime and bNode RBPD Resistance connected between bNodePrime and dbNode RBPS Resistance connected between bNodePrime and sbNode RBSB Resistance connected between sbNode and bNode RDC Drain contact resistance for per finger device RGATEMOD Gate resistance model selector RGEOMOD
328. er LD or METO but you do not specify CGDO then simulation calculates CGDO CGSO F 1 0p TFT gate to source overlap capacitance 40 F m Gate to source overlap capacitance 39 If you specify TOX but you do not specify either LD or METO and you do not specify CGSO then simulation calculates CGSO CHI 0 5 Temperature exponential part 40 CJ F m2 0 Source drain bulk zero bias junction 39 capacitance CJSW F m 0 Sidewall junction capacitance 39 CLM GDS 0 0 Selects a channel length modulation 6 7 8 equation COX F m2 3 453e 4 Oxide capacitance per unit gate area If yu 1 2 3 do not specify COX simulation calculates it from TOX CSC F m 10u Space charge capacitance 40 110 HSPICE MOSFET Models Manual X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 25 Basic MOSFET Model Parameters Continued Name Alias Units Default Description Level DEFF 2 0 Drain voltage effect for the TFT leakage 40 current DERIV 1 Derivative method selector 3 39 DERIV 0 analytic DERIV 1 finite difference DP um 1 0 Implant depth depletion model only 5 38 ECRIT V cm 0 0 Critical electric field for the carrier velocity 2 8 ESAT saturation From Grove electrons 6e4 holes 2 4e4 Zero indicates an infinite value The ECRIT equation is more stable than VMAX Simulation estimates ECRIT as ECRIT 100 VMAX UO V cm 0 0 Drain source critical field Zero indicates an 6 7 infinite value typical
329. er for seff 0 seffwe Exponent for w dependence for seff 0 sefft Temperature adjustment coefficient for seff 0 jro S D diode SCR recombination coef A cm 2 1e6 jroll I dependence parameter for jro 0 jrole Exponent for I dependence for jro 0 jrowl W dependence parameter for jro 0 jrowe Exponent for w dependence for jro 0 jrot Temperature adjustment coefficient for jro calculated m S D diode recombination slope factor 2 0 jgo S D diode SCR generation coefficient A cm 2 0 jgoll I dependence parameter for jgo 0 jgole Exponent for I dependence for jgo 0 jgowl Exponent for I dependence for jgo 0 jgowe Exponent for w dependence for jgo 0 jgot Temperature adjustment coefficient for jgo 0 mg S D diode generation slope factor 2 0 seffs S diode QNR recombination velocity asymmetrical cm s seff seffsll I dependence parameter for seffs seffll seffsle Exponent for dependence for seffs seffle seffswl W dependence parameter for seffs seffwl HSPICE MOSFET Models Manual X 2005 09 513 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Table 147 8 SSIMSOI Model Parasitic Parameters Continued Name Parameter Units Default seffswe Exponent for w dependence for seffs seffwe seffst Temperature adjustment coefficient for seffs sefft jros S diode SCT recombination coefficient asymmetrical A cm 2 jro jrosll I dependence parameter for jro jroll jrosle Exponent for I dependence for jro jrole jroswl W dependence parameter for jr
330. er of parameters 12 11 17 18 28 Physical parameters 50 55 12 11 7 Continuous derivatives no no yes yes yes Positive GDS yes yes no yes no Monotonic GM IDS no no no yes yes Outline of Optimization Procedure 1 Extract XL LD XW WD TOX RSH CGSO CGDO CGBO Cu Mu CJSW and MJSW from resistor and capacitor data and plots of Beta versus W and L 2 For each W L device e Extract VT versus VBS from IDS versus VGS data e Calculate ETA from log IDS versus VGS plots at VDS 0 1 5 0 e Fit VT parameters to the VT versus VBS data e Optimize other parameters except L and W sensitivity to IDS GDS and GM versus VGS VDS and VBS data 3 For each W L device calculate L and W sensitivity parameters from the optimized parameters of nearby devices 4 Fitthe models together into one model using the Lmin Lmax Wmin Wmax feature of the MOSFET models HSPICE MOSFET Models Manual 579 X 2005 09 B Comparing MOS Models Examples of Data Fitting Examples of Data Fitting 580 The following plots fit Levels 2 3 13 28 and 39 to data from a submicron device fabricated using a modern CMOS process All models were optimized to the same data Similar optimization files were used optimizing different parameters All models except LEVEL 39 use the impact ionization model with ALPHA and VCR parameters Level 39 has its own impact ionization parameters To avoid negative GDS LEVEL 13 uses improved optimization of the para
331. erature Equations Saturation Current Temperature Equations MOS Diode Capacitance Temperature Equations Surface Potential Temperature Equations Threshold Voltage Temperature Equations HSPICE MOSFET Models Manual X 2005 09 71 71 71 72 72 73 73 74 77 77 78 78 79 79 80 80 81 82 83 83 84 85 86 88 92 93 95 95 95 95 95 96 98 98 100 101 101 102 102 104 105 Contents Mobility Temperature Equations 200 0c eee eae 105 Channel Length Modulation Temperature Equation 105 Calculating Diode Resistance Temperature Equations 106 3 Common MOSFET Model Parameters aaaaaaaa 107 Basic MOSFET Model Parameters 000 0c eee een eee 108 4 Standard MOSFET Models Level 1 to 40 127 LEVEL 1 IDS Schichman Hodges Model liuius 128 LEVEL 1 Model Parameters 00 00 cece eee eee eee 128 LEVEL 1 Model Equations aana 128 IDS EQuations d ose dp EV Ei 128 Effective Channel Length and Width 0 0005 129 LEVEL 2 IDS Grove Frohman Model 000 0c eee eee 130 LEVEL 2 Model Parameters 0 000 cece eee eee 130 LEVEL 2 Model Equations 0 0 00 cece eee 130 IDS Equations o vs paces eh oe aaa sede vee ho yo 130 Effective Channel Length and Width isses 131 Threshold Voltage Vij cee I 131 Saturation Voltage
332. ers Level 1100 Continued Name Description Units NMOS PMOS THER2 Denominator of the gate voltage V 1 1 dependent part of the series resistance for the reference transistor THESATR Velocity saturation parameter due to V 1 0 5 0 2 optical acoustic phonon scattering for the reference transistor at the reference temperature SLTHESAT Coefficient of the length dependence of 1 1 0SAT THESATEXP Exponent of the length dependence of 1 1 0SAT ETASAT Exponent of the temperature dependence 1 04 0 86 of OSAT SWTHESAT Coefficient of the width dependence of m 0 0 0SAT THETHR Coefficient of the self heating for the V 3 1E 3 1E 3 reference transistor at the reference temperature THETHEXP Exponent of the length dependence of 1 1 0TH SWTHETH Coefficient of the width dependence of m 0 0 0TH SDIBLO Drain induced barrier lowering parameter V 1 2 2E 3 1E 3 for the reference transistor SDIBLEXP Exponent of the length dependence of 1 35 1 35 oDIBL MOR Parameter for the short channel 0 0 subthreshold slope for the reference transistor HSPICE MOSFET Models Manual X 2005 09 283 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 62 Level 68 MOS11 Parameters Level 1100 Continued Name Description Units NMOS PMOS MOEXP Exponent of the length dependence of mO 1 34 1 34 SSFR Static feedback parameter for the V 1 2 6 25E 3 6 25E 3 reference transistor SLSSF Coefficient of the l
333. ers for this current which you can use for all levels The BSIM2 model includes a complex impact ionization model with AIO AIB BIO and BIB parameters In the Berkeley SPICES release this current was all assigned to the IDS drain to source current Using the ALPHA and VCR parameters in the MOSFET models simulation assigns the impact ionization HSPICE MOSFET Models Manual 577 X 2005 09 B Comparing MOS Models Ability to Simulate Process Variation current to IDB which is essential for cascode simulation The IIRAT parameter allows the model to divide the current between IDS and IDB if needed Ability to Simulate Process Variation Usually simulation performs full model parameter extraction or optimization only on a small number of test wafers For a large number of wafers in fabrication measurements gather statistical data about process variation for example TOX and simple electrical measurements for example VT This statistical data uses a worst case Monte Carlo or Taguchi methodology to represent the process variation Before you run this simulation you must modify models to account for variations in TOX thresholds line widths and sheet resistance Different MOSFET model levels use these parameters in similar ways All of the models discussed here accept the following parameters TOX DELVTO XL XW and RSH The DELVTO model parameter shifts the threshold Forthe LEVEL 2 and 3 models setting DELVTO 0 1 is equ
334. est BSIM type MOSFET models supported by HSPICE This chapter describes seven of the newest Berkeley Short Channel IGFET BSIM type MOSFET models that HSPICE supports Level 47 BSIM3 Version 2 MOS Model Level 49 and 53 BSIM3v3 MOS Models Level 54 BSIM4 Model Level 57 UC Berkeley BSIM3 SOI Model Level 59 UC Berkeley BSIM3 SOI FD Model Level 60 UC Berkeley BSIM3 SOI DD Model Level 65 SSIMSOI Model These models are all based on models developed by the University of California at Berkeley You can find documentation on BSIM3 and BSIM4 at this website http www eigroup org cmc cmos default htm For descriptions of older BSIM models that Synopsys supports see Chapter 6 BSIM MOSFET Models Levels 13 to 39 HSPICE MOSFET Models Manual 381 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model Level 47 BSIM3 Version 2 MOS Model The BSIMS version 2 0 MOS model from UC Berkeley is available as the Synopsys Level 47 model Table 91 MOSFET Level 47 Model Parameters Name Unit Default Description VTHO V 0 7 Threshold voltage of the long channel at V5 0 and small Vas 0 7 for n channel e 0 7 for p channel K1 y 0 53 First order body effect coefficient K2 0 0186 Second order body effect coefficient K3 80 0 Narrow width effect coefficient K3B 1 V 0 Body width coefficient of the narrow width effect KT1 V 0 11 Temperature coefficient for the threshold voltage KT2 0 022 Body bias coeffi
335. eter Description CLM 0 No channel length modulation default CLM 1 One sided step depletion layer drain field equation CLM 2 Frohman s electrostatic fringing field equation CLM 3 One sided step depletion layer drain field equation with carrier velocity saturation CLM 4 Wang s equation linearly graded depletion layer CLM 5 Synopsys channel length modulation CLM 6 Synopsys AL equations The following sections describe these equations and the associated model parameters CLM 0 No Channel Modulation Default AL 0 176 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model This is the default channel length equation representing no channel length modulation it corresponds to MSINC GDS 0 0 Table 35 CLM 1 Step Depletion Equation for MOSFET Level 6 Name Alias Units Default Description KL 0 0 Empirical constant saturation voltage LAMBDA LAM LA cm 2 1 137e 4 Channel length modulation If you do not specify s simulation calculates it from NSUB The default LAMBDA corresponds to the default NSUB value KL AL LAMBDA vds vdsar t2 184 vsat If you do not specify LAMBDA simulation calculates it as 2 8gsi Jue q DNB LAMBDA This is a one sided step depletion region formulation by Grove AL varies with the depletion layer width which is a function of the difference between the effective saturation voltage vd
336. eter Unit Default Bin Description SA 0 0 Distance between S D diffusion edge to poly gate edge instance from one side If not given or if lt 0 the stress effect is parameter turned off SB 0 0 Distance between S D diffusion edge to poly gate edge instance from the other side If not given or if lt 0 the stress parameter effect is turned off SD 0 0 Distance between neighboring fingers For NF 1 If not instance given or lt 0 stress effect is turned off parameter STIMOD 0 0 STI model selector which gives priority to the instance Also parameter instance 0 No STI effect parameter 1 UCB s STI model 2 TSMC s STI model SAREF M 1e 06 No Reference distance for SA gt 0 0 SBREF M 1e 06 No Reference distance for SB gt 0 0 WLOD M 0 0 No Width parameter for stress effect KUO M 0 0 No Mobility degradation enhancement coefficient for stress effect KVSAT M 0 0 No Saturation velocity degradation enhancement parameter for stress effect 1 0 lt kvsat lt 1 0 TKUO 0 0 No Temperature coefficient of KUO LKUO 0 0 No Length dependence of KUO WKUO 0 0 No Width dependence of KUO LLODKUO 0 0 No Length parameter for UO stress effect 50 HSPICE MOSFET Models Manual X 2005 09 405 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 93 Supported HSPICE BSIM3v3 STI LOD Parameters Continued Parameter Unit Default Bin Description WLODKUO 0 0 No Width parameter for
337. evice eecs berkeley edu bsim3 Note Version 3 11 was introduced in HSPICE97 4 This version represents Berkeley Version 3 1 Dec 1996 with HSPICE bug fixes This model maintains backward compatibility Starting with HSPICE98 2 Version 3 1 and 3 11 are identical and represent Version 3 1 with Berkeley June 1998 bug fixes Version 3 2 Berkeley release June 16 1998 Simulation invokes this version if you specify VERSION 3 2 and HSPVER 98 2 Version 3 2 1 Berkeley release April 20 1999 Simulation invokes this version if you specify VERSION 3 21 and HSPVER 99 2 Version 3 2 2 Berkeley release April 20 1999 Simulation invokes this version if you specify VERSION 3 22 and HSPVER 99 2 For the latest HSPICE improvements use VERSION 3 24 and HSPVER 02 2 HSPICE MOSFET Models Manual 399 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models 400 Note Versions 3 2 1 and 3 2 2 are identical except BSIM3v3 2 1 uses a bias dependent Vfb and BSIM3v3 2 2 uses a bias independent Vfb for the capMod 1 and 2 capacitance models Versions 3 23 and 3 2 4 provide various model fixes compared to Version 3 22 Table 92 Parameter Settings for Berkeley Releases MOSFET Levels 49 53 Berkeley Release VERSION HSPVER Version 3 0 October 1995 3 0 98 0 Version 3 0 with June 1998 bug fixes 3 0 98 2 Version 3 1 December 1996 3 1 98 0 Version 3 1 with June 1998 bug fixes 3 1 98 2 Version 3
338. ewall of the source drain which is not under the gate 10 Philips MOS11 has its own LMIN parameter which has a different definition from that of HSPICE To avoid the conflict with LMIN in simulation Synopsys changed the LMIN parameter in the Level 63 MOSFET model to LLMIN Description of Parameters Table 62 Level 63 MOS11 Parameters Level 1100 Name Description Units NMOS PMOS LEVEL Level of this model 63 VERSION Version of this model 1100 HSPICE MOSFET Models Manual 279 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 62 Level 68 MOS11 Parameters Level 1100 Continued Name Description Units NMOS PMOS LER Effective channel length reference m 1E 6 1E b transistor WER Effective channel width reference m 1E 5 1E 5 transistor LVAR Difference between the actual and m 0 0 programmed poly silicon gate length LAP Effective channel length reduction per m 4E 8 4E 8 side due to lateral diffusion of the source drain dopant ions WVAR Difference between the actual and the m 0 0 programmed field oxide opening WOT Effective reduction of channel width per m 0 0 side due to the lateral diffusion of channel stop dopant ions TR Temperature at which simulation 21 21 determines the parameters for the reference transistor VFBR Flat band voltage for the reference V 1 05 1 05 transistor at the reference temperature STVFB Coefficient of temperature dependence V K 5E 4 5E 4 f
339. f M PS WMLT SCALE Calculating Effective Saturation Current For ACM 0 simulation calculates the MOS diode effective saturation currents as follows Source Diode Saturation Current Define val JSscaled ASeff JSWscaled PSeff If val gt O then isbs val Otherwise isbd M IS Drain Diode Saturation Current Define val JSscaled ADeff JSWscaled PDeff If val gt O then isbd val Otherwise isbd M IS Calculating Effective Drain and Source Resistances For ACM 0 simulation calculates the effective drain and source resistances as follows Source Hesistance Define val NRS RSH val 4 RSC If val 0 then RSeff val eff M HSPICE MOSFET Models Manual 45 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models 46 RS RSC Otherwise RSeff 37 Drain Resistance Define val 2 NRD RSH If val gt 0 then RDeff LDAP Otherwise RDeff KOLA Using an ACM 1 MOS Diode If you specify the ACM 1 model parameter simulation uses ASPEC style diodes and does not use the AD PD AS and PS parameters The JS and CJ units differ from SPICE style diodes ACM 0 Figure 12 ACM 1 MOS Diode Source Drain Gate gt k LD A k LDIF Contact Example Table 11 lists parameter value settings for a transistor with the following parameter values LD 0 5 um W 10 um L 3 um
340. fect the common MOS parameters such as XL LD XW WD CJ CUSW JS and JSW Level 47 uses the common Synopsys MOS parasitic models which ACM specifies Level 47 uses the common Synopsys MOS noise models which NLEV specifies You can use DELVTO and DTEMP on the element line with MOSFET Level 47 The PSCBE1 and PSCBE2 model parameters determine the impact ionization current which contributes to the drain source current Impact ionization does not contribute to the bulk current HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model Leff and Weff Equations for BSIM3 Version 2 0 The standard equations for Leff and Weff in the Synopsys model are Lag L XL 2 LD Wop W XW 2 WD BSIM3 uses the following UCB SPICE3 equations Lag L 2 DL Wer The units for these parameters are meters with defaults of zero W 2 DW Simulation uses the standard Synopsys model equation for both cases and accepts DL DW as the value for LD WD If you specify both LD WD and DL DW in a MODEL statement simulation uses the LD WD value If you specify LDAC and WDAC in the MODEL statement then Leff L XL 2 LDAC Weff W4 XW 2 WDAC This model uses the LD DL and WD DW values to generate defaults for CGSO CGDO and CGBO You can also use these values with the RS and RD parameters for ACM gt 0 Example The following two models return the same simulation results
341. fects at this level Use LEVEL 5 and LEVEL 38 for depletion MOS devices LEVEL 2 models consider bulk charge effects on current LEVEL 3 models require less simulation time provides as much accuracy as LEVEL 2 and have a greater tendency to converge LEVEL 6 models are compatible with models originally developed using ASPEC Use LEVEL 6 models to model ion implanted devices Selecting Models HSPICE MOSFET Models Manual X 2005 09 To describe a MOS transistor in your netlist use both an element statement and a MODEL statement The element statement defines the connectivity of the transistor and references the MODEL statement The MODEL statement specifies either an n or p channel device the level of the model and several user selectable model parameters 1 Overview of MOSFET Models Selecting Models Example The following example specifies a PMOS MOSFET PCH is a model reference name The transistor is modeled using the LEVEL 13 BSIM model Select the parameters from the MOSFET model parameter lists in this chapter M3 3 2 1 0 PCH lt parameters gt MODEL PCH PMOS LEVEL 13 lt parameters gt Selecting MOSFET Model LEVELs MOSFET models consist of private client and public models that you select using the LEVEL parameter in the MODEL statement Synopsys frequently adds new models to the MOSFET device models Not all MOSFET models are available in the PC version Table 1 shows what is available fo
342. ffusion drawn dimension for nitride 1 2 4 7 Nitride layer width after etch 3 1 Periphery of the diode CD represents the cut from the source to the drain Figure 5 on page 31 and includes the contacts 30 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models Field Effect Transistors Figure 5 Isoplanar MOSFET Construction Part B Intermediate Oxide Contact 10 73 l Center I Gate Gate Oxide onde _ Drain A N Field Implant mi baga LONGUM dE I IDI 901 2 5 6 7 8 Drawn channel length L 2 5 Actual poly width after etching L XL where XL O 3 4 Effective channel length after diffusion L XL LD 4 5 Lateral diffusion LD 9 10 Diffusion periphery for diode calculations 5 6 Gate edge to center contact for ACM 1 and ACM 2 calculations The planar process produces parasitic capacitances at the poly to field edges of the device The cut along the width of the device demonstrates the importance of these parasitics Figure 6 on page 32 The encroachment of the field implant into the channel narrows the channel width and increases the gate to bulk parasitic capacitance HSPICE MOSFET Models Manual 31 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Equivalent Circuits Figure 6 Isoplanar MOSFET Width Cut Oxide 2 Sh Internediate Oxide l l Polysilic
343. field mobility at T TREF TNOM 250 pmos UA m V 2 25e 9 Yes First order mobility degradation coefficient 416 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 97 Basic Model Parameters MOSFET Levels 49 53 Continued Name Unit Default Bin Description UB m2 V2 5 87e 19 Yes Second order mobility degradation coefficient UC 1 V 4 65e 11 or Yes Body bias sensitivity coefficient of mobility 0 0465 4 65e 11 for MOBMOD 1 2 or 0 0465 for MOBMOD 3 AO 1 0 Yes Bulk charge effect coefficient channel length AGS 1 V 0 0 Yes Gate bias coefficient of Abulk BO m 0 0 Yes Bulk charge effect coefficient channel width B1 m 0 0 Yes Bulk charge effect width offset KETA 1 V 0 047 Yes Body bias coefficient of the bulk charge effect VOFF V 0 08 Yes Offset voltage in the subthreshold region VSAT m sec 8e4 Yes Saturation velocity of the carrier at T TREF TNOM Al 1 V 0 Yes First nonsaturation factor A2 1 0 Yes Second nonsaturation factor RDSW ohm um 0 0 Yes Parasitic source drain resistance per unit width PRWG 1N 0 Yes Gate bias effect coefficient of RDSW PRWB 1172 0 Yes Body effect coefficient of RDSW WR 1 0 Yes Width offset from Weff for the Rds calculation NFACTOR 1 0 Yes Subthreshold region swing CIT F m 0 0 Yes Interface state capacitance HSPICE MOSFET Models Manual X 2005 09 417 7 BSIM MOSFET Models Levels 47 to 65 Lev
344. for GIDL BGIDL 2 3e9V m Yes Exponential coefficient for GIDL CGIDL 0 5V3 Yes Parameter for the body bias effect on GIDL DGIDL 0 8V Yes Fitting parameter for band bending for GIDL Table 119 Gate Dielectric Tunneling Current Model Parameters MOSFET Level 54 Parameter Default Binnable Description AIGBACC 943 E Yes Parameter for ly in the accumulation BIGBACC 9 954 ES miv Yes Parameter for gp in the accumulation CIGBACC 9 975 Yes Parameter for gpin the accumulation NIGBACC 1 0 Yes Parameter for gp in the accumulation AIGBINV 0 35 Fs me Yes Parameter for Igp in the inversion BIGBINV 0 03 Fs mY Yes Parameter for gp in the inversion CIGBINV 900006V Yes Parameter for Igp in the inversion EIGBINV 1 1V Yes Parameter for Igp in the inversion NIGBINV 3 0 Yes Parameter for gp in the inversion AIGC 0 054 NMOS Yes Parameter for lyes and lyca and 0 31 PMOS poit m 452 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 119 Gate Dielectric Tunneling Current Model Parameters MOSFET Level 54 Continued Parameter Default Binnable Description BIGC 0 054 NMOS Yes Parameter for hcs and lyca and 0 024 PMOS Fie mv CIGC 0 075 NMOS and Yes Parameter for ycs and lyca 0 03 PMOS V AIGSD 0 43 NMOS Yes Parameter for Ig and Iya and 0 31 PMOS F 2 95 m BIGSD 0 054 NMOS 0 024 Yes Parameter for Ig and Iya PMOS F 2 95 mv CIGSD 0 075 NMOS and Yes
345. for rshpls 1 K 0 rshbody Sheet res of intrinsic body ohm sq 3000 rshbodyext Sheet res of extrinsic body ohm sq 3000 tcbody Temperature coefficient for rshbody and rshbodyext 1 K 0 cfr Gate to S D fringing capacitance F micron O cfrs Gate to s fringing capacitance asymmetrical F micron cfr foc Bias dependence of overlap capacitance 1 0 VOC Bias dependence of overlap capacitance V 0 cfrb Gate to body overlap capacitance F micron O cfrbox Back gate to S D fringing capacitance F micron O odifact SID active diffusion micron 0 HSPICE MOSFET Models Manual X 2005 09 515 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Table 147 8 SSIMSOI Model Parasitic Parameters Continued Name Parameter Units Default odifbc Body contact diffusion micron 0 ccp Contact to poly capacitance F micron O ccpr Ccp error in RCE netlist F micron O ccc Contact to contact capacitance F micron O cccr Ccc error in RCE netlists F micron O CCX Contact to soisub capacitance F micron O cpx Poly to soisub capacitance F micron O wir Additional contact width micron 0 nlev Flicker noise equation level 0 kf Flicker noise coefficient 0 0 af Flicker noise exponent 1 0 cexp Flicker noise cox exponent 1 0 fexp Flicker noise frequency exponent 1 0 516 HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Describes a Synopsys program interface you can use to add your own proprietary MOSFET models into the HSP
346. for the geometry x 1 35 3 75 independent nph part PLTETAPH Coefficient for the length dependent 0 0 part of nph PWTETAPH Coefficient for the width dependent 0 0 part of nph PLWTETAPH Coefficient for the length times width 0 0 nph dependent part POTETAMOB Coefficient for the geometry K 1 0 0 independent part of ST nph PLTETAMOB Coefficient for the length dependent K 1 0 0 part of nph PWTETAMOB Coefficient for the width dependent K 1 0 0 part of nph PLWTETAMOB Coefficient for the length times width K 1 0 0 dependent part of ph NU Exponent of the field dependence of 2 2 the mobility model at the reference temperature 304 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS POTNUEXP Coefficient for the geometry 5 25 3 23 independent vexp part PLTNUEXP Coefficient for the length dependent 0 0 part of vexp PWTNUEXP Coefficient for the width dependent 0 0 part of vexp PLWTNUEXP Coefficient fore the length times width 0 0 dependent part of vexp POTETAR Coefficient for the geometry x 0 95 0 4 independent nR part PLTETAR Coefficient for the length dependent 0 0 part of nR PWTETAR Coefficient for the width dependent 0 0 part of nR PLWTETAR Coefficient for the length times width 0 0 dependent part of nR POTETASAT
347. forward and reverse normalized currents The charge formulation also expresses the effective mobility dependence of a local field HSPICE MOSFET Models Manual 225 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Bulk Reference and Symmetry Voltages are all referred to the local substrate Vg Voz Intrinsic gate to bulk voltage Vs Vsp Intrinsic source to bulk voltage Vp Vor Intrinsic drain to bulk voltage Vs and Vp are the intrinsic source and drain voltages so the voltage drop over the extrinsic resistive elements must already be accounted for externally Vp is the electrical drain voltage where Vp Vs Bulk reference handles the model symmetrically with respect to source and drain This symmetry is inherent in common MOS technologies excluding non symmetric source drain layouts Note Intrinsic model equations are presented for an N channel MOSFET P channel MOSFETs are dealt with as pseudo N channel That is simulation reverses the polarity of the voltages Va Vs Vp VFB VTO and TCV before computing the P channel current which has a negative sign No other distinctions are made between N channel and P channel except the n factor for calculating the effective mobility Figure 35 Level 55 Equivalent Circuit Gate Intrinsic EKV model elements e Cgsov Cgbov Cobi Cosi Cgai _ Cgdov Sou
348. from the PHIO VFBO K1 K2 ETAO X2E X3E GAMMN and ETAMN model parameters and from their respective length and width sensitivity parameters xbs zphi vbs 1 2 xeta zeta zx2e vbs zx3e vds vth zvfb zphi zk1 xbs zk2 xbs2 xeta vds This equation is quadratic in xbs and vds It is joined to linear equations at d vth d xbs zgammn and at d vth d vds zetamn which prevents the quadratics from going in the wrong direction Both gammn and etamn default to zero and typically do not affect behavior in the normal operating region Effective Mobility The effective model parameter values for mobility after you adjust the device size are zmuz zx2m zx3m zx33m zu0 and zx2u0 Simulation calculates these values from the MUZ X2M X3m X33M U00 and X2U0 model parameters and from their respective length and width sensitivity parameters Vest 7 Yes Vth Pop zmuz zx2m Vp zx ms muz zx33m v cx3ms gsi 172 1 ex3ms VDDM v4 VDDM VDDM vgs vq xu0 zuO zx2u0 v HSPICE MOSFET Models Manual 355 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modified BSIM Model Wi L eff m elf beta ug COX eff 1 xu0 Vest Saturation Voltage vdsat The effective model parameter values for the saturation voltage after you adjust the device size are zu1 zx2u1 and zx3u1 Simulation calculates these values from the U1 X2U1 and X3U1 model parameters and from their leng
349. g constants have been added to the original Huang Taylor mechanism to account for the effects caused by nonuniform channel implants and also to make up for an oversight in the average capacitance construct The enhancement region uses a significantly more elaborate surface mobility model Body effect in LEVEL 38 is calculated in two regions Bulk body effect vsb vsbc 0 With sufficiently high and negative substrate bias exceeding vsbc the depletion region at the implanted channel substrate junction reaches the Si oxide interface Under such circumstances the free carriers can accumulate only at the interface as in an enhancement device and the bulk doping level determines the body effect Implant dominated body effect vsb vsbc 0 Before reaching vsbc and as long as the implant dose overwhelms the substrate doping level the deeply buried transistor due to the implant dominates the body effect of the depletion mode device The y body effect coefficient is proportional to both the substrate doping and to the first order implant depth In Level 38 the BetaGam empirical parameter amplifies the body effect due to deep implant Model parameters that start with L or W represent geometric sensitivities In the model equations three model parameters determine the zX quantity X is the variable name arge and wide channel case value X Length sensitivity LX Width sensitivity WX The model calculates thes
350. g current 57 59 QSRCO LX55 Total Source charge Charge 49 53 57 59 Conservation QS QG QD QB CQs LX56 Source charge current 57 59 CGEBO LX57 CGEBO dQg dVe intrinsic gate to 57 59 substrate capacitance CSSBO LX58 CSSBO dQs dVs intrinsic source 57 59 capacitance CSGBO LX59 CSGBO dQs dVg intrinsic source to 57 59 gate capacitance CSDBO LX60 CSDBO dQs dVd intrinsic source to 57 59 drain capacitance CSEBO LX61 CSEBO dQs dVe intrinsic source to 57 59 substrate capacitance weff LX62 Effective channel width 54 HSPICE9 MOSFET Models Manual 21 X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level leff LX63 Effective channel length 54 weffcv LX64 X Effective channel width for CV 54 leffcv LX65 Effective channel length for CV 54 igbo LX66 Gate to Substrate Current Igb Igoacc 54 Igbinv igcso LX67 Source Partition of Igc 54 igcdo LX68 Drain Partition of Igc 54 iimi LX69 Impact ionization current 54 igidlo LX70 Gate induced drain leakage current 54 igdt LX71 Gate Dielectric Tunneling Current lg 2lgs 54 Igd Igcs Igcd lgb igc LX72 Gate to Channel Current at zero Vds 54 igbacc LX73 Determined by ECB Electron tunneling 54 from the Conduction Band significant in the accumulation igbinv LX74 Determined by EVB Electron tunneling 54 from the Valence Band signific
351. g of short channel devices The approach is empirical rather than physical It frequently uses polynomials This makes it easier to write a parameter extraction program but the polynomials often behave poorly For example this model uses a VDS quadratic function for mobility Parameters specify the values at VDS 0 and 5 and the slope at VDS 5 unfortunately values that look reasonable can produce a quadratic that is non monotonic causing a GDS 0 problem The Synopsys implementation of BSIM1 as the Level 13 MOSFET device model removed discontinuities in the current function added temperature HSPICE MOSFET Models Manual X 2005 09 B Comparing MOS Models Future for Model Developments parameters and added diode and capacitance models consistent with other models The Berkeley version did not include temperature parameters LEVEL 28 LEVEL 28 is a proprietary Synopsys device model for submicron devices designed to fix the following problems in BSIM1 Negative GDS Poor behavior of some polynomial expressions Akink in GM at threshold LEVEL 28 is based on BSIM1 but some parameters are quite different You cannot use a BSIM1 parameter set as a LEVEL 28 model The LEVEL 28 model is designed for optimization it does not include a simple extraction program It is stable for automatically generating model parameters The LEVEL 28 model routinely optimizes IDS GDS and GM data LEVEL 39 The BSIM2 model was developed by Duster
352. generates a special node for each element This bulk node is named in the form B element names where the element name is the name of the defined element Use this name in any statement such as a PRINT statement to refer to the bulk node in the element Syntax MODEL mname PMOS lt LEVEL 27 gt SOSLEV val t pnamel vall MODEL mname NMOS lt LEVEL 27 gt SOSLEV val t pname vali You can use this MODEL syntax to include a MOSFET Level 27 model in your HSPICE netlist For a general description of the MODEL statement see the HSPICE Command Reference Parameter Description mname Model name PMOS Identifies a p channel MOSFET model NMOS Identifies an n channel MOSFET model LEVEL Model level selector SOSLEV Selects the processing model If you set SOSLEV 1 the default 5um The automatic default 3um pname Parameter model HSPICE MOSFET Models Manual 189 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 27 SOSFET Model 190 LEVEL 27 Model Parameters Table 42 5 um Model Parameters Name Units Default Description Alias CGDO F m Gate drain overlap capacitance per unit channel width Default 3 1e 10 n type 2 2e 10 p type CGSO F m Gate source overlap capacitance per unit channel width Default 3 1e 10 n type 2 2e 10 p type LD m Lateral diffusion The default 0 6p n type 0 3u p type RSH ohm sq Drain and source diffusion sheet resistance T
353. gion W 1 V Listino n Mog Co aa et a ae ae vys Vas AA ip Vis Listino en Sere 1 ds dslin0 Va S Saturation Region Model Parameters litl eta Lip E w D rour Pot P gipit Pippo P cbet Pipe Vasat and Fyag A Ik Vis Esar Legt Vasat Rg Vas Cox Wig v bul a gst 2 Vy S asat 2 Pfactor 1 Ris s V sat C s Wo A bulk Po V E zcpg ydg gst Esat Leg Early Voltage satMod 1 L 1 eta NG m Visart Pu PE P bs i A bulk Ge E at Lot Vost Vi B Visat i Vis B s E litl sat A bulk Esat Left V a m 2 litl E Early Voltage satMod 2 1 ge Va Vasat t Pinge Uy as i G t V aclm adib HSPICE MOSFET Models Manual 393 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model L dd D gom dee s litl A E Let V Veins uk sar ef ST Va Vasat Pin A bulk i E at litl 1 r 1 1 15 Vam a a Ve 2 O out Bag i A bulk Visat Vs p D t L ff D idut Log Bout Paipn exp rx 7 s T 2exp i J HP pipo t 1 V KE P sche2 SE bei litl c Dh mU e s sat Drain Current V Lisat Wi V sat Cox gs Vin 5 dk Visar Pfactor Pfactor A Vost t42 Vic Vos Lis lasar 1 2 re en A ahce Subthreshold Region Model Parameters Nfactor Case Cup Vp CD us etay eta n and DIBL Nfactor 1 034e 10 n 1 Ox L L A Case Caseb Vps EXP
354. gth dependent of V 0 0 PB PWPHIB Coefficient for the width dependent of V 0 0 PB PLWPHIB Coefficient length times width B V 0 0 dependent POBET Coefficient geometry independent AV 2 1 922E 3 3 814E 4 B part PLBET Coefficient for the length dependent of AV 2 0 0 p PWBET Coefficient for the width dependent of B AV 2 0 0 PLWBET Coefficient width over length AV 2 0 0 dependent of p POTHESR Coefficient geometry independent V 1 3 562E 1 7 30E 1 SR part PLTHESR Coefficient for the length dependent V 1 0 0 part of OSR 294 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry X 2005 09 Scaling Continued Name Description Units NMOS PMOS PWTHESR Coefficient for the width dependent V 1 0 0 part of 6SR PLWTHESR Coefficient length times width 6SR V 1 0 0 dependent POTHEPH Coefficient geometry independentOph V 1 1 290E 2 1 0E 3 part PLTHEPH Coefficient for the length dependent of V 1 0 0 Oph PWTHEPH Coefficient for the width dependent of V 1 0 0 Oph PLWHEPH Coefficient length times width eph V 1 0 0 dependent POETAMOB Coefficient geometry independent 1 40 3 nmob part PLETAMOB Coefficient length dependent mob 0 0 part PWETAMOB Coefficient width dependent ri mob 0 0 part PLWETAMOB Coefficient length times width 0 0 dependent nmob part POTHER Coefficient for
