Home

Printed Circuit Board (PCB) Test Methodology User Guide

1. Push the Calibrate softkey Push the Calibrate plug in softkey Push the Plug in softkey if necessary to toggle the selection to the plug in that will be used for measurements 1 and 2 or 3 and 4 Push the Start cal softkey Follow the on screen instructions using the softkeys or numeric keypad as necessary Make sure a high quality 50 Q termination is used for the module calibration process e g the HP 909D APC 3 5 50 Q termination 62 User Guide Printed Circuit Board PCB Test Methodology intel 6 2 2 Hardware Setup and Initialize Figure 37 Basic instrument setup diagram HP 54750A ve NUMERIC KEYPAD H HP 8120 4977 RF DISPLAY Si CABLE TO TDR ni CHANNEL 1 Q TOR Plug in N1020A TDR PROBE TDR Channel MANIPULATOR TOR Channel 2 HP 54754A USE HP 8710 1765 TORQUE WRENCH Note All measurements must be made on a nonmetallic surface Metal tables or other metal surfaces may corrupt the TDR measurement The user must not touch the probe or trace that is being measured Procedure e Connect cable and probe to TDR module See Figure 37 e Push the Setup front panel key Select the Default setup softkey e Push the TDR TDT Setup key located on the TDR module that is being used Select the Stimulus softkey and press again until the desired TDR channel is
2. To complete differential impedance measurements two TDR sources need to be injected across two differentially routed signal lines simultaneously The TDR output edges must be phase aligned so that the output edges have zero time delay skew between them and opposite in polarity odd mode switching Edge alignment can be completed by setting the polarity the same and adjusting the head time delays until no visible difference in the time base is observed between the two After aligning the edges reverse the polarity on one source and apply both signals to the differential line pair It is critical to have proper edge alignment and probe placement so the electrical switching characteristics along the two lines will match in time If this is not confirmed distorted TDR responses will occur A simple method to check the setup and measurements is to apply one signal to measure the single line impedance which should be similar to a typical TDR impedance response Note the impedance value and next connect the second source with the inverse response Once the second source is applied it should be visible that the impedance for the single line should drop The amount of difference will be dependent upon the design of the traces The illustration for Figure 18 is an example of a 56 Q differential impedance measurement for Direct Rambus This shows both step responses of which either could be used to extract the impedance User Guide intel Printed Ci
3. intel Verification and Calibration at the Probe Tip In situations where it is required to compensate for measurement errors introduced by the microstrip probe the instrument may be calibrated directly at the probe tip as described below 5 4 5 1 Verification 1 Select Diagnostics from the Utilities menu 2 Connect the working probe cable and the microstrip probe to the MAIN SMA front panel connector 3 Press the Learn Cable Length button The CITS will determine the length of the cable and probe and display the time difference from the nominal length If the time difference exceeds 10ps press Apply Correction to adjust the system calibration 4 Attach an Airline PCB adaptor Polar part number ACC257 for 28 Ohm ACC258 for 50 Ohm to the airline 5 Press the Check button and enter the exact impedance of the reference airline 6 Press the microstrip probe into the test pad on the adaptor board Ensure that the probe pins are fully compressed 7 Press OK and the system will now display the measurement error at this impedance Ensure that the probe is not moved during this step 8 Repeat as necessary for other impedance calibration points 9 If necessary perform the Calibration described in the next section 10 Press OK to exit 5 4 5 2 Calibration Caution Caution Incorrect use of Calibration mode may invalidate the CITS500s calibration 1 Select the Diagnostics command from the Utilities menu 2 Connect the working
4. Printed Circuit Board PCB Test Methodology User Guide Revision 1 6 January 2000 Order Number 298179 001 Printed Circuit Board PCB Test Methodology Information in this document is provided in connection with Intel products No license express or implied by estoppel or otherwise to any intellectual property rights is granted by this document Except as provided in Intel s Terms and Conditions of Sale for such products Intel assumes no liability whatsoever and Intel disclaims any express or implied warranty relating to sale and or use of Intel products including liability or warranties relating to fitness for a particular purpose merchantability or infringement of any patent copyright or other intellectual property right Intel products are not intended for use in medical life saving or life sustaining applications Intel may make changes to specifications and product descriptions at any time without notice Designers must not rely on the absence or characteristics of any features or instructions marked reserved or undefined Intel reserves these for future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them The Printed Circuit Board PCB Test Methodology may contain design defects or errors known as errata which may cause the product to deviate from published specifications Current characterized errata are available on request Contact y
5. Measurement verification should be completed daily Example comparison of a 28 precision thick film resistor e Measured value day 1 gt 28 04 Ohms e Measured value day 2 gt 28 06 Ohms e Environmental drift gt 0 02 Ohms Verification measurements should be plotted on a control chart to determine if re calibration re normalization or diagnostics are appropriate 7 Example reference verification standards precision alumina substrate preferred Available from Hewlett Packard precision 28 ohm airline available from Maury Microwave User Guide 75 Printed Circuit Board PCB Test Methodology intel 6 3 1 Precision Normalization and Verification Standards Precision Alumina Thick Film Substrate preferred Substrate specifications TBD e 28 Ohms 0 14 Ohms DC to gt 2 GHz e 50 Ohms 0 14 Ohms DC to gt 2 GHz e 75 Ohms 0 30 Ohms DC to gt 2 GHz Precision 28 Ohm Airline for non normalized offset technique e Airline specifications TBD 28 Ohms 0 1 Ohms DC to gt 2 GHz 7mmor3 5 mm 15 cm min length 30 cm max length Nist Certified Precision 25 Ohm Airlines for non normalized offset technique Calibration to 25 Ohms can be completed by utilizing two commonly available 50 Ohm airlines connected in parallel as illustrated in the figure below Both airlines must be identical to each in length and type Disadvantage of the 25 ohm calibration is accuracy while the advanta
6. Test structures without dummy traces or floods will not represent the actual bus characteristic impedance and will also exhibit more etch variation thus more Zo variation in the coupon than the actual bus Example SMA Test Structure Including Flood Dummy Traces Structure No Flood Dummy trace Structure Flood Dummy traces Ca CR mmm GND Dummy Traces SMA_Test_F D_Traces Printed Circuit Board PCB Test Methodology 2 1 1 2 1 2 10 intel Impedance Test Coupon General Guidelines Slight variations in impedance coupon design will depend on the probing method used for measurement The items listed below are general guidelines for both normal and differential impedance measurements that can be adapted for different types of probing The differences and descriptions of these different probing techniques are covered in the measurement techniques section of this document e Trace geometry must replicate the design requirements Structures must include ground shielding around the test trace if they are used in the design in order to comprehend the effects of ground shielding on impedance and etch variation e Traces with ground shielding must both reference their corresponding plane either VSS or VCC e Traces should be single ended with no pad or via at the end of the line to avoid perturbations from the reflected TDR pulse e Route the traces straight without no bends Bends can induce ringing on t
7. e 50 Q 0 5 Q termination e SMA male to male cable Tektronix part 174 1120 00 Sin Tektronix part 174 1341 00 1m e Tektronix handheld probe with solid ground spanner Tektronix Probe part 206 0398 00 Tektronix solid gnd spanner part 131 4474 00 e Torque wrench 5in lb of torque SMA amp 3 5mm connector s Notes 1 Required probe and ground spanner that must be cut to match probe 2 Torque wrench should be used for all SMA amp 3 5 mm connections User Guide 31 Printed Circuit Board PCB Test Methodology 4 2 intel General Setup Prior to initialization put on the grounding strap and connect the cable and probe to the TDR channel that will be used for measurements using the torque wrench Measurements need to be completed with the board laying on nonconductive surface with microstrip traces faced up air While probing the user must not touch the probe or trace that is being measured Figure 22 Basic instrument setup diagram D D D eS Push Buttons TDR Head Touch Screen 4 Large Knobs L basic_instr_setup Figure 23 Illustration of probe and properly cut ground spanner to match the probe dimensions 32 Probe Cut To match probe Note Wear ground strap at all times The test environmental conditions must follow the Tektronix equipment specified environmental operating conditions User Guide intel 4 2
8. 1 Printed Circuit Board PCB Test Methodology Initialize Procedure Connect cable and probe to Sampling head Push Utility Button Select initialize touch Screen Select initialize under Verify touch screen Push Trigger Button Select Source touch screen Select Internal Clock touch screen Push Select Channel button on TDR Head Yellow light should begin flashing to indicate head is on and flat line should appear on the screen Push Waveform Button Select Head Fcn s touch screen Select channel touch screen Select TDR TDR to ON touch screen Select Smoothing to ON touch screen Select Graticules touch screen Select rho touch screen Select Acquire Desc touch screen Select Avg to ON touch screen Select Set AVGN touch screen Use upper Knob to set avgn 8 Push Auto Set Button Screen should look similar to Figure 24 Figure 24 TDR Screen Step Response After Initialization User Guide 33 Printed Circuit Board PCB Test Methodology intel 4 2 2 Instrument Calibration For high volume manufacturing certification and testing the instrument must follow the long term stability procedures This is necessary to determine instrument control charts Note The TDR sampling head must be on for a minimum of 30 minutes prior to any calibration and measurements Procedure e Calibration should be completed only after a control c
