Using the MSC121x as a High-Precision Intelligent Temperature
Contents
1. 2 105 5 D m 105 4 9 a oO 2 105 3 5 VDD V Figure 9 Self Heating Effect from Power Supply Voltage Change 14 Using the MSC121x as a High Precision Intelligent Temperature Sensor We TEXAS INSTRUMENTS SBAA100 Heat generated by sensors when power is applied causes self heating When the MSC121x operates as a temperature sensor the applying power introduces self heating and affects the calibration coefficient The device heat dissipation is proportional to the power supply voltage operation frequency and on chip operating features Figure 8 and Figure 9 show the self heating effect on slope and intercept coefficients when the device operation frequency or supply voltage is increased If the coefficients are calibrated the self heating effect is removed with the calibration process However if the coefficients are not calibrated or the application environment is very different from the environment when the coefficients are prepared minimum heat dissipation from the device is preferred Minimum heat dissipation requires a lower power supply voltage and clock frequency The lowest heat dissipation condition recommended for the MSC121x is 1 MHz clock frequency with 3V power supply 6 Example Code The MSC1210 DAQ EVM 5 board is used for running the example code Raisonance IDE is used for the code development The clock frequency is running at 1 8432MHz The power supply is at 3V Appendix B lists the source code u2. The self heating effect will be discussed in Section 5 H Galden ZT 180 4 a heat transfer fluid is used in the bath It is a non electric conducting thermally and chemically stable fluid To increase the effectiveness of the bath a 1A stirring fan with a PWM speed control circuit is used to break the H Galden fluid ventilation and to maintain constant temperature over the bath The temperature bath is placed inside a programmable oven to perform calibration Using the MSC 121x as a High Precision Intelligent Temperature Sensor 13 vB TEXAS SBAA100 INSTRUMENTS 4 4 Package Junction Temperature The MSC121x device is packaged in a 64 lead TQFP package This package has junction to case thermal resistance 6 jc of 4 3 C W This application controls MSC121x power dissipation within 3 6mW 3V x 1 2mA at 1 8MHz This power dissipation introduces a mere 0 015 C 3 6mW x 4 3 C W of temperature gradient between the silicon die and the temperature bath fluid which is removed by the temperature calibration process High device power dissipation should be avoided for precise temperature measurement 5 Self Heating Effect 382 5 108 0 382 0 107 5 381 5 A 381 0 a gt F 106 5 S 380 5 E 106 0 8 2 380 0 z f D 3705 105 5 379 0 105 0 378 5 104 5 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 Cik Freq MHz Figure 8 Heating Effect from Clock Frequency Change 106 0 105 9 105 8 105 7 105 6 2 3S
3. 0 1 C can be achieved Inspecting the coefficient of the T 2 term in Table 2 this value is much smaller than the other two coefficients of the same equation When third or higher order polynomial curve fittings for the 15 measurement points are used See Appendix A no significant curve fitting error reduction can be obtained It can be seen that third order polynomial curve fittings may slightly reduce measurement error However using a third order polynomial increases computational requirements Using the MSC 121x as a High Precision Intelligent Temperature Sensor We TEXAS INSTRUMENTS SBAA100 Equation 5 shows the inverse function of the second order polynomial Viemp f T Table 2 Equation 5 is useful to calculate To from the ADC acquired voltage Vtemp Similar to the inverse function of the linear equation the coefficients for this equation can be generated by using curve fitting with Viemp at the x axis and Te at the y axis T aV bV temp te ng T Equation 5 Second Order Inverse Function of Vtemp f Tc 3 Temperature Sensor Calibration Methods The curve fitting method is very effective for computing the linear or polynomial equation coefficients This method requires measuring the Vtemp Tc points The precision of the equation increases as more points are included However increasing measuring points increases equipment manufacturing cost and complexity Reducing the number of calibration measurement points is de