355. h sq WLN offset DWG m V 0 0 Yes Coefficient of the gate dependence for Weff DWB qgyy 2 0 0 Yes Coefficient of the substrate body bias dependence for Weff LINT m 0 0 No Length offset fitting the parameter from the I V without the bias LL mLLN 0 0 No Coefficient of the length dependence for the length offset LLN 1 0 No Power of the length dependence of the length offset LW mLWN 0 0 No Coefficient of the width dependence for the length offset LWN 1 0 No Power of the width dependence of the length offset LWL mN 0 0 No Coefficient of the length and width cross term for the length m LLN offset DLC m LINT No Length offset fitting parameter from CV DWC m WINT No Width offset fitting parameter from CV Table 100 Temperature Parameters MOSFET Levels 49 53 Name Unit Default Bin Description KT1 V 0 11 Yes Temperature coefficient for Vth KTIL m V 0 0 Yes Temperature coefficient for the channel length dependence of Vth KT2 0 022 Yes Body bias coefficient of the Vth temperature effect UTE 1 5 Yes Mobility temperature exponent 420 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 100 Temperature Parameters MOSFET Levels 49 53 Continued Name Unit Default Bin Description UA1 m V 4 31e 9 Yes UB1 m v 7 61e 318 Yes UC1 mv 5 69e 11 Yes AT m sec 3 3e4 Yes PRT ohm um 0 Yes XTI 3 0 No Temperature coefficient for UA Temperature coefficient for UB Temper
356. h as MOSFET Level 3 or Level 28 If you specify the L and W values in microns rather than meters for example L 1 rather than L 1y or 1e 6 set OPTION SCALE 1e 6 HSPICE MOSFET Models Manual 205 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 40 HP a Si TFT Model 206 The SCALM option is disabled in the LEVEL 40 model For standard MOSFET models Such as LEVEL 3 SCALM affects the scale of model parameters such as XL XW LD WD CJ and CJSW Because the LEVEL 40 model ignores the SCALM option you can mix LEVEL 40 models in a simulation with other models that use SCALM In general netlists for Synopsys simulators should be as standard as possible Also you should convert L and W to meters scale instead of microns scale so that you can use the netlist without OPTION SCALE 1E 6 If you follow these recommendations a system level simulation can use I O sub circuits from different vendors Noise Model The LEVEL 40 model uses the standard NLEV 0 noise model inherited from other Synopsys MOSFET models DELVTO Element You can use DELVTO and DTEMP on the element line with LEVEL 40 Device Model and Element Statement Example MODEL nch nmos LEVEL 40 UO 0 4229 VTO 1 645 PHI 1 25 NSS 0 t NFS 2 248E 21 VMAX 1231 THETA 0 01771 ETA 0 0002703 T1 22 6E 07 T2 0 E1 3 9 E2 0 GO 9 206E 15 NU 0 K2 2 CHI 0 5 t PSI 1E 20 VTIME 0 01 TREF 1 5 CGSO 5 203E 14 CGDO 4 43E 14 CSC 0 0001447 RD 5097 t RS 5097 FREQ 1E 06 DEFF
357. h sensitivity 2 3 6 7 8 HSPICE MOSFET Models Manual 121 X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 27 Threshold Voltage Parameters Continued Name Alias Units Default Description Level WNO um 0 0 NO width sensitivity 2 3 6 7 WVT V um 0 0 VT dependence on the channel width 38 WVTO Use curve fitting to determine the mobility parameters Generally you should set UTRA between 0 0 and 0 5 Nonzero values for UTRA can result in negative resistance regions at the onset of saturation Table 28 Mobility Parameters Name Alias Units Default Description Level BEX 1 5 Surface channel mobility 38 39 temperature exponent BFRC s cm V 0 0 Field reduction coefficient 38 variation due to the substrate bias FACTOR Mobility degradation factor 6 7 Default 1 0 FEX 0 Temperature exponent for 39 velocity saturation FRC A s cm 0 0 Field reduction coefficient 5 38 FRCEX 0 0 Temperature coefficient for 38 F1EX FRC FSB V1 2 s cm 0 0 Lateral mobility coefficient 5 38 HEX TUH 1 5 Implant channel mobility 38 temperature exponent KBeta1 1 0 Effective implant channel 38 mobility modifier 122 HSPICE MOSFET Models Manual X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 28 Mobility Parameters Continued Name Alias Units Default Description Level KIO KIO LBFRC LFRC LKBeta1
358. hannel resistance effect for intrinsic input resistance XRCRG2 1 Y Parameter to account for the excess channel diffusion resistance for intrinsic input resistance Gate Resistance Equivalent Circuit RGATEMOD 0 No gate resistance default Ed ot Lo 480 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIMS3 SOI Model RGATEMOD 1 Constant gate resistance Hgeltd ot Lo RGATEMOD 2 Variable resistance with Rii model Rgeltd Ri oS Lo RGATEMOD 3 Rii model with two nodes Hgelt Cgso i Cgd SE Rgeltd Poly gate electrode resistance bias independent Rii Intrinsic input gate resistance reflected to the gate from the intrinsic channel region It is bias dependent and a first order non quasi static model for RF and rapid transient MOSFET operations HSPICE MOSFET Models Manual 481 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 59 UC Berkeley BSIM3 SOI FD Model Enhanced Binning Capability Model parameters for the following components are now binnable for better accuracy and scalability Junction depth QGate tunneling current Temperature dependence of threshold voltage mobility saturation velocity parasitic resistance and diode currents Bug Fixes The BSIMSOI3 1 from UC Berkeley fixes the following bugs in the BSIMSOI3 0 model The model now takes NSEG into account when calculating the gate channe
359. hared Library 1 Follow the steps in the preceding section to modify the model routine files and the configuration file 2 Manually set the HSPICE CMI environment variable to the working Customer CMI directory see Creating the Directory Environment on page 522 setenv HSPICE CMI home useri userx model cmi 3 Usea single make operation to compile both the model routines and the shared library 4 To check the syntax of your C functions before you launch the compilation process enter the following command o make f makecmi lint This command lists any syntax errors in your model routines 5 To invoke the compilation process enter the following 9 make f makecmi The simulator creates the new libCMImodel shared library in the lib subdirectory It also generates all object files in the obj subdirectory Note During compilation Customer CMI creates files makefile SUN on SUN makefile HP on HP and subdirectories obj and lib in the Customer CMI working directory Do not modify these generated files Choosing a Compiler To use any functional compiler set the CC environment variable to the location of the compiler the auto generated makefile makefile SUN or makefile HP uses CC Set the appropriate compiler and link flags properly so the final HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Adding Proprietary MOS Models Customer CMI library build is in
360. he default 25 nj type 100 p type SOSLEV 1 Model index TOX m 7 0e 8 Oxide thickness UO cm2 V s Surface mobility Default 350 n type 220 p type VTO V Threshold voltage Default 1 25 n type 1 25 p type Table 43 3 um Model Parameters Name Units Default Description Alias A m V 0 4um Channel length shortening coefficient 2nd effect ALPHA V m CAPOP Threshold voltage length dependence Default 0 1 5p n type 0 18 p type Capacitance model selector HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 27 SOSFET Model Table 43 3 um Model Parameters Continued Name Units Default Description Alias CGDO F m Gate drain overlap capacitance per unit channel width Default 4 6e 10 n type 3 6e 10 p type CGSO F m Gate source overlap capacitance per unit channel width The default 4 6e 10 n type 3 6e 10 p type EC V m Critical electric field for velocity saturation 2nd effect The default 3 0e6 n type 7 5e6 p type FB Body effect coefficient 2nd effect Default 0 15 n type O p type LD m Lateral diffusion Default 0 3u n type 0 2u p type LEVEL 27 Model level selector RSH ohm sq Drain and source diffusion sheet resistance Default 25 n type 80 p type SOSLEV 2 Model index THETA 1 V Mobility degradation coefficient 2nd effect Default 0 055 n type 0 075 p type TOX m 3 4e 8 Oxide thickness
361. he UCB BSIM1 and the Synopsys LEVEL 13 model parameters Units in this table are in brackets This comparison uses the model parameter name only if it differs from the SPICE name The model specifies parameter units only if they differ from SPICE units These aliases are in parentheses Some parameter aliases match the SPICE names An asterisk in front of a UCB SPICE name denotes an incompatibility between the parameter name in the Synopsys Level 13 MOSFET device model and the UCB SPICE name that is the parameter alias does not match or the units are different Even if the parameter name in this model is not the same as in SPICE the corresponding L and W sensitivity parameter names might not differ Table 86 HSPICE MOSFET Models Manual 341 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 342 lists the L and W sensitivity parameters only for the few cases where the parameters are different Table 86 Comparing Synopsys Model Parameters amp UCB SPICE 2 3 UC Berkeley SPICE 2 3 Synopsys Device Model VEB V PHI V K1 V12 K2 ETA MUZ cm V s DL um DW um UO 1 V U1 W V X2MZ cm V s LX2MZ um cm V s WX2MZ um cm V s X2E 1 V X3E 1 V X2U0 1 V X2U1 um V MUS cm V s VFBO VFB PHIO same same ETAO same DLO DWO UOO same X2M X2MZ X2M LX2MZ WX2M WX2MZ same same same same same HSPICE MOSFET Model
362. he next section The HSPICE Documentation Set Chapter Description Chapter 1 Overview of Provides an overview of MOSFET model types MOSFET Models and general information on using and selecting MOSFET models Chapter 2 Technical Describes the technology used in all HSPICE Summary of MOSFET MOSFET models Models Chapter 3 Common Lists and describes parameters that are common MOSFET Model to several or all MOSFET model levels Parameters Chapter 4 Standard Lists and describes standard MOSFET models MOSFET Models Level 1 to Levels 1 to 40 40 HSPICE MOSFET Models Manual xiii X 2005 09 About This Manual The HSPICE Documentation Set Chapter Description Chapter 5 Standard MOSFET Models Levels 50 to 64 Chapter 6 BSIM MOSFET Models Levels 13 to 39 Chapter 7 BSIM MOSFET Models Levels 47 to 65 Chapter 8 Customer Common Model Interface Appendix A Finding Device Libraries Appendix B Comparing MOS Models Lists and describes standard MOSFET models Levels 50 to 64 Lists and describes three of the earliest BSIM type MOSFET models supported by HSPICE Lists and describes seven of the newest BSIM type MOSFET models supported by HSPICE Describes a Synopsys program interface you can use to add your own proprietary MOSFET models into the HSPICE or HSPICE RF simulator Describes how to use the HSPICE automatic model selector to find the proper model for each transistor size Comp
363. her at high gate voltages Due to this variation in gain the enhancement models cannot accurately represent a depletion device The physical model for a depletion device is basically the same as an enhancement model except that a one step profile with DP depth approximates the depletion implant Due to the implant profile simulation calculates the drain current equation by region The MOSFET Level 5 model has three regions depletion enhancement and partial enhancement Depletion Region Vgs Vip lt O The bulk channel dominates the low gate voltage region Enhancement Region Vgs Vip gt O Vas lt Vgs Vib High gate voltage and low drain voltage define the enhancement region In this region both channels are fully turned on Partial enhancement region Vgs Vip gt O Vas gt Vgs Vf The region has high gate and drain voltages so the surface region is partially turned on and the bulk region is fully turned on IDS Equations Depletion Model LEVEL 5 Depletion Vgs Vip lt 0 2 Ij pl 4 NI vde cav o vde Ye i cav y vde v 2 vy 92 Enhancement Vus Vip vde 0 3 gs 2 Ti pl a NI vde 3 Pcav by P vde v P tep vde p UE pug 150 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model Partial Enhancement Vgg Vip lt vde iss 24 lis Bio NT nde ec vg Vp vde 2S E cav Y vde v P W3
364. hold voltage shift This parameter is type 39 sensitive For example DELVTO gt 0 increases the magnitude of the n channel threshold decreases the magnitude of the p channel threshold and adds to the element line DELVTO parameter LATD LD um 1 7 XJ Lateral diffusion on each side 5 38 LDAC m This parameter is the same as LD but if you 1 2 3 specify LDAC in the MODEL statement it 6 7 8 replaces LD in the Leff calculation for the AC 13 28 gate capacitance 38 39 LMLT 1 0 Gate length shrink factor 1 2 3 5 6 7 8 13 28 38 39 Scale MOSFET drawn length 54 LD DLAT m Lateral diffusion into the channel from the 1 2 3 LATD source and the drain diffusion 6 7 8 If you do not specify LD and XJ 13 28 LD Default 0 0 If you specify XJ but you do not specify LD simulation calculates LD as LD default 0 75 XJ LDgcaleq LD SCALM LD m 0 Lateral diffusion under the gate per side of the 39 HSPICE MOSFET Models Manual X 2005 09 S D junction Use this parameter to calculate Les only if DL 0 LDgcaleg LD SCALM 115 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 26 Effective Width and Length Parameters Continued Name Alias Units Default Description LREF m OXETCH um Px x Puu WD m WDAC m WMLT 116 0 0 0 0 o 0 0 1 0 1 0 Channel length reference LREF gcaieg LREF SCALM If the Level 13 model does not define LREF a
365. hort channel modulation of GAMMA 6 7 T1 m 280n First thin film thickness 40 T2 m 0 0 Second thin film thickness 40 TDVSBC V K 0 0 Body effect transition voltage shift due to the 38 temperature 120 HSPICE MOSFET Models Manual X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 27 Threshold Voltage Parameters Continued Name Alias Units Default Description Level UFDS 0 0 High field FDS 6 7 UPDATE 0 0 Selects different versions of the LEVEL 6 6 7 model For the UPDATE 1 or 2 alternate saturation voltage simulation modifies the MOB 3 mobility equation and the RS and RD series resistances so they are compatible with ASPEC UPDATE 1 is a continuous Multi Level GAMMA model VFDS V 0 0 Reference voltage for selecting FDS or UFDS 6 7 Uses FDS if vds lt VFDS Uses UFDS if vds VFDS VSH V 0 0 Threshold voltage shifter for reducing the zero 6 7 bias threshold voltage VTO as a function of the ratio of LD to Leff VT VTO V 0 0 Extrapolated threshold voltage 5 38 VTO VT V 0 0 Zero bias threshold voltage If you do not 1 2 3 specify VTO simulation calculates it 6 7 8 40 WBetaGam um 0 0 BetaGam dependence on the channel width 38 WDVSBC V um 0 0 Adjusts the W dependent body effect transition 38 voltage WETA um 0 0 Channel width dependent drain induced 38 barrier lowering WEX Weak inversion exponent 6 7 WIC 0 0 Subthreshold model selector 2 3 6 7 8 WND um V 0 0 ND widt
366. hysical Geometry Scaling Continued Name Description Units NMOS PMOS KOV Body effect factor for the Source Drain V1 2 2 5 2 5 overlap extensions IGOVR Gain factor for Source Drain overlap AV 2 0 0 tunneling current for reference transistor TOX Thickness of the gate oxide layer M 3 2E 9 3 2E 9 COL Gate overlap capacitance per unit channel F 3 2E 16 3 2E 16 length GATENOISE In exclusion flag of induced gate thermal 0 0 noise NT Thermal noise coefficient at the actual J 1 656E 20 1 656E 20 temperature NFAR First coefficient of the flicker noise forthe V 1m 4 1 573E23 3 825E24 reference transistor NFBR Second coefficient of the flicker noise for V 1m 2 4 752E9 1 015E9 the reference transistor NFCR Third coefficient of the flicker noise for the V 1 0 7 3E 8 reference transistor MOO Parameter for short channel subthreshold 0 0 slop AGIDLR Gain factor for gate induced drain leakage AV 3 0 0 current for a channel width of 1 um BGIDL Probability factor for gate induced drain V 41 0 41 0 leakage current at the reference temperature 292 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 63 Level 63 MOS11 Parameters Level 11010 Physical Geometry Scaling Continued Name Description Units NMOS PMOS CGIDL Factor for the lateral field dependence of 0 0 the gate induce drain leakage current STBGIDL Coefficient of the temperature
367. i and k1 are chosen so that vth 1v The AC sweep is chosen so that the last point is 2 u freq le6s Input File The input file is located in the following directory Sinstalldir demo hspice mos calcap sp Calculations Leff 0 6u 0X 103F m tox Leff Weff e0x _ cp Asp Cap tox BSIM equations for internal capacitance in saturation with xqc 0 4 1 KI bod 1605 1 EL To Mero 1 744 0 8364 CHIESE PHIO vsb 1 zs o5 1 D 1 744 0 8364 1 3062 1 C ecce Cap 0 7448 44 69F CEE ap G sony ap ced 0 cdg 4 Cap 16F cdd 0 HSPICE MOSFET Models Manual 71 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Gate drain overlap d meto Weff ma 20e 15F Adding the overlaps ceg 44 60F 2 20F 84 69F ced 20F cdg 36F cdd 20F Drain bulk diode cap cj ad 0 5e 4 200e 12 10F Adding the diodes cgg 84 69F ced 20F cdg 36F cdd 30F Results subckt element O m model O nch cdtot 30 0000f cgtot 84 6886f cstot 65 9999f cbtot 43 4213f cgs 61 2673f cgd 20 0000f The calculation and simulation results match Plotting Gate Capacitances The following input file shows how to plot gate capacitances as a function of bias Set OPTION DCCAP to turn on capacitance calculations for a DC sweep The model used is the same as for the previous gate capacitance calculations Example This example netlist
368. iculties for typical forward body bias operation in floating body MOSFETs Side wall and areal junction depletion capacitances are removed and replaced with the appropriate SOI back oxide capacitances HSPICE MOSFET Models Manual 505 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model The SSIM HCI degradation oxide reliability and electromigration models are removed Additional parameters are added to account for parasitics related to device contacts Additional parameters are added to account for asymmetric S D parasitics diode current and capacitance gate overlap tunneling currents GIDL GISL currents and S D resistances Asymmetry can dramatically impact PDSOI floating body effects Using Level 65 with Synopsys Simulators To simulate using the SSIMSOI model 1 2 Set Level 65 to identify the model as the SSIMSOI model Set the correct simulator room temperature The default room temperature is 25C in Synopsys circuit simulators but is 27 oC in most other simulators When comparing to other simulators use TEMP 27 or OPTION TNOM 27 to set the simulation temperature to 27 in the netlist Set DTEMP on the element line You can use DTEMP with this model to increase the temperature of individual elements relative to the circuit temperature If you do not specify DTEMP simulation extracts TRISE from the model card If you do specify DTEMP it overrides TRISE in the model card General S
369. ied BSIM Model 352 Table 87 Transistor Process Parameters Continued Name Alias Units Default Description WX2U1 um V 0 0 Width sensitivity X33M cm V s 0 0 Gate field reduction of X8MS LX33M um cm V s 0 0 Length sensitivity WX33M um em2 V2 s 0 0 Width sensitivity X3E 1 V 0 0 Vds correction to the linear vds threshold coefficient LX3E um V 0 0 Length sensitivity WX3E um v 0 0 Width sensitivity X3MS cm2 V s 5 0 Vds correction for the high drain field mobility LX3MS um cm V s 0 0 Length sensitivity WX3MS um cm2 V2 s 0 0 Width sensitivity X3U1 1 V 0 0 Vds reduction to the drain field mobility reduction factor LX3U1 um V 0 0 Length sensitivity WX3U1 um V 0 0 Width sensitivity XPART 1 0 Selects the coefficient for sharing the gate capacitance charge Notes 1 When you read parameter names be careful about the difference in appearance between the capital letter O and the number zero 0 Use NMOS conventions to specify all LEVEL 28 parameters even for PMOS for example ETAO 0 02 not ETAO 0 02 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modified BSIM Model 3 Youcanuse the WL product sensitivity parameter for any parameter with an L and W sensitivity Replace the leading L of the L sensitivity parameter name with a P Table 88 Temperature Parameters Name Alias Units Default Description BEX 1 5 Temperature exponent fo
370. ient of the channel length dependency for the diffusion capacitance pbswg V 0 7 Built in potential of the source drain gate side sidewall junction capacitance tt secon 1ps Diffusion capacitance transit time coefficient d vsdfb V cal Flatband voltage for the source drain bottom diffusion capacitance vsdth V cal Threshold voltage for the source drain bottom diffusion capacitance xpart 0 Charge partitioning rate flag Table 132 MOSFET Level 57 Temperature Parameters Parameter Unit Default Description at m sec 3 3e4 Temperature coefficient for Ua cthO m C W s 0 Normalized thermal capacity 474 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model Table 132 MOSFET Level 57 Temperature Parameters Continued Parameter Unit Default Description kt V 0 11 Temperature coefficient for the threshold voltage kt2 0 022 Body bias coefficient of the threshold voltage temperature effect ktil V m 0 Channel length dependence of the temperature coefficient for the threshold voltage Ntrecf 0 Temperature coefficient for Nrect Ntrecr 0 Temperature coefficient for Nec prt Qn 0 Temperature coefficient for Rdsw rthO m C W 0 Normalized thermal resistance tcjswg 1 K 0 Temperature coefficient of Ciswg tnom 0G 25 Temperature at which simulation expects parameters tpbswg V K 0 Temperature coefficient of Pyswg ual m V 4 31e 9 Temperature coefficient for Ua ub1 m V
371. ight not be available HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Estimation and Limits of Static Intrinsic Model Parameters If you do not have access to a parameter extraction facility simulation can roughly estimate the EKV intrinsic model parameters from the SPICE level 2 3 parameters as indicated in Table 48 Pay attention to the units of the parameters This estimation method generally returns reasonable results Nevertheless be aware that the underlying modeling in SPICE Level 2 3 and in the EKV model is not the same even if the names and the function of several parameters are similar Therefore it is preferred if parameter extraction is made directly from measurements Lower and upper limits indicated in the table should give an idea on the order of magnitude of the parameters but do not necessarily correspond to physically meaningful limits nor to the range specified in the parameter tables These limits may be helpful for obtaining physically meaningful parameter sets when using nonlinear optimization techniques to extract EKV model parameters Table 48 Static Intrinsic Model Limits Name Unit Default Example Lower Upper Estimation COX F m 0 7E 3 3 45E 3 s eox TOX XJ m 0 1E 6 0 15E 6 0 01E 6 1E 6 XJ VTO V 0 5 0 7 0 2 VTO GAMMA JN 1 0 0 7 0 2 Bass NSUB COX PHIP V 0 7 0 5 0 3 2 2V In NSUB n KP AN 50E 6 150E 6 10E 6 UO COX EO V m 1 0E
372. imulation HSPICE or HSPICE RF calls this routine during both an outer loop over all models and an inner loop over all instances that are associated with each model Syntax int CMI FreeInstance char pmodel char pinst Parameter Description pmodel Pointer to the model pinst Pointer to the instance HSPICE MOSFET Models Manual 545 X 2005 09 8 Customer Common Model Interface Interface Variables Example int ifdef STDC CMImos3FreeInstance char pmodel char ptr else CMImos3FreeInstance pmodel ptr char pmodel char DE ey endif CMI_ENV penv MOS3instance ptran ptran MOS3instance ptr free memory allocated for the model data The CMImos3freeinstance source code is not shown m void CMImos3freeinstance MOS3model pmodel ptran return 0 int CMImos3FreeInstance CMI WriteError If model evaluation detects an error then this routine writes error messages that you define to standard output You define these error messages All Customer CMI functions return an error code and pass it to CMI WriteError nCMI WriteError you define an error statement and copy it to err str based on the error code value CMI WriteError returns the error status e ferr status 0 then HSPICE or HSPICE RF issues an error and aborts e If err_status 0 then HSPICE or
373. imulation extracts parameters This parameter defaults to the TNOM option which defaults to 25 C Sensitivity Factors of Model Parameters To denote the L channel length and W channel width sensitivity factors of a basic electrical parameter in a transistor add L and W characters at the start of the name For example VFBO sensitivity factors are LVFB and WVFB If AO is a HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model basic parameter then LA and WA are the corresponding L and W sensitivity factors of this parameter Do not use the SCALM option to scale LA and WA The Level 13 MOSFET model uses the following equation to obtain this parameter value A2 A0 LA LL amp WA Ll Leff LREFeff Weff WREFeff Specify LA and WA in units of microns times the units of AO The left side of the equation represents the effective model parameter value after you adjust the device size All effective model parameters are in lower case and start with the z character followed by the parameter name Example VFBO 0 350v LVFB 0 1vu WVFB 0 08v Leff 1 10 9m Ip Weff 2 10 m 2u LREFeff 2 10 6m 2u WREFeff 1 10 5m 10u zvfb VFBO LVFB 4 E LI WVFB 4 E LI Leff LREFeff Weff WREFeff LI Eu b 035v4 0 1v 41 1 008 1 1 zvfi v veiw TET veiw 2 TOn zvfb 0 35v 0 05v 0 032v zvfb 0 368v MODEL VERSION Changes to BSIM Models You can use the
374. in to source Current For P channel devices Ips has a negative sign HSPICE MOSFET Models Manual 239 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model This drain current expression is a single equation valid in all operating regions e weak moderate and strong inversion e non saturation saturation This current is therefore not only continuous among all of these regions but also continuously derivable Transconductances Simulation derives the transconductances from the drain current Ems gy Sms gy 5nd gy e In the following relationships the source for the derivatives is a reference 8m 3V 8mg mbs JVs 8ms 7 amp mg Emad 8ds OV s 8md This model includes the analytic derivatives Impact lonization Current l IBA IBB Lc Ds pg Vib exp e l 0 for V lt 0 for V gt 0 Ipg Note The factor 2 in the Vip expression accounts for the fact that the numerical value of Vpss is half of the actual saturation voltage The substrate current is a component of the total extrinsic drain current flowing from the drain to the bulk This model therefore expresses the total drain current as The substrate current therefore also affects the total extrinsic conductances especially the drain conductance 240 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Quasi static Model Equations MOSFET Le
375. ing at the actual geometry from source to drain Figure 2 shows a perspective of the nonisoplanar MOSFET 28 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models Field Effect Transistors Figure 2 Field Effect Transistor Geometry m Metal e i Gate Field Gate Oxide Oxide T g Source n Drain 5 l i 6 7 8 9 10 13 11 12 Drawn metal gate channel length Drawn oxide cut Effective channel length Etched channel length Lateral diffusion Drawn diffusion edge Actual diffusion edge O000 mNm Aas eie to o D 00 Co A To visualize the construction of a silicon gate MOSFET observe how a source or drain to field cuts Figure 3 Cut A B shows a drain contact Figure 4 HSPICE MOSFET Models Manual 29 X 2005 09 2 Technical Summary of MOSFET Models Field Effect Transistors Figure 3 Isoplanar Silicon Gate Transistor x x 6 Source Gate Drain B Drawn pattern for nitride definition and subsequent source drain diffusion formation Polysilicon definition where the poly crosses the source drain diffusion on MOS gate is formed Es Source drain to metal contact Figure 4 Isoplanar MOSFET Construction Part A Intermediate Contact Oxide Center N Field Implant NI Field Diffusion Tm Y Oxide N Ig TRIPS LI 345 678 Di
376. ing rate flag 0 cgso Non LDD region source gate overlap capacitance F m calculated nC 1 per channel length cgdo Non LDD region drain gate overlap capacitance F m calculated nC 2 per channel length cgeo Gate substrate overlap capacitance per unit F m 0 0 channel length cjswg Source Drain gate side sidewall junction F m2 1e 10 Capacitance per unit width normalized to 100nm Tg pbswg Built in potential for the Source Drain gate side V YA J sidewall junction capacitance mjswg Grading coefficient for the Source Drain gate V 0 5 side sidewall junction capacitance tt Coefficient for the diffusion capacitance transit second 1ps time vsdfb Flatband voltage for the source drain bottom V calculated nC 3 diffusion capacitance vsdth Threshold voltage for the source drain bottom V calculated nC 4 diffusion capacitance csdmin Minimum capacitance of the source drain bottom V calculated nC 5 diffusion asd Smoothing parameter for the source drain bottom 0 3 diffusion csdesw Source drain sidewall fringing capacitance per unit F m 0 0 length HSPICE MOSFET Models Manual X 2005 09 501 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model Table 142 MOSFET 60 AC and Capacitance Parameters Continued SPICE Description Unit Default See Symbol Table 144 cgs1 Overlap capacitance for the lightly doped source F m 0 0 gate region cgd1 Overlap capacitance for the lightly doped
377. ion Vsatn Vsdtb Csdmin Positive nsub means the same type of doping as the body e Negative nsub means opposite type of doping Level 59 Template Output For a list of output template parameters in the MOSFET models and which parameters this model supports see Table 4 on page 14 Level 60 UC Berkeley BSIM3 SOI DD Model 492 The UC Berkeley SOI model BSIM3SOI supports Fully Depleted FD Partially Depleted PD and Dynamically Depleted DD SOI devices BSIM3DD2 2 for DD SOI devices is Level 60 in the Synopsys MOSFET models For a description of this model see the BSIM3DD2 1 MOSFET MODEL User s Manual at http www device eecs berkeley edu bsim3soi BSIM3DD2 1 includes many advanced concepts for dynamic and continuous transition between PD and FD operation These concepts are collectively named Dynamic Depletion HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model Model Features Simulation applies dynamic depletion to both l V and C V Tbox and Tsi continuously scale the charge and drain current Supports external body bias and backgate bias a total of 6 nodes Real floating body simulation in both l V and C V Diode and C V formulation properly bind the body potential Improved self heating mproved impact ionization current model Various diode leakage components and parasitic bipolar current Depletion charge m
378. ion depletion regions Separate set of length width dependence parameters for the CV model LLC LWC LWLC WLC WWC and WWLC parameters A Additional parameter checking Bug fixes Note If you use the defaults for all new Version 3 2 parameters Version 3 2 and Version 3 1 with June 1998 bug fixes return identical DC results However transient and AC results generally differ This discrepancy arises only from differences in the flatband voltage calculations used in the intrinsic charge capacitance models These differences occur in all CAPMOD models 1 3 Level 53 resets HSPVER lt 98 0 to 98 0 HSPVER 98 2 resets to 98 2 if VERSION gt 3 2 for Levels 49 53 Version 3 0 3 1 and 3 11 in HSPICE do not support NQSMOD and CAPMOD 3 Only Version 3 2 supports them Version 3 24 added Rds noise to the thermal noise model Simulation smooths out the unified flicker noise from the linear region to the saturation region You might need to re extract the parameters for the unified flicker noise model For more information about the Berkeley releases see the BSIM3 web site http www device eecs berkeley edu bsim3 HSPICE MOSFET Models Manual 401 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models 402 Nonquasi Static NQS Model You can also select the Berkeley NonQuasi Static NQS model for Levels 49 and 53 This model provides a first order correction to the quasi static
379. ion due to the source drain electric field NU 1 0 Mobility reduction due to the source drain electric field UPDATE Parameter for LEVEL 6 and LEVEL 7 The general form of the ly equation for LEVEL 6 is the same as the LEVEL 2 MOS model However the small size effects mobility reduction and channel length modulation are included differently Also you can use the multi level GAMMA capability to model MOS transistors with ion implanted channels The LEVEL 6 model can represent the ASPEC MSINC or ISPICE MOSFET model Use the UPDATE model parameter to invoke different versions of the LEVEL 6 model UPDATE 0 This is the original Synopsys Level 6 MOSFET device model which is not quite compatible with the ASPEC model It has some discontinuities in the weak inversion mobility equations MOB 3 and multi level GAMMA equations UPDATE 1 This enhanced version of the LEVEL 6 model contains improved multi level GAMMA equations The saturation voltage drain source current and conductances are continuous HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model UPDATE 2 This version of the LEVEL 6 model is compatible with the ASPEC model The multi level GAMMA model is not continuous as it is in the ASPEC program See ASPEC Compatibility on page 180 Set UPDATE to 1 0 to use changes to the device equations Set UPDATE to 1 0 or 2 to use the default RS and RD hand
380. ion inserts vinth into the ids vsat and mobility equation in place of vgs except for vgs in the exponential term of the subthreshold current The following equation computes the inversion threshold voltage vinth at a specified vsb vinth vfb LN _ ysbh DVIN zETA vds cox 200 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 38 IDS Cypress Depletion Model Saturation Voltage vdsat This equation determines the vsat saturation voltage 2 51 2 vsat veces P 1o Ve vash veh vsb Phd Y J vdsat vsat Simulation modifies vsat to include the carrier velocity saturation vdsat vsat vc vsat vc 2 The following equation calculates the vc value used in the preceding equation vc ECV Leff Mobility Reduction UBeff The UB surface mobility depends on the terminal voltages 1 1 P zFRC zVFRC vde zBFRC vsb vgs vfb pa vde 2FSB vsb zUO TOX VST Le UBeff The following equations calculate Le for the preceding equation Le Leff Linear region Le Leff AL Saturation region At elevated temperatures the following equation calculates the zFRC value used in the preceding equation FRCEX zFRC t zFRC tnom tnor AL is the channel length modulation effect defined in the next section Vip assumes the role of vip in the LEVEL 5 mobility equation The degradation parameters are semi empirical and are grouped according
381. is an exponent in an expression for mobility temperature dependence 9 Other models use the BEX parameter for similar mobility temperature dependence expressions The HP a Si TFT TREF model parameter is not the same as the TREF reference temperature in other models The reference temperature for the HP a Si TFT model is 312 K or 38 85 C you cannot modify it Experimental results from TFT manufacturers indicate that amorphous silicon materials are most stable at this temperature 10 The default room temperature is 25 C in Synopsys circuit simulators but is 27 C in most other simulators When comparing to other simulators set the nominal simulation temperature to 27 C by using either TEMP 27 or OPTION TNOM 27 in the netlist Although the reference temperature of the HP a Si TFT model is fixed at 312 K or 38 85 C the behavior of the model adjusts to other simulation temperatures that you specify or that are defaults in Synopsys circuit simulators In the Level 40 MOSFET model temperature dependency is enabled 11 The default CAPOP value is 40 which is the HP a Si TFT non charge conserving capacitance model CAPOP values of 0 1 2 3 4 5 9 12 or 13 have not been thoroughly tested 12 The DERIV default is zero which selects the analytical method Set DERIV to 1 to select the finite difference method Effect of SCALE and SCALM OPTION SCALE has the same effect for LEVEL 40 as for other Synopsys device models suc