9. 2 Er DT NEEN 42 5 3 Ene Length AA 42 5 4 Section 1 CITS500s 32 bit eee neee 43 5 4 1 Horizontal Standardisation ii 43 54 2 clestliles i fol ilo ita iaia 43 A Tested Areal EEN 43 9 43 POSTING ion anna a ai alare li ai 44 5 4 3 1 Datalogging and SPC Analysis ii 44 5 4 3 2 Guidelines when using the CITGb ts 45 5 4 4 Verification at 28 Ohm Impedance i 47 5 4 4 1 Verification at 28 ORMS eee eee 48 5 4 4 2 Calibration at 28 Oms 48 5 4 5 Verification and Calibration at the Probe mp 50 54 E En Wed EE 50 5 4 5 2 ee e TT 50 5 5 Section 2 CITS5DOOS 16 bit 51 5 5 1 Horizontal Standardisation ii 51 5 5 2 Test File ege feel 53 5 5 2 1 TESTED Ane EE 53 E E TOSUN EE 54 5 5 3 1 Use of Macro Test Files and Datalogooimg 54 5 5 3 2 Guidelines when using the CTTSROU US see eee eee 55 5 5 4 Verification at 28 Ohm Impedance eee eee eee 57 5 5 4 1 Verification Procedure sese esse ener ener ennenen 58 5 5 4 2 Alternative Meibod A 59 6 Appendix C Hewlett Packard Direct Rambus Impedance Measurement Procedure 61 6 1 EQUIP MO ci a 61 6 2 General Setup ino a tina einen 61 6 2 1 Instrument Plug in Module Calibration ii 62 6 2 2 Hardware Setup and Initialize eee eee eee eee 63 6 2 3 Display Adjustment toco it tddi 64 6 2 3 1 Measurement Overvi
10. 7 T T T r 7 r r 1 0 00E 00 2 00E 10 4 00E 10 6 00E 10 8 00E 10 1 00E 0 1 20E 09 1 40E 09 1 60E 09 1 80E 09 2 00E 09 Note Difference 0 15 Time cur_pos_TDR_response Averaging Mean Method Preferred The averaging mean method is similar to the previous technique except that a measurement of a pre defined region along the line is used to determine the mean impedance value in that region This is accomplished by setting the instrument acquisition in averaging mode n gt 8 Documenting the correct scaling vert horiz test structure and pre defined mean region will provide the best means for correlation between different equipment and users The averaging mean method eliminates error due to user dependency and response ringing This is particularly important for instances where excessive ringing occurs on the TDR response and the difference between peaks and valleys are a large percentage of the specification window Taking the mean and selecting the proper region in which to take measurements minimizes a large portion of these errors User Guide Note Printed Circuit Board PCB Test Methodology The response in Figure 16 illustrates an averaged pulse response and a good general measurement region that is used to determine the impedance for a 28Q measurement Current IPC spec is to measure between 25 85 of response This is based on previous common 50 70 2 measurements Low impedance measurements such as Direct R
11. Calibration eee eee eee eee eee 16 3 3 1 Direct RambusStandard Example eee eee eee eee 17 3 3 2 Airline 28 Ohm Preferred ui 17 3 9 9 O nen iena i a 17 3 4 Probing TOCNMIQUOS ia aa T 18 3 44 Handheld Probs como aio 19 34 2 SMA CONNEC OTS eseu bilia 20 3 4 3 Controlled Impedance Microprobes i 21 3 5 Impedance Measurement Techniques esse esse ee sese esse essere nenen neee nee 23 3 5 1 Display eT Un ra 711 iii 23 3 5 2 Cursor Positioning Method 24 3 5 3 Averaging Mean Method Preferred sss 24 3 5 4 Filter Options EE 25 3 5 5 Differential Impedance i 26 3 6 Velocity Measurement Techniques sese sees essere sees neee nenen 27 e Hp TDR Method PTT 27 3 6 2 TDT Methods idi a AiO ARRARAS 28 Appendix A Intel Direct RambusImpedance Measurement Procedure for Tektronix 31 4 1 Equipment RequirO EE 31 4 2 ce o 32 4 21 MAMAN TEE 33 4 2 2 Instrument Calibration eee eee eee eee eee eee eee 34 4 3 Display TEE ul 34 4 3 1 Measurement ea 36 4 4 TDR Direct Rambus 28 Ohm Impedance Calibration eee eee eee eee eee eee 37 4 4 1 Airline 28 Ohm Preferred sss sss 39 4 4 2 Alrline 25 Ohm iaia 39 3 Printed Circuit Board PCB Test Methodology 5 Appendix B CITS500s Procedure for Rambus Impedance Measurement 41 5 1 Equipment ica 41 5
12. Document This is the second issue of a test procedure for Rambus boards using the Polar Instruments CITS500s Controlled Impedance Test System Familiarity with the operation of the CITS500s is assumed in this procedure For further information refer to the CITS500s Operator Manual Procedures are included for both CITS500s 32 bit software and CITS500s 16 bit software Please direct any suggestions or comments on the procedure to Polar Instruments Ltd 5 1 Equipment Polar Instruments CITS500s Controlled Impedance Test System Polar Instruments CITS500s 32 bit software Polar Instruments IP 28 28 Ohm Microstrip Probe Probe cable Polar part no WMA238 pair Torque wrench 5 in lb 0 312 Hex Personal Computer IBM AT or compatible 28 Ohm Airline NIST NPL certified for verification 30cm long 7mm diameter with APC7 connectors APC7 to APC3 5 female SMA adaptor Airline PCB adaptor4 Polar part no ACC257 28 Ohm ACC258 50 Ohm Notes 1 Alternative is CITS500s 16 bit software 2 IP 28 is recommended for testing of short traces Alternative is IP 50 50 Ohm microstrip probe 3 Alternative is to use two 50 Ohm airlines connected in parallel with an SMA T connector 4 Optional only required for calibration at probe tip User Guide 41 Printed Circuit Board PCB Test Methodology intel 5 2 Connection The Polar Instruments CITS500s is connected as shown below For single ended measurement
13. Typical Test File Editor display using IP 28 sese eee 53 Figure 32 Typical Test File Editor display using IP 54 Figure 33 Open circuit display USING IP 28 55 Figure 34 Typical measurement USING IP 28 eee eee 56 Figure 35 Open circuit display USING 1P 50 seene 56 Figure 36 Typical measurement USING IP 50 57 Figure 37 Basic instrument setup diagram sees eee eee eee eee 63 Figure 38 TDR Step Response after Initialization i 64 Figure 39 Display showing two rising edges or StepS 65 Figure 40 Horizontal position coarse adiusiment sees eee eee eee eee 65 Figure 41 Horizontal position fine adjUustment i 66 Figure 42 Final horizontal adjustments completed eee 67 Figure 43 Probing locations 68 Figure 44 Example short zero Q measurement eee i 69 Figure 45 Example load 50 Q measurement eee eee eee eee eee 69 Figure 46 Measurement of 28 Q thick film 28 ohm verification resistor eee eee 71 Figure 47 Reflected voltage from 28 Q standard i 72 Figure 48 28 ohm coupon measurement essere sese esse esse e eee e neee 73 Figure 49 Coupon measurement for offset technique i 74 Figure 50 Low impedance verification probing onto a standard 75 User Guide 5 Printed Circuit Board PCB Test Methodology This page is intentionally
14. left blank 6 User Guide 1 1 User Guide Printed Circuit Board PCB Test Methodology Introduction The primary focus of this document is to detail the measurement procedures and techniques necessary to accurately characterize Printed Circuit Board PCB trace impedance and propagation velocity By using the methodologies described herein PCB suppliers and customers should expect to have significantly better correlation on the key PCB electrical parameters that affect high speed digital bus designs Overview With existing PCB technology the evolution of system bus design speed requires critical attention to design trade offs for delivering high performance These needs can be met by delivering strict PCB layout guidelines and rules that ensure requirements while utilizing low cost FR4 dielectric PCBs The combination of proper bus trace geometry and PCB test methodology are essential in understanding what existing PCB technology can deliver As PCB tolerances decrease both proper coupon design and measurement procedures are essential for obtaining accurate results This document presents recommendations for implementing test coupons and transmission line measurements to achieve PCBs with 10 impedance variation Printed Circuit Board PCB Test Methodology This page is intentionally left blank 8 User Guide 2 1 Note Figure 1 User Guide Printed Circuit Board PCB Test Methodology Test Structures Test str
15. near the bottom of the display Figure 48 28 ohm coupon measurement User Guide 73 Printed Circuit Board PCB Test Methodology intel 6 2 4 2 Non normalized offset alternate Measurements If using the 28 ohm airline offset technique the offset must be applied to the measurement Also the measurements will need to be converted from reflected voltage to impedance e Press the Run front panel key e Probe the coupon under test Hold the probe carefully onto the pads during the measurement period or use the HP N1020A e Press the Clear Display front panel key Wait for the histogram database to finish collecting data After 8 traces averages the histogram statistics will display information about the measurement e Press the Stop front panel key The mean reflected voltage is indicated near the bottom of the display See Figure 49 for an example Figure 49 Coupon measurement for offset technique e Add the offset From the example above the offset calculated was 5 6612 mV This must be added to the mean reflected voltage indicated in this case 147 7430 mV 5 6612 mV 142 0818 mV e Calculate impedance D AROSA 1 0 2896 and 200mV y 50 02899 _ 77 54400 1 0 2896 74 User Guide intel Printed Circuit Board PCB Test Methodology 6 3 TDR Rambus Calibration Re Verification and Drift Impedance measurements significantly different than 5
16. standard This is especially useful when measuring air line standards when probing the connector can be difficult for periodic checking The options outlined in the next section use a known good reference low impedance standard 2connected at the end of the cable to be used for measurements to calibrate against as illustrated in Figure 4 Figure 4 Low impedance calibration error offset using standard Standard Instrument Cable Probe Total Error Limp_CalErr_Offset 2 Airline Standard Available from Maury Microwave User Guide Note 3 3 2 3 3 3 User Guide Printed Circuit Board PCB Test Methodology Direct RambusStandard Example Calibration against a standard is completed to determine the offset between instrument measured and actual PCB impedance Complete this by following the average mean method described in the measurement section of this document After determining the difference between the known standard value and the instrument reading use this value as a fixed offset to obtain the actual PCB impedance Calibration against standard should be completed daily Example e Calibrated standard 25 Q 0 1 e Measured value 27 0 e Offset value 25 27 2 0Q Therefore the measured values in this example will need to subtract 2 Q to obtain the actual real PCB impedance For this example a measured value of 33 Q will correspond to an actual 31 Q impedance Again for this example th