4. file to calculate the coefficients OR 0T Q per C defines the requirement of measurement accuracy For example at 25 C OR dT 99 Q C R is around 2 2kQ To measure 0 05 C a DMM needs to have resolution of 5 Q 99 0 05 for the range setting for 2kQ The dR dT values for different temperatures are supplied with the YSI attached Excel file Reference Sensor Measurements Source Sense Sense Source HV Shield Figure 7 RTD 5 Wires Configuration An HP3458A Digital Multimeter is used in this report to measure the RTD resistance Figure 7 shows the 5 wires connection used for RTD resistance measurement the meter supports Sense Sense connection pair is for RTD voltage sensing inputs This pair carries almost zero current to avoid measuring any voltage drop caused by lead connections resistance Source Source connection pair is for excitation current for resistance measurement The current from the meter is less than 100uA to minimize the RTD self heating effect caused by I R drop This meter is able to measure better than 0 05 C over the RTD resistance range within its corresponding temperature The HV Shield connection Figure 7 protects the measurement from noise interference and external high voltage discharge Temperature Bath A stable temperature bath helps to reduce both the self heating effect and the temperature gradient caused by the device package and provides a constant temperature for calibration
5. 100 1 Temperature Sensing System The MSC121x has an on chip precision temperature sensor 24 bit delta sigma ADC an 8 bit 8051 CPU communication and input output peripherals Figure 1 These single chip embedded functions can be used to compose a temperature sensing system for intelligent temperature monitoring temperature measurements for IPC or cold junction corrections Usage examples for such temperature measurement results are e Provide feedback to other systems via communication peripherals e Generate status and or response to other control via input output peripherals e Provide temperature logging record with 32KB Flash storage e Detect system temperature drift to calibrate measurement offset and gain accuracy for external analog sources The device communication peripherals include a variety of industrial standards UART dual SPITM and I2C MSC1211 only Communication Peripherals External Analog Signal Sources 32KB Flash Memory Temperature Sensor 24 bit Delta Sigma ADC 8051 CPU VO Peripherals Figure 1 MSC121x Intelligent Temperature Sensor Using the MSC 121x as a High Precision Intelligent Temperature Sensor 3 4 TEXAS SBAA100 INSTRUMENTS 1 1 Temperature Sensor Circuit The MSC121x has a pair of temperature sensing diodes D1 and D2 as shown in Figure 2 The differential inputs of the on chip ADC converter are selectable via input multiplexers When the ADC multiplexer control SFR ADM
6. UX is set to FFy the ADC inputs are connected to the temperature diode outputs The differential voltage of the diode pair provides a temperature reading AINO To ADC AIN7 l AINCOM l Figure 2 Block Diagram of Temperature Sensor 4 Using the MSC121x as a High Precision Intelligent Temperature Sensor We TEXAS INSTRUMENTS SBAA100 1 2 Temperature Sensor Parameters The theory of using a silicon diode to measure temperature is discussed in the TI Application Report Measuring Temperature with the ADS1216 ADS1217 or ADS1218 2 Most of the dis cussion from this bulletin applies to the MSC121x design Equation 1 shows the PN junction ideal diode equation Table 1 shows the parameter description for the diode equation In the MSC121x implementation Ipjope is from a constant current source Is is the only temperature dependent parameter To eliminate ls differential measurement is used nkT In 2100 I Ss Vorope Equation 1 Ideal Diode Equation Diode emission coefficient typical value 1 n k Boltzman constant 1 8806503E 23 J K Charge of an electron 1 602176462E 19 C Diode saturation current function of diode design and temperature Absolution Temperature K Table 1 Diode Equation Parameters The diode voltages for D1 and D2 in Figure 2 are Vp and Vppa respectively The difference of the diode voltages is Vtemp The constant current applied to the diodes Ip and Ip2 is design