382. istance either internal or external to the intrinsic MOSFET Accepts either the electrical or physical gate oxide thickness as the model input in a physically accurate manner A quantum mechanical charge layer thickness model for both IV and CV Amore accurate mobility model for predictive modeling Agate induced drain leakage GIDL current model not available in earlier BSIM models An improved unified flicker 1 f noise model which is smooth over all bias regions and which considers the bulk charge effect Different diode IV and CV characteristics for the source and drain junctions A junction diode breakdown with or without current limiting Adielectric constant of the gate dielectric as a model parameter OP now prints the total capacitances instead of just the intrinsic capacitances for the BSIM4 Level 54 MOSFET model BSIM4 2 1 has the following improvements over BSIM4 2 0 Anew GISL Gate Induced Source Leakage current component corresponds to the same current at the drain side GIDL The warning limits for effective channel length channel width and gate oxide thickness have been reduced to avoid unnecessary warnings if you use BSIM4 aggressively beyond the desired model card application ranges The DELTOX parameter in the MOS active element M models the relative variation on the transconductance oxide thickness of the MOS in Monte Carlo analysis OPTION LIST can now print an el
383. istance improves the modeling accuracy Binning enhances the model flexibility BSIM PD version 2 21 updates the PD version 2 2 for bug fixes and S D swapping for the gate current components BSIM PD version 2 22 updates the 2 21 version for bug fixes and enhancements The major features are e FRBODY instance parameter e Improved temperature dependence of the gate direct tunneling model e Two new model parameters VEVB and VECB e UC Berkeley code no longer supports the NECB and NEVB model parameters Version 2 22 accepts these parameters for backwards compatibility but they have no effect OPTION LIST prints an element summary for the MOSFET Level 57 model bjtoff BJT on off flag Turn off BUT if equal to 1 rthO Thermal Resistance per unit width cthO Thermal Capacitance per unit width nrb Number of squares for the body series resistance frbody Coefficient of the distributed body resistance effects nbc Number of body contact isolation edge nseg Number of segments for channel width partitioning pdbcp Parasitic perimeter length for the body contact at the drain side psbcp Parasitic perimeter length for the body contact at the source side agbcp Parasitic gate to body overlap area for the body contact HSPICE MOSFET Models Manual 477 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model 478 aebcp vbsusr analysis Parasitic body to substrate overlap area for the body contact Op
384. itical gate bulk electric field at which mobility reduction becomes significant UEXP F2 0 0 Mobility exponent Use a factor of 0 36 for n channel and 0 15 for p channel UTRA F3 factor 0 0 Source drain mobility reduction factor VMAX VMX cm s 0 0 Maximum drift velocity of carriers The VMAX setting determines which calculation scheme vdsat uses Zero indicates an infinite value The mobility reduction equation MOB 2 produces good results for high gate voltages and drain fields with constant back bias Typically you can use this equation for p channel pull ups and n channel pull downs The VMAX value selects the proper vdsat calculation scheme MOB 2 SPICE default corresponds to MSINC UN 2 factor ey COX vgs vbi F3 vde vde is the same in this equation as in the MOB 1 equation Table 33 MOB 3 Normal Field Equation Name Alias Units Default Description F1 1 V 0 0 Low field mobility multiplier F4 1 0 Mobility summing constant UEXP F2 0 0 Mobility exponent HSPICE MOSFET Models Manual 173 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model 174 Table 33 MOB 3 Normal Field Equation Continued Name Alias Units Default Description UTRA F3 1 V 0 0 High field mobility multiplier VF1 V 0 0 Low to high field mobility voltage switch This equation is the same as MSINC UN 1 ves vth f2 VF1 1 F4 Fl vgs vth
385. itivity simulations Because negative values for both KP and GAMMA are not physically meaningful simulation clips them at zero GAMMA GAMMA Reverse Short channel Effect RSCE EN L C 4 22x10 C 008 amp C 10 1 2 00 1 OVRSCE 790K 4 fos ae eee t C Effective Gate Voltage Including RSCE Vo Vg VTO AVpsep PHI GAMMA PHI HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Effective substrate factor including charge sharing for short and narrow channels Pinch off voltage for narrow channel effect GAMMA GAMMA Vg PHI GAMMA Vg for V SO Vpo 2 2 PHI for Vg 0 Effective substrate factor accounting for charge sharing 1 Vyp PHD 4v Vs 5 Vs PHI 4 Vs py PHI 4 4V Note This equation prevents a negative value in the square roots argument in the subsequent code FLETA 3 WETA GAMMA cox NULLO ee MT PHI i Cox LX As An YE pot 1 5 EZ yrz 5 fy 0 1 V Note This equation prevents the effective substrate factor from becoming negative Pinch off Voltage Including Short Channel and Narrow Channel Effects 2 PHI for Vo lt 0 Vp Note The pinch off voltage accounts for channel doping effects such as the threshold voltage and the substrate effect For long channel devices Vp is a function of gate voltage for short channel
386. ity Reduction Due to Vertical Field NP W tf Bo KP _ La Note The NP or M device parameter returns accurate results for simulating parallel devices Using NS or N for series devices is only approximate Leg accounts for multiple NS series device numbers i 1 2 for NMOS WG 1 3 for PMOS 4B0 GAMMA PHI 0 7 Po UU 0 450 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Bo 3 V lag n qj pe COX EQ 2 i 1 For the definition of the qg normalized depletion and the q inversion charges refer to the Normalized Intrinsic Node Charges on page 241 Use By to ensure that B By when q lt lt q The formulation of D arises from the integration of the local effective field as a function of depletion and inversion charge densities along the channel You do not need to specify the substrate bias dependency because the model includes the depletion charge Note The resulting mobility expression also depends on Vps Simple Mobility Reduction Model For compatibility with the former EKV model versions before v2 6 you can choose the simpler mobility reduction model which uses the THETA parameter If you do not specify the EO model parameter see Parameter Preprocessing on page 231 simulation uses the simpler mobility model Vp HUE Vp 2Vj Bo P TITHETA Vy Specific Current IL 2 n B V Dra
387. ivalent to adding 0 1 to VTO For the LEVEL 13 28 39 models DELVTO 0 1 is equivalent to adding 0 1 to VFBO The XL and XW parameters represent the line width variation The equation for effective channel length is Leff L XL 2 LD The Berkeley BSIM1 and BSIM2 models use Leff L DL The MOSFET models support the DL and DW parameters DLO DWO for BSIM1 for compatibility use XL LD XW and WD instead In the MOSFET models the geometry parameters XL LD XW and WD and the parasitic parameters CJ MJ CJSW MJSW and RSH remain simple and level independent to consistently use process variation information Gate Capacitance Modeling 578 LEVEL 2 and 3 were released in Berkeley SPICE with the Meyer model for gate capacitance This model is non charge conserving and sets dQG dVD dQD dVG Although it is not valid in a real device it provides an adequate response for most digital simulations Berkeley BSIM1 and BSIM2 models are charge conserving non symmetric capacitance models HSPICE MOSFET Models Manual X 2005 09 B Comparing MOS Models Gate Capacitance Modeling The range of capacitance model choices and the default values depend on which model you choose The default for LEVELs 2 and 3 is still the Meyer model but you can also select a charge conserving Ward Dutton model Table 150 Comparison of Parameters for MOSFET Model Levels LEVEL 2 3 13 28 39 Number of parameters 13 12 60 63 99 Minimal numb
388. l MOB 3 Mobility Reduction AA Rl e elf 1421e 8 vgs vth egfet o 6 TOX In the preceding equation egfet is the silicon energy gap at the analysis temperature 7 02e 4 t t 1 16 egfe T 1108 In the preceding equation t is the temperature in degrees Kelvin If VMAX 1 P eff u 1 a eee vde VMAX Lo MOB 6 Mobility Reduction For UEXP gt 0 nng Esi UCRIT vgs vth gt COX ej UCRIT UEXP UO aa n COX vgs vth then upg UTRA 1 vde Lo otherwise Wem UO ef BE 1 TAS vde eff For UEXP 0 TE UO eff 1 UCRIT vgs vth 1 pene vde eff UCRIT for UEXP 0 has a dimension of 1 V HSPICE MOSFET Models Manual 185 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 8 IDS Model 186 MOB 7 Mobility Reduction _ UO Uef 2 vde 1 UTRA ves vbi n QUE body The following equation calculates the body value used in the preceding equation body y vde vsb 2 vsb Channel Length Modulation The CLM equation selector controls the channel length modulation equations In the LEVEL 8 model set CLM to 6 7 or 8 Default 7 CLM 6 SPICE Channel Length Modulation If LAMBDA 0 xd vds vdsat KN _ Yas vasat up Eee YOO NP TETE qu Sm ii 4 Otherwise A LAMBDA Then V Lg vds 1 ZAMI La Note The LEVEL 2 model has no LAM1 term This model modifies the curre
389. l Table 65 Level 63 JUNCAP Parameters Continued Name Description Units Default JSDSR Sidewall saturation current density due to back contact diffusion Am 1E 03 JSGGR Gate edge saturation current density due to generating an electron Am 1E 03 hole at V Vp JSDGR Gate edge saturation current density due to back contact diffusion Am 1E 03 NB Emission coefficient of the bottom forward current 1 NS Emission coefficient of the sidewall forward current 1 NG Emission coefficient of the gate edge forward current 1 CJBR Bottom junction capacitance at V Vp Fm 1E 12 CJSR X Sidewall junction capacitance at V Vp Fm 1E 12 CJGR Gate edge junction capacitance at V Vp Fm 1E 12 VDBR Diffusion voltage of the bottom junction at T Tp V 1 VDSR Diffusion voltage of the sidewall junction at T2 Tg V 1 VDGR Diffusion voltage of the gate edge junction at T Tp V 1 PB Bottom junction grading coefficient 0 4 PS Sidewall junction grading coefficient 0 4 PG Gate edge junction grading coefficient 0 4 Note All symbols refer to Unclassified Report NL UR 2001 813 Example 1 model nch nmos level 63 VERSION 1100 LER 1E 06 WER 1E 05 LAP 1 864E 08 TR 21 VFBR 1 038 SLPHIB 1 024E 08 SL2PHIB 1 428E 14 KOR 5 763E 01 SLKO 2 649E 08 308 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50
390. l GTE style 5 Rutherford 5 Cypress 5 195 197 Dallas Semiconductor 5 Empirical 136 EPFL EKV 6 224 231 Fluke Mosaid 5 Frohman Bentchkowski equations 173 GE CRD Franz 5 GE Intersil 5 Grove Frohman 4 HP a Si TFT 204 206 Hspice junction 402 HSPICE PC version 4 IDS Cypress depletion 6 LEVEL 5 144 equations 150 LEVEL 6 equations 159 168 example 164 166 LEVEL 7 181 LEVEL 8 182 183 Lattin Jenkins Grove 4 LEVEL 61 circuit 263 LEVEL 62 272 levels 4 HSPICE MOSFET Models Manual X 2005 09 Index LEVELs 49 and 53 equations 431 modified BSIM LEVEL 28 347 354 MOS 143 571 MOS2 4 MOS3 4 MOS9 214 Motorola 6 National Semiconductor 6 Philips MOS9 214 quasi static equations 241 RPI Poli Si TFT 265 Schichman Hodges 4 128 130 SGS Thomson 6 Sharp 6 Siemens 5 6 Sierra 1 5 Sierra 2 5 Siliconix 5 SOSFETSs 5 188 statement 13 STC ITT 5 Taylor Huang 4 TI 6 University of Florida SOI 7 249 user defined 5 VTI 6 n channel specification 13 noise 96 model equations 96 model parameters 96 summary printout 97 p channel specification 13 RPI a Si TFT model 260 saturation current 102 sensitivity factors 353 SPICE compatibility 4 surface potential 104 temperature coefficient model parameters 100 effects parameters 98 parameters 98 template input 492 threshold voltage model parameters 58 temperature equations 105 transconductance 35 Motorola model 6 Multi Level Gamma model example 164 166 605 Index N N m
391. l tunneling current The GMIN connecting gate and drain is now multiplied by 1e 6 to reduce false leakage current A swapping error in the source drain overlap capacitance stamping The bulk charge effect coefficient Ap cv is now corrected for Qj and its derivatives in CAPMOD 2 Level 59 UC Berkeley BSIM3 SOI FD Model The UC Berkeley SOI BSIM3 SOl Fully Depleted FD model is Level 59 in the Synopsys MOSFET models For a description of this model see the BSIM3SOI FD2 1 MOSFET MODEL User Manual at http www device eecs berkeley edu bsim3soi The general syntax for including a BSIM3 SOI FD MOSFET element in a netlist is Mxxx nd ng ns ne lt np gt mname lt L val gt t W val lt M val gt AD val AS val lt PD val gt lt PS val gt t lt NRD val gt lt NRS val gt lt NRB val gt lt RTHO val gt lt CTHO val gt t off BJToff val IC Vds Vgs Vbs Ves Vps gt 482 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 59 UC Berkeley BSIM3 SOI FD Model Parameter Description Mxxx SOI MOSFET element name Must begin with M followed by up to 1023 alphanumeric characters nd Drain terminal node name or number ng Front gate node name or number ns Source terminal node name or number ne Back gate or substrate node name or number np Optional external body contact node name or number mname MOSFET model name reference L SOI MOSFET
392. l 1 X 2005 09 1 Overview of MOSFET Models both the time required to create and simulate a netlist and the risk of errors compared to fully defining each element within your netlist One type of model that you can use as a template to define an element in your netlist is a Metal Oxide Semiconductor Field Effect Transistor MOSFET device This manual describes the MOSFET models supplied for use with HSPICE A MOSFET device is defined by the MOSFET model and element parameters and two submodels selected by the CAPOP and ACM model parameters The CAPOP model parameter specifies the model for the MOSFET gate capacitances The ACM Area Calculation Method parameter selects the type of diode model to use for the MOSFET bulk diodes Parameters in each submodel define the characteristics of the gate capacitances and bulk diodes MOSFET models are either p channel or n channel models they are classified according to level such as LEVEL 1 or LEVEL 50 This manual describes Design model and simulation aspects of MOSFET models Parameters of each model level and associated equations Parameters and equations for MOSFET diode and MOSFET capacitor models Over the years Synopsys has developed or adapted 64 different versions or levels of MOSFET models for use with HSPICE Currently Synopsys fully and officially supports 32 different MOSFET models The MOSFET models described in this manual are the most currently
393. l 60 UC Berkeley BSIM3 SOI DD Model Table 141 MOSFET Level 60 DC Parameters Continued SPICE Description Unit Default See Symbol Table 144 cii First Vdsatii parameter for the Vds dependence 0 0 dii Second Vdsatii parameter for the Vds V 1 0 dependence alphaO First parameter of the impact ionization current m V 0 0 alpha1 Second parameter of the impact ionization 1 V 1 0 current beta0 Third parameter of the impact ionization current V 30 Agidl GIDL constant W 1 0 0 Bgidl GIDL exponential coefficient V m 0 0 Ngidl GIDL Vds enhancement coefficient V 1 2 ntun Reverse tunneling non ideality factor 10 0 Ndiode Diode non ideality factor 1 0 Isbjt BJT injection saturation current A m2 1e 6 Isdif Body to source drain injection saturation current A m2 0 0 Isrec Recombination in the depletion saturation A m2 1e 5 current Istun Reverse tunneling saturation current A m2 0 0 Edl Electron diffusion length m 2e 6 Kbjt1 Parasitic bipolar early effect coefficient m V 0 Rbody Intrinsic body contact sheet resistance ohm m2 0 0 Rbsh Extrinsic body contact sheet resistance ohm m2 0 0 rsh Source drain sheet resistance in ohm per O square 0 0 square 500 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model Table 142 MOSFET 60 AC and Capacitance Parameters SPICE Description Unit Default See Symbol Table 144 xpart Charge partition
394. l card If you specify RTHO it overrides RTHO in the model card CTHO Thermal capacitance per unit width Ifyou do not specify CTHO simulation extracts it from the model card If you specify CTHO it overrides CTHO in the model card OFF Sets the initial condition to OFF for this element in DC analysis BJTOFF Turns off BUT if equal to 1 IC Initial guess in the order drain front gate internal body back gate external voltage Simulation ignores Vps in a 4 terminal device Use these settings if you specify UIC in the TRAN statement The IC statement overrides it Level 60 Model Parameters Table 139 MOSFET Level 60 BSIMSOI Model Control Parameters SPICE Description Unit Default See Symbol Table 144 shMod Flag for self heating 0 0 2 no self heating e 1 self heating mobmod Mobility model selector 1 capmod Flag for the short channel capacitance model 2 nl 1 noimod Flag for the noise model 1 HSPICE MOSFET Models Manual 495 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model Table 140 MOSFET Level 60 Process Parameters SPICE Description Unit Default See Symbol Table 144 Tsi Silicon film thickness m 10 7 B Tbox Buried oxide thickness m 3x10 Tox Gate oxide thickness m 1x108 Nch Channel doping concentration 1 cm3 17x10 7 Nsub Substrate doping concentration 1 cm 6x1016 nl 2 ngate Poly gate doping concentration 1 cm3 0
395. l parameters are in all upper case Roman These expressions assume that you have already adjusted all model parameters for geometry and that you have already adjusted parameters without a trailing O for the bias as appropriate The exceptions are U1 and N for which the following equations explicitly calculate the bias dependences Threshold voltage Vp Vi Vy KU PHI V KJ PHI Vy ETA Vi The following equation calculates the vp value used in the preceding equation Vj VFB PHI HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model Strong inversion Vos 5 Vin VGHIGH Linear region Vas Vasat drain source current Ips i a B Vu Sv Vis Ips T 1 UA Va UB V V4 UL Va The following equations calculate values used in the preceding equation V Vee Vin dsat a JK k LtV 4 1 2V TE Ur V Uis V Vin co 2 37 all UAV V4 UB V Va UID V V y UI Pag Van Na AD der 2 Visat In the preceding equations x is the usual unit step function Vi B os B tanh MU2 7 B3 Vas Ba V3 sa W eff Bo T MU Coy Loff p Bs Bo B VDD B VDD W K1 g B mui C i S 3 4 pelt i L di 2 PHI V PONES NS 7 771344 0 8364 PHI V HSPICE MOSFET Models Manual 363 X 2005 09 6 BSIM MOSFET Mo
396. ld coefficient LETAMN um 0 0 Length sensitivity WETAMN um 0 0 Width sensitivity GAMMN y1 2 0 0 Minimum root vsb threshold coefficient LGAMN V2 um 0 0 Length sensitivity WGAMN V 2 um 0 0 Width sensitivity K1 y1 2 0 5 Root vsb threshold coefficient LK1 V 2 m 0 0 Length sensitivity WK1 V 2 um 0 0 Width sensitivity K2 0 0 Linear vsb threshold coefficient LK2 um 0 0 Length sensitivity WK2 um 0 0 Width sensitivity MUZ cm2 V s 600 Low drain field first order mobility LMUZ um cm V s 0 0 Length sensitivity WMUZ um cm2V g 0 0 Width sensitivity NO 200 Low field weak inversion gate drive coefficient value of 200 for NO disables the weak inversion calculation LNO um 0 0 Length sensitivity HSPICE MOSFET Models Manual 349 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modified BSIM Model Table 87 Transistor Process Parameters Continued Name Alias Units Default Description WNO um 0 0 Width sensitivity NBO 0 0 Vsb reduction to the low field weak inversion gate drive coefficient LNB um 0 0 Length sensitivity WNB um 0 0 Width sensitivity NDO 0 0 Vds reduction to the low field weak inversion gate drive coefficient LND um 0 0 Length sensitivity WND um 0 0 Width sensitivity PHIO V 0 7 Two times the Fermi potential LPHI V um 0 0 Length sensitivity WPHI V um 0 0 Width sensitivity TOXM TOX um m 0 02 Gate oxide thickness if TOXM or TOX gt 1 uses Angstroms U00 1 V 0 0 Gate field mobility re
397. ld mobility reduction MOB 5 Universal field mobility reduction with an independent drain field HSPICE MOSFET Models Manual 171 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model Parameter Description MOB 6 Modified MOB 3 equations lateral field effect included MOB 7 Modified MOB 3 equations lateral field effect not included The following sections describe these equations MOB 0 Default No Mobility factor 1 0No mobility reduction Table 31 MOB 1 Gm Equation Name Alias Units Default Description F1 1 V 0 0 Gate field mobility reduction UTRA F3 factor 0 0 Source drain mobility reduction factor Use the MOB 1 equation for transistors with constant source to bulk voltage because the factor does not contain a vsb term This equation sometimes over estimates mobility for small gate voltages and large back bias such as depletion pull ups 1 t m ux SCO T CRISE a FS Bude vde min vds vdsat 172 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model Note In the alternate saturation model vde is different if UPDATE 0 than if UPDATE 1 See Alternate DC Model ISPICE model on page 168 Also if VMAX gt 0 then vde min vds vsat If you do not specify VMAX then vde min vds vdsat Table 32 MOB 2 Frohman Bentchkowski Equation Name Alias Units Default Description F1 V cm 0 0 Cr
398. le ETA0 0 02 not ETAO 0 02 Table 83 Transistor Parameters MOSFET Level 13 Name Alias Units Default Description LEVEL 1 MOSFET model level selector 13 is the BSIM model CGBOM CGBO F m 2 0e 10 Gate to bulk parasitic capacitance F m of length CGDOM CGDO F m 1 5e 9 Gate to drain parasitic capacitance F m of width CGSOM CGSO F m 1 5e 9 Gate to source parasitic capacitance F m of width DLO um 0 0 Difference between drawn poly and electrical DWO um 0 0 Difference between drawn diffusion and electrical DUM1 0 0 Dummy not used DUM2 0 0 Dummy not used ETAO 0 0 Linear vds threshold coefficient LETA mm 0 0 Length sensitivity WETA um 0 0 Width sensitivity K1 y 0 5 Root vsb threshold coefficient HSPICE MOSFET Models Manual X 2005 09 325 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 326 Table 83 Transistor Parameters MOSFET Level 13 Continued Name Alias Units Default Description LK1 V 2 um 0 0 Length sensitivity WK1 V 2 um 0 0 Width sensitivity K2 0 0 Linear vsb threshold coefficient LK2 um 0 0 Length sensitivity WK2 um 0 0 Width sensitivity MUS cm V s 600 High drain field mobility LMS LMUS um cm V s 0 0 Length sensitivity WMS WMUS um cm V s 0 0 Width sensitivity MUZ cm V s 600 Low drain field first order mobility LMUZ um cm V s 0 0 Length sensitivity WMUZ um cm V s 0 0 Width sensitivity NO 0 5 Low field weak inversion gate dri
399. le cdgb double cddb double cdsb noise parameters double nois irs Source noise current 2 double nois ird Drain noise current 2 double nois idsth channel thermal shot noise current 2 double nois idsfl 1 f channel noise current 2 double freq ac frequency extended model topology char topovar topology variables double leff effective channel length double weff effective channel width CMI VAR The CMrenv global variable defines the nominal and device temperature The pCMIenv pointer to the global CMlenv structure global variable accesses the CMlenv structure The CMI_ENV type in the include CMldef h file defines the structure for CMlenv environment variables typedef struct CMI env double CKTtemp simulation temperature double CKTnomTemp nominal temperature double CKTgmin GMIN for the circuit int CKTtempGiven temp setting flag hspice specific options follow double aspec double spice double scalm CMI ENV model parameters for JFET amp MESFET JFET amp MESFET model parameter for CMI VAR in CMIdef h double gg gate conductance double cigs gate to source current double gigs gate to source conductance double cigd gate to drain current double gigd gate to drain
400. lectron charge q k 1 3807 x 10 UK Boltzmann constant T m 300 15 K Reference temperature Taom K Nominal temperature of model parameters T K Model simulation temperature VAT a Thermal voltage T ET 1 16 0 000702 T 110 eV Energy gap nf 145x 106 3 T E Tref E T m Intrinsic carrier x T1 2 V T 2 VAT concentration Parameter Preprocessing Handling of Model Parameters for P Channel MOSFETs For P channel devices simulation reverses the sign of VFB VTO and TCV before processing Therefore VTO and TCV are usually positive and VFB is usually negative for N channel and vice versa for P channel MOSFETs Initializing Intrinsic Parameters The basic intrinsic model parameters are related to the fundamental process parameters as in early SPICE models COX TOX GAMMA and PHI 3NSUB VTO gt VFB KP UO UCRIT VMAX HSPICE MOSFET Models Manual 231 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model 232 For statistical circuit simulation you should introduce parameter variations on the level of the latter parameters You can also use these dependencies to analyze device scaling and to obtain parameter sets from other MOSFET models Therefore you can use the following relations If you do not specify COX simulation initializes it as E TOX for TOX5 O COX default otherwise If you do not specify GAMMA simulation ini
401. lects 2 2 11 3 2 12 13 28 39 13 13 others 2 11 LEVELs 49 and 53 use the Berkeley CAPMOD capacitance model parameter Proprietary models and LEVELs 5 17 21 22 25 27 31 33 49 53 55 and 58 use built in capacitance routines HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models Selecting Models Selecting MOS Diodes The ACM Area Calculation Method model parameter controls the geometry of the source and drain diffusions and selects the modeling of the bulk to source and bulk to drain diodes of the MOSFET model The diode model includes the diffusion resistance capacitance and DC currents to the substrate Parameter Description ACM 0 SPICE model Element areas determine the parameters ACM 1 ASPEC model The element width determines the parameters ACM 2 Synopsys model combination of ACM 0 1 with provisions for lightly doped drain technology ACM 3 Extension of ACM 2 model that deals with stacked devices shared source drains and source drain periphery capacitance along the gate edge For more about ACM see MOSFET Diode Models on page 39 Searching Models as Function of W L Model parameters are often the same for MOSFETs that have width and length dimensions within specific ranges To take advantage of this create a MOSFET model for a specific range of width and length These model parameters help the simulator to select the appropriate model for the specified width and length
402. ling These values and changes provide a more accurate ASPEC model UPDATE 1 or 2 TOX 690 UO UB 750 cm V s N ch UTRA F3 0 0 UPDATE 0 TOX 1000 UO UB 750 cm2 V s N ch UTRA F3 0 0 If you do not specify LDIF then the RD and RS values change in the MOSFET UPDATE 1 or 2 and LDIF 0 RD NRD RL M RD Rs RS NRS RL M Note The ASPEC program does not use the M multiplier HSPICE MOSFET Models Manual 157 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model LDIF 0 LATDscaled LDIFscaled RL RD NRD Weff M LATDscaled LDIFscaled RL So RS NRS I Weff M The vde value in the mobility equations change for the alternate saturation model RD RS vde vds min Ke vsat UPDATE 1 or2 vde min vds vfa vsat UPDATE 0 The impact ionization equation calculates the saturation voltage vdsat vfa vsat UPDATE 1 or 2 vdsat vsat UPDATE 0 The MOB 3 mobility equation changes UPDATE 1 or 2 and vgs vth gt VF1 ueff eB F4 4 F1 F3 VF1 F3 vgs vth UPDATE 0 and vgs vth F2 gt vF1 UB ue EE e f F4 F3 vgs vth 158 HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model LEVEL 6 Model Equations UPDATE 0 2 IDS Equations ids p oe ybi de vde by P PHI
403. lt Description THETA M V 0 Mobility degradation parameter ISUBMOD 0 Channel length modulation selector 0 uses LAMBDA 1 uses LS and VP LAMBDA 1 V Channel length modulation parameter LS 3 50E 008 Channel length modulation parameter VP V 0 2 Channel length modulation parameter INTDSNOD 1 if VERSION 1 Extrinsic series resistance mode 0 if VERSION 2 selector 0 uses internal approximation for drain and source resistances 1 uses drain and source resistances as extrinsic elements VSIGMAT V VST if specified 1 7 VGS dependence parameter otherwise VSIGMA V VSI if specified 0 2 VGS dependence parameter otherwise Table 60 Self Heating Parameters Name Unit Default Description SHMOD 0 Self heating flag RTHO m9c N o0 Thermal resistance per unit width CTHO Ws m9C 1 00E 005 Thermal capacitance per unit width WTHO m 0 Width offset for thermal resistance and capacitance scaling 270 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model Table 61 ACM Parameters for Drain and Source Resistanance Calculus Specific to HSPICE Name Unit Default Description ACM 5 Area calculation method LD m 0 Lateral diffusion into channel from source and drain diffusion WD m 0 Lateral diffusion into channel from bulk along width HDIF m 0 Length of heavily doped diffusion LDIF m 0 Length of heavily doped region adjacent to gate RDC ohm 0 Addition
404. ltages Q1 V1 V2 V3 V4 Q4 V1 V2 V3 V4 The derivatives form a four by four matrix dQi dVj i 1 4 j 1 4 Simulation directly interprets this matrix as AC measurements If you apply an AC voltage signal to the j terminal and you ground the other terminals to AC and if you measure AC current into the terminal then the current is the imaginary constant times 2 pi frequency times dQi dVj HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Because the charges add up to zero each column of this matrix must add up to zero Because the charges can depend only on voltage differences each row must add up to zero In general the matrix is not symmetrical dQi dVj need not equal dQj dVi This is not an expected event because it does not occur for the two terminal case For two terminals because the rows and columns must add up to zero dQ1 dV2 must equal dQ2 dV1 For three or more terminals this relation does not generally hold The terminal input capacitances are the diagonal matrix entries Cii dQi dVi tel ra The transcapacitances are the negative of off diagonal entries Cij dQi dVj i not equal to j All of the C values are normally positive Figure 17 MOS Capacitances Source CSS HSPICE MOSFET Models Manual 67 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models 68 In Figure 17 Cij determines the current transfe
405. luding the device size effects and the terminal voltages vth vfb Bd vch y Phid vsb vcrit The following equations calculate values used in the preceding equation vfb zVTO zETA vds Bd vch y Phid verit v L Phid vsbc for vsb gt vsbc O otherwise zBetaGam Bd UH cav vch LMI zUO cox cav _ 2 esi qnal Wc EIS Cav nal d DNB ad NI nd DNB DP 1e 4 The following equation computes the effective y including small device size effects y Y for vsb vsbc and g otherwise zBetaGam Y Yo 1 scf 19 ncf HSPICE MOSFET Models Manual 199 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 38 IDS Cypress Depletion Model The following equations calculate values used in the preceding equations If SCM lt 0 then scf 0 Otherwise m 1 2 ue De pae SCM vds vsb Phid 1 Leff L XJ If NWM s0 then ncf 0 Otherwise NWM xd Phid Weff This equation calculates the xd value used in the preceding equation a xd q DNB The following equation calculates the body effect transition point ncf Vsbc ape NI _ _ ___ DNB zDVSBC TDVSBC t tnom Phid 2esi DP 1e 4 If vgs lt vth this model inverts the surface and includes a residual DC current If vsb is large enough to make vth vinth then vth is the inversion threshold voltage To determine the residual current simulat
406. ly 40 000 V cm ECV V um 1000 Critical field 5 38 FEFF 0 5 Frequency effect constant 40 FREQ Hz 400 Frequency of the device 40 GO ohm 10e 15 Conductance of the TFT leakage current 40 IIRAT 0 Impact ionization source bulk current 39 partitioning factor One corresponds to 100 source Zero corresponds to 100 bulk JS Am 0 Source drain bulk diode reverse saturation 39 current density K2 2 0 Temperature exponential part 40 HSPICE MOSFET Models Manual X 2005 09 111 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 25 Basic MOSFET Model Parameters Continued Name Alias Units Default Description Level KAPPA vi 0 2 Saturation field factor The channel length 3 modulation equation uses this parameter KCS 2 77 Implant capacitance integration constant 38 KP BET A N Intrinsic transconductance parameter If yu 1 2 3 BETA specify UO and TOX but you do not specify KP simulation computes the parameter from KP UO COX Level 1 default 2 0718e 5 NMOS 8 632e 6 PMOS Level 2 3 default 2 0e 5 LAMBDA y1 0 0 Channel length modulation 2 8 LAM LA MJ 0 5 Source drain bulk junction grading 39 coefficient MJSW 0 33 Sidewall junction grading coefficient 39 NEFF 1 0 Total channel charge fixed and mobile 2 coefficient NI cm 2e11 Implant doping depletion model only 5 38 NU 0 0 First order temperature gradient 40 PB V 0 8 Source drain bulk junction potential 39 PB
407. ly depletion Table 71 Level 64 Short Channel Parameters Parameter Default Description PARL1 1 0 Strength of the lateral electric field gradient PARL2 2 2e 8m Depletion width of the channel contact junction SC1 135v X Short channel coefficient 1 SC2 1 8V Short channel coefficient 2 SC3 0 0V m Short channel coefficient 3 SCP1 0 0 v Short channel coefficient 1 for the pocket SCP2 0 0V Short channel coefficient 2 for the pocket SCP3 0 0v2m Short channel coefficient 3 for the pocket Table 72 Level 64 Narrow Channel Parameters Parameter Default Description WFC 0 0m F cm Voltage reduction MUEPH2 0 0 Mobility reduction WO 0 0log cm Minimum gate width HSPICE MOSFET Models Manual X 2005 09 315 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model 316 Table 72 Level 64 Narrow Channel Parameters Continued Parameter Default Description WVTHSC 0 0 Short channel effect at the shallow trench isolation STI edge if VERSION gt 110 NSTI 0 0cm3 Substrate impurity concentration at the shallow trench isolation STI edge if VERSION gt 110 WSTI 0 0m Width of the high field region at the shallow trench isolation STI if VERSION gt 110 Table 73 Level 64 Mobility Parameters Parameter Default Description VDSO 0 05V Drain voltage for extracting low field mobility MUECBO 300 0cm Vs Coulomb scattering MUECB1 30 0cm Vs Coulomb scattering MUEPHO 0 295 Phonon scattering MUEPH
408. me Default Description muluO 1 0 Low field mobility UO multiplier mulua 1 0 First order mobility degradation coefficient UA multiplier mulub 1 0 Second order mobility degradation coefficient UB multiplier When HSPICE prints back a MOSFET element summary OPTION LIST it identifies the BSIM3V3 MOSFET and prints back these three additional instance parameters Using BSIM3v3 Note the following points when you use BSIM3v3 with a Synopsys circuit simulator Use either the Level 49 or Level 53 model Level 53 fully complies with the Berkeley BSIM3v3 release In most cases Level 49 returns the same results as Level 53 runs as fast or faster shows better convergence and allows a wider range of parameter specifications Explicitly set all Berkeley specific BSIM3 model parameters in the model card This minimizes problems resulting from version changes and compatibility with other simulators You do not explicitly set all lwp binning parameters HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models To match results with simulations from previous HSPICE versions use the HSPVER YY N model parameter such as HSPVER 97 4 Do not use the full year specification such as 1997 4 The patch version number format is HSPVER YY NN for example HSPVER 98 21 is release 98 2 1 Levels 49 and 53 support the TNOM model parameter name as an alias for TREF The conve