17. the connector can be difficult for periodic checking 37 Printed Circuit Board PCB Test Methodology intel The options outlined in the next section use a known good reference low impedance standard 3connected at the end of the cable to be used for measurements to calibrate against as illustrated in Figure 29 Figure 29 Low Impedance Calibration Offset using Standard Standard Instrument Cable Probe Total Error Llmp_CalErr_Offset Calibration against a standard is completed to determine offset between instrument measured and actual Complete this by following the previous measurement procedure described in the prior measurement section and obtain a reading Determine the difference between the known standard value and the reading and use that value as a fixed offset to obtain actual impedance Note Calibration against standard should be completed daily Example e Calibrated standard 25 Q 0 1 e Measured value 270 e Offset value 25 27 2Q Therefore measured values will need to subtract 2 2 to obtain the actual impedance For this example a measured value of 33Q will correspond to a 31 Q actual Measurement spec window for Rambus will be 27 2 Q to 32 8 Q in order to meet a 25 242 to 30 8 Q actual Different options available for Rambus calibration will vary on accuracy and setup time Pricing and availability are the major differences regarding each 3 Airline Standard Available from M
18. the only method that can be used to extract small PCB variations but is not well suited for high volume manufacturing Figure 12 and Figure 13 illustrate the TDR pulse response comparison between SMA and microprobes for the same test structure The microprobe curve exhibits significantly less ringing of the pulse and 21 Printed Circuit Figure 12 Board PCB Test Methodology intel much improved edge response at the open circuited end of the test coupon It can be seen that measurement points along the SMA curve deviate from the microprobe curve These are points at which one could take measurement data and attribute it to PCB process variation when in reality it is due to errors associated with the measurement technique Comparison of Microprobe and SMA TDR Responses TDR comparison of SMA vs Cascade microprobe connector Micro strip structures identical except for SMA connector or Cascade probe 0 2 0 1 Cascade probe structure 11 Improved ringing response with 0 2 Cascade microprobe Lab Dela 0 3 SMA Connector structure 1 of SMA 0 5 5 00E 10 7 00E 10 9 00E 10 1 10E 09 1 30E 09 1 50E 09 1 70E 09 1 90E 09 2 10E 09 2 30E 09 Time seconds Comp_MProbe SMATDR Figure 13 Comparison of Microprobe and SMA Edge Rate Degradation TDR comparison of SMA vs Cascade microprobe connector Micro strip structures identical except for SMA connector or Cascade probe Waveforms shif
19. 0 Ohms are subject to errors caused by connection repeatability and environment drift The normalization process removes the other systematic errors such as cable and probing frequency response errors losses and system variability A simple means to verify the normalization and ensure negligible environmental drift has occurred is to use a known reference standard close to the characteristic impedance of the line under test For most instances the error due to probing will be small typically less than 0 1 Q when using normalization The following options outlined in the next section use a known reference standard at the end of the probe to be used for measurements to verify as illustrated in Figure 50 Figure 50 Low impedance verification probing onto a standard Note NUMERIC KEYPAD DISPLAY SOFTKEYS TDR Plug in Instrument Cable Probe Total Error NOTES The errors within the red circles are characterized and removed from the measurements during the normalization process Verification against a standard is completed to determine whether the environmental factors in the measurement setup have changed and provides an intuitive confirmation that the normalization is valid over time Probe onto the standard and measure the impedance Calculate the difference between readings taken at different time intervals to determine the environmental drift
20. 8905 ns Reference left center Time base 1 000 ns div 49 8905 ns windowing 1 100 mU div 200 mu User Guide 65 Printed Circuit Board PCB Test Methodology intel Select the Scale key and adjust the time div to 250 ps div Select the Position softkey Adjust the knob so the rising edge is approximately one division from the left edge of the display The display should be similar to Figure 41 Figure 41 Horizontal position fine adjustment e Probe the coupon and re adjust the horizontal scale and position using the time base keys if necessary until the launch edge is aligned near the first graticule and the reflected edge is aligned near the ninth graticule Horizontal scaling will be dependent on coupon length A standard coupon should require the time base scale to be set to about 200 to 300 ps div The display should look similar to Figure 42 e Push the module setup button on the TDR module Select the offset softkey Set the offset to 143 6mv using the numeric keypad Select the scale softkey Set the offset to 35mv per division using the numeric keypad Note Do not change the vertical scale during the test process 66 User Guide intel Printed Circuit Board PCB Test Methodology Figure 42 Final horizontal adjustments completed 6 2 3 1 User Guide Measurement Overview There are two methods for making measurements on co
21. Press TDR TDT Setup on the plug in module Press the Normalize Response softkey Press the Establish normalization and reference Plane softkey Follow the directions on the upper left corner of scope display Carefully probe the Short and then the 50 Q thick film resistor on the calibration substrate when requested by the instrument Hold the probe carefully onto the pads during the measurement period or use the HP N1020A The small pad attached to the 50 Q standard is for the center or signal contact while the large pad is for the outer or ground contact The accuracy of all measurements is dependent upon these calibration devices See Figure 43 for the correct probing locations on the calibration substrate See Figure 44 for an example short and Figure 45 for an example 50 Q measurement display during normalization Figure 43 Probing locations kick Hte Callier Y erint Rewtalor rehe Tiga Lenpnnd or Signal Pad User Guide intel Figure 44 Example short zero 2 measurement Figure 45 Example load 50 Q measurement User Guide Printed Circuit Board PCB Test Methodology 1 225 0 ps div 100 mU div 200 mu Response 1 Risetime 780 ps TDR TDT TOR II TOR normalize off on Establish normalization 5 ref plane 49 5321 ns 1 225 0 ps div 100 mU divw 200 mu Response 1 Risetime 780 ps TDR TDT TOR A TDR normalize
22. acting the reflected delay differences between the two structures The purpose of two test structures is to null out determining the point where the pulse enters the test structure 27 Printed Circuit Board PCB Test Methodology 3 6 2 28 intel To make this measurement maximize the TDR time base and position the cursors at the point where the reflected pulse begins to rise as illustrated in Figure 19 The AT value as shown will be twice the actual delay difference due to reflection delay down and back The delay per unit length is calculated by Delay unit length AT 2 length difference Figure 19 Example of a Velocity Measurement Using the TDR Method TDR of 3 and 6 Coupon Microprobe of 28 Ohm traces 80 70 AT 60 50 4 Short Long Ohms 40 30 20 10 r T T T 0 00E 00 5 00E 10 1 00E 09 1 50E 09 2 00E 09 Time seconds TDT Method The best accuracy for measuring velocity characteristics with a TDR is by using the TDR in TDT mode The TDT is completed by launching the pulse on one end of the test coupon with a 50 Q probe and capturing the signal at both the launch point and open end with a low capacitance high impedance probe as illustrated in Figure 20 The advantage of the TDT over the TDR is that the captured signal has propagated only once down the coupon yielding an improved rise time response This improves the guess work
23. ad Typically ground connections have square pads to distinguish them from round signal pads 4 The CITS500s displays the trace s impedance against the test program limits If the waveform remains between the Test Limits over the whole of the tested region the CITS500s records a PASS 5 Repeat for all the test traces on the coupon Datalogging and SPC Analysis If required test results may be datalogged for analysis purposes The optional CITS Datalog Report Generator DRG may be used to analyse the data Alternatively the data may be accessed by third party SPC programs Refer to the CITS500s Operator Manual for further details User Guide Printed Circuit Board PCB Test Methodology 5 4 3 2 Guidelines when using the CITS500s Wear the Wrist Strap at all times when using the CITS500s Use the Torque Wrench to connect the cable and probe or airlines to the CITS500s Never touch the tip of the Microstrip Probe When testing a PCB do not touch the trace being measured Caution The CITS500s is an extremely sensitive measuring instrument To prevent damage to the instrument observe static precautions at all times Figure 23 Open circuit display using IP 28 HRS Board cl CTSS LL Hal Fla cad Dasep Wirman Vere Limo Hee p uj ART e Deoecaphon K mret GO 22 00 T oct SCHOUL trad Daraga L Svane Dee P MOTs STs 1 sod hs Finacy Tezer 10 Can irom 4 Logged teri G Loge User G
24. ambus 28 Q need to follow the illustration below to yield accurate results Figure 16 Example of Averaging Mean Method for a 28 Q Coupon 3 5 4 User Guide TDR Response 30 Ohm 6 coupon 65 Window 55 50 70 Window 45 Coupon Impedance ohms Measurement Region Settling Region 2 5E 10 7 5E 10 1 25E 09 1 75E 09 2 25E 09 Time Filter Option The filtered TDR response can be useful under conditions where excessive ringing occurs on the TDR response The filtered response can be used for designs where the minimum fastest edge rate is much larger slower than the TDR edge rate Enabling the filter function eliminates unnecessary high frequency content on the TDR response This will provide a smoother TDR response making it easier to determine the impedance characteristics of the trace under test Figure 17 is an example of a filtered TDR response in comparison with the normal unfiltered TDR response 25 Printed Circuit Board PCB Test Methodology Figure 17 Comparison on Unfiltered vs Filtered 200 ps Response 3 5 5 26 TDR 28 Ohm filtered 200ps vs Unfiltered Response 55 45 C E lt e g L L Filter_200p S Normal o o e 35 A Seli n i r e a O ne 25 4 2 50E 10 7 50E 10 1 25E 09 1 75E 09 2 25E 09 Time Differential Impedance
25. arisons with any non normalized measurements the X1 position should be set to the 5 horizontal graticule See Figure 47 for an example Select the X2 position softkey Use the knob to adjust the X2 position to the 7 5 horizontal graticule See Figure 46 for an example If using a lower bandwidth probe or making comparisons with any non normalized measurements the X2 position should be set to the 7 horizontal graticule See Figure 47 for an example Select the Y1 position softkey Use the numeric keypad to set the Y1 position to 0 ohms Select the Y2 position softkey Use the numeric keypad to set the Y2 position to 100 ohms Probe one of the 28 Q precision calibration resistors on the calibration substrate Hold the probe carefully onto the pads during the measurement period or use the HP N1020A Press the Clear Display front panel key This resets the histogram database and begins collecting data for the current measurement Wait for the histogram database to finish collecting data After 4 traces averages the histogram statistics will display information about the measurement Press the Stop front panel key The histogram s displayed mean value should be 28 ohms 0 25 ohms See Figure 46 This completes the setup and verifies the normalization User Guide intel Printed Circuit Board PCB Test Methodology Figure 46 Measurement of 28 2 thick film 28 ohm verification res
