• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 45 subscribers
  • Views 796 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

ADS124S08: Four 4-wire RTD measurement circuit with ADS124S08

teannen
Prodigy 40 points
Part Number: ADS124S08
Other Parts Discussed in Thread: ADS114S08, ADS114S08B, TMUX1208

Hello engineers

I'm currently working on a project for university where one task is to measure ten Pt100 in the temperature range of 0°C to 200°C. The goal is to reach a accuracy of 0.1°C, even though I think that's a little bit enthusiastic. Anyway, I created a schematic using the ADS124S08 and would like to have an opinion to check if I correctly interpreted the information available in the datasheet and the application notes/cookbooks. 

I intended to cycle IDAC1 through AIN0, AIN3, AIN6 and AIN9. This way it should be possible to measure four Pt100 with one single ADS124S08, right? 

The following conditions apply for the design:

  • Excitation current IDAC1=1mA
  • Less than 10 SPS expected
  • I calculated a gain value of 8 for the PGA
  • The filter calculation results are added to the schematic, see green text
  • Start and reset will be performed over the serial interface

The corresponding schematic sheet is attached to this post. Any tips would be highly appreciated, especially from somebody that has used the IC before. If more information is needed from my side, let me know. 

Thanks in advance,ThomasADS124S08_Schematic_Draft.pdf

  • 0
    TI__Guru** 112895 points

    Hi Teannen,

    Welcome to the E2E forum!  Your design appears to be 4-wire RTD and 3.3V analog supply.  It appears that you are understanding and following the RTD Guide and Cookbook circuits.  I don't see any issues with your design.

    You may find the ADS124S08 Design Calculator helpful to verify any changes you may make:

    http://www.ti.com/tool/ADS124S08-EXCEL-CALC-TOOL

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 40 points

    Dear Bob,

    Thanks for your response and for the link to the Design Calculator, I wasn't aware that it exists. Three questions arose in the mean time:

    1. Am I correct with the assumption that four RTDs is the most I can manage to measure with one single IC in 4-wire config, If I want to be able to remove single RTDs from the measurement? I thought about rearranging them to in series, but that would lead to the problem that if I disconnect one RTD, all sensors in series will not work anymore. Also the lead resistance would add up, which might lead to a violation of the input requirements.
    2. Is the 0.1°C accuracy and resolution goal feasible when I decide to use the 16Bit version (ADS114S08)? I know that the accuracy is mainly influenced by the tolerances of the parts (reference resistor, RTD, etc.). Regarding the resolution: according to the Calculator-Sheet, the LSD size would be 6.3 uV, so I should be able to detect resistance changes as low as 6.3 mOhm (with IDAC = 1 mA). This would lead to a resolution of roughly 0.016 °C, am I correct?
    3. The resolution seems to be almost too high for my requirements, but it seems like there is no similar IC with the same number of channels and integrated current source that makes it perfect for the RTD, right? The only one I found was the ADS114S08.

    Thanks and best regards, Thomas

  • 0 in reply to teannen
    TI__Guru** 112895 points

    Hi Thomas,

    See my responses below.

    Best regards,
    Bob B

    teannen said:

    Dear Bob,

    Thanks for your response and for the link to the Design Calculator, I wasn't aware that it exists. Three questions arose in the mean time:

    1. Am I correct with the assumption that four RTDs is the most I can manage to measure with one single IC in 4-wire config, If I want to be able to remove single RTDs from the measurement? I thought about rearranging them to in series, but that would lead to the problem that if I disconnect one RTD, all sensors in series will not work anymore. Also the lead resistance would add up, which might lead to a violation of the input requirements. [Bob B] One other option would be to add an analog mux to direct the current.  You could use the AINCOM as the current output and direct the current to the appropriate RTD from the mux.  This configuration would allow 2 more RTDs to be connected for a total of 6, 4-wire RTDs to the ADS124S08.
    2. Is the 0.1°C accuracy and resolution goal feasible when I decide to use the 16Bit version (ADS114S08)? I know that the accuracy is mainly influenced by the tolerances of the parts (reference resistor, RTD, etc.). Regarding the resolution: according to the Calculator-Sheet, the LSD size would be 6.3 uV, so I should be able to detect resistance changes as low as 6.3 mOhm (with IDAC = 1 mA). This would lead to a resolution of roughly 0.016 °C, am I correct? [Bob B] Not quite.  You have to also consider the total noise of conversion.  For the ADS114S08 and low-latency filter using a gain of 8 there is 9.5uV noise (from noise table 3 in the ADS114S08 datasheet).  This would reduce the noise-free bits slightly to around 14.4 bits due to the smaller reference voltage as the tables are based on a 2.5V reference.  The resistance resolution now becomes 9.5mOhm.
    3. The resolution seems to be almost too high for my requirements, but it seems like there is no similar IC with the same number of channels and integrated current source that makes it perfect for the RTD, right? The only one I found was the ADS114S08. [Bob B] There is also noise and drift of the ADC to consider along with external system noise and drift of the reference resistor.  These will all directly impact your level of precision.  It also depends on whether or not you are calibrating your system.  The ADC may have a high level of precision and be able to resolve to a small value, but the accuracy may be something quite different if the system is uncalibrated.  You need to consider the tolerance of the RTD, offset, and gain errors throughout the range of temperature the ADC and reference resistor will operate.  You may find you need the high level of precision to be able to maintain the overall goal of 0.1 deg C.  You could also consider the ADS114S08B which is a slightly lower performing device at lower cost.

    Thanks and best regards, Thomas

  • 0 in reply to Bob Benjamin
    Prodigy 40 points

    Hi Bob

    Thanks for the quick and detailed reply.

    I have to check the approach with the Analog MUX first, I looked at the TMUX1208 which looks like it would be a good fit. However, I think the muxing of the current might then lead to some timing related challenges when I have to align the muxing and the ADC. Guess I have to think this through first. 

    You wrote: 

    Bob Benjamin said:
    This configuration would allow 2 more RTDs to be connected for a total of 6, 4-wire RTDs to the ADS124S08.
    Am I right that this is only possible if I get rid of the reference resistor? 

    Thanks for the correction regarding my assumption for the resolution (I thought that's too good to be true) and the comment regarding the overall precision.I will leave the question open until the remaining question regarding the number of RTDs if I choose to use multiplexing is definitively solved. 

    Thanks, Thomas

  • 0 in reply to teannen
    TI__Guru** 112895 points

    Hi Thomas,

    There is a dedicated reference input pair REF0 which you would use for the reference input from the reference resistor.  The analog IDAC current would be provided from AINCOM and then to a mux.  That leaves the 12 remaining analog inputs as 6 differential input pairs.

    You are correct in that there will be some settling delay whens switching the mux channels, but there would be some settling delay anyway with your present design as there will need to be a period of time when switching IDAC channels for the input caps to settle once the RTD is excited.  For this reason there is a configuration register setting that allows you to enter a Programmable Conversion Delay time.  This is discussed on page 61 of the datasheet in section 9.4.4.3.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 40 points

    Hi Bob

    Thanks again for the help, I'll mark the thread as resolved. 

    Kind regards, Thomas