409. meter values However Level 13 does not eliminate the GDS discontinuity in LEVEL 3 or the GM discontinuity in LEVEL 2 The next section provides model versus data plots for drain and gate sweeps followed by close up plots of the models using a small step size to show GM and GDS problems in the individual levels LEVEL 28 2 3 Ids Model vs Data ds versus Vds at Vgs 1 2 3 4 5 Vbs 0 Fits to IDS only not GDS and GM might look better for these plots but would not be acceptable for analog design Figure 48 LEVEL 2 Ids Model versus Data Curves 12 00 E C MODEL Param Lin PDW Lin HSPICE MOSFET Models Manual X 2005 09 B Comparing MOS Models Examples of Data Fitting Figure 49 LEVEL 28 Ids Model versus Data Curves 12 0U 10 0U 8 0U 6 0U Param Lin 20U C MODEL PDW Lin 3 0 4 0 5 0 Figure 50 LEVEL 3 Ids Model versus Data Curves 12 0U 10 0U 8 0U 6 0U Param Lin 40U 7 MODEL c Ec DATA tri AUS 2 0U 2 0 PDW Lin 3 0 4 0 5 0 HSPICE MOSFET Models Manual X 2005 09 581 B Comparing MOS Models Examples of Data Fitting LEVEL 13 28 39 Ids Model vs Data Ids versus Vds at Vgs 1 2 3 4 5 Vbs 0 Figure 51 LEVEL 13 Ids
410. model DLC LINT m No Channel length offset parameter for the CV model DWC WINT m No Channel width offset parameter for the CV model VFBCV 1 0V Yes Flat band voltage parameter for CAPMOD 0 only NOFF 1 0 Yes CV parameter in Vgctett cy for weak to strong inversion VOFFCV 0 0V Yes CV parameter in Voster cy for weak to strong inversion 454 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 120 Charge Capacitance Model Parameters MOS 54 Continued Parameter Default Bin Description nab le ACDE 1 0m V Yes Exponential coefficient for the charge thickness in CAPMOD 2 for the accumulation and depletion regions MOIN 15 0 Yes Coefficient for the gate bias dependent surface potential Table 121 High Speed RF Model Parameters MOS Level 54 Parameter Default Binnable Description XRCRG1 12 0 Yes Parameter for the distributed channel resistance effect for both intrinsic input resistance and charge deficit NQS models XRCRG2 1 0 Yes Parameter to account for the excess channel diffusion resistance for both intrinsic input resistance and charge deficit NQS models RBPB 50 0ohm No Resistance between bNodePrime and bNode RBPD 50 0ohm No Resistance between bNodePrime and dbNode RBPS 50 0ohm No Resistance between bNodePrime and sbNode RBDB 50 0ohm No Resistance between dbNode and dbNode RBSB 50 0ohm No Resistance between sbNode and bNode GBMIN 1 0e 12mho No Conductance in pa
411. model is based on the UC Berkeley BSIM4 MOS model BSIM4 4 is fully supported in this release For details see the BSIM 434 web site http www device eecs berkeley edu General Form Mxxx nd ng ns nb mname lt L val NRS val NR RBPS val RB D val lt W val gt lt M val gt lt AD val gt lt AS val gt lt PD val gt lt PS val gt lt RGATEMOD val gt lt RBODYMOD val TRNOSMOD val ACNOSMOD val GEOMOD val lt RGEOMOD val gt lt RBPB val gt lt RBPD val gt MULUO val x l DELK WNFLAG val MIN val lt RDC val lt lt Vgs DB val RBSB val lt NF val gt lt RSC val gt lt I val gt lt DELNFCT val gt DELTOX val gt lt OFF gt lt IC Vds DELVTO val gt Vbs Parameter Description ACNQSMOD AC small signal NQS model selector AD Drain diffusion area AS Source diffusion area DELK1 Shift in body bias coefficient K1 DELNFCT Shift in subthreshold swing factor NFACTOR DELTOX Shift in gate electrical and physical oxide thickness TOXE and TOXP That is the difference between the electrical and physical gate oxide insulator thicknesses DELVTO Shift in the VTHO zero bias threshold voltage DELVTO HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Paramet
412. model selector e for VERSION lt 110 COISUB 0 yes otherwise COISUB 1 no COIIGS 0 Selects the gate tunneling current model COIIGS 0 yes COIIGS 1 no VERSION 111 does not support this model COGIDL 0 Selects the gate induced drain leakage GIDL current model e COGIDL 0 yes e COGIDL 1 no VERSION 111 does not support this model CONOIS 0 1 f noise model selector CONOIS 0 no e CONOIS 1 yes COISTI 0 Selects the shallow trench isolation STI leakage current COISTI 0 no e COISTI 1 yes only if VERSION gt 110 312 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model Table 66 Level 64 Model Selectors Continued Parameter Default Description NOISE 5 Channel thermal and flicker noises combination selector e NOISE 1 Channel thermal noise SPICE2 model Flicker noise SPICE2 model e NOISE 2 Channel thermal noise HiSIM1 model for the BSIM3 model Flicker noise HiSIM1 model e NOISE 3 Channel thermal noise SPICE2 model Flicker noise HiSIM1 model e NOISE 4 Channel thermal noise HiSIM1 model for the BSIM3 model Flicker noise SPICE2 model e NOISE 5 Channel thermal noise NONE Flicker noise HiSIM1 model Table 67 Level 64 Technological Parameters Parameter Default Description TOX 3 6e 9m Oxide thickness XLD 0 0m Gate overlap length XWD 0 0m Gate overlap width XPOLYD 0 0m Differe
413. n Param Lin CAPOP1 SVO 30 0F C68 VOSPO05 A C68 VOSP05 D N o o T G88 VOSP05 z amp G d o o o T A CAPOP1 SVO abe z C68 VOSP05 A C68 VOSP05 G 20 0F 7 Volts Lin CAPOP 3 Gate Capacitances Simpson Integration The CAPOP 3 model uses the same set of equations and parameters as the CAPOP 2 model Simulation obtains the charges using Simpson numeric integration instead of the box integration found in the CAPOP 1 2 and 6 models Gate capacitances are not constant values with respect to voltages The incremental capacitance best describes the capacitance values In the preceding equation q v is the charge on the capacitor and v is the voltage across the capacitor HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models The formula for calculating the differential is difficult to derive Furthermore the voltage is required as the accumulated capacitance over time The timewise formula is i x co LO The charge is qv C v dv 0 To calculate the current i LO E foa 0 For small intervals Vin 1 _ dq v _ 1 WE GA J Vin C v dv In SPICE the following equation approximates the integral Vin 1 cer pus 1 CLM CRI m Katy 2 This last formula is the trapezoidal rule for integration over two points The charge is approximate
414. n Angstroms TEMP C Not used in Level 39 see Compatibility Notes on page 366 VDD V 5 Drain supply voltage NMOS convention VGG V 5 Gate supply voltage NMOS convention VBB V 5 Body supply voltage NMOS convention DL m 0 Channel length reduction DW m 0 Channel width reduction VGHIGH V 0 Upper bound of the weak strong inversion transition region VGLOW V 0 Lower bound of the weak strong inversion transition region VFB V 0 3 Flat band voltage PHI V 0 8 Surface potential K1 vi 0 5 Body effect coefficient K2 0 Second order body effect coefficient for nonuniform channel doping ETAO 0 Drain induced barrier lowering coefficient ETAB vi 0 Sensitivity of the drain induced barrier lowering coefficient to Vhs 359 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model 360 Table 89 BSIM2 Model Parameters Continued Name Alias Units Default Description MUO cm2 V s 400 Low field mobility MUOB ene ngo a 0 Sensitivity of low field mobility to Vps MUSO cm2 V s 600 High drain field mobility MUSB cm2 V2 5 O Sensitivity of the high drain field mobility to Vps MU20 0 Empirical parameter for the output resistance MU2B vi 0 Sensitivity of the empirical parameter to Vps MU2G y1 0 Sensitivity of the empirical parameter to Vg MUSO cm V s O Empirical parameter for the output resistance MUSB cm2 V3 5 O Sensitivity of the empirical parameter to Vps MU3G cm2 V3 s O Sensitivity of the empirical parameter
415. n at a time However the performance difference between dynamic loading and static binding is usually less than 5 The Customer CMI includes several source code examples for integration of MOS JFET and MESFET models in simulation They are standard Berkeley SPICE MOSFET models LEVEL 1 2 3 BSIM1 2 3 and JFET MESFET models To minimize the effort required for adding models Synopsys provides installation scripts which automate the shared library generation process If you derive your proprietary models from SPICE models the integration process is similar to the examples with minimal modifications Note HSPICE or HSPICE RF includes equations and programs for bias calculation numerical integration convergence checking and matrix loading You do not need to use these programs to complete a new model integration so the source code examples do not include them The Customer CMI supports the following platforms Sun Solaris 2 5 2 7 2 8 m HP UX 10 20 11 0 m RedHat Linux6 2 Linux7 0 Linux7 1 Windows2000 Windows NT and Windows XP HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Directory Structure Directory Structure Figure 40 shows the structure of the Customer CMI distribution for Unix Sun and HP and Linux platforms Figure 41 on page 520 shows the structure of the Customer CMI distribution for the PC Windows 95 Win98 Win2000 Windows NT and Windows XP platforms You mu
416. n calculates the Leff value for each model differently and saves this value in the corresponding model section The Weff calculation is not the same as the weff value in the LEVEL 1 2 3 6 7 and 13 models Weff M Wscaled WMLT XWscaled The 2 WDscaled factor is not subtracted CAPOP 0 SPICE Meyer Gate Capacitances Definition cap COXscaled Weff Leff Gate Bulk Capacitance cgb Accumulation vgs lt vth PH1 cgb cap vth vgs Depletion vgs vth cgb cap PHI HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Strong Inversion vgs 2 vth cgb 0 Gate Source Capacitance cgs Accumulation vgs vth EH cgs 0 Depletion vgs lt vth cgs CF5 cap tap cn Strong Inversion Saturation Region vgs vth and vds 2 vdsat cgs CF5 cap Strong Inversion Linear Region vgs gt vth and vds lt vdsat L2 vdsat vsb vds vsb vdsat vds a cgs CF5 cap LS e Gate Drain Capacitance cgd The gate drain capacitance has value only in the linear region Strong Inversion Linear Region vgs gt vth and vds lt vdsat i ni cgd CF5 cap 2 vdsat vsb vds vsb Example The netlist for this example is located in the following directory Sinstalldir demo hspice mos capop0 sp HSPICE MOSFET Models Manual 79 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Figure
417. n model is more charge conserving than the original Ward Dutton model in SPICE 2G6 Calculating the automatic diode area and the resistance estimates the junction capacitance saturation current and resistance as a function of the transistor width Use the VNDS and NDS parameters for a piecewise linear approximation to reverse the junction current characteristics Figure 32 Non Fully Depleted SOI Model source Isolation Isolation Insulator Silicon Substrate Example The netlist for this example is located in the following directory Sinstalldir demo hspice mos ssoi sp HSPICE MOSFET Models Manual 193 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 27 SOSFET Model 194 Figure 33 LEVEL 3 Floating Bulk Model SOI SP LEVEL 3 FLOATING SUBSTRATE MODEL NON FULLY DEPLETED 14 MAY 2003 16 51 37 2 20M 2 0M 1 80M gt 1 60M 7 1 40M 7 1 20M 7 10M 800 0U 7 600 0U 400 0U 7 200 0U 7 E 2 0 VOLTS LIN Fully Depleted SOI Model Considerations Fully depleted transistors require additional modeling equations The first order effects are Threshold sensitivity to the substrate No kink current m Silicon thickness limits the minimum depletion capacitance Lack of these effects does not seriously affect an inverter because the source to substrate voltage does not move Digital circuits with good gate drive are not seriously affected because a la
418. n parallel The M setting affects all channel widths diode leakages capacitances and resistances Default 1 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIMS3 SOI Model Parameter Description AD Drain diffusion area Overrides OPTION DEFAD statement Default DEFAD AS Source diffusion area Overrides OPTION DEFAS statement Default DEFAS PD Drain junction perimeter including channel edge Overrides OPTION DEFPD PS Source junction perimeter including channel edge Overrides OPTION DEFPS NRD Number of squares of drain diffusion for the drain series resistance Overrides OPTION DEFNRD NRS Number of squares of source diffusion for the source series resistance Overrides OPTION DEFNRS NRB Number of squares for the body series resistance FRBODY Coefficient of the distributed body resistance effects Default 1 0 RTHO Thermal resistance per unit width If you do not specify RTHO simulation extracts it from the model card f you specify RTHO it overrides RTHO in the model card CTHO Thermal capacitance per unit width f you do not specify CTHO simulation extracts it from the model card If you specify CTHO it overrides CTHO in the model card NBC Number of body contact isolation edge NSEG Number of segments for partitioning the channel width PDBCP Parasitic perimeter length for the body contact a the drain side PSBCP Parasitic perimeter le
419. nce between the gate poly and the design lengths TPOLY 0 0m Height of the gate poly Si RS 0 0ohm m Source contact resistance RD 0 0ohm m Drain contact resistance HSPICE MOSFET Models Manual 313 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model Table 67 Level 64 Technological Parameters Continued Parameter Default Description NSUBC 5 94e 17cm Substrate impurity concentration NSUBP 5 94e417cm Maxim pocket concentration VFBC 0 722729V Flat band voltage LP 0 0m Pocket penetration length XJ 0 0m Junction depth if VERSION 110 XQY 0 0m Distance from the drain junction to the maximum electric field point if VERSION gt 110 Table 68 Level 64 Temperature Dependence Parameters Parameter Default Description BGTMP1 903e 5evK Bandgap narrowing BGTMP2 305e 7evK2 Bandgap narrowing Table 69 Level 64 Quantum Effect Parameters Parameter Default Description QME1 0 0mV Coefficient for the quantum mechanical effect QME2 0 0V Coefficient for the quantum mechanical effect QME3 0 0m Coefficient for the quantum mechanical effect 314 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model Table 70 Level 64 Poly Depletion Parameters Parameter Default Description PGD1 0 0V Strength of the poly depletion PGD2 0 0V Threshold voltage of the poly depletion PGD3 0 0 Vas dependence of the po
420. nce coefficient of SAT 0 0 THETHR Coefficient of self heating for the V 3 1E 3 0 5E 3 reference transistor at the reference temperature HSPICE MOSFET Models Manual X 2005 09 289 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 63 Level 63 MOS11 Parameters Level 11010 Physical Geometry Scaling Continued Name Description Units NMOS PMOS THETHEXP Exponent of the length dependence of 1 1 0TH SWTHETH Coefficient of the width dependence of 0 0 0TH SDIBLO Drain induced barrier lowering parameter V 1 2 1E 4 1E 4 for the reference transistor SDIBLEXP Exponent of length dependence of oDIBL 1 35 1 35 MOR Parameter for short channel subthreshold 0 0 slope for the reference transistor MOEXP Exponent of the length dependence of mO 1 34 1 34 SSFR Static feedback parameter reference V 1 2 6 25E 3 6 25E 3 transistor SLSSF Length dependence coefficient of oSF 1 0 1 0 SWSSF Coefficient of the width dependence of 0 0 oSF ALPR Factor of the channel length modulation 1E 2 1E 2 for the reference transistor SLALP Coefficient of the length dependence ofa 1 1 ALPEXP Exponent of the length dependence ota 1 1 SWALP Coefficient of the width dependence ofa 0 0 VP Characteristic voltage of the channel V 5E 2 5E 2 length modulation LLMIN Minimum effective channel length in M 1 5E 7 1 5E 7 technology calculates smoothing factor m 290 HSPICE MOSFET Models Manual X 2005 09 5 Stan
421. nces 0 a Noise Model ree eal ieee RR ae edt Performance Improvements 0000e eee eae Reduced Parameter Set BSIM3v3 Model BSIM3 lite Parameter Binning 0 cee eee eee eee Charge Models cee eects VEBPLEAGS naaa ddan ieee nad Iced Rie PrintDack uc gece rd nant Gerth PEE Re EIS VENTE E Melee be C Mobility Multiplier B BBBIRIR Using BSIMI VIZ 2 cir yoke Ginga KGG ANAN wate tend Parameter Range Limits llle Level 49 53 Equations 0000 cece ees MODEL CARDS NMOS Model 200000 ee eee PMOS Model 0 000 cece eee ee Level 54 BSIM4 Model 222 eee ees General FOM un es ERE Ree ew eh ds Improvements Over BSIM3v3 0000 cee eee eee TSMC Diode Model 0 000 cc ccc eee eee eee BSIM4 STILOD osx peter RR e e e Dao LMLT and WMLT in BSIM4 llle lessen HSPICE Junction Diode Model and ACM Level 54 BSIM4 Template Output List Level 57 UC Berkeley BSIM3 SOI Model General Syntax for BSIM3 SOI 00 Level 57 Model Parameters 000002 eee eeaee Level 57 Template Output 00000 e eee eee eee Level 57 Updates to BSIM3 SOI PD versions 2 2 2 21 and 2 22 HSPICE MOSFET Models Manual X 2005 09 Contents 396 397 398 398 400 402 402 404 404 406 407 407 407 410 411 411 412 412 412
422. nd WREF their value is infinity Reference channel length to adjust the length of the BSIM model parameters For Berkeley compatibility LREF eo use LREF 0 LREF csed LREF SCALM Oxide etch Px is a Synopsys proprietary WL product sensitivity parameter where x is a model parameter with length and width sensitivity Lateral diffusion into the channel from the bulk along the width WD caled WD SCALM Channel stop lateral diffusion under the gate per side Use this parameter to calculate Wor only if DW 0 WD scaled WD SCALM This parameter is the same as WD but if you specify WDAC in the MODEL statement it replaces WD in the Weff calculation for the AC gate capacitance Diffusion layer and width shrink factor Diffusion and gate width shrink factor Scale MOSFET drawn width 39 5 38 54 HSPICE MOSFET Models Manual X 2005 09 3 Common MOSFET Model Parameters Basic MOSFET Model Parameters Table 26 Effective Width and Length Parameters Continued Name Alias Units Default Description WREF XJ XL DL LDEL XL XLREF XW DW WDEL XW HSPICE MOSFET Models Manual X 2005 09 m um 0 0 0 0 1 5 0 0 0 0 0 0 Channel width reference WREF calea WREF SCALM If the Level 13 model does not define LREF and WREF their value is infinity Reference device width to adjust the width of the BSIM model parameters For Berkeley
423. nd the LEVEL 28 model is monotonic decreasing HSPICE MOSFET Models Manual X 2005 09 589 590 B Comparing MOS Models Examples of Data Fitting Figure 66 LEVEL 2 gm lds versus Vgs Curves 110 0 100 0 90 0 80 0 A a rus Teo 60 0 AG joue p 3 50 0 40 0 70 0 Param Lin 300 E 20 0 a Coe Q Claus SS 3 C MODEL 500 0M 1 0 1 50 2 0 PGD Lin 2 50 Figure 67 LEVEL 28 gm lds versus Vgs Curves 110 0 100 0 900 800 70 0 60 0 MD M 50 0 N 40 0 Param Lin 30 0 E 20 0 10 0 gt QE do ae ec Ua tie AE F 3 C MODEL eT C DATA p te 500 0M 1 0 1 50 2 0 PGD Lin 2 50 3 00 HSPICE MOSFET Models Manual X 2005 09 Figure 68 LEVEL 3 gm lds versus Vgs Curves B Comparing MOS Models Examples of Data Fitting 100 0 Param Lin a o O Hk 3 DI MODEL kai DI DATA 500 0M 1 0 1 50 2 0 PGD Lin 2 50 3 00 LEVEL 13 28 39 Gm lds Model versus Data gm lds vs Vgs at Vds 0 1 Vbs 0 2 and 39 are monotonic HSPICE MOSFET Models Manual X 2005 09 LEVEL 13hasa kink at Vth which is not visible at this resolution LEVEL 28 591 B Comparing MOS Models Examples of Data Fitting Figure 69 LEVEL 13 gm Ids versus Vgs Curves 110 0 z DI MODEL
424. nductance of the diode and a TT model parameter representing HSPICE MOSFET Models Manual 55 X 2005 09 2 Technical Summary of MOSFET Models MOS Diode Equations 56 the transit time of the diode The depletion capacitance depends on which ACM you choose To calculate bias dependent depletion capacitance define COBS COBD COBS SW and COBD SW intermediate quantities These depend on geometric parameters such as ASeff and PSeff calculated under various ACM specifications For ACM 3 the COBS SW and COBD SW intermediate quantities include an extra term to account for CJGATE ACM 2 includes the CJGATE parameter for backward compatibility Therefore the default behavior of CJGATE makes the COBS SW and COBD SW intermediate quantities the same as for previous versions The default patterns are f you do not specify CJSW or CJGATE both default to zero f you do not specify CJGATE it defaults to CJSW which defaults to zero f you specify CJGATE but you do not specify CJSW then CJSW defaults to zero Simulation calculates the COBS COBS SW COBD and COBD SW intermediate quantities as follows COBS CJscaled ASeff COBD CJscaled ADeff f ACM O or 1 then COBS SW CJSWscaled PSeff COBD SW CJSWscaled PDeff If ACM 2 and PSeff lt Weff then COBS SW CJGATEscaled PSeff If ACM 2 and PSeff gt Weff then COBS_SW CJSWscaled PSeff Weff CJGATEscaled Weff f ACM 2 and P
425. nel length modulation effect Channel Length Modulation The LEVEL 5 model modifies the lgs current to include the channel length modulation effect Dis Ig e 1 AL Log The following equation calculates the AL value used in the preceding equation 1 3 AL 1le4 _2 73e3 XI v4 Vasat t PHD PHI 1 c DNB In a AL is in microns if XJ is in microns and DNB is in cm3 Subthreshold Current Ig The Fast Surface State FSS characterizes this region of operation if it is greater than 1e10 The following equation then calculates the effective threshold voltage separating the strong inversion region from the weak inversion region von V fast The following equation calculates the fast value used in the preceding equation eee NN EE cox 2 Ope vy In the preceding equations vt is the thermal voltage HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model The following equations calculate lgs Weak Inversion Region Vgs Vi V Jos VON Ij von vde v e fast Strong Inversion Region vgs gt vth Lu z Las gis vde Vsp Note Strong inversion also use the modified threshold voltage von that FSS produces that is the mobility equations use von instead of Vin Depletion Mode DC Model ZENH 0 The LEVEL 5 MOS model uses depletion mode devices as the load element in contemporary standard n channel technologies This model assumes a
426. neling current Le to drain and gate to source 0 for OFF double Igso gate to source tunneling current through G S HSPICE MOSFET Models Manual X 2005 09 overlap double Igdo gate overlap double Igb gate to bulk tun double Igcs gate to source t double Igcd gate u double gigsos Igso transconductan double gigsog Igso transconductan double gigcsg Igcs transconductan double gigcsd Igcs transconductan double gigcsb Igcs transconductan double gigcss Igcs transconductan dIgcs dVd dIgcs dVb double gigdod Igdo transconductan double gigdog Igdo transconductan double gigcdg Igcd transconductan double gigcdd Igcd transconductan double gigcdb Igcd transconductan double gigcds Igcd transconductan dlIgcd dVd dIgcd dVb double gigbg Igb transconductance double gigbd Igb transconductance double gigbb Igb transconductance double gigbs Igb transconductance dlIgb dVd dIgb dVb 8 Customer Common Model Interface neling current unneling current through channel to drain tunneling current through channel ce dIgso dVs ce dIgso dVg ce dIgcs dVg ce dlIgcs dVd ce dIgcs dVb ce dIgcs dVs ce dIgdo dVd ce dIgdo dVg ce dlIgcd dVg ce dIgcd dVd ce dIgcd dVb ce dIgcd dVs dIg
427. ng Annotator TopoPlace TopoRoute Trace On Demand True Hspice TSUPREM 4 TymeWare VCS Express VCSi Venus Verification Portal VFormal VHDL Compiler VHDL System Simulator VirSim and VMC are trademarks of Synopsys Inc Service Marks sv MAP in SVP Caf and TAP in are service marks of Synopsys Inc SystemC is a trademark of the Open SystemC Initiative and is used under license ARM and AMBA are registered trademarks of ARM Limited All other product or company names may be trademarks of their respective owners Printed in the U S A HSPICE MOSFET Models Manual X 2005 09 i HSPICE MOSFET Models Manual X 2005 09 Inside This Manual The HSPICE Documentation Contents Sebi trays et ct an tm adat aa Searching Across the HSPICE Documentation Set Other Related Publications Conventions Customer Support 1 Overview of MOSFET Models 0 0 000005 Overview of MOSFET Model Selecting Models TYPOS bias esr px et Aa Bis ch Saati eps Selecting MOSFET Model LEVELS 0 0 00 cece ene Selecting MOSFET Capacitors 0c eee eee Selecting MOS Diodes Searching Models as Function of W L n a naana a Setting MOSFET Control Options 000s Scaling Units Scaling for LEVEL 25 and 33 a Bypassing Latent Devices 0 0 0 0 c eee eee General MOSFET Model Statement llle
428. ng Proprietary MOS Models 3 Create a subdirectory for the new model under the working Customer CMI directory For example if your MOSFET model is LEVEL 222 copy the subdirectory from the existing MOS model LEVEL 3 as follows cp r mos3 mos222 4 Add the following line in the config configuration file mos222 222 my own MOSFET model Table 148 Configuration File MOS Syntax Parameter Description mos222 model name 222 model LEVEL my own MOSFET model descriptive comment for the model The model name and level must be unique within the configuration file For more information see the in line comment in the configuration file Preparing Model Routine Files In the new mos222 model subdirectory rename mos3 in all filenames to mos222 For example mv CMImos3defs h CMImos222defs h After you rename all files the new model subdirectory should contain the following group of files CMImos222defs h m CMlmos222 c CMImos222Getlpar c CMlmos222Setlpar c CMlmos222GetMpar c CMimos222SetMpar c HSPICE MOSFET Models Manual 523 X 2005 09 8 Customer Common Model Interface Adding Proprietary MOS Models 524 OMImos222eval c OMImos222set c OMlmos222temp c For a detailed description of each routine see Model Interface Routines on page 529 Modify the functions as necessary To add a new model most of the work required is modifying these files Compiling the S
429. ng point information so it has a non zero value in weak inversion Consequence This enhancement improves design information Corrections from EPFL R11 March 1999 Equation 45 Equation 53 Equation 54 and Equation 58 were corrected for multiple series device behavior by using the NS parameter Corrections from EPFL R12 July 30 1999 EPFL released the following corrections Correction 1 99 07 30 mb r12 corrected dGAMMAprime dVG narrow channel An error in the analytical model derivatives of the GAMMAprime variable affected the transconductances and transcapacitances Correction 2 99 07 30 mb r12 prevents PHI from being smaller than 0 2 both at init and after updating the temperature For some CMOS technologies PHI parameter values are as low as 400mV required to account for particular process details If you increase the temperature from room temperature PHI decreases due to its built in temperature dependence As a result PHI attains very low values or even becomes negative when it reaches 100degC For the model to function at these temperatures PHI has a lower limit of 200mV The usual range for this parameter is well above this value 600mV to 1V HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 58 University of Florida SOI Correction 3 99 06 28 mb r12 fixed COX KP initialization rg Correction 4 99 05 04 mb r12 completed parameter initialization for XQC DL and D
430. ngth and width cross terms for the C wwn V channel width offset HSPICE MOSFET Models Manual X 2005 09 425 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Notes 1 If you do not specify Cogo simulation calculates it as follows e f you specify a dlc value that is greater than 0 0 then cgso pl max 0 dlc cox cgs1 e Otherwise cgso 0 6 xj cox 2 If you do not specify Cggo simulation calculates it as follows e ifyou specify a dlc value that is greater than 0 0 then cgdo p2 max 0 dlc cox cgdl e Otherwise cgdo 0 6 xj cox 3 If you do not specify C simulation calculates it using 2 ox 4x10 C Pang T OX 4 If you do not specify Vipo in the MODEL statement simulation calculates it with Vg 1 using Vino V tO K o 5 f you do not specify K and Ko simulation calculates it using K GAMMA 2K fb Vp _ GAMMA GAMMA f Vrs 9 2 Jo JO z Vini 2 NO Vym 6 If you do not specify Nen but you specify GAMMA then simulation calculates nop from 2 GAMMA Cox Don 2qe If you do not specify no or GAMMA then nch defaults to 1 7e17 per cubic meter and simulation calculates GAMMA from Men 426 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models 7 If you do not specify PHI simulation calculates it using kg T 2 o n i i ta jofc T
431. ngth for the body contact at the source side AGBCP Parasitic gate to body overlap area for the body contact AEBCP Parasitic body to substrate overlap area for the body contact HSPICE MOSFET Models Manual 465 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model Parameter Description VBSUSR Optional initial value of Vbs that you specify for transient analysis DELTOX Shift in gate oxide thickness TOX That is the difference between the electrical and physical gate oxide insulator thicknesses TNODEOUT Temperature node flag indicating the use of the T node OFF Sets the initial condition of the element to OFF in DC analysis BJTOFF Turning off BJT if equal to 1 IC Initial guess in the order drain front gate internal body back gate external voltage ignores Vps for 4 terminal devices Use these only if you specify UIC in the TRAN statement The IC statement overrides it f you do not set TNODEOUT you can specify four nodes for a device to float the body Specifying five nodes implies that the fifth node is the external body contact node with a body resistance between the internal and external terminals This configuration applies to a distributed body resistance simulation f you set TNODEOUT simulation interprets the last node as the temperature node You can specify five nodes to float the device Specifying six nodes implies body contact Seven nodes is a body con
432. nning To support parameter binning Berkeley BSIM3v3 specifies LWP parameters To bilinearly interpolate a subset of model parameters over 1 Leff and 1 Weff you specify four terms Xo parameter X length term Xw width term Xp product term Simulation then interpolates the parameter value at a specified L W X Xo Xl Leff Xw Weff Xp Leff Weff See Parameter Range Limits on page 428 to determine whether you can bin a parameter Simulation adds the LMIN LMAX WMIN WMAX LREF and WREF parameters to allow multiple cell binning LMIN LMAX WMIN WMAX 410 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models define the cell boundary LREF and WREF are offset values that provide a convenient interpolation scheme Simulation uses the LREF and WREF offsets if you define both values and you specify the BINFLAG gt 0 9 model parameter Simulation then interpolates the parameter value at a specified L W X Xo Xl 1 Leff 1 LREF Xw 1 Weff 1 WREF Xp 1 Leff 1 LREF 1 Weff 1 WREF To select micron units for the lwp geometry parameters set the BINUNIT 1 model parameter For other choices of BINUNIT the lengths are in units of meters Simulation handles the XL XLREF XW and XWREF parameters in a manner consistent with other Synopsys MOSFET models and they produce shifts in parameter
433. nsitivity parameters Simulation calculates the weak inversion current when zn0 is less than 200 and adds it to the strong inversion current V Itotal Istrong Iweak 1 exp 4 In deep subthreshold xn zn0 znb v znd vg KT v yt vtherm Tax xweak 5 xn vtherm Iweak const exp xweak zwfac and zwfacu control the modification of this formula near the threshold Just above threshold the device is in saturation 2 Istrong const xweak lweak needs an xweak term to cancel the kink in gm at the threshold Then lweak goes to zero for xweak gt A0 which is at a small voltage above the threshold Iweak has four regions 1 xweak lt zwfac A0 Iweak const exp xweak HSPICE MOSFET Models Manual 357 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model 2 zwfac A0 lt xweak lt 0 Iweak const exp xweak const wf In the preceding equation wf is the integral with respect to the xweak value of xweak zwfac A0 1 xweak zwfac AO 1 zwfacu xweak zwfac A0 3 0 xweak AO dwf Iweak same formula as in region 2 const xweak 4 A0 xweak Iweak 0 AO and the constants in the preceding equations are not model parameters Continuity conditions at the boundaries between regions uniquely determine these constants LEVEL 39 BSIM2 Model BSIM Berkeley Short Channel IGFET Model 2 is the LEVEL 39 MOSFE
434. nt type vbi vto ratio If vrIMEx1 then uo uo ratio and kp kp ratio Note TREF is an exponent in adjusting the temperature It is not the reference temperature of this device model vfb vto 0 5 PHI 0 5 1 4 eg vbi vfb 0 5 PHI ratio vto vbi printback definition phi phi ratio printback definition HSPICE MOSFET Models Manual 207 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 40 HP a Si TFT Model 208 vfb vbi phi printback definition vdsat 0 beta kp W L vth vbi ETA vds If NU 0 K2 0 PSI 0 and VrIME 1 then vth vth f vgs vds NU K2 PSI CHI VTIME TEMP von vth If NFS20 then q NFS 10 ion xn 14 Cfm von f vth vt xn Cutoff Region NFS 0 vgs lt von If NFS 0 and vgs lt von then Cgdi 0 Cgsi 0 Ids GO f vgs DEFF vds gm GO gds GO DEFF Noncutoff Region NFS 0 i If vgs von then Vgsx vgs E If vgs Svon then vgsx von Mobility modulation by vgs ueff f uo n vgs THETA If vMAx 0 then vdsat vgsx vth vdsc JE vth vdsc a AC im CSC y OM eC HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 40 HP a Si TFT Model Cfmlw is the series combination of the dielectric and space charge capacitance for the MIS structure If vds lt vdsat then T2 71 vdsx vds epsfm Cfm
435. nt for the channel length modulation effect in the entire regions ids 1 AL Lo CLM 7 Intersil Channel Length Modulation If CLM 7 this model computes AL only for the saturation region ids HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 8 IDS Model vds gt vdsat LAMBDA Log 4 4 T LAMI Lay MAS VASGE ids ids L L Lo CLM 8 If CLM 8 this model computes AL only for the saturation region vds gt vdsat AL La 1 LAM1 L 1 vde LAMBDA vds vde ids ids AL L eff Subthreshold Current Ids The LEVEL 8 model has different subthreshold current equations depending on the value of the CAV model parameter Define kon gr cs es a unu EE L COX 2 vsh 2 y COX CAV 0 von vth CAV fast Subthreshold Region vgs lt von If vgs gt vth lesa fast 2 CAV fast CA a a ja 3 e ids ids von vde vsb e HSPICE MOSFET Models Manual 187 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 27 SOSFET Model If vgssvth CAY ves sh ids ids von vde vsb e 2 fast CAV 0 If CLM 8 von vth 3 fast otherwise von vth 2 fast Subthreshold Region vgs lt von ues If WIC 8 the next equation calculates the ids subthreshold current ids ids vgs vde vsb isub NOeff NDeff vgs vds NOeff and NDeff are functions of effective device width and length ids ids von vde vsb e LEVEL 27 SOS
436. nt gate work function difference WKB V calc Back gate work function difference TAUO S calc 107 102 Carrier lifetime in lightly doped regions BFACT 0 3 0 1 0 5 Vps averaging factor for mobility degradation 258 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 58 University of Florida SOI Table 57 Optional MOSFET Level 58 Parameters Continued Parameter Unit Default Typical Value Description FVBJT 0 0 0 1 BJT current directional partitioning factor 0 for lateral 1D flow RHOSD ohm sq 0 0 50 Source drain sheet resistance Notes The model line must include Level 58 and NFDMOD 0 for FD or NFDMOD 1 for NFD devices Specifying VFBF turns off the narrow width effect defined by NQFSW which can be positive or negative and the reverse short channel effect defined by LRSCE and NBH or NHALO if specified the latter effect is also turned off if you specify WKF For floating body devices CGFBO is small you should set it to O JRO and SEFF influence the gain of the BJT but LDIFF affects only bipolar charge storage in the source drain If you specify THALO then NBH and NHALO also influence the BJT gain Loosely correlate the TAUO value with JRO in accord with basic pn junction recombination generation properties Its default value is based on JRO which is appropriate for short L for long L body generation predominates over that in the junctions so specify TAUO The non
437. ntional terminology in HSPICE is TREF which all Synopsys model MOS levels support as a model parameter Both Levels 49 and 53 support the TNOM alternative name for compatibility with SPICES The default room temperature is 25 C in Synopsys circuit simulators but is 27 C in SPICES If you specify the BSIM3 model parameters at 27 C add TNOM 27 to the model so that simulation correctly interprets the model parameters To set the nominal simulation temperature to 27 add OPTION TNOM 27 to the netlist when you test the Synopsys model versus SPICES You can use DELVTO and DTEMP on the element line with Levels 49 and 53 The following equation converts the temperature setup between the Synopsys model and SPICE3 SPICE3 OPTION TEMP 125 MODEL NCH NMOS Level 8 TNOM 27 Synopsys Model TEMP 125 MODEL NCH NMOS Level 49 TNOM 27 To automatically calculate the drain source area and perimeter factors for Berkeley junction diode models use ACM 12 with CALCACM 1 Normally ACM 10 13 defaults the area and perimeter factors to O To override this value for ACM 12 specify CALCACM 1 Define the HSPICE specific parameter HDIF in the model card If you do not want parasitic Rs and Rd with the BSIM3v3 internal Rsd either do not specify the RSH RSC RDC RS and RD HSPICE parameters default is 0 or set them to 0 Simulation and analysis either warns or aborts with a fatal error if certain m