26. aury Microwave 38 User Guide l Printed Circuit Board PCB Test Methodology intel 4 4 1 Airline 28 Ohm Preferred Calibration to a 28 Q open ended airline to determine offset e Disconnect probe from cable connect airline to cable and measure e Airline specifications 28 Q 0 1 DC to gt 2 GHz 7 mmor 3 5 mm 15 cm min length 30 cm max length NIST Certified 4 4 2 Airline 25 Ohm Calibration to 25 2 can be completed by utilizing two commonly available 50 open ended airlines connected in parallel as illustrated in Figure 30 Both airlines must be identical in electrical delay and model type The disadvantage of the 25 calibration is slightly reduced accuracy while having significant advantages with cost and availability e Disconnect probe from cable connect airlines and measure e Airline specifications 502 0 1 DC to gt 2GHz 7mmor3 5 mm SMA 50 Q MACOM 2041 6204 00 15 cm min length 30 cm max length NIST Certified Figure 30 25 Q Calibration with Two 50 Airlines Airline 50 Ohm SMA 50 Ohm Measurement STO Cable Airline 50 Ohm Note ALL 7 mm airlines will need a 3 5 mm to 7 mm adapter User Guide 39 Printed Circuit Board PCB Test Methodology This page is intentionally left blank 40 User Guide Printed Circuit Board PCB Test Methodology 5 Appendix B CITS500s Procedure for Rambus Impedance Measurement About This
27. autions at all times Figure 33 Open circuit display using IP 28 RAMBUS TST Polas CITS500s Ee each Dese pions Mamo Help Nei e utca 3 min USE 0 28 28 OHM MICROS TRIP PROBE User Guide 55 Printed Circuit Board PCB Test Methodology 56 Figure 34 Figure 35 Typical measurement using IP 28 gt RAMBUS TST Polar IT Site Ele Wvelom isting QDintions Macro Help Aug 2412 SD 010 Mn 2836 Maw 295 1 d d 4 5 E r a Sed Posch futocal 1 em USE IP 29 26 DM HICAOSTAIF PROBE Open circuit display using IP 50 RAMBUS 151 Pola 0115505 Ele Weeelom isting Dions Macro Help aa HU e ma Avg D 1 2 2 4 5 Li d Rasad dusocal 19 man USE 1P 50 GO 0884 MICROSTRIP PROBE User Guide intel Printed Circuit Board PCB Test Methodology Figure 36 Typical measurement using IP 50 5 5 4 User Guide Y RAMBUS TST Polar CITS 5005 Eis daci Datalog Dplons Macs Help Beg s RI SOs 012 Mr 2347 Mara 23 50 Aboca 18 min USE IP 50 50 OHM MICROS TRIP PROBE Verification at 28 Ohm Impedance When making very accurate measurements using the CITS500s the effects of the connecting cable must be considered Losses in the cable will result in the system measuring a higher impedance value than actual This effect is especially significant when measuring low values of impedance Use of a traceable standard airli
28. d in Figure 21 Store both the transmitted and received signals and measure the difference between curve and curve 2 to get the transmitted delay Figure 20 Example of a Basic TDT Setup Figure 21 User Guide High Impedance probes gt sample points 1 amp 2 D Launch Open Example Delay Measurement using TDT Approach TDT coupon launch and open 0 8 0 6 gt Launch Open Voltage o A a H N TO 0 2 0 2 0 0E 00 2 0E 10 4 0E 10 6 0E 10 8 0E 10 1 0E 09 1 2E 09 1 4E 09 Time The response from Figure 21 shows that the TDT has an improved edge rate over the TDR reducing the error associated with setting the delay points This may not be evident with a quick look comparing Figure 21 vs Figure 19 however by looking at the time base between the two for the region of reflection for Figure 19 in comparison to Figure 21 the uncertainty is reduced dramatically 29 Printed Circuit Board PCB Test Methodology This page is intentionally left blank 30 User Guide Printed Circuit Board PCB Test Methodology intel 4 Appendix A Intel Direct Rambusimpedance Measurement Procedure for Tektronix 4 1 Equipment Required e Tektronix 11801 A B C e Tektronix SD24 Sampling Head TDR Head
29. e apparent measurement spec window for Direct Rambuswould be 27 2 Q to 32 8 Q in order to meet the actual 25 2 Q to 30 8 Q requirement Airline 28 Ohm Preferred Calibration against a 28 open ended airline provides the most accurate calibration This will require a custom made airline The primary disadvantages of air lines are availability and cost e Disconnect probe from cable connect airline and measure e Airline specifications 28 Q 0 1DC to gt 2 GHz 7mmor3 5 mm 15 cm min length 30 cm max length NIST Certified Airline 25 Ohm Calibration to 25 2 can be completed by utilizing two commonly available 50 open ended airlines connected in parallel as illustrated in Figure 5 Both airlines must be identical in electrical delay and model type The disadvantage of the 25 calibration is slightly reduced accuracy while having significant advantages with cost and availability e Disconnect probe from cable connect airlines and measure e Airline specifications 50 0 1 DC to gt 2GHz 7mmor3 5 mm SMA T 50 Q MACOM 2041 6204 00 15cmmin length 30cm max length NIST Certified 17 Printed Circuit Board PCB Test Methodology Figure 5 25 Q calibration with two 50 Q airlines 3 4 Airline 50 Ohm 3 5 mm 50 Ohm Measurement T Cable Airline 50 Ohm Probing Techniques The most commonly used probe techniques in use today are handheld SMA and
30. e following test file used 25 OHM AIRLINE SES TWO 50 OHM AIRLINES The procedure is then identical to that for the 28 Ohm airline 59 Printed Circuit Board PCB Test Methodology This page is intentionally left blank 60 User Guide Printed Circuit Board PCB Test Methodology intel 6 Appendix C Hewlett Packard Direct Rambus Impedance Measurement Procedure 6 1 Equipment HP 54750A Digitizing Oscilloscope HP 5475449 Differential TDR Module HP 909D 50 Ohm Termination HP 54121 68701 RF Accessory Kit HP N1020A TDR Probe HP 8710 1582 Torque Wrench HP Precision Calibration Verification Substrate P N TBD The torque wrench should be used for all SMA amp 3 5 mm connections 6 2 General Setup Warning The HP 54754A and HP 54753A are very static sensitive Always wear static protection 4 Or HP 83480A with the HP 54755A option 5 Or HP 54753A single TDR module User Guide 61 Printed Circuit Board PCB Test Methodology intel 6 2 1 Instrument Plug in Module Calibration Instrument Plug in module calibration should be performed after each power on cycle or after any control chart violations during statistical process monitoring Note The HP 54750A must be on for a minimum of 30 minutes prior to any calibration and measurements Procedure e Put on anti static protection e Disconnect any cables and probes from the instrument e Push the Utility front panel key
31. electing the proper ground spanner is critical to minimizing the discontinuity due to the probe ground loop Minimizing the ground loop reduces the inductance spike high impedance as the pulse launches into the load under test The benefits of minimizing the ground loop are improved settling time response smoothness and probe loss Figure 7 Handheld Ground Mechanisms Illustrating Ground Loop Differences Large Ground Small Loop Ground Loop Signal Signal PCB PCB Figure 8 Handheld Probe short vs long ground loop measuring the same test coupon User Guide Handheld Long Short ground return loop 75 65 _ Settling Time ps 5 Difference 1 Ohm B short loop 5 Long loop K o 45 35 25 I i I K 0 5E 10 0 000000001 1 5E 09 0 000000002 Time Handheld Probe_LoopM 19 Printed Circuit Board PCB Test Methodology 3 4 2 20 intel The illustration in Figure 8 shows the difference between two different ground spanners for the same probe The difference between ground mechanisms were determined by using two different ground spanners provided with the Tektronix P6150 kit that are very similar to the drawings in Figure 7 Figure 8 shows how ground loop differences affect settling time response smoothness and the impedance measured Figure 9 SMA Probe Example SMA Connectors SMA connectors solde
32. est accuracy for calibration Air dielectric loads can be used to obtain the reference voltage used to calculate load impedances Air dielectrics should be used to obtain the highest accuracy gt 0 4 Q User Guide 15 Printed Circuit Board PCB Test Methodology 3 3 intel TDR Direct Rambus 28 2 Impedance Calibration Impedance measurements significantly different than 50 Q can result in large errors between measured and actual This systematic error is very common when completing Direct Rambus28 2 measurements where measured values can easily be off by 2 3 Q from the actual impedance The primary sources of error include but are not limited to instrument bias probes and cable loss A simple means to account for these affects is to use a known reference standard close to the characteristic impedance of the line under test Reference standards are used to determine the systematic offset between measured and actual PCB impedance Measuring the standard with the same probe and cable that will be used for measurements connected to the instrument provides the additive effective error of the system to the probe tip When the probe used is a known good probe error due to probing will be small 0 2 Q in most instances To verify the probe effects compare the TDR response between the probed reference vs direct cable connection If it is deemed that probe effects are negligible measurements can be completed by direct cable connection to the
33. ew ee ee ee i 67 6 2 3 2 Normalization Recommended Procedure 68 6 2 3 3 Offset alternate Procedure sese 71 6 2 4 Coupon Measurement ii 73 6 2 4 1 Normalized preferred Measurements eee eee eee eee eee 73 6 2 4 2 Non normalized offset alternate Measurement 74 6 3 TDR Rambus Calibration Re Verification and Dt 75 6 3 1 Precision Normalization and Verification Standards sese eee eee 76 4 User Guide Printed Circuit Board PCB Test Methodology intel Figures Figure 1 Example SMA Test Structure Including Flood Dummy Traces sese esse eee eee ee 9 Figure 2 Sample Impedance and Velocity Test Structures for 28 Q Microstrip 11 Figure 3 Sample Impedance and Velocity Test Structure for 28 Q Stripline Design 14 Figure 4 Low impedance calibration error offset using standard i 16 Figure 5 25 Q calibration with two 50 Q airlines eee eee eee eee 18 Figure 6 Example Handheld Probe with Ground Spanner i 18 Figure 7 Handheld Ground Mechanisms Illustrating Ground Loop Differences 19 Figure 8 Handheld Probe short vs long ground loop measuring the same test coupon 19 Figure 9 SMA Probe Example gola aa 20 Figure 10 Comparison of Handheld and SMA Probing TechniQues eee ee ee ee 21 Figure 11 Microprobe Example ii 21 Fig