7. a o ga aE E AEE aE Ea aa EE eed 5 2 Temperature Sensor Accuracy 62ic22 62si0se cere setdnveidventeidedsanpdadeasideiess 6 2 1 near Curve Fitting errasse asemana nni adaa ieii ear aie a E 6 2 2 Polynomial Curve FUmO 3c icceieieseedeerecdieaticeedebectsde tte dnde deeded cbse 7 3 Temperature Sensor Calibration Methods 00 cece cece eee eee eens 8 3 1 Two Points Calibration ct css lt vencimecinetescarauhasuenier ge aeieneenndeenectaatsne 9 3 2 Three Points Galibrati M ee ctenxeccturctst ence eai a e a a e ei aaa i 9 3 3 No Calibration and One Point Calibration 0 ccc eee 10 3 3 1 Average Coefficient Values 0 0 cece ees 10 3 3 2 One Point Calibration lt gcc cccnees evereaeatncene diverted idasdagetezeacniewes 11 3 3 3 Meas rement SOUNDS 4010 24 idunexd tite traded deere e aE a ES 11 4 CGallbtation Setups c ccs22 602 tose ieee eee cet todd bee Eee ie ee toe 11 41 Referente SENSO creatine elvan heat ees eee BR eed es 11 4 2 Reference Sensor Measurements 00 cece eee eee eee eae 12 4 3 Temperature Ban c i2c0 ei aitare aiee i iiia aani Lede ental tatedwteades 12 4 4 Package Junction Temperature ssassaaaa nunnan naene 13 5 Self Heating EN Cl cccccc2ccsccchccencicecassianssesseedtaietaed E E a E 13 6 Ex Mmple Code nenrno eoe A O R Oa 14 References iiss 0b tiedie inian n E E cee eee EEA E E ewe EE E EE 15 Appendix A Sample Part Measurement Data 0 cece eee eee eee e
8. e routine print is added instead of the library printf printf with a floating point requires over 1kB additional memory while print and its calling routine need only 234 bytes References 16 1 MSC1210 MSC1211 MSC 1212 Product Data Sheets SBAS203A SBAS267 SBAS278 2 Measuring Temperature with the ADS1210 ADS1217 or ADS1218 SBAA073 3 YSI Precision Temperature Group http ww ysi com 4 Solvay Solexis http www solvaysolexis com pdf h_gald_zt180 pdf 5 MSC1210 DAQ EVM User s Guide and Examples SBAU083 Using the MSC 121x as a High Precision Intelligent Temperature Sensor Wy TEXAS INSTRUMENTS SBAA100 Appendix A Sample Part Measurement Data IBE 0007714854 Table A 1 Sample Part Measurement Data Using the MSC121x as a High Precision Intelligent Temperature Sensor 17 4 TEXAS STDZ001A INSTRUMENTS Appendix B Temperature c Source Listing include lt REG1210 H gt include lt stdlib h gt include lt math h gt Vref 1 25V bipolar results give 1 25 2 2 24 define LSB 1 25 8388608 define MO 2586 67098545498 define CO 271 7529 define A1 1 08436244851484e3 define B1 2 83158996601634e3 define C1 285 1293 define WINDOWSIZE 16 signed long summer void extern void putspace4 void extern void tx_byte char Print a message void putstr char code data msg while msg 0 tx_byte unsigned char msg if msg n tx byte r msg recursive digit di
9. ed such that Ip2 80 Ip1 The two diodes are also designed such that diode saturation currents Is_ of the two diodes are identical Identical Is are achieved by closely matching the two diode fea ture sizes and applying silicon layout techniques to reduce process variation for components on the same silicon die Therefore as shown in Equation 2 the diode pair output voltage Vtemp is proportional to the absolute temperature T where amp is a temperature independent constant temp a ee n2 me j iy sa q 4 s s q D2 Equation 2 Differential Diode Equation The MSC121x device Vtemp Measurement results do not exactly match with Equation 2 Exam ples for the sources of deviation from Equation 2 are e Diode emission coefficient n in Equation 1 is a function of Ipiope therefore the assumption that n equals 1 should be corrected e Perfectly matching D1 and D2 is not possible therefore the saturation currents Is of D1 and D2 are not identical Using the MSC121x as a High Precision Intelligent Temperature Sensor 5 k TEXAS SBAA100 INSTRUMENTS e The ratio of Ip and Ipa is not temperature independent therefore when the diode pair differential voltage is calculated from the temperature in Equation 2 the resulting voltage has a high order term of T for example Viemp aT2 bT C e The ratio of Ip and Ip2 is not identical across each device therefore a single value cannot be used for all devices e The coefficients a