438. o jrowl jroswe Exponent for w dependence for jro jrowe jrost Temperature adjustment coefficient for jro jrot ms S diode recombination slope factor asymmetrical m jgos S diode SCR generation coefficient asymmetrical A cm 2 jgo jgosll I dependence parameter for jgo jgoll jgosle Exponent for I dependence for jgo jgole jgoswl W dependence parameter for jgo jgowl jgoswe Exponent for w dependence for jgo jgowe jgost Temperature adjustment coefficient for jgo jgot mgs S diode generation slope factor mg gidla GIDL pre exponential parameter A m off gidlb GIDL exponential parameter m V 3 0e9 gidlc GIDL bulk dependence parameter V43 8 0 gidle GIDL bandgap V calculated gidlt GIDL temperature dependence parameter calculated 514 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Table 147 8 SSIMSOI Model Parasitic Parameters Continued Name Parameter Units Default gisla GISL pre exponential parameter A m gidla gislb GISL exponential parameter m V gidlb gislc GISL bulk dependence parameter V 3 gidlc gisle GISL bandgap V gidle gislt GISL temperature dependence parameter gidlt rshmin Sheet res of s d gate overlap ohm sq 0 tcmin Temperature coefficient for rshmin 1 K 0 rshmins Sheet res of s gate overlap asymmetrical ohm sq rshmin tcmins Temperature coefficient for rshims asymmetrical 1 K tcmin rshpls Sheet res of heavily doped S D ohm sq 0 tcpls Temperature coefficient
439. ociated with Igb current 2 for circiut element summary O t noise associated with Igdo Igcd double cdsat drain to bulk saturation current double cssat source to bulk saturation current double capbd zero biased junction capacitance on the drain side double capbs zero biased junction capacitcance on the source side double nf number of device figures double min flag to determine the number of drain or source diffusions for even number fingered device double rbpd resistance between ib and db double rbdb resistance between xb and db double rbpb resistance between ib and xb double rbps resistance between ib and sb double rbsb resistance between xb and db double geomod geometry dependent parasitics model selector double delvto zero bias threshold voltage shift double mulu0 low field mobility multipler double delk1 first order body bias coefficient shift double delnfct subthreshold swing factor shift for template output double templti1 output template 1 double templt2 output template 2 double templt3 output template 3 double templt4 output template 4 double templt5 output template 5 double t
440. odel EBCI for better accuracy in predicting capacitive coupling The BSIM3v3 based model is also improved Dynamic depletion can suit different requirements for SOI technologies Single l V expression as in BSIM3v3 1 to assure continuities of Ids Gds Gm and their derivatives for all bias conditions Syntax The general syntax for a BSIM3SOI MOSFET element in a netlist is Mxxx nd ng ns ne np mname lt L val gt W val lt M val gt t lt AD val gt lt AS val gt lt PD val gt lt PS val gt lt NRD val gt lt NRS val gt t lt NRB val gt RTHO val lt CTHO val gt off BJToff val t IC Vds Vgs Vbs Ves Vps gt Parameter Description Mxxx SOI MOSFET element name Must begin with M followed by up to 1023 alphanumeric characters nd Drain terminal node name or number ng Front gate node name or number ns Source terminal node name or number HSPICE MOSFET Models Manual 493 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model 494 Parameter Description ne np mname AD AS PD PS NRD NRS NRB RDC Back gate or Substrate node name or number External body contact node name or number MOSFET model name reference SOI MOSFET channel length in meters This parameter overrides DEFL in an OPTION statement Default DEFL with a maximum of 0 1m SOI MOSFET channel width in meters This parameter overrides D
441. odel pmodel ptran pslot vgs pslot vds pslot vbs pslot ibs ptran MOS3ibs pslot ibd ptran MOS3ibd pslot gbs ptran MOS3gbs pslot gbd ptran MOS3gbd pslot capbs ptran gt MOS3capbs pslot capbd ptran MOS3capbd pslot qbs ptran gt MOS3qbs pslot qdb ptran gt MOS3qbd return 0 int CMImos3DiodeEval CMI_ Noise Based on the bias conditions temperature and model instance parameter values this routine evaluates the noise model equations It then returns the noise characteristics via the CMI_VAR variable HSPICE or HSPICE RF passes values to u ps LOT t nois irs Thermal noise associated with the parasitic source resistance expressed as a mean square noise current in parallel with Rs Ww wos LOT t nois ird Thermal noise associated with the parasitic drain resistance expressed as a mean square noise current in parallel with Rd HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Interface Variables pslot nois idsth Thermal noise associated with a MOSFET expressed as a mean square noise current referenced across the MOSFET channel pslot gt nois_idsfl Flicker noise associated with a MOSFET expressed as a mean square noise current referenced across the MOSFET channel HSPICE or HSPICE RF also passes the frequency into CMI Noise via pslot gt freq Syntax int
442. odel Parameters Contents Specifying XQC and XPART for CAPOP 4 9 11 12 13 Overlap Capacitance Equations CAPOP 0 SPICE Meyer Gate Capacitances Gate Bulk Capacitance cgb Gate Source Capacitance cgs Gate Drain Capacitance cgd CAPOP 1 Modified Meyer Gate Capacitances Gate Bulk Capacitance cgb Gate Source Capacitance cgs Gate Drain Capacitance cgd CAPOP 2 Parameterized Modified Meyer Capacitance Gate Bulk Capacitance cgb Gate Source Capacitance cgs Gate Drain Capacitance cgd CAPOP 3 Gate Capacitances Simpson Integration CAPOP 4 Charge Conservation Capacitance Model CAPOP 5 No Gate Capacitance CAPOP 6 AMI Gate Capacitance Model CAPOP 11 Ward Dutton model specialized CAPOP 12 Ward Dutton model specialized CAPOP 13 BSIM1 based Charge Conserving Gate Capacitance Model IEEVEED AA LEVEL 3 CAPOP 39 BSIM2 Charge Conserving Gate Capacitance Model Calculating Effective Length and Width for AC Gate Capacitance Noise Models sse enn Temperature Parameters and Equations Temperature Parameters MOS Temperature Coefficient Sensitivity Parameters Temperature Equations Energy Gap Temp
443. odel parameter 40 narrow width effect 131 National Semiconductor model 6 NB model parameter 41 NDS model parameter 40 noise CMI Noise 542 MOSFETs 96 equivalent circuits 37 models 243 parameters BSIM3v3 422 NonQuasi Static NQS model 402 parameters 424 Normal Field equations 173 NSUB model parameter 41 O operating point capacitance printout 68 Early voltage 244 model internal variables 243 Overdrive voltage 244 saturation non saturation flag 244 saturation voltage 244 SPICE like threshold voltage 244 transconductance efficiency factor 244 OPTION MBYPASS 12 output conductance 62 overlap capacitors 77 P parameters noise 96 voltage 58 parasitic diode MOSFETs LEVEL 39 372 generation 49 MOSFETs LEVEL 13 344 PB model parameter 41 PHA model parameter 41 PHD model parameter 41 Philips MOS9 model 214 PHP model parameter 41 PHS model parameter 41 606 PMOS model 433 PRD model parameter 42 print CMI PrintModel 544 PRS model parameter 42 Q quasi static model equations 241 R RD model parameter 42 RDC model parameter 42 regions charge equations 337 resistance 42 MOSFETs model parameters 42 RL model parameter 42 RPI a Si TFT model 260 circuit 263 Poli Si TFT model 265 272 RS model parameter 42 RSC model parameter 42 RSH model parameter 42 S saturation carrier velocity 178 current temperature 102 voltage vdsat 335 voltage equations MOSFETs LEVEL 1 129 LEVEL 13 335 LEVEL 2 132 LEVEL 28 35
444. odel parameter values are out of a normal range To view all warnings you might need to increase the OPTION WARNLIMIT value default 1 To turn on full parameter range checking set the PARAMCHK 1 model parameter default is 0 If you use PARAMCHK 0 simulation checks a smaller set of parameters See Parameter Range Limits on page 428 for more details about parameter limits Use the APWARN 1 model parameter default 0 to turn off PS PD lt Weff warnings Use NQSMOD only with Version 3 2 and specify it only in the model card HSPICE MOSFET Models Manual 413 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Level 49 53 Model Parameters The following tables describe all Level 49 and Level 53 model parameters including parameter name units default value whether you can bin the parameter adescription These tables are a superset of the BSIM3v3 model parameter set and include HSPICE parameters These HSPICE parameters are noted in the description column and always default for Level 53 to maintain compliance with the BSIM3v3 standard These parameters also apply to Level 49 with the following exceptions ACM default value 0 XPART default value 1 CAPMOD default value 0 Table 96 Model Flags for MOSFET Levels 49 53 Name Unit Default Bin Description VERSION z 3 2 No Selects from BSIM3 Versions 3 0 3 1 3 2 Issues a warning if you do not explicitly se
445. odels MOSFET Diode Models 52 Drain Diode Saturation Current Define val JSscaled ADeff JSWscaled PDeff If val O then isbd val Otherwise isbd M IS Calculating Effective Drain and Source Resistances For ACM 2 simulation calculates the effective drain and source resistances as follows Source Hesistance If you specify NRS then LDscaled LDIFscaled Mae RSH RS Seff Weff S r Otherwise Rseff RSC HDIFeff RSH LDscaled LDIFscaled RS M Weff Drain Resistance If you specify NRD then LDscaled LDIFscaled GP RSH RD A O T RD zz in Weff M Otherwise RDeff RDC HDIFeff RSH LDscaled LDIFscaled RD M Weff Using an ACMz3 MOS Diode Use ACM 3 to properly model MOS diodes of stacked devices You can also use the CJGATE parameter to model the drain and source periphery capacitances separately along the gate edge Therefore the PD and PS calculations do not include the gate periphery length CJGATE defaults to the CJSW value which in turn defaults to O HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models The AD AS PD and PS calculations depend on the device layout as determined by the value of the GEO element parameter You can specify the following GEO values in the MOS element description GEO 0 other devices do not share the drain and source of the device default GEO 1 another device shares the drain G
446. odels Manual X 2005 09 383 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model Table 91 MOSFET Level 47 Model Parameters Continued Name Unit Default Description UB m V 5 87e 19 Second order mobility degradation coefficient UB1 m V 7 61e 18 Temperature coefficient of UB UC 1 V 0 0465 Body bias sensitivity coefficient of mobility UC1 1 V 0 056 Temperature coefficient of UC VSAT cm sec a 8e6 Saturation velocity of the carrier at T TREF AT m sec 3 3e4 Temperature coefficient of VSAT RDSW ohm um 0 0 Source drain resistance per unit width RDSO ohm 0 0 Source drain contact resistance LDD m 0 0 Total length of the LDD region ETA 0 3 Coefficient of the drain voltage reduction ETAO 0 08 DIBL Drain Induced Barrier Lowering coefficient for the subthreshold region ETAB 1 V 0 07 Subthreshold region DIBL coefficient EM V m 4 1e7 Electrical field in the channel above which the hot carrier effect dominates NFACTOR 1 0 Subthreshold region swing VOFF V 0 11 Offset voltage in the subthreshold region LITL m Characteristic length Default is LITL si OX VGLOW V 0 12 Lower bound of the weak strong inversion transition region 384 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model Table 91 MOSFET Level 47 Model Parameters Continued Name Unit Default Description VGHIGH V 0 12 Upper bound of the weak strong inver
447. odels Manual X 2005 09 2 Technical Summary of MOSFET Models Noise Models You can model the channel thermal noise and the flicker noise as the ind current source which the following equation defines ind channel thermal noise flicker noise If the NLEV model parameter is less than 3 then N 1 2 channel thermal noise Lem The preceding formula used in both saturation and linear regions can lead to wrong results in the linear region For example at VDS 0 channel thermal noise becomes zero because gm 0 This calculation is physically impossible If you set the NLEV model parameter to 3 simulation uses a different equation which is valid in both linear and saturation regions See Tsivids Yanis P Operation and Modeling of the MOS Transistor McGraw Hill 1987 p 340 For NLEV 3 1 2 l ata 2 channel thermal noise 82 B vgs vth a GDSNOI The following equations calculate the a value used in the preceding equation vds a Linear region vdsat a 0 Saturation region Use the AF and KF parameters in the small signal AC noise analysis to determine the equivalent flicker noise current generator which connects the drain to the source s AF 1 2 NLEV 0 SPICE flicker noise a COX Leff For NLEV 1 Leff in the above equation is replaced by Weff Leff KF gm 1 2 2 NLEV 2 3 flicker noise NETTE eJ Lejff HSPICE MOSFET Models Manual 97 X 2005 09 2 T
448. oefficient of the width dependence of 0 0 SR ETAMOBR Effective field parameter depletion 1 4 3 inversion charge dependence for reference transistor STETAMOB Temperature dependence coefficient of K 1 0 0 nMOB SWETAMOB Width dependence coefficient of nMOB s 0 0 288 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 63 Level 63 MOS11 Parameters Level 11010 Physical Geometry Scaling Continued Name Description Units NMOS PMOS NU Exponent of field dependence mobility 2 2 model minus 1 such as v 1 at reference temperature NUEXP Exponent of the temperature dependence 5 25 3 23 of parameter v THERR Series resistance coefficient for reference V 1 0 155 0 08 transistor at reference temperature ETAR Exponent of temperature dependence of 0 95 0 4 R SWTHER Coefficient of the width dependence of OR 0 0 THER1 Numerator of gate voltage dependent part V 0 0 of series resistance for reference transistor THER2 Denominator of gate voltage dependent V 1 1 part of series resistance for the reference transistor THESATR Velocity saturation parameter due to V 1 0 5 0 2 optical acoustic phonon scattering for the reference transistor at the reference temperature SLTHESAT Length dependence coefficient of 6SAT 1 1 THESATEXP Exponent of length dependence of OSAT 1 1 ETASAT Exponent of temperature dependence of 1 04 0 86 0SAT SWTHESAT Width depende
449. of IGINV PWIGINV Coefficient for the width dependent AV 2 0 0 part of IGINV PLWIGINV Coefficient for the length times width AV 2 0 0 dependent part of IGINV POBINV Coefficient for the geometry V 48 48 independent part of IGINV PLBINV Coefficient for the length dependent V 0 0 part of IGINV PWBINV Coefficient for the width dependent V 0 0 part of IGINV PLWBINV Coefficient for the length times width V 0 0 dependent part of IGINV HSPICE MOSFET Models Manual 299 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 64 Level 68 MOS11 Parameters Level 11011 Binning Geometry Scaling Continued Name Description Units NMOS PMOS POIGACC Coefficient for the geometry AV 2 0 0 independent part of IGACC PLIGACC Coefficient for the length dependent AV 2 0 0 part of IGACC PWIGACC Coefficient for the width dependent AV 2 0 0 part of IGACC PLWIGACC Coefficient for the length times width AV 2 0 0 dependent part of IGACC POBACC Coefficient for the geometry V 48 87 5 independent part of BACC PLBACC Coefficient for the length dependent V 0 0 part of BACC PWBACC Coefficient for the width dependent V 0 0 part of BACC PLWBACC Coefficient for the length times width V 0 0 dependent part of BACC VFBOV Flatband voltage for the source drain V 0 0 overlap extensions KOV Body effect factor for the source drain V1 2 2 5 2 5 overlap extensions POIGOV Coefficient for the geometry AV 2 0 0 in
450. of width dependence of eavfac 2 0 eavexp Exponent of mobility field function 1 0 ubvds Drain dependence of eff field 0 5 vsat Channel carrier saturation velocity cm sec 1 0e 7 esat0 Vsat divisor velocity field model 2 esati Divisor carrier velocity at sat 1 lc00 Mult For channel length modulation 0 2 Ic01 Length dependence of Ic00 1 micron 0 Ic1 Bias dependence of channel length modulation 1 V 0 510 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Table 146 SSIMSO Model Intrinsic Parameters Mobility and Saturation Output Conductance Continued Name Parameter Units Default wimod Mult for channel width modulation 0 dv2 Par for lin sat transition region 0 05 dv3 Length dependence of dv2 0 exb Temperature exponent of ubref 1 5 Table 147 SSIMSO Model Parasitic Parameters Name Parameter Units Default aimp0 Impact ionization parameter 0 aimpl Length dependence of aimpO micron 0 aimpw Width dependence of aimpO micron 0 aimpt Temperature dependence of aimpO 1 K 0 bimpO Exponent for impact ionization 1 V 28 0 bimpl Length dependence of bimpO micron 0 bimp2 Width dependence of bimpO micron 0 fsatt Bias dependence of impact ionization 1 V 0 onkink Voltage adjustment for onset of the kink V 0 gtundeltox Intrinsic region delta tox electrical vs physical angstrom off gtundtoxovl S D overlap region delta tox electrical vs physical angstrom off gtunstoxov S
451. om 27 0 nch 5 73068E 16 tox 1 00000E 08 xj 1 00000E 07 lint 8 195860E 08 wint 1 821562E 07 vth0 86094574 k1 341038 k2 2 703463E 02 k3 12 24589 dvt0 767506 dvt1 2 65109418 dvt2 2 0 145 nlx 1 979638E 07 w0 1 1e 6 k3b 2 4139039 vsat 60362 05 ua 1 348481E 09 ub 3 178541E 19 uc 1 1623e 10 rdsw 498 873 u0 137 2991 prwb 1 2e 5 a0 3276366 keta 1 8195445E 02 a1 2 0232883 a2 9 voff 6 623903E 02 nFactor 1 0408191 cit 4 994609E 04 cdsc 1 030797E cdscb 2 84e 4 eta0 0245072 etab 1 570303E 03 dsub 24116711 pclm 2 6813153 pdiblcl 4 003703E 02 pdiblc2 2 00329051 pdiblcb 2 e 4 drout 1 pvag 1 prwg 0 dvt0wz 0 3 380235 pscbel 0 pscbe2 1 e 28 6370527 001 ags 1 2 58 dvtlw 5 3e6 dvt2w 0 0032 kt2 03 prt 76 4 4 31E 09 ubl 7 61E 18 uc1 2 378e 10 al TEED 1 0 cle wr 1 cgd bO 7 bl le 7 dwg 5e 8 dwb 2e 8 delta 0 015 le 10 cgsl le 10 cgbo 1e 10 xpart 0 0 0 6 cgdo 0 4e 9 cgso 0 4e 9 clc20 1e 6 ckappa 0 6 HSPICE MOSFET Models Manual 433 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Level 54 BSIM4 Model The UC Berkeley BSIM4 model explicitly addresses many issues in modeling sub 0 13 micron CMOS technology and RF high speed CMOS circuit simulation The Level 54
452. on Gate Field E B DN N NG T Field Implant Ep a St Gate Oxide Substrate d 34 56 1 2 Drawn width of the gate W 3 4 Depleted or accumulated channel parameter WD 4 5 Effective channel width W XW 2 WD 3 6 Physical channel width W XW MOSFET Equivalent Circuits Equation Variables This section lists the equation variables and constants Table 5 Equation Variables and Constants Variable Definition Quantity cbd Bulk to drain capacitance cbs Bulk to source capacitance cbg Gate to bulk capacitance cgd Gate to drain capacitance cgs Gate to source capacitance 32 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Equivalent Circuits Table 5 Equation Variables and Constants Continued Variable Definition Quantity f Frequency gbd Bulk to drain dynamic conductance gbs Bulk to source dynamic conductance gds Drain to source dynamic conductance controlled by vds gdb Drain to bulk impact ionization conductance gm Drain to source dynamic transconductance controlled by vgs gmbs Drain to source dynamic bulk transconductance controlled by vsb ibd Bulk to drain DC current ibs Bulk to source DC current ids Drain to source DC current idb Drain to bulk impact ionization current ind Drain to source equivalent noise circuit inrd Drain resistor equivalent noise circuit inrs Source resistor equivalent noise circuit rd Drain resistanc
453. on VTHO V 0 7 NMOS Yes Threshold voltage of the long channel device VTHO 0 7 PMOS at Vg 0 and small V4 NSUB cm 6 0e16 Yes Substrate doping concentration HSPICE MOSFET Models Manual 415 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 97 Basic Model Parameters MOSFET Levels 49 53 Continued Name Unit Default Bin Description NCH cm3 See 6 1 7e17 Yes Peak doping concentration near the interface NLX m 1 74e 7 Yes Lateral nonuniform doping along the channel K1 y1 2 0 50 Yes First order body effect coefficient K2 0 0186 Yes Second order body effect coefficient K3 80 0 Yes Narrow width effect coefficient K3B 1 V 0 Yes Body width coefficient narrow width effect WO m 2 5e 6 Yes Narrow width effect coefficient DVTOW 1 m 0 Yes Narrow width coefficient 0 for Vth small L DVT1W 1 m 5 3e6 Yes Narrow width coefficient 1 for Vth small L DVT2W 1 V 0 032 Yes Narrow width coefficient 2 for Vth small L DVTO s 2 2 Yes Short channel effect coefficient O for Vin DVT1 i 0 53 Yes Short channel effect coefficient 1 for Vin DVT2 1 V 0 032 Yes Short channel effect coefficient 2 for Vin ETAO 0 08 Yes DIBL drain induced barrier lowering coefficient for the subthreshold region ETAB 1 V 0 07 Yes DIBL coefficient for the subthreshold region DSUB a DROUT Yes DIBL coefficient exponent in the subthreshold region VBM V 3 0 Yes Maximum substrate bias for calculating Vin UO cm2 V sec 670nmos Yes Low
454. on calculates the reduction of the saturation voltage due to the carrier velocity saturation effect vdsat vsat vc vsat vc 2 In the preceding equation determines vc if the ECRIT model parameter gt 0 or VMAX gt 0 and KU lt 1 If you specify both ECRIT and VMAX then simulation uses only the VMAX equation However this model does not use the VMAX equation if MOB 4 or MOB 5 because these mobility equations already contain a velocity saturation term _ VMAX Leff IF ueff Because vsb gt VBO y switches from y to yp and the ids vsat and conductance values are not continuous as in the following example To correct this discontinuity problem specify the UPDATE 1 model parameter The next section discusses this improvement vc ECRIT Leff or vc Example An example of a multi level gamma model with UPDATE O is located in the following directory Sinstalldir demo hspice mos tgam2 sp Figure 28 Variation of IDS VTH and VDSAT for UPDATE 0 TGAM2 SP MULT LEVEL GAMMA MODEL UPDATE 0 14 MAY 2003 15 22 48 550U TS 4 TGAM2 0 SPO a i i l IDS L 50 0U gt are N 45 0U gt 40 0U Finis j TGAM2_0 SP0 L 720 0M 7 VTH 700 0M N 680 0M F tate oba ARR LO x qur qd o pai E qua uc TGAM20 SPO 260 0M i AA DAN L pape 240 0M dun i Eee CN P ILE us N E Leere d na a EPIS 1 eeu 1 050 1 10 1 150 1 0 VOLTS LIN HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Le
455. or 0 006 slgamoo 0 etagamr 2 0 mor 0 5 stmo 0 simo 0 t etamr 2 0 zetlr 1 0 etazet 0 5 slzetl 0 vsbtr 2 5 slvsbt 0 t alr 10 stal 0 slal 0 swal 0 a2r 30 sla2 0 swa2 0 a3r 0 8 sla3 0 swa3 0 tox 15 00e 9 col 0 3e 9 ntr 2 0e 20 nfr 5 0e 11 acm 2 hdif lu js le 3 cj le 3 tmj 0 5 pb 0 8 t cjsw le 9 cjgate 1e 9 t mjsw 0 3 php 0 8 Level 55 EPFL EKV MOSFET Model 224 The EPFL EKV MOSFET model is a scalable and compact simulation model built on fundamental physical properties of the MOS structure This model is dedicated to the design and simulation of low voltage low current analog and mixed analog digital circuits using submicron CMOS technologies The intrinsic part of the MOSFET includes the equations and parameters used to simulate the EPFL EKV MOSFET model The extrinsic part of the MOSFET model includes series resistances of the source and drain diffusions junction currents and capacitances Single Equation Model The EPFL EKV MOSFET model is a single expression which preserves continuity of first order and higher order derivatives with respect to any terminal voltage in the entire range of validity of the model This section describes the analytical expressions of first order derivatives as transconductances and transcapacitances you can use them in simulation HSPICE MOSFET Models Manual X 2005 09 5 St
456. or VFB KOR Body effect factor for the reference V1 2 0 5 0 5 transistor SLKO Coefficient of the length dependence of V1 2m 0 0 kO SL2KO Second coefficient of the length V1 2m2 0 0 dependence of kO SWKO Coefficient of the width dependence of kO V1 2m 0 0 280 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 62 Level 68 MOS11 Parameters Level 1100 Continued Name Description Units NMOS PMOS KPINV Inverse of the body effect factor poly V 1 2 0 0 silicon gate PHIBR Surface potential at the onset of strong V 0 95 0 95 inversion at the reference temperature SLPHIB Coefficient of the length dependence of op Vm 0 0 SL2PHIB Second coefficient of the length Vm2 0 0 dependence of op SWPHIB Coefficient of the width dependence of 68 Vm 0 0 BETSQ Gain factor for an infinite square transistor AV 2 3 09E 4 1 15E 4 at the reference temperature ETABET Exponent of temperature dependence 1 3 0 5 gain factor FBET1 Relative mobility decrease of the first 0 0 lateral profile LP1 Characteristic length of the first lateral m 8E 7 8E 7 profile FBET2 Relative mobility decrease due to the 0 0 second lateral profile LP2 Characteristic length of the second lateral m 8E 7 8E 7 profile THESRR Mobility reduction coefficient due to the V 1 0 4 0 73 surface roughness scattering for the reference transistor at the reference temperature SWTHESR Coefficient of the
457. os level 63 VERSION 11010 LVAR 0 000000E 00 LAP 1 864000E 08 WVAR 0 000000E 00 WOT 0 000000E 00 TR 2 100000E 01 VFB 1 038000E 00 STVFB 0 000000E 00 SLPHIB 1 024000E 08 SL2PHIB 1 428000E 14 SWPHIB 0 000000E 00 KOR 5 763000E 01 SIKO 2 649000E 08 SL2KO 1 737000E 14 SWKO 0 000000E 00 KPINV 2 200000E 01 PHIBR 8 500000E 01 BETSQ 1 201000E 04 ETABETR 1 300000E 00 FBET1 3 741000E 01 IP1 2 806000E 06 FBET2 0 000000E 00 LP2 1 000000E 10 THESRR 7 109000E 01 SWTHESR 0 000000E 00 THEPHR 1 000000E 03 ETAPH 1 750000E 00 SWTHEPH 0 000000E 00 ETAMOBR 2 825000E 00 STETAMOB 0 000000E 00 SWETAMOB 0 000000E 00 NU 1 000000E 00 NUEXP 3 228000E 00 THERR 1 267000E 01 ETAR 4 000000E 01 SWTHER 0 000000E 00 THER1 0 000000E 00 THER2 1 000000E 00 THESATR 6 931000E 02 SLTHESAT 1 000000E 00 THESATEXP 2 000000EK 00 HSPICE MOSFET Models Manual 309 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model ETASAT 8 753000E 01 SWTHESAT 0 000000E 00 SSFR 2 304000E 03 SLSSF 1 002000E 06 SWSSF 0 0
458. ot used in the Level 39 MOSFET model use OPTION DEFW in the netlist instead Specify all BSIM2 parameters according to the NMOS convention even for a PMOS model Examples VDD 5 not 5 VBB 5 not 5 and ETA0 0 02 not 0 02 See Compatibility Notes on page 366 HSPICE MOSFET Models Manual X 2005 09 361 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model 362 The Level 39 MOSFET model also includes the JSW A m source drain bulk diode sidewall reverse saturation current density Other Device Model Parameters that Affect BSIM2 You must specify the following MOSFET model parameters before you can use some Synopsys enhancements such as _LDD compatible parasitics Adjusts the model parameter geometry relative to a reference device mpactionization modeling with bulk source current partitioning Element temperature adjustment of the key model parameters This is a partial list For complete information see the following Calculating Effective Length and Width for AC Gate Capacitance on page 95 Drain and Source Resistance Model Parameters on page 42 Impact lonization Model Parameters on page 61 Temperature Parameters and Equations on page 98 MODEL VERSION Changes to BSIM2 Models on page 368 describes how the VERSION parameter in the MODEL statement changes the BSIM2 model depending on the model version number LEVEL 39 Model Equations In the following expressions mode
459. ou specify XW or WD then Weff Wscaled WMLT XWscaled 2 PWDscaled WREFeff WREFscaled WMLT XWscaled 2 PWDscaled IDS Equations Process oriented model parameters model the device characteristics Simulation maps these parameters into model parameters at a specific bias voltage The ids equations are as follows Cutoff Region vgs lt vth ids O see subthreshold current On Region vgs gt vth For the vds lt vdsat triode region aan AREA E i _ body 2 ids EET ogs vth vds 7 Pvds For the vdssvdsat saturation region TCR MM NA T MEER NU Im vgs vth The following equations calculate values used in the preceding equation Weff B ueff COX Leff _ uo uejf 14 xuO vgs vth xuO zu0 zx2u0 Pvsb Simulation uses quadratic interpolation through three data points to calculate the uo carrier mobility uo _ MUZ zx2mz Pvsb vds 0 uo as VDDM mus zx2ms Pvsb HSPICE MOSFET Models Manual T X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model Simulation also calculates the sensitivity of uo to vds at vds VDDM which is zx3ms The following equation calculates the body factor aaa 2 zphi vsb The following equation calculates the g value used in the preceding equation 1 ila a See i 1 744 0 8364 zphi 4 vsb The following equation calculates the arg term in the saturation region arg T 1 vce 1 2 vc
460. pacitance Models The LD parameter uses units in meters to obtain its scaled value simulation multiplies the value of LD by SCALM If the units are in meters squared simulation multiplies the parameter by SCALM f the units are in reciprocal meters the parameter s value is divided by SCALM For example because CGBO is in farads meter the value of CGBO is divided by SCALM f the units are in reciprocal meters squared then the parameter is divided by SCALM For the scaling equations specific to each CAPOP level see the individual CAPOP subsections MOS Gate Capacitance Model Parameters Table 18 Basic Gate Capacitance Parameters Name Units Default Description Alias CAPOP 2 0 Capacitance model selector COX CO F m2 3 453e 4 Oxide capacitance If you do not specify COX simulation calculates it from TOX The default corresponds to the TOX default of 1e 7 COXscaled COX SCALM TOX m 1e 7 Oxide thickness calculated from COX if you specify COX The program uses the default if you do not specify COX For TOX gt 1 simulation assumes that the unit is Angstroms A level dependent default can override it See specific MOSFET levels in this manual 74 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Table 19 Gate Overlap Capacitance Model Parameters Name Alias Units Default Description CGBO Fim 0 0 Gate bulk overlap capaci
461. pacitance Models Cascode Circuit Example iirat gout ac rout 0 0 8 86E 6 113 K 0 5 4 30E 6 233 K 1 0 5 31E 8 18 8 Meg The input file for this example is located in the following directory Sinstalldir demo hspice mos cascode sp MOS Gate Capacitance Models 64 You can use capacitance model parameters with all MOSFET model statements Three fixed capacitance parameters CGDO CGSO and CGBO represent gate to drain gate to source and gate to bulk overlap capacitances to model charge storage use fixed and nonlinear gate capacitances and junction capacitances The algorithm used for calculating nonlinear voltage dependent MOS gate capacitance depends on the value of the CAPOP model parameter To model MOS gate capacitances as a nonlinear function of terminal voltages use Meyer s piecewise linear model for all MOS levels The charge conservation model is also available for MOSFET model Levels 2 through 7 13 and 27 For LEVEL 1 you must specify the TOX model parameter to invoke the Meyer model The next three sections describe the Meyer Modified Meyer and Charge Conservation MOS Gate Capacitance models Some of the charge conserving models Ward Dutton or BSIM can cause timestep too small errors if you do not specify other nodal capacitances Selecting Capacitor Models When you select a gate capacitance model you can choose various combinations of capacitor models and DC models You can incrementally update old
462. parameter 2 CAPMOD 0 Model capacitance selector zero recommended CGDO F m 0 Gate drain overlap capacitance per meter channel width CGSO F m 0 Gate source overlap capacitance per meter channel width DASAT 1 C 0 Temperature coefficient of ASAT DD m 1400 Vas field constant DELTA 4 0 Transition width parameter DG m 2000 Vgs field constant DMU1 xu icoxX O Temperature coefficient of MU1 DVT V 0 The difference between VON and the threshold voltage DVTO V C 0 Temperature coefficient of VTO EB EV 0 68 Barrier height of the diode ETA 7 Subthreshold ideality factor ETACO ETA Capacitance subthreshold ideality factor at zero HSPICE MOSFET Models Manual X 2005 09 drain bias 267 5 Standard MOSFET Models Levels 50 to 64 Level 62 RPI Poli Si TFT Model Table 58 MOSFET Level 62 Model Parameters Continued Name Unit Default Description ETACOO 1 V 0 Capacitance subthreshold coefficient of the drain bias 10 A m 6 0 Leakage scaling constant 100 A m 150 Reverse diode saturation current KSS 0 Small signal parameter zero is recommended LASAT M 0 Coefficient for length dependence of ASAT LKINK M 19E 6 Kink effect constant MC 3 0 Capacitance knee shape parameter MK 1 3 Kink effect exponent MMU 3 0 Low field mobility exponent MUO emis 100 High field mobility MU1 cm Vs 0 0022 Low field mobility parameter MUS cm2 Vs 1 0 Subthreshold mobility RD u 0 Drain resistance RDX Q 0 Resistan
463. ped drain diffusion about 2000 RS Resistance ohm square of lightly doped source diffusion about 2000 RSH Diffusion sheet resistance about 35 50 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models Calculating Effective Areas and Peripheries For ACM 2 simulation calculates the effective areas and peripheries as follows If you do not specify AD then ADeff 2 HDIFeff Weff Otherwise ADeff M AD WMLT SCALE f you do not specify AS then ASeff 2 HDIFscaled Weff Otherwise ASeff M AS WMLT SCALE f you do not specify PD then PDeff M 4 HDIFeff 2 Weff Otherwise PDeff M PD WMLT SCALE f you do not specify PS then PSeff M 4 HDIFeff 2 Weff Otherwise PSeff M PS WMLT SCALE The following equations calculate values used in the preceding equation Weff Wscaled WMLT XWscaled HDIFeff HDIFscaled HDIFscaled HDIF SCALM WMLT The Weff value is not the same as the Weff value in the LEVEL 1 2 3 and6 models The 2 WDscaled term is not subtracted Calculating Effective Saturation Currents For ACM 2 simulation calculates the MOS diode effective saturation currents as follows Source Diode Saturation Current Define val JSscaled ASeff JSWscaled PSeff If val gt O then isbs val Otherwise isbs M IS HSPICE MOSFET Models Manual 51 X 2005 09 2 Technical Summary of MOSFET M
464. pendence of Weg m V 0 0 dwb Coefficient substrate body bias dependence m V 1 0 0 Wer 498 HSPICE MOSFET Models Manual X 2005 09 Table 141 MOSFET Level 60 DC Parameters Continued 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model SPICE Description Unit Default See Symbol Table 144 voff Offset voltage in the subthreshold region for V 0 08 large W and L values nfactor Subthreshold swing factor 1 eta0 DIBL coefficient in the subthreshold region 0 08 etab Body bias coefficient for subthreshold DIBL 1 V 0 07 effect dsub DIBL coefficient exponent 0 56 cit Interface trap capacitance F m2 0 0 cdsc Drain Source to the channel coupling F m 2 4e 4 capacitance cdscb X Body bias sensitivity of Cys F m 0 cdscd Drain bias sensitivity of Cg F m 0 z pclm Channel length modulation parameter 1 3 pdibl1 Correction parameter for the DIBL effect ofthe 39 first output resistance pdibl2 Correction parameter for the DIBL effect ofthe 0 086 second output resistance drout L dependence coefficient of the DIBL correction 0 56 parameter in Rout pvag Gate dependence of the Early voltage 0 0 delta Effective Vgs parameter 5 0 01 aii First Vdsatii parameter for the Leff dependence 1 V 0 0 bii Second Vdsatii parameter for the Leff m V 0 0 dependence HSPICE MOSFET Models Manual X 2005 09 499 7 BSIM MOSFET Models Levels 47 to 65 Leve