34. ge will be cost and availability e Disconnect probe from cable connect airlines and measure e Airline specifications 50 Ohms 0 1 DC to gt 2GHz 7mmor3 5 mm 15 cm min length 30 cm max length Nist Certified Airline 50 Ohm 3 5 mm 50 Ohm Measurement V Cable Airline 50 Ohm 76 User Guide
35. hart violation e Disconnect cable and probe from instrument e Push Utility button Select Page to Enhance Accuracy touch screen Select Offset touch screen Select Auto Cal touch screen Terminate TDR head with 50 Ohm load Select Proceed touch screen Select Store Const touch screen e Select TDR Amplitude touch screen Select Auto Cal touch screen Terminate TDR head 50 Ohm load Select Proceed touch screen Select Store Const touch screen 4 3 Display Adjustment Procedure e Reconnect cable and probe The screen should look similar to the left side box of Figure 25 Select the horizontal positioning arrow lt gt located at the top left of the touch screen Use both upper and lower knobs located on the front panel to adjust time per division and horizontal position so the displayed response is similar to the right side box of Figure 25 Set time div to Ins div Note selecting the Main Size touch screen a menu will appear to select coarse medium fine step sizes for both horizontal and vertical adjustments This is helpful to fine tune the scaling to meet the figures shown 34 User Guide Printed Circuit Board PCB Test Methodology intel Figure 25 Horizontal Positioning horiz_posit Select the vertical scaling positioning arrows located at the middle left side of the touch screen Use both the upper and lower knobs located o
36. he TDR pulse reducing measurement accuracy e Required coupon line length is dependent upon the type of probing and equipment used for measurements Handheld probe with ground spanner minimum of 3 inches long Tektronix Hewlett Packard minimum of 6 inches long Polar with IP50 probe minimum of 4 inches long Polar with 1P28 probe Fixed pitch controlled impedance microprobes minimum of 1 inch Tektronix Hewlett Packard e Microstrip structures must provide signal and reference plane pads e Stripline structures must provide a signal pad and a pad for each reference plane e Pads for handheld probes should be a minimum of 0 025 inches in diameter with 100 to 150 mils 2 54 to 3 81mm spacing between signal and ground e Microprobe pad dimensions are shown in the following section Propagation Velocity Test Coupon General Guidelines Measurement of velocity or propagation delay is in general more difficult than measuring impedance For velocity the structure delay is determined by measuring the difference in time it takes the pulse to propagate through the structure Measurement points for propagation delay are not as simple as for impedance and accuracy is extremely dependent upon the probing technique The most accurate delay measurements require advanced probing techniques utilizing controlled impedance microprobes with the TDR in Time Domain Transmission TDT mode This improved accuracy comes at a cost in terms
37. hi lighted for example 1 only or 2 only Press the Enter softkey Select the Preset TDR TDT softkey e Push the Acquisition front panel key Select the Number of averages softkey Set the number of averages to 4 using the numeric keypad The display should look similar to Figure 38 User Guide 63 Printed Circuit Board PCB Test Methodology Figure 38 TDR Step Response after Initialization 6 2 3 Display Adjustment Procedure e Push the Time base front panel button Select the Scale softkey Press the down arrow key to increase the time base scale until two rising edges can be seen on the display similar to Figure 39 The first rising edge or step represents the internal step generator of the TDR system The second rising edge represents the end of the probe or fixture 64 User Guide Printed Circuit Board PCB Test Methodology intel Figure 39 Display showing two rising edges or steps Time base Units bit period Scale 1 000 ns div Position 40 9705 ns Reference left center Time base 1 000 ns div 40 9705 ns windowing 1 100 mU div 200 mu Select the Position softkey Adjust the knob so the second step representing the end of the probe is near the left edge of the display See Figure 40 Figure 40 Horizontal position coarse adjustment Time base 1 000 nsi Position 49
38. ical length of the microstrip probe and its cable Usually a slightly smaller value is used so that poor quality probe to test trace connections can be viewed in the display screen The Test From and Test To limits define the tested area of the coupon trace see below This is usually the flattest part of the waveform and is referred to as the undisturbed interval Typical sets of parameters are shown on the following pages These may need to be adjusted depending on the type of cable and probe being used and the length of the coupon trace under test Tested Area The tested area is usually the flattest portion of the coupon waveform and is usually referred to as the undisturbed interval Ignore test connection aberrations and open circuit termination effects Select the Test From and Test To limits to test the undisturbed interval 43 Printed Circuit Board PCB Test Methodology Figure 22 Typical Test File Editor display 5 4 3 5 4 3 1 44 Testing 1 Allow the CITS500s to warm up for 60 minutes before making measurements 2 Load the test file created in the previous section 3 Connect the Microstrip probe across the test impedance trace and press the foot switch or click the Test button Take care not to touch the trace on the coupon during testing Note Ensure the Microstrip probe signal pin is connected to the test trace signal pad and the probe ground pin to test coupon ground p
39. involved for determining the measurement positions on the response curves i e the voltage level at which the delay measurement will be taken The TDT requires a 50 launch probe and an high impedance probe to capture the transmitted signal It is recommended that the high impedance probe should be a low capacitance 10X or 20X microprobe to provide minimum loading and maximum bandwidth Complete measurements by connecting the 50 Q probe to a sampling head with the TDR TDT mode ON The setup is the same as with TDR measurements and is only used to launch the signal driver The high impedance probe should be connected to a separate channel and with the head function s TDR TDT setting mode to OFF This enables the high impedance probe to act like a normal oscilloscope probe receiver with the sole purpose to capture the launched pulse Once the instrument is setup position the high impedance probe as close as possible to the launch point from the TDR probe The TDR 50 Q probe should be launching the pulse and the captured response from the high impedance User Guide Printed Circuit Board PCB Test Methodology probe should be visible on the screen Adjust the scaling on the screen to position the high impedance response to the left side and maximize the voltage scale The time base may need to be adjusted to the minimum time scale that will allow both the transmitted and received signals to be displayed on the same screen as illustrate
40. istor 6 2 3 3 User Guide Offset alternate Procedure e Push the Acquisition front panel key Select the Number of averages softkey Set the number of averages to 8 using the numeric keypad To turn on the histogram statistics Press the shift blue key and the Histogram front panel key Shifted Display key Press the Mode softkey and activate Waveform histograms Press the Axis softkey to hi light and specify a vertical histogram Press the Histogram Window softkey Press the Default Window softkey Select the X1 position softkey Use the knob to adjust the X1 position to the 5 horizontal graticule See Figure 47 for an example Select the X2 position softkey Use the knob to adjust the X2 position to the 7 horizontal graticule See Figure 47 for an example Select the Y1 position softkey Use the numeric keypad to set the Y1 position to 0 mV Select the Y2 position softkey Use the numeric keypad to set the Y2 position to 400 mV Probe one of the 28 Q precision calibration resistors on the calibration substrate preferred or the 28 Q airline alternate Hold the probe carefully onto the pads during the measurement period or use the HP N1020A 71 Printed Circuit Board PCB Test Methodology intel e Press the Clear Display front panel key This resets the histogram database and begins collecting data f
41. le created in the previous section 3 Connect the microstrip probe across the test impedance trace and press the foot switch or click the Test button Take care not to touch the trace on the coupon during testing Note ensure the microstrip probe signal pin is connected to the test trace signal pad and the probe ground pin to test coupon ground pad Typically ground connections have square pads to distinguish them from signal pads 4 The CITS500s displays the trace s impedance against the test program limits If the waveform remains between the Test Limits over the whole of the tested region the CITS500s records a PASS 5 5 3 1 Use of Macro Test Files and Datalogging Where multiple traces are to be tested on a coupon a Macro File may be used to simplify testing Refer to the CITS500s Operator Manual for further details Tf required test results may be datalogged for analysis purposes Refer to the CITS500s Operator Manual for further details 54 User Guide Printed Circuit Board PCB Test Methodology intel 5 5 3 2 Guidelines when using the CITS500s e Wear the Wrist Strap at all times when using the CITS500s e Use the Torque Wrench to connect the cable and probe or airlines to the CITS500s e Never touch the tip of the Microstrip Probe e When testing a PCB do not touch the trace being measured Caution The CITS500s is an extremely sensitive measuring instrument To prevent damage to the instrument observe static prec
42. microprobe The primary factors that need to be considered when choosing the probe technique that will be used are accuracy and the amount of work involved to complete the measurements The difference between probes will play a critical role in accuracy correlation and repeatability The probe s purpose is to provide the medium for injecting the pulse from the TDR s reference 50 Q output to the particular load under test Anytime the probe is not exactly 50 Q an impedance discontinuity will occur between the TDR output and the load under test This discontinuity induces ringing and reflections on the TDR response which would ideally be flat Minimizing the discontinuity involves matching the probe as close as possible to 50Q through the probe to the tip The degree to which the probe meets 50 Q is a primary factor on measurement accuracy Understanding the key principle of proper probing is critical to determine the possible sources of measurement errors due to probing Figure 6 Example Handheld Probe with Ground Spanner 18 User Guide intel 3 4 1 Printed Circuit Board PCB Test Methodology Handheld Probes The easiest most commonly used probe is the handheld probe with ground spanner This probe is useful for quick process variation checks but is not the best for accurate repeatable measurements A common problem with handheld probing is measurement variations due to the ground spanner mechanism of the probe S