10. ee eee 16 Appendix B Temperature c Source Listing 00 2c eee teen eee eens 17 All trademarks are the property of their respective owners 4 TEXAS SBAA100 INSTRUMENTS List of Figures Figure 1 MSC121x Intelligent Temperature Sensor 00 eee eee cece eee 2 Figure 2 Block Diagram of Temperature Sensor ceee cece eee eee eee eens 3 Figure 3 MSC Viemp and Measurement Accuracy 0c eee cece eee eee eee eens 7 Figure 4 Increasing Accuracy with Calibration Complexity 0cee eee eee eee 9 Figure 5 Slope and Intercept Coefficient Histograms for 6 Samples 0e000 10 Figure 6 Temperature Measurement Error for 6 Samples 002cceeee ee eee eens 10 Figure 7 RTD 5 Wires Configuration 2 c 0icccesccteensesevecseease basses wetretbaeeseces 12 Figure 8 Heating Effect from Clock Frequency Change 0eeceeeee eee eeeeeees 13 Figure 9 Self Heating Effect from Power Supply Voltage Change 0eeeeeeeeees 13 List of Tables Table 1 Diode Equation Parameters 0 0c cece eee eee eee eee 4 Table 2 Temperature Sensor Slope and Intercept Coefficients 0c cece eee e eens 5 Table 3 Temperature Sensor Calibration Complexities vs Precision 0eeeeeeeee 8 Table A 1 Sample Part Measurement Data 0 00 cece eee eee eee eee eens 16 2 Using the MSC 121x as a High Precision Intelligent Temperature Sensor We TEXAS INSTRUMENTS SBAA
11. k TEXAS Application Report INSTRUMENTS SBAA100 July 2003 Using the MSC121x as a High Precision Intelligent Temperature Sensor Hugo Cheung Data Acquisition Products Microsystems ABSTRACT The MSC121x 1 is an embedded controller with a high precision high stability temperature sensor a 24 bit delta sigma AX analog to digital converter ADC and an enhanced 8051 CPU These circuit functions are the major building blocks for an intelligent temperature sensor After temperature calibration the on chip temperature sensor measurement accuracy can reach 0 1 C with a resolution of 0 01 C over an operating range of 40 C to 85 C Applications for precision temperature measurements include but are not limited to thermocouple cold junction correction measurement industrial process control and system temperature monitoring This article introduces the fundamental concept and operation of the on chip temperature sensor This report also describes the calibration procedures to turn the device into a high precision intelligent temperature sensor This application note uses the term MSC 121x to refer to the MSC1210 MSC1211 MSC1212 as a series of devices Contents 1 Temperature Sensing System 0 cece cece eee eee eee eee eens 2 121 Temperature Sensor CircUit sirere stscass Bite Bhat ede alld delta dite e 3 1 2 Temperature Sensor Parameters 000 cece eee ete ees 4 1 3 Ideal Diode Equation mpini iiaia aiie ia
12. nsor calibration process This calibration process eliminates offset c and gain m error in the complete signal chain Example errors in the chain are interconnection offset signal conditioning circuit gain offset ADC gain offset reference voltage accuracy temperature draft and silicon die to package temperature difference 140 0 5 0 4 130 0 3 120 0 2 0 1 a 1D 110 Vtemp mV 0 0 5 Linear Err deg C ai Poly Err deg C 100 0 1 0 2 90 0 3 80 0 4 44 40 30 20 9 1 11 21 31 41 51 60 71 80 85 Temperature deg C Figure 3 MSC Vtemp left y axis and Measurement Accuracy right y axis Polynomial Curve Fitting The linear ideal equation Equation 3 includes only a first order effect from Tc Vtemp is actually a function of a higher order effect of Te When the higher order component of Te is considered a new level of temperature measurement accuracy can be achieved The Viemp data points Figure 3 are curve fitted with a second order polynomial The Polynomial Curve Fitted row of Table 2 shows the second order polynomial and its coefficients The error of temperature measurement that occurs when this polynomial is used appears on the Poly Err curve in Figure 3 The curve fitting error is reduced to 0 045 0 048 C an error range of 0 093 C showing improvement over the linear curve fitting method by a factor of five When measurement system error is considered overall system accuracy of