465. psys Device Model Enhancements Synopsys modified the standard SPICE models to satisfy the needs of customers The modifications are in the areas of Drawn dimensions corrected for photolithography and diffusion Corrections for optical shrink HSPICE MOSFET Models Manual X 2005 09 571 B Comparing MOS Models History and Motivation 572 Model independent process variation parameters Uniform subthreshold equations Charge conserving capacitance equations Impact ionization with selectable source bulk partitioning of the excess drain current Enhanced temperature relationships LEVEL 2 The LEVEL 2 model is an enhanced Grove equation It is the most common MOS equations in all simulators The basic current equation with the 3 2 power terms was developed by Ihantola and Moll in 1964 Reddi and Sah added channel length modulation in 1965 Crawford added vertical field reduction in 1967 Klassen added the ECRIT parameter in 1978 LEVEL 3 The LEVEL 3 model was developed by Liu in 1981 It is computationally more efficient replacing the 3 2 power terms with a first order Taylor expansion It includes the drain induced barrier lowering effect ETA parameter The LEVEL 3 models is impressively physical modeling two dimensional effects based on junction depth and depletion depths LEVEL 13 BSIM The BSIM1 model was developed by Sheu Scharfetter Poon and Hu at Berkeley in 1984 for higher accuracy modelin
466. ptran pname value char ptran char pname double value endif int param CMImos3GetIpar pname amp param CMImos3SetIpar param value MOS3instance ptran return 0 int CMImos3AssignIP 538 HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Interface Variables CMI SetupModel After you specify all model parameters where pmodel is a pointer to the model this routine sets up a model Syntax int CMI SetupModel char pmodel In the preceding syntax pmodel points to the model Example int ifdef STDC__ CMImos3SetupModel har pmodel el e se CMImos3SetupModel pmodel char pmodel endif CMImos3setupModel MOS3model pmodel return 0 int CMImos3SetupModel CMI SetupInstance After you specify all instance parameters this routine sets up an instance HSPICE or HSPICE RF typically processes the temperature and the geometry Syntax int CMI SetupInstance char pinst In the preceding syntax pinst points to the instance Example int ifdef STDC CMImos3SetupInstance char pmodel char ptran else CMImos3SetupInstance pmodel ptran char pmodel char ptran endif HSPICE MOSFET Models Manual 539 X 2005 09 8 Customer Common Model Interface Interface Variables temperature modified parameters CMImos
467. r PC users The third column indicates models supported on all platforms including PC The last column lists models supported on all platforms except the PC Table 1 MOSFET Model Descriptions All Platforms All Platforms Level MOSFET Model Description including PC except PC 1 Schichman Hodges model X 2 MOS2 Grove Frohman model SPICE 2G X 3 MOS3 empirical model SPICE 2G X 4 Grove Frohman LEVEL 2 model derived X from SPICE 2E 3 5 AMI ASPEC depletion and enhancement X Taylor Huang 6 Lattin Jenkins Grove ASPEC style X parasitics 7 Lattin Jenkins Grove SPICE style X parasitics 8 advanced LEVEL 2 model X 4 HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models Selecting Models Table 1 MOSFET Model Descriptions Continued All Platforms All Platforms Level MOSFET Model Description including PC except PC gr AMD X 10 AMD X 11 Fluke Mosaid model X 12 CASMOS model GTE style X 13 BSIM model X 14 Siemens LEVEL 4 X 15 user defined model based on LEVEL 3 X 16 not used 17 Cypress model X 18 Sierra 1 X 19 Dallas Semiconductor model X 20 GE CRD FRANZ X 21 STC ITT X 22 CASMOS GEC style X 23 Siliconix X 24 GE Intersil advanced X 25 CASMOS Rutherford X 26 Sierra 2 X 27 SOSFET X 28 BSIM derivative Synopsys proprietary X model HSPICE MOSFET Models Manual X 2005 09 1 Overview of M
468. r Unit Default Description DvtO 2 2 First coefficient of the short channel effect on Vth dvtOw 0 First coefficient of the narrow width effect on Vth for a small channel length dvt1 0 53 Second coefficient of the short channel effect on Vth dvtiw 5 3e6 Second coefficient of the narrow width effect on Vth for a small channel length dvt2 1 V 0 032 Body bias coefficient of the short channel effect on Vth dvt2w 1 V 0 032 Body bias coefficient of the narrow width effect on Vth for a small channel length dwb m V 1 2 0 0 Coefficient of the substrate body bias dependence of Weff dwbc m 0 0 Width offset for the body contact isolation edge dwg m V 0 0 Coefficient of the gate dependence of Weff esati V m 1 e7 Saturation channel electric field for the impact ionization current eta0 0 08 DIBL coefficient in the subthreshold region etab 1 V 0 07 Body bias coefficient for the DIBL effect in the subthreshold region fbjtii 0 0 Fraction of the bipolar current affecting the impact ionization Isbjt A m 1 0e 6 BJT injection saturation current Isdif Alm 0 Body to source drain injection saturation current Isrec A m2 1 0e 5 Recombination in the depletion saturation current Istun Am 0 0 Reverse tunneling saturation current HSPICE MOSFET Models Manual X 2005 09 469 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model Table 130 MOSFET Level 57 DC Parameters Continued Parameter Unit Default Desc
469. r the MUZ X2M X3MS and X33M mobility parameters FEX 0 0 Temperature exponent for the U1 mobility reduction factor TCV VPK 0 0 Flat band voltage temperature coefficient Sensitivity Factors of Model Parameters For transistors drop the O from the end of the parameter name and add one of the following product sensitivity factors for a basic electrical parameter L channel length W channel width WL width and length For example the VFBO sensitivity factors are LVFB WVFB and PVFB If AO is a basic parameter LA WA and PA are the corresponding sensitivity factors for this parameter you cannot use the SCALM option to scale LA WA and PA Then the model uses the following general formula to obtain the parameter value The left side of the equation represents the effective model parameter value after you adjust the device size All effective model parameters are in lower case and start with the z character followed by the parameter name 1 1 E A0 amp LA NE MES Se Weff WREFeff WA du REE Leff LREFeff Mere meo HSPICE MOSFET Models Manual 353 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modified BSIM Model 354 Specify LA and WA in units of microns times the units of AO Specify PA in units of square microns times the units of AO If you set LREF or WREF 0 you effectively set the parameter value to infinity This is the default Example VFBO 0 350v LVFB
470. rallel with each of the five substrate resistances to avoid potential numerical instability due to an unreasonably large substrate resistance HSPICE MOSFET Models Manual 455 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 122 Flicker and Thermal Noise Model Parameters MOS Level 54 Parameter Default Binnable Description NOIA 6 25e41 No Flicker noise parameter A ev 1s EFm3 for NMOS 6 188e40 eV tst EFm for PMOS NOIB 3 125e26 No Flicker noise parameter B ev s Etm for NMOS 1 5e25 eV 151EFm for PMOS NOIC 8 75 eVy 1S Fm No Flicker noise parameter C EM 4 1e7V m No Saturation field AF 1 0 No Flicker noise exponent EF 1 0 No Flicker noise frequency exponent KF 0 0 AZ FFgt EFE No Flicker noise coefficient NTNOI 1 0 No Noise factor for short channel devices for TNOIMOD 0 only TNOIA 1 5 No Coefficient of the channel length dependence of the total channel thermal noise TNOIB 3 5 No Channel length dependence parameter for partitioning the channel thermal noise 456 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 123 Layout Dependent Parasitics Model Parameters MOSFET Level 54 Parameter Default Binnable Description DMCG 0 0m No Distance from the S D contact center to the gate edge DMCI DMCG No Distance from the S D contact center to the isolation edge in the channel length direction DMDG 0 0m No Same as DMCG bu
471. rasitic diode current and capacitor currents Different MOSFET model levels support different subsets of these output parameters Table 4 on page 14 lists all parameters in the MOSFET output templates and indicates which model levels support each parameter Table 4 Parameters in MOSFET Output Templates Name Alias Description MOSFET Level L LV1 Channel Length L All This is also the effective channel length for all MOSFET models except Level 54 W LV2 Channel Width W All This is also the effective channel width for all MOSFET models except Level 54 AD LV3 Area of the drain diode AD All AS LV4 Area of the source diode AS All ICVDS LV5 Initial condition for the drain source All voltage VDS ICVGS LV6 Initial condition for the gate source All voltage VGS ICVBS LV7 Initial condition for the bulk source voltage All except 57 VBS 58 59 ICVES LV7 Initial condition for the substrate source 57 58 59 voltage VES 14 HSPICE MOSFET Models Manual X 2005 09 1 Overview of MOSFET Models MOSFET Output Templates Table 4 Parameters in MOSFET Output Templates Continued Name Alias Description MOSFET Level LV8 Device polarity All e 1 forward e 1 reverse not used after HSPICE release 95 3 VTH LV9 Threshold voltage bias dependent All VDSAT LV10 Saturation voltage VDSAT All PD LV11 Drain diode periphery PD All PS LV12 Source diode periphery PS All RDS LV13 Drain resistance squares RDS
472. rate current coefficient 3 HSPICE MOSFET Models Manual 317 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model Table 76 Level 64 Gate Current Parameters Parameter Default Description GLEAK1 gga v 3 2 G Gate current coefficient 1 GLEAK2 0 0 Gate current coefficient 2 GLEAK3 0 0 Gate current coefficient 3 Table 77 Level 64 GIDL Current Parameters Parameter Default Description GIDL1 0 0A m v2 2 GIDL current coefficient 1 GIDL2 0 0V 1 2 cm GIDL current coefficient 2 GIDL3 0 0 GIDL current coefficient 3 Table 78 Level 64 1 f Noise Parameters Parameter Default Description NFALP 2 0e 15 Contribution of the mobility fluctuation NFTRP 1 0611 Ratio of trap density to the attenuation coefficient CIT 0 0F cm2 Capacitance caused by the interface trapped carriers AF 1 0 SPICE2 flicker noise exponent KF 0 0 SPICE2 flicker noise coefficient EF 0 0 SPICE2 flicker noise frequency exponent 318 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model Table 79 Conserving Symmetry at V4520 for Short Channel MOSFETS Parameter Default Description VZADDO PZADDO 1 0e 2V Symmetry conservation coefficient 1 0e 3V Symmetry conservation coefficient Table 80 MOS DIODE Parameter Default Description JSO 1 0e 4Am Saturation current density JSOSW 0 0Am Sidewall saturation current density
473. rce za eam e D 5 1 e Fgert DS Roe Drain AL al LO e Ez Csbj Cabi Cabi Ue Cabj Bulk 226 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model Figure 35 represents the intrinsic and extrinsic elements of the MOS transistor For quasi static dynamic operation this figure shows only the intrinsic capacitances from the simpler capacitances model However you can also use a charge based transcapacitances model in simulation Table 47 Device Input Variables Name Unit Default Description L m Channel length W m Channel width MorNP 1 0 Parallel multiple device number NorNS 1 0 Series multiple device number EKV Intrinsic Model Parameters Name Unit Default Range Description COX F m 0 7E 3 Gate oxide capacitance per unit area XJ m 0 1E 6 gt 1 0E 9 Junction depth Dw m 0 i Channel width correction DL m 0 Channel length correction a This model can calculate the default value of COX as a function of TOX b DL and DW parameters are usually negative see the effective length and width calculation Name Unit Default Range Description vro V 05 Long channel threshold voltage GAMMA N 1 0 gt 0 Body effect parameter HSPICE MOSFET Models Manual 227 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model 228 Name Unit pefauit Range Description PHI V 0 7 gt 0 1 Bulk Fermi
474. re offset of JUNCAP element relative to Ta VR V 0 0 Voltage at which simulation determines the parameters JSGBR Am2 1 00e 3 Bottom saturation current density due to electron hole generation at V Vp JSDBR Am2 1 00e 3 Bottom saturation current density due to diffusion from back contact JSGSR Am 1 00e 3 Sidewall saturation current density due to electron hole generation at V Vp JSDSR Atm 1 00e 3 Sidewall saturation current density due to diffusion from back contact JSGGH A q 1 00e 3 Gate edge saturation current density due to electron hole generation at V Vp JSDGR Am 1 00e 3 Gate edge saturation current density due to diffusion from back contact NB 1 00 Emission coefficient of the bottom forward current NS 1 00 Emission coefficient of the sidewall forward current NG 1 00 Emission coefficient of the gate edge forward current CJBR F m 1 00e 12 Bottom junction capacitance at V Vp HSPICE MOSFET Models Manual X 2005 09 221 5 Standard MOSFET Models Levels 50 to 64 Level 50 Philips MOS9 Model 222 Table 46 JUNCAP Model Parameters MOSFET Level 50 Continued Name Unit Default Description CJSR F m 1 00e 12 Sidewall junction capacitance at V Vp CJGR F m 1 00e 12 Gate edge junction capacitance at V2 Vg VDBR V 1 00 Diffusion voltage of the bottom junction at T Tp VDSR V 1 00 Diffusion voltage of the sidewall junction at T Tp VDGR V 1 00 Diffusion voltage of the gate edge junction at T
475. reas effective substrate doping threshold voltage effective mobility sm channel length modulation subthreshold current LEVEL 8 Model Parameters MOSFET Level 8 uses the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters This level also uses the parameters described in this section which apply only to MOSFET Level 8 Table 41 Channel Length Modulation Parameters MOSFET Level 8 Name Alias Units Default Description Al 0 2 Channel length modulation exponent CLM 8 CLM 7 Channel length modulation equation selector LAM1 1 m 0 0 Channel length modulation length correction LAMBDA LAM LA 0 0 Channel length modulation coefficient HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 8 IDS Model LEVEL 8 Model Equations This section lists the LEVEL 8 model equations IDS Equations LEVEL 8 ids equations are the same as in the LEVEL 2 model see LEVEL 2 Model Equations on page 130 Effective Channel Length and Width The Level 8 model calculates the effective channel length and width from the drawn length and width see LEVEL 2 Model Equations on page 130 Effective Substrate Doping nsub The SNVB model parameter varies the substrate doping concentration linearly as a function of vsb nsub NSUB SNVB vsb The preceding equation computes y 6 and xd parameters for nsub 2 u 4q nsub Y COX
476. rge gate voltage renders a small Vth shift to a small change in the IDS current The substrate threshold sensitivity can affect circuits such as analog amplifiers that include transistors at back bias and low gate voltages HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 38 IDS Cypress Depletion Model LEVEL 38 IDS Cypress Depletion Model The LEVEL 38 Cypress Depletion MOSFET model Cypress Semiconductor Corporation is a further development of the Synopsys Level 5 MOSFET device model Level 38 features BSIMestyle length and width sensitivities Degraded body effect at high substrate bias second GAMMA Empirical fitting parameters for Ids current calculations in the depletion mode of operations Acomprehensive surface mobility equation Drain induced barrier lowering At the default parameter settings the LEVEL 38 model is basically backwards compatible with LEVEL 5 ZENH 0 0 with the exception of the surface mobility degradation equation see the discussion on the next page Refer to the documentation for LEVEL 5 for the underlying physics that forms the foundation for the Huang Taylor construct In LEVEL 38 the temperature compensation for threshold is ASPEC style concurring with the default in LEVEL 5 This section describes the model parameters that are unique to this depletion model It also describes additional temperature compensation parameters LEVEL 38 lets you
477. ription k1 y1 2 0 6 First order body effect coefficient kiwi m 0 First order effect width dependent parameter k1w2 m 0 Second order effect width dependent parameter k2 0 Second order body effect coefficient k3 0 Narrow coefficient k3b 1 V 0 Body effect coefficient of k3 kb1 1 Backgate body charge coefficient keta 1 V 0 6 Body bias coefficient of the bulk charge effect Ketas V 0 0 Surface potential adjustment for the bulk charge effect LbjtO m 0 2e 6 Reference channel length for the bipolar current lii 0 Channel length dependence parameter for the impact ionization current lint m 0 0 Length offset fitting parameter from I V without bias Ln m 2 0e 6 Electron hole diffusion length Nbjt 1 Power coefficient of the channel length dependency for the bipolar current NdioDE 1 0 Diode non ideality factor nfactor 1 Subthreshold swing factor Ngidl V 1 2 GIDL Vg enhancement coefficient nlx m 1 74e 7 Lateral non uniform doping parameter NrecfO 2 0 Recombination non ideality factor at the forward bias 470 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIMS3 SOI Model Table 130 MOSFET Level 57 DC Parameters Continued Parameter Unit Default Description NrecrO 10 Recombination non ideality factor at the reversed bias Ntun 10 0 Reverse tunneling non ideality factor pclm 1 3 Channel length modulation parameter PDIBLC1 0 39 Correction parameter
478. rred out of the i node from a voltage change on the j node The arrows representing direction of influence point from node j to node i A MOS device with terminals named D G S B provides _dQD dVG CGG represents input capacitance a change in gate voltage requires a current equal to CGGxdVG dt into the gate terminal CGG 228 CGD 428 CDG dVG dVD CGD represents Miller feedback a change in drain voltage creates a current equal to CGGxdVG dt out of the gate terminal CDG represents Miller feedthrough capacitive current out of the drain due to a change in gate voltage To show how CGD might not equal CDG the following example is a simplified model with no bulk charge with a gate charge as a function of VGS only and with the 50 50 channel charge partitioned into QS and QD QG Q vgs QS 0 5 Q vgs QD 0 5 Q vgs OB 0 Consequently _ 4QG dQD 5 dQ CGD war 0 CDG dVG 0 5 dvgs Therefore this model has Miller feedthrough but no feedback Operating Point Capacitance Printout The operating point printout reports six capacitances Table 16 Operating Point Capacitance Capacitance Value cdtot dQD dVD cgtot dQG dVG cstot dQS dVS HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Table 16 Operating Point Capacitance Continued Capacitance Value cbtot dQB dVB cgs dQG dVS cgd dQG dVD These capacitances in
479. rrent To use LDAC and WDAC enter XL LD LDAC XW WD and WDAC in the MODEL statement The model uses the following equations for DC current calculations Leff L XL 2 LD Weff 2 W XW 2 WD HSPICE MOSFET Models Manual 95 X 2005 09 2 Technical Summary of MOSFET Models Noise Models The model parameters also use the following equations to calculate the AC gate capacitance Leff L4XL 2 LDAC Weff 2 W XW 2 WDAC The noise calculations use the DC Weff and Leff values Use LDAC and WDAC with the standard XL LD XW and WD parameters Do not use LDAC and WDAC with other parameters such as DLO and DWO Noise Models 96 This section describes how to use noise models Table 22 Noise Parameters Name Units Default Description Alias AF 1 0 Flicker noise exponent KF 0 0 Flicker noise coefficient Reasonable values for KF are in the range 1e 19 to 1e 25 VF NLEV 2 0 Noise equation selector GDSNOI 1 0 Channel thermal noise coefficient use with NLEV 3 The MOSFET model noise equations have a selector parameter NLEV that selects either the original SPICE flicker noise or an equation proposed by Gray and Meyer You can model thermal noise generation in the drain and source resistors as two sources inrd and inrs units amp Hz as shown in Figure 10 on page 37 The following equations calculate the values of these sources 4 amp 5 NG t0 He inrs inrd rs rd HSPICE MOSFET M
480. s HSPICE MOSFET Models Manual 553 X 2005 09 8 Customer Common Model Interface Extended Topology Figure 44 Gate Tunneling Current Components gs tunnel gcs tunnel l Polysilicon Gate gb tunnel lgd tunnel gcd tunnel Substrate Oxide In this figure loastunnel Igsstunnel is the tunneling current between gate and drain source through overlaps locd tunnel Igcs tunnel represents the tunneling current from gate to channel and then to drain source lobstunnel IS the tunneling current component that takes place between gate and bulk Generalized Customer CMI modeling capability assumes that the above current components each have bias dependence similar to those in BSIM4 lod tunnei depends only on vgd los tunner depends only on vgs You can use the Customer CMI option custcmi for backward compatibility To activate gate tunneling current modeling include an OPTIONS custcmi 1 statement in the netlist To turn off gate tunneling current modeling include an OPTIONS custcmi 0 statement in the netlist The following is the list of variables being added into headfile CMldef h to support gate direct tunneling current modeling double Igbmod flag to control GBMOD 1 for ON and O for OFF double Igcmod flag to control currents IGCMOD 1 for ON and 554 ga ty ga Ca to bulk tun
481. s 1 vth 0 4 0 65 0 5 1 ETA vds 0 75 0 001 vds 0 745 1 wur 61963 1 744 0 8364 1 body 14 829 14025 9 1153116 2 1 ve Oarg 1 vdsat Was vth _ G 9 745 _ 3 60000 body J arg body At vds VDDM default VDDM 5 mobilityzmus 700 2 ea ep ape E Ge 2 body arg TM ey E ids 7 1 153111 cox 54953 36 cox ids 1 09907e 2 These calculations agree with the above simulation results HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model Compatibility Notes Model Parameter Naming The following names are HSPICE specific U00 DLO DWO PHIO ETAO NBO and NDO A zero was added to the SPICE names to avoid conflicts with other standard parameter names For example you cannot use UO because it is an alias for UB the mobility parameter in many other levels You cannot use DL because it is an alias for XL a geometry parameter available in all levels You can use DLO and DWO with this model but you should use XL LD XW and WD instead noting the difference in units Watch the units of TOX It is safest to enter a number greater than one which simulation always interprets as Angstroms To avoid negative gds 1 Set X3U1 LX3U1 and WX3UI to zero 2 Check that zx3ms gt 0 where zx3ms X3MS with L W adjustment 3 Check that zmuz VDDM zx3ms zmus SPICE Synopsys Model Parameter Differences Table 86 compares t
482. s except for the LEVEL 5 model If you do not specify PHI then PHT Dap in 2248 If you do not specify GAMMA then 2 q si NSUB COX GAMMA The following equations determines the energy gap eg and intrinsic carrier concentration used in the above equations tnom eg llo Oe FEET TEL 1 mom 54 300 tnom ee 300 1 cm ni 1 45 10 HSPICE MOSFET Models Manual 59 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Impact lonization In the preceding equation tnom TNOM 273 15 If you do not specify VTO then for Al Gate TPG 0 the following equation determines the ms work function PE eens die 7 type 5 0 05 ms In the preceding equation type is 1 for n channel or 1 for p channel For Poly Gate TPG 1 the following equations determine the work function If you do not specify the NGATE model parameter then ms type TPG eg PH 2 If you specify NGATE then ms type TPG vt BESTE If you do not specify the VTO model parameter then the following equation determines the VTO voltage VTO vfb type GAMMA PHI PHI q NSS In the preceding equation vfb ms Cox DELVTO If you specify VTO then VTO VTO DELVTO MOSFET Impact lonization 60 Impact ionization current is available for all MOSFET levels ALPHA VCR and IIRAT are the controlling parameters IIRAT sets the fraction of the impact ioniz
483. s Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model Table 86 Comparing Synopsys Model Parameters amp UCB SPICE 2 3 Continued UC Berkeley SPICE 2 3 Synopsys Device Model LMUS um cm2 V s WMUS um cm V s X2MS cm V s X3MS cm V s X3U1 um V TOX um TEMP C VDD V CGDO F m CGSO F m CGBO F m XPART NO NB ND RSH ohm sq JS A m PB V MJ PBSW V HSPICE MOSFET Models Manual X 2005 09 LMS LMUS WMS WMUS same same same TOXM u TOX m TREF VDDM CGDOM CGDO CGSOM CGSO CGBOM CGBO same same NBO NDO RSHM RSH IJS JS PJ PB MJO MJ PJW PHP 343 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 344 Table 86 Comparing Synopsys Model Parameters amp UCB SPICE 2 3 Continued UC Berkeley SPICE 2 3 Synopsys Device Model MJSW MJW MJSW CJ F m CJM CJ CJSW F m CCJW CJSW WDF m DELL m In UCB SPICE you must specify all BSIM model parameters The Synopsys model provides default values for the parameters Parasitics ACM s0 invokes parasitic diode models ACM 0 default is SPICE style Temperature Compensation The default TNOM model reference temperature is 25 C unless you use OPTION SPICE to set the default TNOM value to 27 C This option also sets some other SPICE compatibility parameters You set TNOM in an OPTION line in the netli
484. s conductance gm Substrate trans conductance gmbs Turn on voltage von Saturation voltage vsat Gate overlap capacitances cgso cgdo cgbo Intrinsic MOSFET charges ag ad qs Intrinsic MOSFET capacitances referenced to bulk cggb cgdb cgsb cbgb cbdb cbsb cdgb cddb cdsb Parasitic source and drain conductances gs gd Substrate diode current ibd ibs Substrate diode conductance gbd gbs Substrate diode charge qbd qbs Substrate diode junction capacitance capbd capbs Substrate impact ionization current isub Substrate impact ionization trans conductances gbgs dIsub dVgs gbds dIsub dVds gbbs dIsub dVbs Current for the source resistance noise squared nois irs Current for the drain resistance noise squared nois ird Current for the noise from the Thermal or Shot channel squared nois Tdsth Current for the source resistance noise squared nois idsfl You cannot use Meyer capacitance models in HSPICE or HSPICE RF The CMr VAR variable type in the include CMldef h file transfers the transistor biases and the output characteristics between the Customer CMI and the model interface routines The vds vgs and vbs entries provide bias conditions The other entries carry the results from evaluating the model equations must be consistent with its counterpart in HSPICE Fa typedef struct CMI var in in do device
485. s device models All MOSFET models follow this convention You can use OPTION SCALE with the LEVEL 5 model however you cannot use the SCALM option due to the difference in units You must specify the following parameters for MOS LEVEL 5 VTO VT TOX UO UB FRC and NSUB DNB IDS Equations Cutoff Region vgs lt vth Ij O See Subthreshold Current Ids on page 148 On Region vgs gt vth vde 2 3 2 _ 3 2 Lx p Cae dec KA vde v p Pet vob 1 The following equations calculate values used in the preceding equation vde min vq Vasat Werf D UB cox f Lo DNB P 2v In ni HSPICE MOSFET Models Manual 145 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 5 IDS Model 146 The following equation calculates the gate oxide capacitances per unit area Box F AM T TOXCIEHO 77 Effective Channel Length and Width The following equations determine the effective channel length and width in the LEVEL 5 model We Wecatea WMLT OXETCH L L 4 LMLT 2 LATD DEL scale eff Threshold Voltage Vj The VTO model parameter is an extrapolated zero bias threshold voltage for a large device The following equation calculates the effective threshold voltage including the device size effects and the terminal voltages Vin Vut Op yy The following equations calculate values used in the preceding equation 2 Ey q DNB Y bi COX Note You must specify DNB an
486. s only capacitors with interconnects Diffusion model compatible with SPICE BSIM diffusion models To set Level 13 model parameters either Enter model parameters as numbers as in SPICE or Assign the model parameters If you convert from SPICE to the Synopsys models use the S keyletter for SPICE BSIM or M for the Synopsys model see IDS and VGS Curves for PMOS and NMOS on page 347 BSIM Model Features Vertical field dependence of the carrier mobility Carrier velocity saturation Drain induced barrier lowering Depletion charge sharing by source and drain Non uniform doping profile for ion implanted devices Channel length modulation JSubthreshold conduction Geometric dependence of electrical parameters LEVEL 13 Model Parameters MOSFET Level 13 uses the generic MOSFET model parameters described in Chapter 3 Common MOSFET Model Parameters It also uses the parameters described in this section which apply only to MOSFET Level 13 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model Note When you read parameter names be careful about the difference in appearance between the upper case letter O the lower case letter o and the number zero 0 For reference purposes only simulation obtains the following default values from a medium size n channel MOSFET device To specify LEVEL 13 parameters use NMOS conventions even for PMOS for examp
487. s to gather in the cuts making etching difficult The inability to accurately control the metal width necessitates very conservative design rules and results in low transistor gains HSPICE MOSFET Models Manual 27 X 2005 09 2 Technical Summary of MOSFET Models Field Effect Transistors Planar techonology In planar technology the oxide edges are smooth with a minimal variance in metal thickness Shifting to nitride is accomplished using polysilicon gates Adding a chemical reactor to the MOS fabrication process enables depositing silicon nitride silicon oxide and polysilicon The ion implanter is the key element in this processing by using implanters with beam currents greater than 10 milliamperes Because implanters define threshold voltages diffusions and field thresholds processes require a minimum number of high temperature oven steps This enables low temperature processing and maskless pattern generation The new wave processes are more similar to the older nonplanar metal gate technologies Field Effect Transistors The metal gate MOSFET is nonisoplanar as shown in Figure 1 and Figure 2 on page 29 Figure 1 Field Effect Transistor Source drain put into the field oxide Metal used to form the MOS gate as les well as contacting the source and drain c Source Drain F Thin oxide cut Source drain to metal contact Look
488. sat and the drain to source channel voltage vds Typically you can use this equation for long channels and high dopant concentrations This corresponds to GDS 1 in MSINC Table 36 CLM 2 Electrostatic Fringing Field Name Alias Units Default Description A1 0 2 First fringing field factor gate drain A2 0 6 Second fringing field factor gate vdsat AL E vds vdsat T COX Al vds vgs vbi A2 vgs vbi vdsat HSPICE MOSFET Models Manual 177 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model 178 You can use the fringing field equation or electrostatic channel length reduction developed by Frohman Bentchkowski to model short channel enhancement transistors In MSINC the equivalent equation is GDS 2 Table 37 CLM 3 Carrier Velocity Saturation for MOSFET Level 6 Name Alias Units Default Description KA 1 0 vds scaling factor for velocity saturation KCL 1 0 Exponent for vsb scaling factor KU 0 0 Velocity saturation switch If KU lt 1 simulation uses the standard velocity saturation equation LAMBDA cm V 2 1 137e 4 Channel length modulation If you do not LAM LA specify LAMBDA simulation calculates it from NSUB The default LAMBDA corresponds to the default NSUB value MAL 0 5 vds exponent for velocity saturation MCL 1 0 Short channel exponent AL vfu MCL LAMBDA vds vfa Pvsat KCL vsb PHI KCL vsb PHI This eq
489. sb Phid 3 vsb Phid 3 eri vgs von 2 B von vfb vde d e fast Depletion vgs vfb 0 E 24 ids B1 osa NI vde eav vgs vfb vde YEE 2 vgs von Cas Gan y vde vsb Phid vsb Phid tes ug fast Example Model File file Depstor mod MODEL DEPSTOR NMOS LEVEL 38 PARASITIC ELEMENTS ACM 1 LD 0 15u WD 0 2u S for LEFF AND WEFF c C J 0 3E 16 MJ 0 4 PB 0 8 JS 2 0E 17 INTRINSIC DIODE JSW 0 MJSW 0 3 BULK 98 DEFAULT NODE FOR SUBSTRATE THRESHOLD VTO 2 5 LVT 0 25 WVT 0 leta 0 02 eta 0 0 weta 0 0 TCV 0 003 S TEMPERATURE COEFFICIENT MISC DVIN 0 5 PHI 0 75 NFS 2e10 DNB 3 0E16 Mobility Model UH 1300 UO 495 FRC 0 020 FSB 5e 5 VFRC le 4 BFRC 0 LUO 100 LFRC 03 LFSB 1e 5 LVFRC 002 LBFRC le 3 HSPICE MOSFET Models Manual 203 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 40 HP a Si TFT Model WUO 30 WFRC 0 01 WFSB 5e 5 WVFRC 0 00 WBFRC 0 4e 3 KIO 9 KBETA1 5 LKI0 20 16 LKBETA1 2 0 15 WKI0 20 0 WKBETA1 0 0 t BEX 1 3 TUH 1 0 Frcex 1 0 Body Effect NWM 0 5 SCM 1 DVSBC 0 1 t TDVSBC 002 BetaGam 0 9 LBetaGam 2 WBetaGam 1 Hy DVSBC 0 WDVSBC 0 Saturation ECV 2 9 VST 8000 UHSAT 0 CHANNEL LENGTH MODULATION t XJ 0 1 OXIDE TH
490. sical returning 12 physical parameters For LEVEL 28 only PHIO and MUZ are physical returning 11 physical parameters For LEVEL 39 only PHIO and MUZ are physical returning 7 physical parameters Robustness and Convergence Properties A discontinuity in the GM GDS and GMBS derivatives can cause convergence problems Also because real devices have continuous derivatives a discontinuity leads to a large inaccuracy in the derivatives near that region This can be annoying to an analog designer looking at a plot of gain versus bias for example The most common important discontinuities are GDS at vds vdsat and GM at vgs vth The LEVEL 2 and 3 models contain these discontinuities but the LEVEL 13 28 and 39 models do not However the LEVEL 13 model BSIM1 often produces a negative GDS which is obviously inaccurate and causes oscillation which can lead to convergence failure or a timestep too small error A LEVEL 13 model can avoid negative GDS but it depends on complex relationships between the MUZ X2M MUS X2MS X3MS U1 X2U1 and X3U1 parameters Usually you can set X3MS 0 to remove a negative GDS but this lowers the accuracy of the model in the linear region The LEVEL 39 BSIM2 model can also produce negative GDS unless you select parameters carefully The LEVEL 28 model does not create a negative GDS The BSIM1 model has a continuous GM at vgs vth but a plot of GM IDS versus VGS shows a kink when dat
491. sion transition region CDSC F m 2 4e 4 Drain source and channel coupling capacitance CDSCB F Vm2 0 Body coefficient for CDSC CIT F m 0 0 Interface state capacitance PCLM 1 3 Coefficient of the channel length modulation PDIBL1 0 39 Coefficient 1 for the DIBL Drain Induced Barrier Lowering effect PDIBL2 0 0086 Coefficient 2 for the DIBL effect DROUT 0 56 Coefficient 3 for the DIBL effect DSUB DROUT DIBL coefficient in the subthreshold region PSCBE1 V m 4 24e8 Exponent 1 for the substrate current induced body effect PSCBE2 m V 1 0e 5 Coefficient 2 for the substrate current induced body effect AO 1 Bulk charge effect Default is 4 4 for PMOS TNOM TREF C 25 Temperature at which simulation extracts parameters This parameter defaults to the TNOM option which defaults to 25 C See 4 and 5 in Notes on page 387 SUBTHMOD 2 Subthreshold model selector SATMOD 2 Saturation model selector KETA 1 V 0 047 Body bias coefficient of the bulk charge effect A1 1 V 0 First nonsaturation factor 0 for NMOS 0 23 for HSPICE MOSFET Models Manual X 2005 09 PMOS 385 7 BSIM MOSFET Models Levels 47 to 65 Level 47 BSIM3 Version 2 MOS Model Table 91 MOSFET Level 47 Model Parameters Continued Name Unit Default Description A2 1 0 Second nonsaturation factor 1 0 for NMOS 0 08 for PMOS UTE 1 5 Mobility temperature exponent KT1L Vm 0 Channel length sensitivity of the temperature coefficient for the threshold volt