43. n the front panel to adjust rho div and vertical position as illustrated in Figure 26 Set rho div to 60 mrho div Figure 26 Vertical Positioning e Probe the test coupon and continue adjusting both the horizontal and vertical scaling until the launch edge is aligned with the first division and the reflected edge is aligned with the last The vertical scaling should be adjusted to 60mrho to maximize the reflection 50 on the screen Horizontal scaling will be dependent upon coupon length The screen should look similar to Figure 27 under probing conditions User Guide 35 Printed Circuit Board PCB Test Methodology Figure 27 Recommended 28 Ohm Probing Display Setup y Launch Region E Reflection Region Reflected Edge 4 3 1 Measurement Procedure e Probe coupon allow for settling time push RUN STOP to stop acquisition e Push Measurement Button Select mean touch screen Select exit touch screen Select the measure box located in the lower left of the touch screen and a menu should appear on the screen Set the left and right limits to 50 and 70 respectively using the upper and lower knobs located on the front panel or the numeric touch screen pad The rho measurement will be the mean value between the two limits and can be used to determine mean impedance This is illustrated in Figure 28 The Tektronix units for rho are displayed in mili rho mrho
44. ne certified by NIST or NPL allows the user to apply an adjustment value when making measurements to ensure precise results It is important to note that the characteristics of the cable will alter during extended use This procedure should be repeated at regular intervals 57 Printed Circuit Board PCB Test Methodology intel 5 5 4 1 Verification Procedure 1 Disconnect the Microstrip Probe from the cable and connect a 28 Ohm precision airline 2 Use the Test File below to measure the impedance of the airline 3 Compare the impedance value measured by the CITS500s Z with the certified impedance value of the airline Z2 The Adjustment Value in Ohms is k Z Z2 The True Value when making a measurement using the CITS500s is therefore Z K Ohms If the value of k is significant then the Nominal Impedance value used in the Test File should be adjusted accordingly This will bias the test limits 4 By recording the Adjustment Value it will be possible to monitor the test system for any long term drift in system characteristics This should provide early indication of change of cable characteristics due to wear as well as verfication of the stability of the CITS500s 58 User Guide intel 5 5 4 2 User Guide Printed Circuit Board PCB Test Methodology Alternative Method If a 28 Ohm airline is not available then two identical 50 Ohm airlines may be connected to the cable using an SMA T connector and th
45. ns in section 3 1 1 e Measurement structures are defined in pairs for each signal layer differing only in length The recommended minimum trace lengths listed below are based on the time required for the TDR pulse to settle The settling time will vary depending on the impedance of the trace Table 1 Trace Length Recommendations for Measuring Propagation Velocity Probe Type Short Trace Long Trace Handheld amp SMA 3 000 0 001 6 000 0 001 Microprobe 1 000 0 001 3 000 0 001 e Stripline structures using microprobes should have a via at each end to route external traces for probing The via size should be no larger than a 25 mil pad with 10 mil finished hole For examples of velocity test structures see Figure 2 and Figure 3 Figure 2 Sample Impedance and Velocity Test Structures for 28 Q Microstrip User Guide 10 mils 18 mils mil F TE X Section gt TDR N Probe T H anual Probe Type 3 amp 6 gt o 133 Microprobe Type ee e O Signal a 2 Gup GND Signal 11 Printed Circuit Board PCB Test Methodology This page is intentionally left blank 12 User Guide Printed Circuit Board PCB Test Methodology 3 Impedance and Velocity Measurement Techniques Proper measurement techniques are essential to obtain accurate PCB characterization results This section describes the techniques required to make accura
46. obing conditions when measuring impedances lower than 50 Q Figure 14 Example Screen Display When Measuring 28 Q Coupons User Guide Launch Region Reflection Region Reflected Edge scr_dsply_Meas28Coupons 23 Printed Circuit Board PCB Test Methodology 3 5 2 3 5 3 24 intel Cursor Positioning Method In the ideal case the waveform response area of interest should contain a large flat region from which to extract impedance data As previously mentioned ringing and reflections will occur on the response depending on probe type inducing spikes in the waveform as shown in Figure 15 Cursor positioning on a line that is not flat will result in measurement uncertainty due to measurement variance Selecting a position along the line after the ringing minimizes this uncertainty This will typically be towards the end of the trace where the line is open circuited This is the most common but least accurate measurement method This method provides useful quick impedance checks but is very user dependent and is not repeatable rendering it unusable for correlation studies Cursor positioning is usually a good way for lab measurements to determine ballpark impedance values for experienced TDR operators Figure 15 Example of the Cursor Positioning Method on a TDR Response TDR Response of 28 Ohm line Rho Cursor Position Series1 On line after ringing 0 05 7
47. odology Test File Use the CITS500s Test File editor to create a test file for the coupon under test Probe Length specifies the total electrical length of the Microstrip probe and its cable Usually a slightly smaller value is used so that poor quality probe to test trace connections can be viewed in the display screen The Test From and Test To limits define the tested area of the coupon trace see below This is usually the flattest part of the waveform and is referred to as the undisturbed interval Typical sets of parameters are shown on the following pages These may need to be adjusted depending on the type of cable and probe being used and the length of the coupon trace under test Tested Area The tested area is usually the flattest portion of the coupon waveform and is usually referred to as the undisturbed interval Ignore test connection aberrations and open circuit termination effects Select the Test From and Test To limits to test the undisturbed interval Figure 31 Typical Test File Editor display using IP 28 User Guide RAM BUS TST Edit SE IP 26 28 OHM MIGROS TRIP PROBE 53 Printed Circuit Board PCB Test Methodology Figure 32 Typical Test File Editor display using IP 50 RAMBEUS TST Edit SE 1850 50 OHM MICROS TRIP PROBE 5 5 3 Testing Allow the CITS500s to warm up for 60 minutes before making measurements Load the test fi
48. of equipment and measurement time Selecting the method for measuring the velocity depends on accuracy desired measurement test time and cost The simplest but least accurate method for measuring propagation delay is using TDR mode to measure delay between two identical test structures of different length Propagation delay can then be calculated by dividing the different structure delay differences by the difference in length Accuracy will be dependent on probes and structure The best case accuracy that can be expected is 8 ps in for the recommendations below If space permits the above approach can be improved by inserting a third test structure of different length than the other two To calculate the velocity graph length vs velocity with a line drawn User Guide Printed Circuit Board PCB Test Methodology connecting the three points Using a least squares method the intercept of the line with the axis gives the measurement error Improved accuracy propagation delay measurements can be completed with the TDR used in TDT mode In TDT mode probes are placed at each end of the test structure A pulse is injected into one end and captured at the other end This approach has less edge rate degradation than the simpler TDR approach resulting in improved accuracy Real results should only be completed with microprobes A proper setup can achieve within 2 ps in accuracy e Traces should follow the guidelines described for impedance coupo
49. off on Establish normalization B ref plane 49 5321 ns 69 Printed Circuit Board PCB Test Methodology 70 To turn on the normalized trace Press TDR TDT Setup on the plug in module Press the Normalize response softkey Press the TDR Normalize softkey and make sure it is set to on Press the Rise time softkey and change the normalized rise time to 200 ps using the numeric keypad This may need to be changed to 250 ps for a lower quality probe To change the display units to ohms Push the module Setup button for the channel in use Press the Alternate scale softkey Press the Units softkey until ohm is hi lighted Press the Enter softkey Press the Done softkey The main channel setup menu is displayed Select the Display softkey to turn off the raw data trace To turn on the histogram statistics Press the shift blue key and the Histogram front panel key Shifted Display key Press the Mode softkey and activate Waveform histograms Press the Axis softkey to hi light and specify a vertical histogram Press the Histogram Window softkey Press the Default Window softkey Select the XI position softkey Use the knob to adjust the X1 position to the 3 horizontal graticule See Figure 46 for an example If using a lower bandwidth probe or making comp
50. or the current measurement Wait for the histogram database to finish collecting data After 8 traces averages the histogram statistics will display information about the measurement e Press the Stop front panel key The histogram s displayed mean value should be indicated See Figure 47 Figure 47 Reflected voltage from 28 standard e Record the calculated offset The nominal voltage from a 28 Q reflection is 143 6 mV For the example displayed here the offset would be 143 6 mV 149 2112 mV 5 6112 mV All measurements of 28 ohm coupons would need to have 5 6612 mV added to their value 6 5 6112 mV corresponds to an offset of 0 0282 rho or 1 7484 Q 72 User Guide Printed Circuit Board PCB Test Methodology intel 6 2 4 Coupon Measurement Measurement involves carefully probing the coupon or other device clearing the display to reset the histogram statistics and recording the mean impedance value on the test records 6 2 4 1 Normalized preferred Measurements e Press the Run front panel key e Probe the coupon under test Hold the probe carefully onto the pads during the measurement period or use the HP N1020A e Press the Clear Display front panel key Wait for the histogram database to finish collecting data After 8 traces averages the histogram statistics will display information about the measurement e Press the Stop front panel key The mean impedance is indicated