13. nt calibration is used with the averaged slope coefficient to calculate the intercept coefficient the error is reduced by more than 4 5 C and reach below 2 C Figure 4 Measurement Setups To minimize self heating errors the measurements use a 1MHz clock frequency and 3V power supply The curves shown in Figure 5 were acquired when an internal reference voltage is used Because the temperature stability of the internal reference voltage is trimmed during the device manufacturing process the slope gain coefficient is very stable The internal reference voltage was able to achieve all the temperature measurement accuracy described in Table 3 Calibration Setups Reference Sensor This report uses an RTD temperature sensor to calibrate the MSC121x measurements YSI Precision Temperature Group manufactures the RTD sensor model 55031 3 The measurement accuracy is trimmed to within 0 05 C The RTD resistance to temperature relationship is shown in Equation 6 T a bIn R cln k Equation 6 RTD Resistance to Temperature Using the MSC 121x as a High Precision Intelligent Temperature Sensor We TEXAS INSTRUMENTS SBAA100 4 2 4 3 R is in ohms a b c are constants and T is in Kelvin Equation 6 coefficients are measurement temperature range dependent A different set of a b and c are used for a specified range of temperature Once the sensing temperature range is defined these factors are constant YSI provided an Excel
14. p accuracy for this report is around 0 05 C System accuracy using linear curve fitting is 0 5 C This level of accuracy outperforms many precision temperature sensors and equipment Each device may have unique values of m and c If enough measurement points of Vtemp Te are acquired for each device and the points are linear curve fitted a system measurement accuracy of 0 5 C should be achievable In addition to high precision measurement another advantage of using linear curve fitting is low computation requirements which need only one multiplication and one addition Computing To from Viemp S simply the inverse function of Equation 3 T Vicma aes nV e g temp m m Equation 4 Linear Inverse Function of Vtemp f To The coefficients m and c for Equation 4 could be calculated from m and c Another possible method for calculating m and c is that instead of using temperature for x axis and Viemp for y axis as in Figure 3 generating the linear curve fitting results again with x axis as Vtemp and y axis as temperature 1 Microsoft Excel functions of regression are used for this report See Microsoft Excel user s manual for details of regression usage 2 Measurement results table is listed in Appendix A Using the MSC 121x as a High Precision Intelligent Temperature Sensor i Vtemp mV 2 2 4 TE XAS SBAA100 INSTRUMENTS When curve fitting results are used to compute Te from Viemp this is the se
15. re calibrated to the temperature range of interest thus extrapolation beyond this range will result in error For instance when T 0 K the voltage is not equal to zero 1 3 Ideal Diode Equation yank In 80 temp q T 273 15 mT c Equation 3 Linear Temperature Equation Slope Coefficient Intercept Coefficient Curve Fitting System Accuracy m uV C c mV ee C lt a z Ideal Diode nk In 80 q 377 6 m 273 15 103 1 Linear Curve Fitted 386 7 104 98 2s 44 0 30 oe 49 0 35 Polynomial Curve Fitted Viemp 6 3595E 5 T 2 0 3842 To 104 90 0 045 0 048 0 095 0 098 Table 2 Temperature Sensor Slope and Intercept Coefficients Converting the voltage to temperature using a linear equation simplifies conversion computation Equation 2 is rearranged to a linear format to give Equation 3 Following are characteristics of Equation 3 e Celsius is used instead of Kelvin Celsius Kelvin 273 15 e Constant current ratio between Ip and Ip2 equals 80 e Tg is the diode temperature in C e mis the temperature slope coefficient e cis the temperature intercept coefficient The slope coefficient m and intercept c represent the gain and offset respectively of the temperature equation When the constants of the ideal diode equation Table 1 are used for Equation 3 coefficients for the ideal equation can be calculated The values are shown in the Ideal Diode row of Table 2 For physical devices m and c a