492. st and you can always use the TREF model parameter to override it locally that is for a model The model reference temperature means that the model parameters were extracted at and are valid at that temperature UCB SPICE does not use TNOM default 27 C for the BSIM models Instead you must specify the TEMP model parameter as both the model reference temperature and the analysis temperature Analysis at TEMP applies only to thermally activated exponentials in the model equations You cannot adjust model parameter values when you use TEMP Simulation assumes that you extracted the model parameters at TEMP because TEMP is both the reference and the analysis temperature In contrast to UCB SPICE s BSIM the Synopsys LEVEL 13 model does provide for temperature analysis The default analysis temperature is 25 C and 27 C in UCB SPICE for all model levels except for BSIM as explained in the HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model previous paragraph Use a TEMP statement in the netlist to change the analysis temperature The LEVEL 13 model provides two temperature coefficients TCV and BEX The following equation adjusts the threshold voltage vth t vth TCV t tnom This model includes two implementations of the BEX factor To select a BEX version use the UPDATE parameter described in the next section The mobility in BSIM is a combination of five quantities M
493. st modify the shaded files or add new ones for new models Figure 40 Customer CMI Directory Structure Unix Linux Platforms mos1 mos2 mos3 b1 b2 b3 JFET usermodel CMI config include link HSPCMI makecmi getarch changes required when adding models doc test lib obj Directory Description HSPCMI Subdirectory containing the utility that processes the configuration files and the makefiles get arch C shell script for identifying platforms config Configuration file doc Customer CMI documentation link Main Customer CMI routines include Customer CMI header files makecmi Master makefile test Model testing example HSPICE MOSFET Models Manual X 2005 09 519 8 Customer Common Model Interface Directory Structure 520 Directory Description lib Shared library directory obj Object code mos1 Model directories mos2 mos3 b1 b2 b3 JFET Figure 41 Customer CMI Directory Structure PC Platforms CMI mos1 cmimodel dsw include link Release test mos2 cmimodel dsp mos3 cmimodel def b1 b2 b3 b3v2 JFET usermodel changes required when adding models Directory Description cmimodel dsw Project workspace file cmimodel dsp Project file cmimodel def Definition file include Customer CMI header files HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Running Simulations
494. stalled in Solaris by default but it is installed in SUNOS 4 1 x by default You must either install the optional language software or use a workable compiler such as cc or acc in the Sun workshop You can also use the gcc compiler with some minor modifications to the makefile HSPICE MOSFET Models Manual 525 X 2005 09 8 Customer Common Model Interface Adding Proprietary MOS Models 526 Runtime Shared Library Path The shared library is now ready to use You must update the shared model search path defined in the hspice lib models environment variable so that the system dynamic loader can find the new Customer CMI shared library Enter the following setenv hspice lib models HSPICE CMI lib Troubleshooting Sometimes even if you successfully build the Customer CMI dynamic library HSPICE or HSPICE RF returns an error message when you run simulation error Unable to load home ant lib models libCMImodel and home ant lib models libCMImodel so If this problem occurs it is usually because one or more symbols are undefined when you run the simulation Note Different compilers usually generate nm output in different formats The following is an example output for undefined symbols using the Sun workshop cc compiler on a SUNOS 5 5 machine nm libCMImodel grep UNDEF 35 0 0 NOTY LOCL 0 UNDEF 34 0 0 NOTY LOC
495. stance CMImos3Getlpar c NN CMImos3Getlpar CMImos3Setlpar CMI_AssignInstanceParm CMImos3Setlpar c CMI_Evaluate een CMImos3GetMpar c CMImos3GetMpar CMImos3SetMpar CMImos3SetMpar c CMI MOSDEF CMI_AssignModelParm CMImos3evaluate Optional Routines CMImos3eval c CMI DiodeEval CMI Noise CMI PrintModel CMI FreeModel CMI Freelnstance CMI WriteError COMI Start CMI_Conclude HSPICE MOSFET Models Manual 551 X 2005 09 8 Customer Common Model Interface Extended Topology Extended Topology 552 In addition to conventional four terminal topoid 0 MOSFET topology HSPICE or HSPICE RF can support other topologies You must assign a unique topoid for each topology To implement BSIM SOI topology Customer CMI assigns topoid 1 If you create your own model named topovar and it is the same as the BSIM SOI model you can specify topoid 1 and use the HSPICE topology structure for stamping information f your model topology is different from either the conventional 4 terminal model or the BSIM SOI then you must specify the topovar structure HSPICE or HSPICE RF assigns a unique topoid for your topology The naming convention for the structure fields is the same as in the BSIM SOI model For detailed information about fields in the structure see the BSIM3PD2 0 MOSFET MODEL User Manual
496. subdirectory This file contains a simple CMOS inverter using MOS LEVEL 3 models Modify the transistor sizes and the model cards as necessary hspice mos3 sp mos3 lis You can then use Avanwaves to inspect the I V and C V characteristics at different biasing conditions Use AvanWaves to carefully check the following aspects Sign and value of the channel current ids m Monotone characteristics of the channel current versus vgs and vds Sign and value of the capacitance cgs cgs cgb csb cdb Refer to the AvanWaves User Guide for more information To verify the Customer CMI integration of your new model run both a DC sweep analysis and a transient analysis on the test netlist Note LEVELs from 100 to 200 are reserved for Customer CMI customer models Choose levels from this range so your custom models do not conflict with existing Synopsys model levels Also add a special prefix or suffix for some of the auxiliary functions used in Customer CMI especially those from the public domain such as the modchk function or dc3p1 from Berkeley Spice3 This ensures that the function names are different from those used in the simulator s core code HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Model Interface Routines After testing if you are satisfied with your Customer CMI library put it in the default Customer CMI library directory installdir SARCH lib models In this syntax SARCH can be
497. substrate overlap capacitance per channel length CGSL F m 0 0 Lightly doped source gate region overlap capacitance CGSO F m calculated Non LDD region source gate overlap capacitance per channel length CJSWG F m2 1 e 10 Source drain gate side sidewall junction capacitance per unit width normalized to 100nm Tsi CKAPPA F m 0 6 Coefficient for lightly doped region overlap capacitance fringing field capacitance CLC m 0 1e 7 Constant term for the short channel model CLE 0 0 Exponential term for the short channel model CSDESW F m 0 0 Source drain sidewall fringing capacitance per unit length CSDMIN V cal Source drain bottom diffusion minimum capacitance DLC m lint Length offset fitting parameter for the gate charge DWC m wint Width offset fitting parameter from C V MJSWG V 0 5 Source drain gate side sidewall junction capacitance grading coefficient PBSWG V 0 7 Built in potential for the source drain gate side sidewall junction capacitance TT second 1ps Diffusion capacitance transit time coefficient 490 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 59 UC Berkeley BSIM3 SOI FD Model Table 137 MOSFET Level 59 AC Capacitance Parameters Continued Parameter Unit Default Description VSDFB V cal Flatband voltage for the source drain bottom diffusion capacitance VSDTH V cal Threshold voltage for the source drain bottom diffusion capacitance XPART 0 Charge partitioning ra
498. sun4 so14 or pa depending on the platform you use to compile your Customer CMI library Model interface routines accept input parameters from Customer CMI For each set of input conditions the model routines must return the transistor characteristics to Customer CMI Model Interface Routines Model interface routines accept the following input parameters from Customer CMI Circuit and nominal model temperatures CKTtemp CKTnomt emp Input biases vds vgs vbs Model parameters level vto tox uo Instance parameters w 1 as ad Mode of the transistor mode 1 for normal 1 for reverse AC frequency freq passes from the simulator to the model code Integration order intorder for transient simulation HSPICE or HSPICE HF returns the following codes e 0 Trapezoidal e 1 1st order Gear e 2 2nd order Gear Transient time step timestep Transient time point timepoint For each set of input conditions the model routines must return the following transistor characteristics to Customer CMI Flag for computing the charge and capacitance 1 for computation 0 for no computation qflag Selector for the charge or capacitance model 0 for Meyer capacitance model 13 for the capop charge based model Channel current ids Channel conductance gds HSPICE MOSFET Models Manual 529 X 2005 09 8 Customer Common Model Interface Model Interface Routines 530 Tran
499. t for merged devices only DMCGT 0 0m No DMCG of the test structures NF 1 No Number of device figures DWJ DWC in CVmodel No Offset of the S D junction width MIN 0 No Minimize the number of drain or source diffusions for even number fingered device XGW 0 0m No Distance from the gate contact to the channel edge XGL 0 0m No Gate length offset due to patterning variations NGCON 1 No Number of gate contacts Table 124 Asymmetric Source Drain Junction Diode Model Parameters MOSFET Level 54 Parameter Default Binnable Description IJTHSREV IJTHSREV 0 1A No Limiting current in the reverse bias region IJTHDREV IJTHDREV No Limiting current in the reverse bias region IJTHSREV IJTHSFWD IJTHSFWD 0 1A No Limiting current in the forward bias region IJHDFWD IJTHDFWD No Limiting current in the forward bias region IJTHSFWD HSPICE MOSFET Models Manual X 2005 09 457 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 124 Asymmetric Source Drain Junction Diode Model Parameters MOSFET Level 54 Continued Parameter Default Binnable Description XJBVS XJBVS 1 0 No Fitting parameter for the diode breakdown XJBVD XJBVD XJBVS No Fitting parameter for the diode breakdown BVS BVS 10 0V No Breakdown voltage BVD BVD BVS No Breakdown voltage JSS JSS 1 0e 4A m2 No Bottom junction reverse saturation current density JSD JSD JSS No Bottom junction reverse saturation current density JSWS JSWS 0 0A m No Isolation
500. t it HSPVER 98 2 No Selects from HSPICE Versions 98 2 97 4 97 2 96 4 96 3 96 1 PARAMCHK 0 No PARAMCHK 1 checks the model parameters for range compliance APWARN 0 No When s0 turns off the warning message for PS PD Weff HSPICE specific BINFLAG 2 0 No Uses wref Iref if you set this flag gt 0 9 HSPICE MOBMOD 1 No Selects a mobility model 414 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 96 Model Flags for MOSFET Levels 49 53 Continued Name Unit Default Bin Description CAPMOD 3 No Selects from the 0 1 2 3 charge models Level 49 CAPMOD defaults to 0 CAPOP No Obsolete for Levels 49 and 53 HSPICE ignores it HSPICE specific in all versions NOIMOD 1 No Berkeley noise model flag NLEV off No The noise model flag non zero overrides NOIMOD HSPICE specific See Noise Models on page 96 for more information NQSMOD 0 off No NQS Model flag SFVTFLAG O off No Spline function for Vth HSPICE specific VFBFLAG O off No VFB selector for CAPMOD 0 HSPICE specific Table 97 Basic Model Parameters MOSFET Levels 49 53 Name Unit Default Bin Description VGSLIM V 0 No Asymptotic Vgs value The Min value is 5V 0 value indicates an asymptote of infinity HSPICE and Level 49 specific TOX m 150e 10 No Gate oxide thickness XJ m 0 15e 6 Yes Junction depth NGATE cm3 0 Yes Poly gate doping concentrati
501. t30 output template 30 double templt31 output template 431 double templt32 output template 432 double templt33 output template 433 double templt34 output template 434 double templt35 output template 435 double templt36 output template 36 double templt37 output template 437 double templt38 output template 438 double templt39 output template 439 double templt40 output template 440 Among the above member variables vges vgms vdbs vdes vses and qdef are biases passed in from the HSPICE engine While all the other components should be calculated through CMI Evaluate and CMI Noise routines among which there are circuit element summary components from cdsat to delnfct in the above list and components for template output from temp1t1 to temp1t 40 in the above list Taking gate tunneling current modeling as an example and assuming that LEVEL 101 is to be used for example as in the example code in extcmi mos101 model evaluations are carried out within the function CMImos101evaluate and the HSPICE MOSFET Models Manual 561 X 2005 09 8 Customer Common Model Interface Extended Topology 562 results here and G are transferred to the HSPICE engine by the gate current related variables using the following pointers
502. tacted case with an accessible internal body node You can use the temperature node to simulate thermal coupling Level 57 Model Parameters Table 128 MOSFET Level 57 Model Control Parameters Parameter Unit Default Description capmod 2 Flag for the short channel capacitance model MOBMOD 1 Mobility model selector 466 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 57 UC Berkeley BSIM3 SOI Model Table 128 MOSFET Level 57 Model Control Parameters Continued Parameter Unit Default Description noimod 1 Flag for the noise model SHMOD 0 Flag for self heating 0 no self heating 1 self heating Table 129 MOSFET Level 57 Process Parameters Parameter Unit Default Description DTOXCV Difference between the electrical and physical gate oxide capmod 3 only thicknesses due to the effects of the gate poly depletion and the finite channel charge layer thickness Nch i cm3 1 7e17 Channel doping concentration Ngate 1 cm3 O Poly gate doping concentration Nsub i cm3 6 0e16 Substrate doping concentration Tbox m 3 0e 7 Buried oxide thickness Tox m 1 0e 8 Gate oxide thickness Tsi m 1 0e 7 Silicon film thickness Xj m S D junction depth Table 130 MOSFET Level 57 DC Parameters Parameter Unit Default Description a0 1 0 Bulk charge effect coefficient for the channel length A1 1 V 0 0 First non saturation effect parameter A2 1 0 Second non saturation
503. tal lt 0 Level 49 reports a warning ETAO lt 0 Warn if paramchk 1 DSUB lt 0 Fatal lt 0 Level 49 reports a warning VBM Ignored if you defined K1 and K2 UO 0 Fatal B1 Weff Fatal B1 Weff lt 107 Warn if paramchk 1 VSAT lt 0 Fatal lt 10 Warn if paramchk Al See a2 conditions on the next line A2 0 01 Warn and reset a2 0 01 if paramchk 1 1 Warn and reset a2 1 a1 0 if paramchk 1 DELTA lt 0 Fatal RDSW lt 0 001 Warn if paramchk 1 and reset rdsw 0 NFACTOR x0 Warn if paramchk 1 CDSC lt 0 Warn if paramchk 1 CDSCD lt 0 Warn if paramchk 1 HSPICE MOSFET Models Manual 429 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 107 Model Parameter Range Limit Levels 49 53 Continued Name Limits Comments PCLM 0 Fatal PDIBLC1 0 Warn if paramchk 1 PDIBLC2 0 Warn if paramchk 1 PS Weff Warn WO Weff Fatal wO Weff lt 107 Warn if paramchk 1 DROUT lt 0 Fatal if paramchk 1 Level 49 reports a warning PSCBE2 lt 0 warn if paramchk 1 CGSO lt 0 Warn and reset to 0 if paramchk 1 CGDO lt 0 Warn and reset to 0 if paramchk 1 CGBO lt 0 Warn and reset to 0 if paramchk 1 ACDE lt 0 4 51 6 Warn MOIN 5 0 gt 25 Warn IJTH 0 Fatal NOFF lt 0 1 gt 4 0 Warn Table 108 Element Parameter Range Limit MOSFET Levels 49 53 Name Limits Comments PD lt Weff Warn PS lt Weff Warn 430 HSPICE MOSFET Models Manual
504. tance per meter channel CGB length If you set WD and TOX but you do not set CGBO then simulation calculates CGBO CGBOscaled CGBO SCALM CGDO Fim 0 0 Gate drain overlap capacitance per meter channel CGD C2 width If you set LD or METO and TOX but you do not set CGDO then simulation calculates CGDO CGDOscaled CGDO SCALM CGSO Fim 0 0 Gate source overlap capacitance per meter CGS C1 channel width If you set LD or METO and TOX but you do not set CGSO then simulation calculates CGSO CGSOscaled CGSO SCALM LD LATD m Lateral diffusion into channel from source and drain DLAT diffusion lf you do not specify either LD or XJ then the LD default 0 0 If you specify XJ but you do not specify LD then simulation calculates LD from XJ LD default 0 75 XJ for all levels except LEVEL 4 for which LD default 0 75 LDscaled LD SCALM LEVEL 4 LDscaled LD XJ SCALM METO m 0 0 Fringing field factor for gate to source and gate to drain overlap capacitance calculation METOscaled METO SCALM WD m 0 0 Lateral diffusion into channel from bulk along width WDscaled WD SCALM HSPICE MOSFET Models Manual 75 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Table 20 Meyer Capacitance Parameters CAPOP 0 1 2 Name Units Default Description Alias CF1 V 0 0 Modified MEYER control for transition of cgs from depletion to weak inversion for CGSO for CAPOP 2 only CF2 V 0 1 Modifie
505. te flag Table 138 MOSFET Level 59 Temperature Parameters Parameter Unit Default Description AT m sec 3 3e4 Temperature coefficient for U4 CTHO m9 C W s 0 Normalized thermal capacity KT1 V 0 11 Temperature coefficient for the threshold voltage KT2 0 022 Body bias coefficient for the temperature effect of the threshold voltage KTIL V m 0 Channel length dependence of the temperature coefficient for the threshold voltage PRT Q um 0 Temperature coefficient for Rasy RTHO m C W 0 Normalized thermal resistance TNOM 0G 25 Temperature at which simulation expects parameters UA1 m V 4 31e 9 Temperature coefficient for U4 UB1 m V 7 61e 18 Temperature coefficient for Up UC1 1 V 0 056 Temperature coefficient for Ug UTE 1 5 Mobility temperature exponent XBJT 1 Power dependence of jp on the temperature HSPICE MOSFET Models Manual X 2005 09 491 7 BSIM MOSFET Models Levels 47 to 65 Level 60 UC Berkeley BSIM3 SOI DD Model Table 138 MOSFET Level 59 Temperature Parameters Continued Parameter Unit Default Description XDIF XBJT Power dependence of jai on the temperature XREC 1 Power dependence of je on the temperature XTUN 0 Power dependence of ji on the temperature Note BSIMFD refers the substrate to the silicon below the buried oxide not to the well region in BSIM3 It calculates the backgate flatband voltage Vibb and the parameters related to the bottom capacitance of the source drain diffus
506. th and width sensitivity parameters xbs zphi vy g lene 1 744 0 8364 xbs body 14g zk1 xul zul vbs zx2ul 2 xbs 2 1 2 rx body zu1 2 body Vost zX3ul 4 Vest 2 Vost Vdsat 7 body rx This vds value generates the partial derivative of Vds E 2x GRE Aao Vase Vos n body 2 vas EG Iul va Vag In the preceding equation vds zero Transition Points The effective model parameter values for the transition points after you adjust the device size are zb1 and zb2 Simulation calculates these values from the B1 and B2 model parameters and from their respective length and width sensitivity parameters Vdsat v 2v zbl dsat l Visat v2 Vasat zb2 Vost 356 HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 28 Modified BSIM Model Strong Inversion Current For vds lt v1 Vas Ij beta v body 2 v TEC CGL S S NE The vds derivative varies approximately linearly between v1 and v2 For vdssv2 ids is a function of beta and vgst only If zb1 and zb2 are both positive their main effect is to increase the saturation current Weak Inversion Current The effective model parameter values for weak inversion current after you adjust the device size are zn0 znb znd zwfac and zwfacu Simulation calculates these values from the NO NDO NBO WFAC and WFACU model parameters and from their respective length and width se
507. the BSIM3v3 scheme as noted in the comments column of Table 107 To control the maximum number of simulation warning messages printing to the output file use OPTION WARNLIMIT f In the preceding OPTION statement is the maximum number of warning messages that simulation reports The default WARNLIMIT value is 1 In some cases as noted in Table 107 simulation checks parameters only if you set the PARMAMCHK 1 model parameter Table 107 Model Parameter Range Limit Levels 49 53 Name Limits Comments TOX 0 Fatal 10 9 Warn if parmchk 1 TOXM 0 Fatal 10 9 Warn if parmchk 1 XJ 0 Fatal NGATE 0 Fatal if 51023 simulation multiplies NGATE by gt 102 Fatal 10 before the other limit checks lt 1018 Fatal if parmchk 1 Level 49 returns 0 Fatal gt 102 Warn lt 101 Warn if paramchk NSUB lt 0 Fatal Ignores NSUB if k1 k2 are defined lt 1014 Warn if parmchk 1 gt 10 Warn if parmchk 1 NLX lt Leff Fatal lt 0 Warn if parmchk 1 428 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models Table 107 Model Parameter Range Limit Levels 49 53 Continued Name Limits Comments NCH 0 de if 5102 simulation multiplies NCH by lt 10 Warn if parmchk 1 10 6 before the other limit checks gt 10 Warn if parmchk 1 DVT1W 0 Fatal lt 0 Level 49 reports a warning DVTO lt 0 Warn if paramchk 1 DVT1 0 Fa
508. the VTHO calculation DVTO 2 2 Yes First coefficient of the short channel effect on Vip DVT1 0 53 Yes Second coefficient of the short channel effect on Vin DVT2 0 032V Yes Body bias coefficient of the short channel effect on Vin DVTPO 0 0m Yes First coefficient of the drain induced Vj shift due to long channel pocket devices DVTP1 0 0 v Yes First coefficient of the drain induced Vyp shift due HSPICE MOSFET Models Manual X 2005 09 to long channel pocket devices 447 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 115 Basic Model Parameters MOSFET Level 54 Continued Parameter Default Binnable Description DVTOW 0 0 Yes First coefficient of the narrow width effect on Vi for a small channel length DVT1W 5 3e6m Yes Second coefficient of the narrow width effect on Vin for a small channel length DVT2W 0 032V Yes Body bias coefficient of narrow width effect for small channel length UO 0 067m Vs Yes Low field mobility NMOS 0 025 m Vs PMOS UA 1 0e 9m V for Yes Coefficient of the first order mobility degradation MOBMOD 0 due to the vertical field and 1 1 0e 15m V for MOBMOD 2 UB 1 0e 19m V Yes Coefficient of the second order mobility degradation due to the vertical field UC 0 0465V for Yes Coefficient of the mobility degradation due to the MOB MOD 1 body bias effect 0 0465e 9 m V for MOBMOD 0 and 2 EU 1 67 NMOS No Exponent for the mobility degradation of 1 0 PMOS MOBMO
509. the element statement generate parasitics These parameters do not have default option values To suppress the diode set IS 0 AD 0 and AS 0 If you set AS 0 in the element and IS 0 in the model simulation suppresses the source diode Use this setting for shared contacts Figure 13 ACM 2 MOS Diode Source Gate Drain HDIF Contact gt lt LD 3 af 9 gt k LDIF Example For a transistor with LD 0 07um W 10 um L 2 um LDIF 1 um and HDIF 4 um Table 12 shows typical MOSFET diode parameter values HSPICE MOSFET Models Manual 49 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models Table 12 ACM 2 MOS Diode Parameters Parameter Description AD Area of drain Default option value for AD is not applicable AS Area of source Default option value for AS is not applicable CJ 1e 4 F m CJSW 1e 10 F m JS 1e 4 A m JSW 1e 10 A m HDIF Length of heavily doped diffusion contact to gate about 2 um HDIFeff HDIF WMLT SCALM LDIF LD Length of lightly doped diffusion about 0 4um NRD Number of squares drain resistance Default value for NRD does not apply NRS Number of squares source resistance Default for NRS does not apply PD Periphery of drain including gate width for ACM 2 No default PS Periphery of source including gate width for ACM 2 No default RD Resistance ohm square of lightly do
510. the following vde and vds values vde min vds vfa vsat UPDATE 0 vds min vds vfa vsat UPDATE 1 2 Subthreshold Current ids is the choice of two different equations selected through The WIC Weak Inversion Choice model parameter characterizes this region of operation Parameter Description WIC 0 No weak inversion default WIC 1 ASPEC style weak inversion WIC 2 Enhanced HSPICE style weak inversion HSPICE MOSFET Models Manual 169 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model 170 In addition to WIC set the NFS parameter NFS represents the number of fast states per centimeter squared Reasonable values for NFS range from 1e10 to 1e12 WIC 0 no weak inversion WIC 1 the vth threshold voltage increases by the fast term von vth fast The following equation calculates the fast value used in the preceding equation q NFS Y COX 2 vsb PHI fast vt L t In the preceding equations vt is the thermal voltage The following equation specifies the ids current for vgs von vgs von ids ids von vde vsb e fast if vgs lt von then ids ids vge vde vsb Note Strong inversion conditions do not use the modified threshold voltage von WIC 2 The subthreshold region is limited between the cutoff region and the strong inversion region If the gate voltage is less than vth PHI this model cannot include any weak inversion conduction However
511. the position independent code PIC Dynamic linking uses PIC For SUNOS 5 4 platforms or later KPIC the automatically generated compiler flag and the G z link flag are for the Sun workshop compiler cc or acc You typically install these compilers in a directory such as usr1 opt SUNWspro SCA 2 bin For HP9000 700 platforms install cc in opt ansic bin For more information type man cc or man acc to display the on line manual page for these commands You can use any optimization flags for the Customer CMI library However for best results use the fast flag for the cc or acc Sun compiler and use the O flag for the cc HP compiler Using the gcc Compiler If you use the gcc compiler modify the makefile makefile SUN or makefile HP to set the compiler flags correctly For gcc set the CC environment variable to gcc Modify the makefile to use the fPIC flag to compile and the r flag to link For example gcc c I include fPIC CMImain c gcc r o lib libCMImodel obj o Using the usr ucb cc Compiler If you use the usr ucb cc compiler modify the makefile makefile SUN or makefile HP to set the compiler flags correctly The usr ucb cc compiler cannot compile C source files until you install the Language Optional Source Package Verify that this source package is installed then set the CC environment variable to usr ucb cc Note The Language Optional Source Package is not in
512. thography and etching process 440 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model steps and finally to the electrical size as a result of subsequent ion implementation and annealing steps Name Alias Default Description LMLT 1 0 Channel length shrinking factor WMLT 1 0 Device width shrinking factor Both LMLT and WMLT must be greater than 0 if not simulation resets them to 1 0 default and issues a warning message To use these two parameters add them in the model cards without any other modifications For example model nmos nmos level 54 1mlt20 85 wmlt 0 9 The drawn channel length L is multiplied by LMLT The drawn channel width W is multiplied by WMLT BSIMA evaluates the effective length and width Leff and Weff as Leff Lnew 2 0 db Weff Wnew 2 0 dw Lnew and Wnew are evaluated as L XL W XW dL and dW are evaluated with Lnew and Wnew TO pow Lnew LIN T1 pow Wnew LWN dL LINT LL TO LW T1 LWL TO T1 T2 pow Lnew WLN T3 pow Wnew WWN dw WINT WL T2 WW T3 WWL T2 T3 Similarly the preceding equations determine the Ldlc Ldlcig LeffCV Wdwc Wdwcig WeffCV WeffCJ grgeltd and Wnew model variables and quantities When multiplied by the LMLT and WMLT parameters Leff and Weff become Leff Lnew 2 0 dL Weff Wnew 2 0 dL HSPIC
513. tializes it as pae NSUB 10 GAMMA cox for NSUB gt 0 default otherwise If you do not specify PHI simulation initializes it as 6 NSUB 10 2V T In for B PHI VAT om a aT or NSUB gt 0 default otherwise If you do not specify VTO simulation initializes it as VFB PHI GAMMA JPHI if you specify VFB VTO default otherwise If you do not specify DP simulation initializes it as ic qe 10 COX for UO gt 0 default otherwise If you do not specify UCRIT simulation initializes it as VMAX UO 10 for VMAX gt 0 UO 5 0 UCRIT default otherwise HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model If you do not specify EO simulation uses a simplified mobility model with the THETA parameter E0 0 if you specify THETA default otherwise Note The EO value is zero indicating to use the simplified mobility model is used in conjunction with THETA instead of the standard mobility model Optional parameters might not be available in all implementations Default Values and Parameter Ranges If you do not define a specific model parameters simulation either initializes it according to the above relations or sets it to its default value Certain parameters restrict their numerical range to avoid numerical problems such as divisions by zero If you specify a parameter value outside the specified range s
514. tional initial value of Vbs which you specify for transient BSIM PD version 2 23 includes several bug fixes and enhancements from version 2 2 e Adds geometric dependency in CV delta L and delta W Fixes a gate body tunneling residue problem in the low bias region e Provides an additional parameter dtoxcv in capMod 3 for flexibility e Other bug fixes Using BSIM3 SOI PD To use BSIM3 SOI PD versions 2 0 2 2 2 21 or 2 22 in simulation apply the VERSION model parameter For example Invokes PD2 0 if VERSION 2 0 Invokes PD2 2 and PD2 21 if VERSION 2 2 nvokes PD2 22 if VERSION 2 22 nvokes PD2 23 if VERSION 2 23 For gate body tunneling set the IGMOD model parameter to 1 Example mckt drain gate source bulk nch L 10 6 W 10 6 model nch nmos Level 57 igmod 1 version 2 2 tnom 27 tox 4 5e 09 tsi 0000001 tbox 8e 08 mobmod 0 capmod 2 shmod 0 paramchk 0 wint 0 lint 2e 08 vth0 2 42 k1 49 k2 1 k3 0 k3b 2 2 nlx 2e 7 dvt0 10 dvt1l 55 dvt2 1 4 dvtOw 0 dvtlw 0 dvt2w 0 nch 4 7e 17 nsub 1e 15 ngate let20 agidl 1E 15 bgidl 1E9 ngidl 1 1 ndiode 1 13 ntun 14 0 nrecf0 2 5 nrecr0 4 vrec0 1 2 ntrecf 1 ntrecr 2 isbjt 1E 4 isdif 1E 5 istun 2E 5 isrec 4E 2 xbjt 9 xdif 9 xrec 9 xtun 0 01 ahli 1e 9 1bjt020 2e 6 ln 22e 6 nbjt 8 ndif 1 aely 1e8 vabjt 0 u0 352 ua 1 3e 11 ub 1 7e 18 uc 4e 10 w
515. tions so it effectively changes the default units in parameters that these options affect Parameter values must be consistent with these scaling factors LEVEL 6 ACM 1 CJ 0 0 IS 0 0 NSUB 1e15 PHI 14 P the Fermi potential TLEV 1 TLEVC 1 Note Do not calculate NSUB from GAMMA if UPDATE 1 or 2 TLEV TLEVC selects the ASPEC method of updating temperatures for the CJ CJSW PB PHP VTO and PHI parameters Note If you explicitly enter PHI this model does not update it for temperature SCALM does not affect how simulation scales parameters for the ASPEC mode If you specify SCALM when you use ASPEC the Level 7 MOSFET model generates an error stating that it ignores SCALM LEVEL 7 IDS Model The LEVEL 7 model is the same as the LEVEL 6 model except for the PHI value If you specify PHI then For LEVEL 6 o where Pz is the surface potential For LEVEL 7 PHI HSPICE MOSFET Models Manual 181 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 8 IDS Model To transform a LEVEL 7 equation to LEVEL 6 make the following substitution PHI 2 PHI To transform a LEVEL 6 model into a LEVEL 7 model make the following substitution PHI Level 7 PHI Level 6 2 LEVEL 8 IDS Model 182 The LEVEL 8 MOSFET model derived from research at Intersil and General Electric is an enhanced version of the LEVEL 2 ids equation LEVEL 2 differs from LEVEL 8 in the following a
516. tive GDS 5 5 a oos iE as ath a Be eee EA 577 Monotonic GM IDS in weak inversion lees 577 Behavior Follows Devices in All Circuit Conditions 577 Ability to Simulate Process Variation llle 578 Gate Capacitance Modeling liiis 578 Outline of Optimization Procedure 0 0 sens 579 Examples of Data Fitting 0 2 e eee eee 580 LEVEL 28 2 3 Ids Model vs Data nananana anaana eee eee 580 LEVEL 13 28 39 Ids Model vs Data 0000s 582 LEVEL 2 3 28 Gds Model vs Data 00 aaa 583 LEVEL 13 28 39 Gds Model versus Data 585 LEVEL 2 3 28 Ids Model versus Data 586 LEVEL 13 28 39 lds Model versus Data 588 LEVEL 2 3 28 Gm Ids Model versus Data 589 LEVEL 13 28 39 Gm lds Model versus Data 591 Gds versus Vds at Vgs 4 Vbs 0 2 593 Gm lds vs Vgs at Vds 0 1 Vbs 0 2 ee 595 Gm Ids versus Vgs at Vds 0 1 Vbs 0 ne 596 Index an PN AA PA RE sa ie fn 597 xvi HSPICE MOSFET Models Manual X 2005 09 About This Manual This manual describes standard MOSFET models that you can use when simulating your circuit designs in HSPICE or HSPICE RF Inside This Manual This manual contains the chapters described below For descriptions of the other manuals in the HSPICE documentation set see t
517. to 64 Level 63 Philips MOS11 Model SL2KO 1 737E 14 KPINV 2 2E 01 PHIBR 0 85 BETSQ 1 201E 04 ETABET 1 3 FBET1 3 741000E 01 IP1 2 806E 06 LP2 1E 10 THESATEXP 2 THESRR 7 109E 01 THEPHR 1E 03 TOX 3 2E 09 ETAPH 1 75E 00 ETAMOBR 2 825 NUR 1 NUEXP 3 228 THERR 1 267E 01 ETAR 0 4 THER2 1 THESATR 6 931E 02 SLTHESAT 1 ETASAT 8 753E 01 SSFR 2 304E 03 VP 5E 02 SLSSF 1 002E 06 ALPR 1 062E 02 SLALP 9 957E 01 ALPEXP 1 039 THETHR 2 413E 03 THETHEXP 1 S 2 DIBLO 1 06E 06 SDIBLEXP 6 756 LLMIN E 07 R 1 05E 03 MOEXP 3 146 R 9 938E 04 STAl 9 3E 02 SLA1 2 805E 03 A2R 4 047E 01 SLA2 1E 15 R 7 54E 01 SLA3 8 705E 08 L 3 2E 10 NTR 1 6237E 20 NFAR 1 NFBR 0 NFCR 0 GATENOISE 0 CJIBR 1 347E 3 CJSR 0 183E 9 CJGR 0 374E 9 JSDBR 0 027E 6 JSDSR 0 040E 12 JSDGR 0 100E 12 VR 0 000 JSGBR 1 900E 6 JSGSR 78 000E 12 JSGGR 54 000E 12 VB 20 000 VDBR 0 828 VDSR 0 593 VDGR 0 500 PB 0 394 PS 0 171 PG 0 193 NB 1 000 NS 1 000 NG 1 000 Example 2 model nch nm
518. to OFF The single exception in Level 49 is that ACM defaults to 0 To achieve Level 49 compliance with Berkeley BSIM3v3 set ACM 10 If you set any of the following parameter values for MOSFET Level 49 and 53 simulation reports a warning Leff lt 5e 8 Weff lt 1e 7 LeffCV lt 5e 8 WeffCV 1e 7 Simulation aborts if you set Leff or Weff lt 0 0 Selecting Model Versions Recommended BSIM3v3 Version The recommended BSIM3v3 model specification is Level 49 VERSION 3 24 This version provides the most stable and up to date representation of the UCB BSIM3v3 2 4 model However do not change the VERSION specification in existing model cards without consulting the foundry or model extraction group that created the model cards As of the 99 2 release there are five official BSIM3v3 releases from Berkeley and several Level 49 releases For additional release information from the UCB group see the BSIM3 home page http www device EECS Berkeley EDU bsim3 To minimize confusion and maintain backward compatibility you can select the VERSION and HSPVER model parameters VERSION selects the Berkeley release version HSPVER selects the Synopsys release version For example HSPVER 97 2 and VERSION 3 1 reproduce results from HSPICE 97 2 using the BSIM3 Version 3 1 model HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models HSPVER defaults to the curr
519. to drain capacitance HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Diode Models MOSFET Diode Models You can use the Area Calculation Method ACM parameter to precisely control bulk to source and bulk to drain diodes within MOSFET models Use the ACM model parameter to select one of three different modeling schemes for the MOSFET bulk diodes This section discusses the model parameters and model equations used for the different MOSFET diode models Selecting MOSFET Diode Models To select a MOSFET diode model set the ACM parameter within the MOSFET model statements f ACM 0 the pn bulk junctions of the MOSFET are modeled in the SPICE style The ACM 1 diode model is the original ASPEC model The ACM 2 model parameter specifies the improved diode model which is based on a model similar to the ASPEC MOSFET diode model The ACM 3 diode model is a further improvement that deals with capacitances of shared sources and drains and gate edge source drain to bulk periphery capacitance If you do not set the ACM model parameter the diode model defaults to the ACM 0 model Jf ACM 0 and ACM 1 models you cannot specify HDIF In the ACM 0 model you cannot specify LDIF The ACM 1 model does not use the AD AS PD and PS geometric element parameters Enhancing Convergence The GMIN option creates a parallel conductance across the bulk diodes and drain source for transi