51. our local Intel sales office or your distributor to obtain the latest specifications and before placing your product order C is a 2 wire communications bus protocol developed by Philips SMBus is a subset of the 1 C bus protocol and was developed by Intel Implementations of the PC bus protocol may require licenses from various entities including Philips Electronics N V and North American Philips Corporation Alert on LAN is a result of the Intel IBM Advanced Manageability Alliance and a trademark of IBM Copies of documents which have an ordering number and are referenced in this document or other Intel literature may be obtained from Intel Corporation www intel com or call 1 800 548 4725 Third party brands and names are the property of their respective owners Copyright O Intel Corporation 2000 2 User Guide intel Printed Circuit Board PCB Test Methodology Contents User Guide laas e VE 7 1 1 ET H Ke 9 2 1 Requirements for Matching Test Coupon to Bus Design 9 2 1 1 Impedance Test Coupon General Guidelines AA 10 2 1 2 Propagation Velocity Test Coupon General Guidelines eee eee eee 10 Impedance and Velocity Measurement TechhiQues i 12 3 1 General Equipment lla elia 13 3 2 General Galibrationi raison iaia tapa atacara 14 3 2 1 Calibration Sequence eee eee eee eee eee eee eee eee ee eee 15 3 2 2 General Calibration Verification i 15 3 3 TDR Direct Rambus 28 Q Impedance
52. probe cable and the microstrip probe to the MAIN SMA front panel connector 3 Double click on the CITS500s icon in the top left corner of the Calibration Window using the right hand mouse button to enable Calibration mode 4 Press the Learn Cable Length button The CITS will determine the length of the cable and probe and display the time difference from the nominal length If the time difference exceeds 10 ps press Apply Correction to adjust the system calibration 5 Attach an Airline PCB adaptor Polar part number ACC257 for 28 Ohm ACC258 for 50 Ohm to the airline 6 Press the Check button and enter the exact impedance of the reference airline 7 Press the microstrip probe into the test pad on the adaptor board Ensure that the probe pins are fully compressed 8 Press OK and the system will now display the measurement error at this impedance Ensure that the probe is not moved during this step If necessary press Apply Correction to adjust the system calibration 9 Repeat as necessary for other impedance calibration points 10 Press OK to exit 50 User Guide intel 5 5 5 5 1 User Guide Printed Circuit Board PCB Test Methodology Section 2 CITS500s 16 bit Horizontal Standardization The horizontal offset of the CITS500s and its cable and probe should be standardized to ensure that any CITS500s system will measure the same region of the coupon waveform and that errors are not introduced due to cable
53. r intervals User Guide 47 Printed Circuit Board PCB Test Methodology 5 4 4 1 5 4 4 2 48 peas intel Verification at 28 Ohms 1 2 3 Select Diagnostics from the Utilities menu Connect the working probe cable without the probe to the MAIN SMA front panel connector Press the Learn Cable Length button The CITS will determine the length of the cable and display the time difference from the nominal length If the time difference exceeds 10 ps press Apply Correction to adjust the system calibration Choose the impedance to be verified 28 ohms and connect the appropriate impedance airline to the probe cable Press the Check button and enter the exact impedance of the reference airline The software will now display the measurement error at this impedance Repeat as necessary for other impedance calibration points If necessary perform the Calibration described in the next section Press OK to exit Calibration at 28 Ohms Caution Incorrect use of Calibration mode may invalidate the CITS500s calibration 1 2 3 Select the Diagnostics command from the Utilities menu Connect the working probe cable without the probe to the MAIN SMA front panel connector Double click on the CITS500s icon in the top left corner of the Calibration Window using the right hand mouse button to enable Calibration mode Press the Learn Cable Length button The CITS will determine the length of the cable and display the
54. rcuit Board PCB Test Methodology Figure 18 Example of a Differential Impedance TDR Response 3 6 Note 3 6 1 User Guide Differential Impedance TDR Response 58 Ohm differential Zdiff 2 Zodd 14 Positive Pulse 0 8 0 6 o o4 29 Ohms E Zodd E gt TDR 0 2 0 Negative Pulse Y 0 2 4E 10 6E 10 8E 10 0 000000001 1 2E 09 1 4E 09 1 6E 09 1 8E 09 0 000000002 Time Velocity Measurement Techniques As mentioned previously velocity measurements are more difficult and susceptible to measurement errors Accuracy is very dependent on test structures setup procedures and probe types Velocity measurements extract the time it takes a pulse to propagate down a given test structure The difficulty to get accuracy within the picosecond domain with these measurements is determining exactly where to take the measurement on the curve rise time of the pulse The faster the rise time of the pulse the less error It is strongly recommended to use microprobes to complete any type of velocity measurement Microprobes provide the fastest rise time launch into a coupon resulting in the highest accuracy measurements TDR Method The easiest most commonly used method to extract velocity is to TDR two different open ended coupons that differ only in length Velocity is determined by subtr
55. red to the board will provide good repeatability but adds test cost due to the SMA connector and reduces throughput time due to soldering These probes are useful for correlation between vendors and customers The plots in Figure 10 illustrate the repeatability differences between SMA connectors and handheld probe techniques The SMA curves show good repeatability between instances of disconnecting and reconnecting the coax cable The handheld probe curves exhibit variation due to different ground points pressure applied on the ground spanner and the angle at which the probe is held User Guide intel Printed Circuit Board PCB Test Methodology Figure 10 Comparison of Handheld and SMA Probing Techniques Figure 11 3 4 3 User Guide SMA vs HandHeld Probing 28 Ohm 4 00E 01 3 00E 01 Handheld 2 00E 01 1 00E 01 SMA 1 SMA 2 Probe 1 Probe 2 Probe 3 Probe 4 Y 0 00E 00 1 00E 01 2 00E 01 3 00E 01 4 00E 01 0 00E 00 2 00E 10 4 00E 10 6 00E 10 8 00E 10 1 00E 09 1 20E 09 1 40E 09 1 60E 09 1 80E 09 x Comp_Hand_SMA_Probe Microprobe Example Controlled Impedance Microprobes The most accurate probing technique uses controlled impedance microprobes shown in Figure 11 for providing a full understanding of PCB characteristics This technique requires specialized costly and setup intensive equipment for obtaining measurements It is
56. s and probes having different physical and electrical lengths 1 Noa 10 11 12 Load the file HORZ500s tst shipped with the CITS500s software A green reference waveform is displayed Optional Turn on the display graticule by selecting Options Screen Colors Click on Graticule and select the color required Click OK then OK again to close the dialog boxes Connect the instrument cable to the CITS500s Ensure nothing is connected to the cable Press the foot switch or click the Test button Use the cursors Options menu to measure the time difference in picoseconds between the CITS system white waveform and the reference green waveform If the white waveform is to the left of the reference then the time offset is positive If it is to the right of the reference then the offset is negative Select Help About Double click on the CITS500s icon and click the Go button Adjust the Horizontal Offset value by the measured offset Click Save Click OK Press the foot switch or click the Test button The white waveform should now be displayed over the green reference waveform 51 Printed Circuit Board PCB Test Methodology 52 Figure 29 Typical display before Horizontal Standardisation Figure 30 Y HOAZS5005 TST Poles CITSS00 Ele Waem aioa Opiiore piaco Helo amaia la DU la AT 7656 DO ul User Guide 5 5 2 1 Printed Circuit Board PCB Test Meth
57. s only one Microstrip Probe is used Foot Switch polar_instr_conn 5 3 Line Length The length of coupon trace that may be tested using the CITS500s depends on the microstrip probe being used e Using an IP 28 28 Ohm Microstrip Probe a minimum trace length of 3 is recommended e Tf an IP 50 50 Ohm Microstrip Probe is used the minimum recommended length is 6 For recommendations regarding coupon design refer to IEC 61188 1 2 IPC D 317A and IPC 2141 42 User Guide 5 4 2 5 4 2 1 User Guide Printed Circuit Board PCB Test Methodology Section 1 CITS500s 32 bit Horizontal Standardization The horizontal offset of the CITS500s and its cable should be standardized to ensure that any CITS500s system will measure the same region of the coupon waveform and that errors are not introduced due to cables having different physical and electrical lengths 1 Select Learn Cable Length from the Utilities menu 2 Connect the working probe cable without the probe to the MAIN SMA front panel connector 3 Press the Learn Cable Length button The CITS will determine the length of the cable and display the time difference from the nominal length 4 Ifthe time difference exceeds 10 ps press Apply Correction to adjust the system calibration Test File Use the CITS500s Test File editor to create a test specification for each test trace on the test coupon Probe Length specifies the total electr
58. st use a known good 50 Ohm load 1 Always wear a ground strap The sampling head can easily be damaged by static electricity 14 User Guide l Printed Circuit Board PCB Test Methodology intel The following section describes procedures for the Tektronix 11801B Digital Storage Oscilloscope with SD24 sampling head Calibration should be performed on a daily basis 3 2 1 Calibration Sequence Gain Adjust loop gain for damping pulse response Actions Auto calibration Connect cable to Cal output Proceed Store Constant Recall User Constant Offset Sets offset of TDR pulse Actions Auto calibration Terminate cable with 50 Ohm load Proceed Store Constant Recall User Constant Actions Auto calibration Terminate cable with 50 Ohm load Proceed Store Constant Recall User Constant 3 2 2 General Calibration Verification After calibration is completed leave the 50 Q load termination on the instrument Turn cursors with TDR vertical scale reading in units of rho and verify that the reading is 50 Ohms TDR sampling heads are susceptible to static discharge damage This damage will not necessarily render the head inoperable but can induce measurement error The following steps can be completed to check for static damage e Connect the 50 Q load termination e Adjust the time scale to 500 ns div The upper portion of the pulse should be flat without sagging Note A 50 Q air dielectric reference load will provide the b