16. re different from the ideal values 6 Using the MSC121x as a High Precision Intelligent Temperature Sensor Wy TEXAS INSTRUMENTS SBAA100 2 2 1 Temperature Sensor Accuracy Linear Curve Fitting To understand if the physical device measurement result is different from that generated by the ideal model a linear curve fitting with least square method is used Vtemp s measured for 15 temperature points within the range of 40 to 85 C Vtemp results are shown on the left y axis of Figure 3 The measurement points have a very linear distribution The linear curve fitted coefficients are shown on the Linear Curve Fitted row of Table 2 The coefficients are within 2 5 of the ideal coefficients The linear curve fitted result also shows the precision of the temperature measurement system When the curve fitted coefficients are used to calculate temperature from measured Viemp the error of the system is shown on the line Linear Err referencing the right y axis of Figure 3 The error of linear curve fitting is within 0 44 and 0 30 C System accuracy is defined as the measurement accuracy when the temperature range of interest is measured repeatedly with the curve fitting coefficients The system accuracy result is listed in the last column of Table 2 The curve fitting precision is affected by the measurement equipment setup accuracy reference temperature sensor specifications and system uncertainties The measurement equipment setu
17. sary to support this warranty Except where mandated by government requirements testing of all parameters of each product is not necessarily performed TI assumes no liability for applications assistance or customer product design Customers are responsible for their products and applications using TI components To minimize the risks associated with customer products and applications customers should provide adequate design and operating safeguards TI does not warrant or represent that any license either express or implied is granted under any TI patent right copyright mask work right or other TI intellectual property right relating to any combination machine or process in which TI products or services are used Information published by TI regarding third party products or services does not constitute a license from TI to use such products or services or a warranty or endorsement thereof Use of such information may require a license from a third party under the patents or other intellectual property of the third party or a license from TI under the patents or other intellectual property of TI Reproduction of information in TI data books or data sheets is permissible only if reproduction is without alteration and is accompanied by all associated warranties conditions limitations and notices Reproduction of this information with alteration is an unfair and deceptive business practice TI is not responsible or liable for such altered documen
18. sing floating point math The size of the code is 3kB which is supported by the Raisonance 4kB Demo Version To minimize power consumption unused peripherals are shut off The PSEN and ALE pins are also turned off PDCON amp 0x0f7 turn on adc PASEL 0xff List 1 Power Control ADC is set to convert at 7SPS Since the DAQ EVM used is calibrated temperature drift error is canceled by the calibration Instead of using an external reference voltage the internal reference is used ACLK 1 ACLK 1 8432MHz 1 1 0 9216MHz DECIMATION Ox7 f ADCONO 0x20 Vref on 1 25V Buff off BOD off PGA 1 ADMUX Oxff ADCON1 0x01 _bipolar auto self calibration offset gain List 2 ADC Setup The on chip summation register is used The 7 SPS conversion is further averaged by 32 times SSCON 0 Clear SUMRO 3 SSCON 0xe4 Set SSCON for ADC 32 summations and div by 32 List 3 On chip summation register setup In order to achieve resolution higher than 0 01 C a moving window of size 32 is used The linear and polynomial equations are calculated with 32 bit floating point math Using the MSC 121x as a High Precision Intelligent Temperature Sensor 15 4 TEXAS SBAA100 INSTRUMENTS tc lin M0 vt CO tc poly Al vt vt Bl vt C1 List 4 32 bit floating point math library calculates the linear and polynomial equations To minimize code memory usage a recursiv