520. to their linearized mathematical dependencies instead of their physical origin to better provide parameter extraction HSPICE MOSFET Models Manual 201 X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 38 IDS Cypress Depletion Model 202 Channel Length Modulation To include the channel length modulation modify the ids current ids 1 AL Leff The following equation calculates AL for the preceding equation ids 1 3 Apatite are PHI Pee nal n 622 nal AL is in microns if XJ is in microns and nat is in cm3 Subthreshold Current ids If device leakage currents become important for operation near or below the normal threshold voltage the model considers the subthreshold characteristics In the presence of surface states this equation determines the effective threshold voltage von von max vth vinth fast The following equation calculates the fast value used in the preceding equation L cox 2 Phid vsb If vgs von then Partial Enhancement 0 lt vgs vfb lt vde 2 ids p1 la zKI0 NI vde cav von vfb vde cav y vde vsb Phid vsb Phid teris vgs von B e f B1 cav von vfb Nile HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 38 IDS Cypress Depletion Model Full Enhancement vgs vfb vde gt 0 ids B1 la zKIO NI vde 5 cav Y vde v
521. trinsic parameters Description This option calculates the electrical parameters as a function of the optional parameters COX and GAMMA TOX PHI NSUB VTO gt VFB KP UO UCRIT VMAX cm is the length unit for the NSUB and UO parameters Consequence These parameters accommodate scaling behavior and allow meaningful statistical circuit simulation due to decorrelation of physical effects If you use these optional parameters this version is compatible with former revisions except for the default calculation of the parameters HSPICE MOSFET Models Manual 247 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model 248 Optional simplified mobility model Description The simple mobility model used in former model versions by using the THETA parameter was reinstated as an option Consequence This mobility model simplifies adaptation from earlier model versions to the current version Parameter synonyms Description You can use EO and QO as synonyms for the EO and QO parameters Consequence This option accommodates some simulators that use only alphabetic characters Operating point information Description This enhancement models the analytical expression for the SPICE like VTH threshold voltage in the operating point information to include the charge sharing and reverse short channel effects This option modifies the analytical expression for the VDSAT saturation voltage in the operati
522. tu pdgr hisim hisim html Since the STARC HiSIM1 1 0 release the Synopsys version of the HiSIM model has included a VERSION number parameter to facilitate backward compatibility Starting in the 2003 03 release Synopsys uses the STARC version control mechanism so you must enter an integer for the VERSION model parameter For example to specify HiSIM version 1 0 0 set the VERSION model parameter to 100 If you do not set the VERSION parameter simulation issues a warning and automatically sets this parameter to 100 You can set the VERSION value to 100 101 102 HiSIM1 0 series 110 111 112 HiSIM1 1 series 120 HiSIM1 2 You can find more details about the differences between the HiSIM versions at the previously mentioned STARC web address Table 66 Level 64 Model Selectors Parameter Default Description LEVEL 64 Model selector VERSION 100 Model version number CORSRD O no Flag Indicates whether to include the Rs and Rd contact resistors and whether to solve equations iteratively CORSRD 1 yes COOVLP O Overlap capacitance model selector COOVLP 1 constant value COOVLP 0 approximating the field linear reduction e COOVLP 1 considering the lateral impurity profile HSPICE MOSFET Models Manual 311 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 64 STARC HiSIM Model Table 66 Level 64 Model Selectors Continued Parameter Default Description COISUB 0 Substrate current
523. ty 1 cm 3 5 0e16 n2 Bulk region doping density 1 cm 3 2 0e16 wbrk Depth of surface region micron 0 2 Vfbll Length dependence parameter of vthO 0 vfble Exponent for length dependence of vthO 1 vfbwl Width dependence parameter of vthO 0 vfowe Exponent for width dependence of vthO 1 dphii Norm error in phi at extro VthO 0 csill I dependence parameter of n1 0 csile Exponent for I dependence of n1 1 cs2ll I dependence parameter of n2 0 cs2le Exponent for I dependence of n2 1 cs1wl W dependence parameter of n1 0 csiwe Exponent for w dependence of n1 1 cs2wl W dependence parameter of n2 0 cs2we Exponent for w dependence of n2 1 dibll I dependence parameter of dibl 0 dible Exponent for I dependence of dibl 2 HSPICE MOSFET Models Manual X 2005 09 509 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Table 145 SSIMSO Model intrinsic Parameters Geometry Modifiers and Threshold Voltage Continued Name Parameter Units Default gp1 Bulk charge coefficient 1 744 gp2 Bulk charge coefficient 1 V 0 8364 shrink Linear size reduction 0 shrink2 Modified areal size reduction 0 Table 146 SSIMSOI Model Intrinsic Parameters Mobility and Saturation Output Conductance Name Parameter Units Default ubref Mobility parameter cm 2 V s 700 nmos 300 pmos ubred Mobility filed reduction factor cm V egvexp 100 eavfac Effective field coefficient 0 5 eavfwl Width dependence of eavfac 0 0 eavfwe Exponent
524. u OD 3 body Qd ES vgs vtho Qs Od 40 60 Channel Charge Partitioning for Drain and Source XPART 0 Triode Region vgs gt vtho vds lt vpof Qg cap vgs xvfb zphi 0 5 Pvds argx vds Qb cap vtho zvfb zphi 1 body 0 5 argx vds Qd cap 0 5 vgs vtho body Pvds body argy vds Qs Qg Qb Od Saturation Region vgs vtho vds vpof ed ae er ni ras viho Qg cap ves zvfb zphi PETITE Ob caps ye eee ha SS OY L 3 body Qd LA vgs vtho Os Qd 0 100 Channel Charge Partitioning for Drain and Source XPART 1 Triode Region vgs vtho vds vpof Qg cap vgs zvfb zphi 0 5 Pvds vds argx Qb cap vtho zvfb zphi 1 body 0 5 argx vds Qd cap 0 5 vgs vtho body Pvds b 0 75 1 5 Pargx Qs Qg Qb Qd HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model Saturation Region vgs gt vtho vds gt vpof cap vgs zwfo zphi P8220 Qg cap vgs zvfb zphi 3 body mS fs Ese N10 Qb cap cub zphi vtho 1 body 3 body Qd 0 Qs Qg Qb Preventing Negative Output Conductance The LEVEL 13 model internally protects against conditions that might cause convergence problems due to negative output conductance The constraints imposed are ND20 MUS 2 MUZ X3MS VDD M 2 This model imposes these constraints
525. uation is an extension of the first depletion layer equation CLM 1 It includes effects of carrier velocity saturation and source to bulk voltage vsb depletion layer width It represents the basic ISPICE equation See Alternate DC Model ISPICE model on page 168 for definitions of vfa and vfu Table 38 CLM 4 Wang s Equation for MOSFET Level 6 Name Alias Units Default Description Al m 0 2 Junction depth A1scaled A1 SCALM DND cm3 1e20 Drain diffusion concentration HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model Linearly Graded Depletion Layer 11 3 Aka tO Ed vds vdsat PHI PHI Kici 167 Wang s equation can include junction characteristics to calculate the channel length modulation The equation assumes that the junction approximates a linearly graded junction and provides a value of 0 33 for the exponent This equation is similar to MSINC GDS 3 Table 39 CLM 5 Channel Length Modulation for MOSFET Level 6 Name Alias Units Default Description LAMBDA amp V 0 Constant coefficient VGLAM 1 V 0 Constant coefficient If CLM 5 the ids current increases by idssat idssat e LAMBDA vds vgs vth 1 VGLAM vgs vth ids ids idssat Note The equation adds the idssat term to ids in all regions of operation Also LAMBDA is a function of the temperature Table 40 CLM 6 AL Equation for MOSF
526. ubstrate Figure 9 Equivalent Circuit MOSFET AC Analysis Source O cgb 7 Gate Drain ES gmbs vbs t cbs gbd Substrate HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOSFET Equivalent Circuits Figure 10 Equivalent Circuit MOSFET AC Noise Analysis Gate cob cos cgd Source ka vas rd Drain 9 CHAM e87 69 HW 99 gm vgs gds ges Te Fea be IAN Se x RE inrs gmbs vbs inrd ind 9 dq cbs gos E gbd gbd cbd Substrate Table 6 MOSFET DC Operating Point Output Quantities Definitions id drain current ibs B S bulk to source current ibd B D bulk to drain current vgs G S gate source voltage vds D S drain source voltage vbs B S bulk source voltage vth threshold voltage vdsat saturation voltage HSPICE MOSFET Models Manual 37 X 2005 09 2 Technical Summary of MOSFET Models MOSFET Equivalent Circuits 38 Table 6 MOSFET DC Operating Point Output Continued Quantities Definitions beta beta value gam eff gamma effective gm DC gate transconductance gds D S drain source conductance gmb B S bulk source conductance cdtot total drain capacitance cgtot total gate capacitance cstot total source capacitance cbtot total bulk capacitance total floating body capacitance for SOI MOSFET cgs total gat to source capacitance cgd total gat
527. ulation extracts it from the model card f you specify CTHO it overrides CTHO in the model card OFF Sets the initial condition to OFF for this element in DC analysis BJTOFF Turns off BJT if equal to 1 IC Initial guess in the order drain front gate internal body back gate external voltage ignores Vps in a 4 terminal device Simulation uses these settings if you specify UIC in the TRAN statement The IC statement overrides them Level 59 Model Parameters Table 134 MOSFET Level 59 Model Control Parameters Parameter Unit Default Description CAPMOD 2 Flag for the short channel capacitance model Level Level 59 for BSIM3SOI MOBMOD 1 Mobility model selector 484 HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 59 UC Berkeley BSIM3 SOI FD Model Table 134 MOSFET Level 59 Model Control Parameters Continued Parameter Unit Default Description NOIMOD 1 Flag for the Noise model SHMOD 0 Flag for self heating 0 no self heating e 1 self heating Table 135 MOSFET Level 59 Process Parameters Parameter Unit Default Description NCH i cm3 1 7e17 Channel doping concentration NGATE 1 cm3 0 Poly gate doping concentration NSUB i cm3 6 0e16 Substrate doping concentration TBOX m 3 0e 7 Buried oxide thickness TOX m 1 0e 8 Gate oxide thickness TSI m 1 0e 17 Silicon film thickness Table 136 MOSFET Level 59 DC Parameters Parameter Unit
528. uration region In the Synopsys MOSFET device models these problems are corrected If you specify the CAPOP 4 model parameter then simulation uses the level dependent recommended charge conservation model The XQC model parameter selects the ratio of channel charge partitioning between drain and source For example if you set XQC 4 then in the saturation region 40 of the channel charge is associated with the drain and the remaining 60 is associated with the source In the linear region the ratio is 50 50 Simulation uses an empirical equation to make a smooth transition from 50 50 linear region to 40 60 saturation region The capacitance coefficients are the derivative of gate bulk drain and source charges and are continuous LEVEL 2 3 4 6 7 and 13 models include a charge conservation capacitance model To invoke this model set CAPOP 4 The following example compares only the CAPOP 4 charge conservation capacitance and the CAPOP 9 improved charge conservation capacitance for the LEVEL 3 model The CGS and CGD capacitances for CAPOP 4 model SPICE2G 6 show discontinuity at the boundary between the saturation and linear regions The CAPOP 9 model does not have discontinuity For comparison the modified Meyer capacitances CAPOP 2 is also provided The shape of CGS and CGD capacitances resulting from CAPOP 9 are much closer to those of CAPOP 2 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Mo
529. urce there is no drain charge XPART 0 5 selects 50 50 partitioning Half of the channel charge in the saturation region is at the source and half is at the drain HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model Define vtho zvfb zphi zk1 zphi vsb cap COX Leff Weff vpof vgs vtho body body vds argx 12 vgs vtho 0 5 Pbody Pvds If ves vtho 0 5 Pbody Pvds 1e 8 then argx 1 6 vgs vtho 0 75 Pbody P vgs vtho Pvds 0 15 body vds argy 5 6 vgs vtho 0 5 Pbody Pvds If vgs vtho 0 5 Pbody Pvds le 8 then 4 argy i Regions Charge Expressions Accumulation Region vgs vtho vgs x zvfb vsb Qg cap vges zvfb vsb Qb qg Qs 0 Qd 0 Subthreshold Region vgs x vtho vgs zvfb vsb l 1 2 Und ESSE Tito etas efie vs EJ Qb qg Qs 0 50 50 Channel Charge Partitioning for Drain and Source XPART 5 Triode Region vgs vtho vds lt vpof Qg cap vgs zvfb zphi 0 5 Pvds vds argx Qb cap vtho zvfb zphi 1 body 0 5 argx vds Qd 0 5 qg qb Qs Qd HSPICE MOSFET Models Manual 337 X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 338 Saturation Region vgs gt vtho vds gt vpof 3 b hi P832 viho Qg cap vgs zvfb zphi SER Qb cap svfb zphi vtho 1 body Q
530. urn 0 int CMImos3Start CMI Conclude After simulation this routine runs the conclude functions that you define Syntax int CMI Conclude void Example int ifdef STDC CMImos3Conclude void else CMImos3Conclude void endif void CMImos3conclude return 0 int CMImos3Conclude Internal Routines Customer CMI Function Calling Protocol Figure 42 illustrates the calling sequence for the interface routines HSPICE MOSFET Models Manual X 2005 09 8 Customer Common Model Interface Interface Variables Figure 42 Calling Sequence for the Interface Routines Y CMI Start a repeated for each alter repeated for each model lt a CMI ResetModel 4 parameter change M CMI_AssignModelParm Y CMI_SetupModel CMI_WriteError Bo repeated for all parameter settings from a model card repeated for each element ret Gull wer Pai parameter change CMI_AssignInstanceParm Y CMI SetupInstance CMI WriteError Y CMI PrintModel OMI DiodeEval CMI WriteError CMI Evaluate CMI WriteError CMI Noise CMI WriteError CM Freelnstance p repeated for all parameter settings from an instance Y CMI_FreeModel CMI_
531. ut capacitance LX19 is the Miller feedback capacitance gate current induced by voltage signal on the drain m LX32 is the Miller feedthrough capacitance drain current induced by the voltage signal on the gate A device operating with node 3 as electrical drain for example an NMOS device with node 3 at higher voltage than node 1 is in reverse mode The LX values are physical but you can translate them into electrical definitions by interchanging D and S CGG reverse CGG LX18 CDD reverse CSS dQS dVS d QG QB QD dVS LX20 LX23 LX34 CGD reverse CGS LX20 CDG reverse CSG dQS dVG d QG QB QD dVG LX18 LX21 LX32 For Meyer models QD and other charges are not well defined The formulas such as LX18 CGG LX19 CGD are still true but the transcapacitances are symmetrical for example CGD CDG In terms of the six independent Meyer capacitances cgd cgs cgb cdb csb and cds the LX printouts are LX18 m CGS CGD CGB LX19 m LX32 m CGD LX20 m CGS LX21 m CGB LX22 m CDB HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models LX23 m CSB LX33 m CGD CDB CDS LX34 m CDS Calculating Gate Capacitance The following input file example shows a gate capacitance calculation in detail for a BSIM model TOX is chosen so that CON Tp Ahi tox In this example VfbO ph
532. values without disrupting the continuity across the bin boundaries Charge Models In BSIM3v3 the BSIM1 capacitance model is CAPMOD 0 Simulation replaces this with a modified BSIM1 capacitance model based on the CAPOP 13 model in Level 49 Level 53 uses the Berkeley BSIM1 capacitance model for CAPMOD 0 Table 95 lists CAPMOD defaults for the Berkeley BSIM3v3 model and for Levels 49 and 53 Table 95 MOSFET Charge Model Versions Version BSIM3v3 Level 49 Level 53 3 0 1 1 1 3 1 2 0 2 3 2 3 3 3 VFBFLAG The CAPMOD 0 capacitance model normally calculates the threshold voltage as Vth vfbc phi k1 sqrt phi vos where vfbc is the VFBCV model parameter This eliminates any dependence on the VTHO parameter To allow capacitance dependence on VTHO set the VFBFLAG 1 model parameter The CAPMOD 0 capacitance model calculates the threshold voltage as Vth vthO k1 sqrt phi vbs k1 sqrt phi The VFBFLAG default value is 0 HSPICE MOSFET Models Manual 411 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models 412 Printback You can printback all model parameters with units The printback also indicates whether Berkeley or Synopsys model junction diodes and noise models are invoked and which parameters are not used for example simulation does not use CJGATE if ACM 0 3 Mobility Multiplier You can define mobility multiplier parameters in the BSIM8V3 instance line Na
533. vds High vds VdS 2 0 1 Accumulation vgs lt vth cgs CF5 cap G Saturation Region vgs lt vth vds cgs CF5 cap Linear Region vgs vth vds vth vds 7 _ f Jf Es vtn vas cgs CF5 cap L2 vgs vth vds HSPICE MOSFET Models Manual 81 X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Gate Drain Capacitance cgd Low vds vds lt 0 1 Accumulation vgs lt vth ced CF5 cap G Dt Weak Inversion vgs lt vth 0 1 ves vgh 0 1 Jj cgd CFS cap D EEE mao 1 557 D Strong Inversion vgs vth 0 1 T vgs vth cgd CF5 cap nay D 1 B vgs vth vds High vds Vas 0 1 Accumulation vgs lt vth cgd CF5 cap G Dt Saturation Region vgs lt vth vds cgd CF5 cap D Strong Inversion vgs vth vds B l Eg vgs vth ced CF5 cap man D 1 a ae TAE eds Example The netlist for this example is located in the following directory Sinstalldir demo hspice mos capopl sp 82 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models Figure 20 CAPOP 1 Capacitances CAPOP1 SVO C68 VOSP05 BA C68 VOSP05 C68 VOSP05 Q 3 A Param Lin CAPOP2 SVO De C68 VOSP05 M amp M C68 VOSP05 G 20 0F Param Lin 100F gt 87ed8 ica ull dee mE if pe px dst ex p mtr aee E 1 0
534. ve Saturation Current Calculations The ACM 3 model calculates the MOS diode effective saturation currents the same as ACM 2 Effective Drain and Source Resistances The ACM 3 model calculates the effective drain and source resistances the same as ACM 2 HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Diode Equations MOS Diode Equations This section describes MOS diode equations DC Current Simulation parallels the drain and source MOS diodes with GMINDC conductance in the DC analysis Simulation parallels the drain and source MOS diodes with GMIN conductance in the transient analysis The total DC current is the sum of the diode current and the conductance current The diode current is calculated as follows Drain and Source Diodes Forward Biased vbs gt 0 ibs isbs eVbs QN vt vbd O ibd isbd ev 4 N _ 1 Drain and Source Diodes Reverse Biased m For OsvbssVNDS ibs gsbs vbs m For vbs lt VNDS ibs gsbs VNDS 55 vbs VNDS For0 gt vbd gt VNDS ibd gsbd vbd For vbd lt VNDS ibd gsbd VNDS 855 wbd VNDS The following equations calculate values used in the preceding equations lesbs isbs ledbd isbd Using MOS Diode Capacitance Equations Each MOS diode capacitance is the sum of diffusion and depletion capacitance Simulation evaluates the diffusion capacitance in terms of the small signal co
535. ve coefficient a value of 200 for NO disables the weak inversion calculation LNO 0 0 Length sensitivity WNO 0 0 Width sensitivity NBO 0 0 Vsb reduction to the low field weak inversion gate drive coefficient LNB 0 0 Length sensitivity WNB 0 0 Width sensitivity HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model Table 83 Transistor Parameters MOSFET Level 13 Continued Name Alias Units Default Description NDO 0 0 Vds reduction to the low field weak inversion gate drive coefficient LND 0 0 Length sensitivity WND 0 0 Width sensitivity PHIO V 0 7 Two times the Fermi potential LPHI V um 0 0 Length sensitivity WPHI V um 0 0 Width sensitivity TREF C 25 0 Reference temperature of model local override of TNOM TOXM TOX um m 0 02 Gate oxide thickness simulation interprets TOXM or TOX gt 1 as Angstroms U00 1 V 0 0 Gate field mobility reduction factor LUO um V 0 0 Length sensitivity WUO um V 0 0 Width sensitivity U1 um V 0 0 Drain field mobility reduction factor LU1 um2 V 0 0 Length sensitivity WU1 um2 V 0 0 Width sensitivity VDDM V 50 Critical voltage for the high drain field mobility reduction VFBO VFB V 0 3 Flatband voltage HSPICE MOSFET Models Manual X 2005 09 327 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 13 BSIM Model 328 Table 83 Transistor Parameters MOSFET Level 13 Continued Name Alias Units Default Des
536. vel 1 to 40 LEVEL 6 LEVEL 7 IDS MOSFET Model Figure 29 Variation of GM GDS and GMBS for UPDATE 0 TGAM2 SP MULT LEVEL GAMMA MODEL UPDATE 0 P 14 MAY 2003 15 33 26 TGAM2_0 SPO 340 0U IX CCELI A cc 320 0U N 300 0U 288004 15 ia 31 8045U 30 0U 28 0U F 2002 TGAM2_0 SP0 GDS P A R A M TGAM2_0 SP0 GMBS Z gt 1 p l 1 LI i l LTE F Ld I bod LI 4 H 1 050 1 10 1 150 1 20 VOLTS LIN Each plot compares IDS VTH VDSAT GM GDS and GMBS as a function of vsb for UPDATE 0 Improved Multi Level Gamma UPDATE 1 As demonstrated in previous sections the regular Multi Level Gamma displays some discontinuities in saturation voltage and drain current This occurs because when Vsp is less than VBO simulation sets y to yi and uses it to calculate iy and vsat This is not correct if vds vsb exceeds VBO then the depletion regions at the drain side expands into the substrate region and the Vsat computation must use yb instead of yi Because Vsat Vgs Vth drain this model uses yi to compute the threshold voltage at the drain for vsb lt VBO As a result the existing model overestimates the threshold voltage yi yb and underestimates the saturation voltage and the drain current in the saturation region This causes a discontinuous increase in the saturation drain current crossing from the vsb lt VBO region to the vsb VBO region The improved Multi Level model upgrades the s
537. vel 55 includes both a charge based model for transcapacitances allowing charge conservation during transient analysis you can select a simpler capacitances based model instead Note The charges model is symmetrical in terms of the forward and reverse normalized currents that is the model is symmetrical for both the drain and source sides The pinch off voltage in the dynamic model provides the short channel effects such as charge sharing and reverse short channel effects Dynamic Model for the Intrinsic Node Charges GAMMA Ng 1 2 Vp PHI 10 Normalized Intrinsic Node Charges Xp tig X iti l E 3x 6x x Ax x di Bi ar qp Ng 2 3 2 2 3 4 3x Ox x Axgx 2x ds m q 15 xp x 2 2 2 7 4 Xp tx H 4 Ist dp Ea n 1 GAMMA JV PHI 10 9 Toe q for V gt 0 a P V n I G t q q 1 Vey for Vg lt 0 t HSPICE MOSFET Models Manual 241 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model 242 4G 41 dox 7B qox is a fixed oxide charge which simulation assumes is zero The preceding equations express the charge conservation among the four nodes of the transistor Total Charges C COX NP W eff QOu p p s G Cox Vi 40 B D 6 Intrinsic Capacitances Transcapacitances Simulation derives the intrinsic capacitances from the node charges for the terminal voltages d C
538. vels 47 to 65 Level 54 BSIM4 Model Table 113 Model Selectors Controllers MOSFET Level 54 Continued Parameter Default Binnable Description MOBMOD 1 NA Mobility model selector RDSMOD 0 NA Bias dependent source drain resistance model selector IGCMOD 0 NA Gate to channel tunneling current model selector IGBMOD 0 NA Gate to substrate tunneling current model selector CAPMOD 2 NA Capacitance model selector RGATEMOD 0 Gate resistance model selector no gate resistance RBODYMOD 0 NA Substrate resistance network model network off selector TRNQSMOD 0 NA Transient NQS model selector ACNQSMOD 0 NA AC small signal NQS model selector FNOIMOD 1 NA Flicker noise model selector TNOIMOD 0 NA Thermal noise model selector DIOMOD 1 NA Source drain junction diode IV model selector PERMOD 1 NA PS PD includes excludes the gate edge perimeter GEOMOD 0 NA Geometry dependent parasitics model isolated selector RGEOMOD 1 NA Source drain diffusion resistance and contact model selector HSPICE MOSFET Models Manual X 2005 09 445 7 BSIM MOSFET Models Levels 47 to 65 Level 54 BSIM4 Model Table 114 MOSFET Level 54 Process Parameters Parameter Default Binnable Description EPSROX 3 9 SiO5 No Gate dielectric constant relative to vacuum TOXE 3 0e 9m No Electrical gate equivalent oxide thickness TOXP TOXE No Physical gate equivalent oxide thickness TOXM TOXE No Tox at which simulation extracts parameters DTOX 0 0m No Defined
539. versus Vds Curves 12 0U 10 0U 8 0U C MODEL A 6 DATA 6 0U Param Lin 40U 20U PDW Lin 5 0 Figure 52 LEVEL 28 Ids versus Vds Curves 12 0U 10 0U 8 0U C MODEL RS M f 6 DATA Bee 6 0U Param Lin PDW Lin 4 0 5 0 582 HSPICE MOSFET Models Manual X 2005 09 B Comparing MOS Models Examples of Data Fitting Figure 53 LEVEL 39 Ids versus Vds Curves 12 0U GC MODEL E LE ana 8 C DATA 10 0U 8 0U Param Lin o o c TD PDW Lin LEVEL 2 3 28 Gds Model vs Data ds vs Vds at Vgs 2 3 4 5 Vos 0 The plot shows that Level 2 and 3 cannot model GDS accurately Figure 54 LEVEL 2 gds versus Vds Curves 00S MODEL Via dada Q0S DATA Duce Param Lin TOO heh oh NAN bee KAN ab redis tell BG desc NR rs al lad 0 PDW Lin HSPICE MOSFET Models Manual 583 X 2005 09 B Comparing MOS Models Examples of Data Fitting Figure 55 LEVEL 28 gds versus Vds Curves Param Lin 00S_MODEL Kz 00S DATA Sil 1005104120353 la xa a Tos a e Ld 0 1 0 2 0 3 0 4 0 PDW Lin Figure 56 LEVEL 3 gds versus Vds Curves 008 MODEL A 00S_DATA nig AZ Param Lin 10 0
540. versus Vgs Curves 35 0 JU 30 0 gt ni 250 3 20 0 Param Lin 15 0 10 0 1 0 1 50 PGD Lin 2 0 Tp 2 50 3 00 Figure 77 LEVELs 13 28 39 gm lds versus Vgs Curves 24 0 22 0 20 0 18 0 16 0 14 0 12 0 Param Lin 10 0 8 0 Z 6 0 4 0 2 0 hee ie a CA 500 0M 500 0M n 1 0 PGD Lin 596 HSPICE MOSFET Models Manual X 2005 09 A AC analysis MOSFETs 36 ACM model parameter 40 MOS diode 43 46 49 52 parameter 9 39 activating generalized customer CMI enhancements 563 alternate saturation model parameters 155 156 AMD models 5 6 AMI gate capacitance model 93 analysis MOSFETs AC 36 37 transient 36 ASPEC AMI model 4 compatibility 3 49 180 option 10 automatic model selection failure 568 multisweep or TEMP effect 568 See also model selection B basic model parameters LEVEL 1 108 LEVEL 2 114 LEVEL 27 190 LEVEL 39 358 LEVEL 47 382 LEVEL 49 53 414 LEVEL 5 114 LEVEL 50 214 LEVEL 57 466 LEVEL 58 250 LEVELs 6 7 114 HSPICE MOSFET Models Manual X 2005 09 Index Berkeley BSIM3 SOI model 463 junction model 402 NonQuasi Static NQS model 402 BJT models 37 BSIM model 324 equations 332 LEVEL 13 5 339 VERSION parameter effects 332 BSIM2 model 358 equations 362 LEVEL 39 6 VERSION parameter effects 368 BSIM3 model equations 390 Leff Weff 389 LEVEL 47 6 SOI FD 482 BSIM3 SOI F
541. von then vgs ren gds m SAS yf exp Mas HO gds feds fp GO DEFF HSPICE MOSFET Models Manual X 2005 09 4 Standard MOSFET Models Level 1 to 40 References Cgd Cgs Cgd Cgdi CGDO Cgs Cgsi CGSO LEVEL 40 Model Topology Figure 34 shows the topology of the LEVEL 40 model Figure 34 LEVEL 40 HP a Si TFT Topology o RD pa CGDO Drain Cgai gt y lds Gate S Source CGSO Kl 8 gsi S References 1 Vladimirescu Andrei and Liu Sally Simulation of MOS Integrated Circuits Using SPICE2 University of California at Berkeley Memorandum No UCB ERL M80 7 February 1980 2 Huang J S and Taylor G W Modeling of an lon Implanted Silicon Gate Depletion Mode IGFET EEE Trans Elec Dev Vol ED 22 pp 995 1000 Nov 1975 3 Frohman Bentchkowski D and Grove A S On the Effect of Mobility Variation on MOS Device Characteristics Proc IEEE 56 1968 HSPICE MOSFET Models Manual 211 X 2005 09 4 Standard MOSFET Models Level 1 to 40 References 212 4 5 6 7 8 9 Fargher H E and Mole P J The Implementation Of A 3 Terminal SOSFET Model In SPICE For Circuit Simulation GEC VLSI Research Laboratory MOS Division Marciniak W et al Comments on the Huang and Taylor Model of lon Implanted Silicon gate Depletion Mode IGFET Solid State Electron Vol 28 No
542. ween bins Because many BSIM3 model parameters account for MOSFET geometry effects these geometry effect parameters are redundant You can eliminate them when you use binning The BSIM3 lite model parameter set was created in response to the question What BSIMS parameters should be excluded when using a binned model To invoke the BSIM3 lite model specify the LITE 1 model parameter in the model card Simulation checks the model card to determine if it conforms to the BSIM3 lite parameter set BSIM3 lite takes advantage of the smaller number of calculations and reduces simulation times by up to 1096 compared to the full parameter set BSIMS model Only Level 49 supports LITE 1 HSPICE MOSFET Models Manual 407 X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 49 and 53 BSIM3v3 MOS Models 408 Table 94 lists model parameters total 49 that the BSIM3 lite model excludes Either exclude all parameters in this list from the model card or explicitly set them to the default value specified in the list In some cases as noted the BSIM3 lite default value differs from the standard BSIM3v3 default value You should also exclude WR ALPHAO and CIT but the BSIM3 lite model card does not require this exclusion Table 94 Parameters Excluded from BSIM3 Lite Parameter Comments mobmod Recommended default or set 1 nqsmod Recommended default or set 0 toxm default tox Il default 0 lin default 1 lw default 0 Iwn
543. width dependence of m 0 0 SR HSPICE MOSFET Models Manual X 2005 09 281 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 62 Level 68 MOS11 Parameters Level 1100 Continued Name Description Units NMOS PMOS THEPHR Coefficient of the mobility reduction dueto V 1 1 29E 2 1E 3 phonon scattering for the reference transistor at the reference temperature ETAPH Exponent of the temperature dependence 1 75 1 75 of OSR for the reference temperature SWTHEPH Coefficient of the width dependence of m 0 0 SR ETAMOBR Effective field parameter for dependence 1 4 3 on depletion inversion charge reference transistor STETAMOB Coefficient of temperature dependence K 1 0 0 nMOB SWETAMOB Coefficient of the width dependence of m 0 0 nMOB NUR Exponent of the field dependence of the 1 1 mobility model minus 1 such as v 1 atthe reference temperature NUEXP Exponent of temperature dependence v 5 25 3 23 parameter THERR Series resistance coefficient for the V 1 0 155 0 08 reference transistor at reference temperature ETAR Temperature dependence exponent of OR 0 95 0 4 SWTHER Coefficient of the width dependence of OR m 0 0 THER1 Numerator of gate voltage dependent part V 0 0 of the series resistance for the reference transistor 282 HSPICE MOSFET Models Manual X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 63 Philips MOS11 Model Table 62 Level 68 MOS11 Paramet
544. xide capacitance cgbx as cgbx cox cgs cgd Depletion capacitance cd is voltage dependent esi E Est Na ug wd cd wd Weff Leff 4 NSUB vc The effective voltage from channel to substrate bulk The following equations show vc under various conditions vgs vsb lt vfb vc 2 0 vgs vsb gt v b vc vgs vsb vfb vgst gt 0 vgs lt vth vgst lt vds ve 2 vth vfb vgst vsb Nl vgst gt 0 vgs lt vth vgst gt vds ve L vth vfb vgst vds vsb vgs gt vth vgst lt vds vc vth vfb vest vsb HSPICE MOSFET Models Manual X 2005 09 2 Technical Summary of MOSFET Models MOS Gate Capacitance Models vgs gt vth vgst gt vds vc vth vfb vds vsb CAPOP 11 Ward Dutton model specialized LEVEL 2 CAPOP 12 Ward Dutton model specialized LEVEL 3 CAPOP 13 BSIM1 based Charge Conserving Gate Capacitance Model See LEVEL 13 BSIM Model on page 324 CAPOP 39 BSIM2 Charge Conserving Gate Capacitance Model See LEVEL 39 BSIM2 Model on page 358 Calculating Effective Length and Width for AC Gate Capacitance For some MOS processes and parameter extraction method AC analysis might need different Leff and Weff values than for DC analysis For AC gate capacitance calculations substitute the LDAC and WDAC model parameters for LD and WD in the Leff and Weff calculations You can use LD and WD in Leff and Weff calculations for DC cu
545. xpression is valid in all regions of operation including for small Vps thermal Flicker Noise The following equation calculates the PSD flicker noise component 5 7 KF LEN flicker NP W v NS Log COX fA In some implementations you can select different expressions Operating Point Information At operating points the following information displays to help in circuit design Numerical values of model internal variables The following are the intrinsic charges and capacitances Va Vs Vp Ips IDB Img Oms Imbs gt Imad Vp n p IS if ir t t0 HSPICE MOSFET Models Manual 243 X 2005 09 5 Standard MOSFET Models Levels 50 to 64 Level 55 EPFL EKV MOSFET Model 244 Transconductance efficiency factor tef Sms V Ips Early voltage VM Ips 8nd Overdrive voltage For P channel devices n Vp Va has a negative sign P S SPICE like threshold voltage VTH VTO AVpgscg Y JV s GAMMA NPHI This expression is the SPICE like threshold voltage the source It also accounts for charge sharing and reverse short channel effects on the threshold voltage For P channel devices VTH has a negative sign Saturation voltage VDSAT 2Vpss 4V For P channel devices VDSAT has a negative sign Saturation non saturation flag i 1 SATLIM SAT or 1 for i LIN or O for IZSATLIM r Note Some simulators implement the operating point differently some information m
546. yntax for SSIMSOI Mxxx nd ng ns ne np mname L val lt W val gt 506 M val AD val AS val PD val PS val t BODYTYPE val IGATE val lt AB val gt lt PB val gt lt LXB val gt WXB val LPE val DTEMP val HSPICE MOSFET Models Manual X 2005 09 7 BSIM MOSFET Models Levels 47 to 65 Level 65 SSIMSOI Model Parameter Description Mxxx SSIMSOI element name Must begin with M followed by up to 1023 alphanumeric characters nd Drain terminal node name or number ng Front gate node name or number ns Source terminal node name or number ne Back gate or Substrate node name or number np External body contact node name or number mname SSIMSO model name reference L SSIMSOI channel length in meters Default is 5 0 um SSIMSOI channel width in meters Default is 5 0 um M Multiplier to simulate multiple SSIMSOIS in parallel Default 1 AD Drain diffusion area Default 0 AS Source diffusion area Default 0 PD Drain diffusion perimeter Default 0 PS Source diffusion perimeter Default 0 BODYTYPE Flag to choose floating 0 or Tgate 2 Default 0 IGATE Flag to turn on off 0 1 gate current calculations Default 1 AB Body diffusion area Default 0 PB Body diffusion perimeter Body Contacted Default 0 LXB Extrinsic Gate Length Body Contacted Default 0 WXB Extrinsic Gate Width Body Contacted Default 0 HSPICE MOSFET Models Manual
547. zero bias parameter then the following equation calculates the P bias dependent parameter P PO PB V PD V PG V The exceptions are the U1 velocity saturation factor and the N subthreshold swing coefficient Modeling Guidelines Removing Mathematical Anomalies on page 373 shows expressions for their bias dependences Compatibility Notes SPICE3 Flag If you specify the SPICE3 0 default model parameter certain Synopsys corrections to the BSIM2 equations are effective If you set the SPICE3 value to 1 the equations are as faithful as possible to the BSIM2 equations for SPICE3E2 Even in this mode certain numerical problems have been addressed and should not normally be noticeable Temperature The default model reference temperature TNOM is 25 C in the Level 39 MOSFET model unless you set OPTION SPICE which sets the TNOM default to 27 C This option also sets some other SPICE compatibility parameters In the Level 39 model you set TNOM in an OPTION line in the netlist to override this locally that is for a model use the TREF model HSPICE MOSFET Models Manual X 2005 09 6 BSIM MOSFET Models Levels 13 to 39 LEVEL 39 BSIM2 Model parameter Reference temperature means that the model parameters were extracted at and are therefore valid at that temperature UCB SPICE 3 does not use TNOM default 27 C for the BSIM models Instead you must specify the TEMP model parameter as both the model referenc

HPSPICE MOSFET Models

Contents

Download Pdf Manuals

Related Search

Related Contents