59. te probing and data extraction measurements using the TDR including proper instrument calibration procedures 3 1 General Equipment The following instruments or equivalents are recommended e Tektronix 11800 series with SD24 TDR module Rambus App Note Appendix A e Hewlett Packard s HP54750A with HP54754 TDR module Rambus App Note Appendix C e Polar Instruments CITS500s Rambus App Note Appendix B Note Impedance coupon general length recommendations are based upon the instruments that will be used for measurements User Guide 13 Printed Circuit Board PCB Test Methodology 3 2 General Calibration Figure 3 Sample Impedance and Velocity Test Structure for 28 Q Stripline Design RIMM 28 Q UNLOADED X Section Microprobe Test Signal Top i Signal Internal Internal TDR Manual Probe Type A GND VIA 25 mil x 25 mil Signal Probe Pad 5 mil x 5 mil SoldermaskOpening 10 mil x 10 mil GND VIA 25 mil x 25 mil Imp Vel_TestStruct_28strip Calibration is essential for ensuring that the equipment being used is measuring correctly and with maximum accuracy This is the first critical procedure necessary to ensure proper instrument setup prior to starting measurements Proper instrument calibration and verification on a regular basis cannot be over emphasized Calibration procedures for each type of equipment should be referenced in the user manual lInstrument calibration mu
60. ted by the SMA delay so that edge rate degradation can be observed 0 2 7 A 0 15 4 Cascade probe structure 11 o1 J Edge rate 8 mili rho nS gt 0 05 4 KH 2 SMA Connector structure 1 Z Edge rate 5 mili rho nS 0 z 0 05 4 F Improved reflection da response with microprobe 0 15 4 0 2 4 1 70E 09 1 75E 09 1 80E 09 1 85E 09 1 90E 09 1 95E 09 2 00E 09 Time seconds Cmp_MProbe SMAEdge 22 User Guide 3 5 1 Printed Circuit Board PCB Test Methodology Impedance Measurement Techniques The TDR provides a simple means for determining PCB impedance and propagation delay characteristics However the actual data extraction from a test structure can be highly dependent on cursor positioning on the TDR pulse This section outlines the general instrument setup necessary to obtain accurate repeatable measurement results independent of probe type and test structure Display Adjustment Display adjustment should be completed to maximize measurement accuracy The horizontal and vertical adjustments should be set under probing conditions It is recommended to adjust both the horizontal and vertical scales until the launch edge is aligned with the first screen division and the reflected edge is aligned with the last The vertical scaling should be adjusted to maximize 50 of the screen between the launch ledge and line under test The instrument screen should look similar to Figure 14 under pr
61. time difference from the nominal length If the time difference exceeds 10 ps press Apply Correction to adjust the system calibration Choose the impedance to be calibrated 28 ohms and connect the appropriate impedance airline to the probe cable Press the Check button and enter the exact impedance of the reference airline The software will now display the measurement error at this impedance If necessary press Apply Correction to adjust the system calibration Repeat as necessary for other impedance calibration points Press OK to exit User Guide intel Figure 27 Typical display during 28 Ohm Verification Printed Circuit Board PCB Test Methodology Very IPS L d u ajian Bes sg Asma adh LA M d ea Srii da DO vsi a DOTE RPA fate te cable Sales VELO VO NE N Della fra Drm i BH N RS RW 206 35 1006 1056 1106 115 rs Edd 24 L hr Fira hirn pmm WO 177000 1 00 ube Tas 1206 1255 1306 1355 m Figure 28 Typical display during 28 Ohm Calibration Cra Las N Arina bred WA Neze n cred ks lt DIP O G ach Bira to cable Select quei vaci n vira Dito J HR SR RR SR BR HS Den Tires e D I m 1006 105 1106 115 1206 12 1106 11 Sigla Ered 23 Dies Fone Varun fi OU 00 UITG YL 9 aree we F Sel interred fiiere Sea Huet er User Guide 49 Printed Circuit Board PCB Test Methodology 5 4 5
62. uctures described in this document are intended for determining trace characteristic impedance and propagation velocity High speed bus designs require improved impedance and coupling control to satisfy voltage and timing budgets This means careful attention to bus design trace geometry in order to develop proper test coupons The following sections provide recommendations for improved test coupon designs Requirements for Matching Test Coupon to Bus Design Test structures should allow you to extract impedance and velocity characteristics that represent actual traces in the bus layout To represent the characteristic impedance and the design accurately test coupons must follow bus layout guidelines including trace to trace spacing and ground shielding Bus designs like the Direct Rambus channel depart from typical designs by operating in pipeline mode to provide data rates up to 800 MT s Sensitivity to trace impedance matching and coupling must be minimized to reduce affects like inter symbol interference ISI Implementing ground shields or floods between signal traces reduces trace to trace parasitic coupling Improved trace to trace impedance control is achieved by inserting ground shields in between signal lines This provides a uniform copper density across the bus to ensure etch characteristics are equivalent for all signal lines These ground shields also add capacitive coupling to ground which affects the trace characteristic impedance
63. uide 45 Printed Circuit Board PCB Test Methodology 46 Figure 24 Typical measurement using IP 28 HSM bos cil CITE 137 But Pie Bagad Dada weston Were Lifter Help Darren Y Smile 29 00 Teu K 2000 Tan P 23 00 Tar End ASIS AaS ole Teri Step al 4 Calin Sem Z MINM band cs CTSS Fla Fond Dasep K mret FDO Tea F aTa F RTs trad Areia Dae 1000 On KR K SE Tez Step Tol 4 Cen 10mm 4 Logged erir User Guide l Printed Circuit Board PCB Test Methodology intel Figure 26 Typical measurement using IP 50 er ee CTS 12 EA Bena EZO Lave Apps P Mme 1 n d Prad Te Step dal 4 Calin Zm 4 Logged lers 5 4 4 Verification at 28 Ohm Impedance The CITS500s 32 bit is specifically calibrated at a number of impedance values including 28 Ohms to ensure measurement accuracy When making very accurate measurements using the CITS500s the effects of the connecting cable must be considered Losses in the cable will result in the system measuring a higher impedance value than actual This effect is especially significant when measuring low values of impedance Use of a traceable standard airline certified by NIST or NPL to calibrate the CITS500s with its cable allows precise measurement It is important to note that the characteristics of the cable will alter during extended use This procedure should be repeated at regula
64. upons The recommended method normalization allows accurate measurements of impedance The alternate method the offset technique characterizes the coupons by comparing them to an alternate standard such as a precision thick film resistor or airline Normalization Before accurate impedance measurements can be performed the frequency response errors and losses caused by the imperfections in the system cables and probing hardware must be removed using normalization Normalization is performed using two calibration standards a short zero Q and a high quality 50 Q thick film termination located on the calibration substrate The process is simple 1 Perform the normalization 2 Measure the coupon s impedance 3 Repeat step 3 for each coupon Offset The offset comparison technique uses a known standard such as a precision thick film resistor or 28 ohm airline With this technique the known standard is characterized and the difference between the known value and the measured value becomes an offset that must be subtracted from each measurement The process is as follows 1 Measure the standard s reflected voltage Calculate the offset as V V offset Vo min al al Measure the coupon s reflected voltage Apply the offset Calculate the impedance Eh M M N Repeat steps 4 5 and 6 for each coupon 67 Printed Circuit Board PCB Test Methodology intel 6 2 3 2 Normalization Recommended Procedure 68 e
65. ure 12 Comparison of Microprobe and SMA TDR Hesponses 22 Figure 13 Comparison of Microprobe and SMA Edge Rate Degradation 22 Figure 14 Example Screen Display When Measuring 28 Q Coupons sese eee eee eee e 23 Figure 15 Example of the Cursor Positioning Method on a TDR Response 24 Figure 16 Example of Averaging Mean Method for a 28 Coupon 25 Figure 17 Comparison on Unfiltered vs Filtered 200 ps Hesponse eee eee 26 Figure 18 Example of a Differential Impedance TDR Hesponse eee eee eee 27 Figure 19 Example of a Velocity Measurement Using the TDR Method 28 Figure 20 Example of a Basic TDT Setup eee 29 Figure 21 Example Delay Measurement using TDT Approach 29 Figure 22 Typical Test File Editor display i 44 Figure 23 Open circuit display using IP 28 ii 45 Figure 24 Typical measurement using IP 28 sese e eee 46 Figure 25 Open circuit display using IP 50 nan cnnrccrnn cnn nr cnn 46 Figure 26 Typical measurement using IP 50 47 Figure 27 Typical display during 28 Ohm Verification i 49 Figure 28 Typical display during 28 Ohm Calibration sss sese eee eee eee eee eee 49 Figure 29 Typical display before Horizontal Standardisation sees eee eee eee eee 52 Figure 30 Typical display after Horizontal Standardisation ii 52 Figure 31
66. where rho 0 001 mrho Zload 50 1 rho 1 rho 36 User Guide intel Printed Circuit Board PCB Test Methodology Figure 28 Mean Measurement Example 4 4 User Guide TDR Direct Rambus 28 Ohm Impedance Calibration Impedance measurements significantly different than 50 Q can result in large errors between measured and actual This systematic error is very common when completing Direct Rambus 28 2 measurements where measured values can easily be off by 2 3 Q from the actual impedance The primary sources of error include but are not limited to instrument bias probes and cable loss A simple means to account for these affects is to use a known reference standard close to the characteristic impedance of the line under test Reference standards are used to determine the systematic offset between measured and actual PCB impedance Measuring the standard with the same probe and cable that will be used for measurements connected to the instrument provides the additive effective error of the system to the probe tip When the probe used is a known good probe error due to probing will be small 0 2 2 in most instances To verify the probe affects compare the TDR response between the probed reference vs direct cable connection If it is deemed that probe effects are negligible measurements can be completed by direct cable connection to the standard This is especially useful when measuring air line standards when probing

Printed Circuit Board (PCB) Test Methodology User Guide

Contents

Download Pdf Manuals

Related Search

Related Contents