19. sired Table 3 shows the summary of different level of calibration complexities Gal Methods Errar Magnitude CO Zero Point Use averaged slope and intercept coefficient Non calibrated M ave 1 Maye 2582 173 C V Cave Cave Mave 269 386 C 1 Point 2 Significant accuracy improvement with only one point calibration 2Points _ 0 62 0 20 Worst case 1 Equation 3 2 points at 10 50 C Linear equ 0 28 0 54 3 points at 40 20 85 C worst case 0 7 Second order equ matching the accuracy as 15 pts Second order equ 0 07 0 14 calibrations worst case 0 2 C 0 1 6 points with third order polynomial 15 points with second order polynomial Table 3 Temperature Sensor Calibration Complexities vs Precision Using the MSC 121x as a High Precision Intelligent Temperature Sensor 9 4 TEXAS SBAA100 INSTRUMENTS 3 1 3 2 10 a Lin Err ipt e Lin Error 2pt a Lin Error 3pt Poly Error 3pt Error deg C Temperature deg C Figure 4 Increasing Accuracy with Calibration Complexity Two Points Calibration The linear equation Equation 3 has two unknown coefficients m and c Two points are needed to solve for m and c When solving the equation an error in m will introduce gain error An error in c will cause offset error The Linear Err curve Figure 3 is a parabola with the most negative error at the bottom of the curve and the most positive errors on the top t
20. splay routine void prt digit unsigned long int i signed char d reentrant unsigned long int j char c j i 10 c i j 10 d if j 0 d amp 0x80 prt digit j d if d 0 tx_byte tx_byte c 48 Display long integer number with sign and decimal void print signed long int i unsigned char d if 1 lt 0 tx_byte i 1 prt_digit i d putspace4 void main void signed long int data sum float data window 16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 float data adres 0 vt tc_lin tc_poly unsigned char fill ptr 1 mod_ptr 0 T2CON 0x34 T2 as baudrate generator RCAP2 65535 57600 Baud 1 8432MHz SCON 0x52 Async mode 1 8 bit UART enable revr TI 1 RI 0 putstr x1b 1 33 46m x1b 20 x1b 12CTemperature Sensor n PDCON amp Ox0f7 turn on adc PASEL Oxff ACLK 1 ACLK 1 8432MHz 14 1 0 9216MHz DECIMATION Ox7ff ADCONO 0x20 Vref on 1 25V Buff off BOD off PGA 1 ADMUX 0xff ADCON1 0x01 bipolar auto self calibration offset gain SSCON 0 Clear SUMRO 3 SSCON 0xe4 Set SSCON for ADC 32 summations and div by 32 sum summer Discard 1st summation result 18 Using the MSC121x as a High Precision Intelligent Temperature Sensor Ww TEXAS INSTRUMENTS STDZOO1A SSCON 0 Clear SUMRO 3 SSCON 0xe4 Set SSCON for ADC 32 summations and div by 32 while 1 sum summer SSCON 0 Clear SUMRO 3 SSCON 0xe4 Set SSCON for ADC 32
21. summations and div by 32 adres adres window mod_ptr Moving Window of 10 results window mod_ptr float sum adres adres window mod_ ptr vt adres float fill _ptr LSB tc_lin MO vt CO print signed long int tc_lin 1000 3 j tc_poly Al vt vt Bl vt C1 print signed long int tc_poly 1000 3 if i11 ptr WINDOWSIZE fill _ptr WINDOWSIZE else fill _ptr if mod_ptr WINDOWSIZE 1 mod_ptr 0 else mod_ptr putstr n if RI 1 Any key will pause any key again to start putstr Pause RI 0 while RI 0 putstr Continue n RI 0 Using the MSC 121x as a High Precision Intelligent Temperature Sensor 19 IMPORTANT NOTICE Texas Instruments Incorporated and its subsidiaries TI reserve the right to make corrections modifications enhancements improvements and other changes to its products and services at any time and to discontinue any product or service without notice Customers should obtain the latest relevant information before placing orders and should verify that such information is current and complete All products are sold subject to Tl s terms and conditions of sale supplied at the time of order acknowledgment Tl warrants performance of its hardware products to the specifications applicable at the time of sale in accordance with Tl s standard warranty Testing and other quality control techniques are used to the extent TI deems neces
22. tation Resale of TI products or services with statements different from or beyond the parameters stated by TI for that product or service voids all express and any implied warranties for the associated TI product or service and is an unfair and deceptive business practice Tl is not responsible or liable for any such statements Following are URLs where you can obtain information on other Texas Instruments products and application solutions Products Applications Amplifiers amplifier ti com Audio www ti com audio Data Converters dataconverter ti com Automotive www ti com automotive DSP dsp ti com Broadband www ti com broadband Interface interface ti com Digital Control www ti com digitalcontrol Logic logic ti com Military www ti com military Power Mgmt power ti com Optical Networking www ti com opticalnetwork Microcontrollers microcontroller ti com Security www ti com security Telephony www ti com telephony Video amp Imaging www ti com video Wireless www ti com wireless Mailing Address Texas Instruments Post Office Box 655303 Dallas Texas 75265 Copyright 2004 Texas Instruments Incorporated
23. ver absolute accuracy is impaired dramatically Using the MSC 121x as a High Precision Intelligent Temperature Sensor 11 4 TEXAS SBAA100 INSTRUMENTS 3 3 2 3 3 3 4 1 12 Slope and intercept coefficients of actual devices were acquired for six MSC1210 samples between the temperature of 40 and 85 C at 1MHz CPU clock frequency clock frequency effects will be discussed later Figure 5 and Figure 6 show the Slope and Intercept chart and the Temperature Error for these six sample parts The average slope m ave and intercept coefficient C aye of the samples are 2582 173 C V and 269 386 C respectively The Slope and Intercept chart shows the wide variations 5 C from 273 5 to 269 5 C of intercept coefficient among all six devices Each line of the Temperature Error chart shows the measurement error of a sample part The chart shows that the error value for each part fluctuates below 2 C These also indicate that the slope variation over the sample devices is low If the averaged slope intercept coefficients are used temperature measurement error is up to 6 5 C This error can easily reduce down to 2 C if the intercept coefficient variation is calibrated One Point Calibration One point calibration for a linear equation with two unknown factors requires either the averaged slope or averaged intercept coefficient As shown in Figure 5 calibrating the intercept coefficient corrects most of the error When one poi
24. wo ends of the curve The error at 10 C and 50 C have the minimum error Using these two temperatures to calibrate the linear equation gives the lowest error These calibration point selections give a measurement accuracy of 0 6 0 2 C that is similar to multiple points calibration Table 3 The worst case error would be lower than the peak to peak of the Figure 3 Linear Err curve that is less than 1 C Three Points Calibration The worst case error for second order polynomial curve fitting would be lower than half of the peak to peak of the Figure 3 Linear Err curve that is less than 0 5 C for most devices Using the MSC 121x as a High Precision Intelligent Temperature Sensor We TEXAS INSTRUMENTS SBAA100 3 3 No Calibration and One Point Calibration 3 3 1 Average Coefficient Values Slope amp Intercept for 6 parts 268 2610 269 2605 N x i 2600 N x 2595 272 2590 Intercept deg C Slope deg C V 273 2585 274 2580 4 Part Figure 5 Slope and Intercept Coefficient Histograms for 6 Samples Error Temp with ave m amp c Error deg C Temperature deg C Figure 6 Temperature Measurement Error for 6 Samples In some applications absolute measurement accuracy is not critical so measuring the calibration point is eliminated For those applications that concentrate only on relative temperature change using the typical coefficient set would be sufficient Howe
Download Pdf Manuals
Related Search