• State TI Thinks Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 4 answers
  • Subscribers 45 subscribers
  • Views 801 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

ADS1220: ADC counter change when the device put into 5-10 oC.

loc nguyen
Prodigy 80 points
Part Number: ADS1220

Hi all,

I used ADS1220 for my project, 4 wire RTD measurement. When i put my device in Room tempearture (25-28 oC), the measurement is correct, when i put my device in 5- 10 oC, the measurement isn't correct (error -2 oC).

My configuration: Normal mode, 90 SPS, Continuous conversion mode, Current sources on, 500 µA, Disables temperature sensor, Internal 2.048-V reference selected, Simultaneous 50-Hz and 60-Hz rejection, : IDAC2 connected to AIN3/REFN1, : Only the dedicated DRDY pin is used to indicate when data are ready,: AINP = AIN0, AINN = AIN1 ,: Gain = 2, : PGA disabled and bypassed.

The schematic below, short SP1 .

Please show my mistake,

Thanks you,

  • 0
    Prodigy 80 points

    HI all,

    I put my device in room temperature and 5 -10 oC, but RTD probe put same tempearature (25oC).

    With room temperature: device return 25.2 oC.

    With 5 -10 oC chamber: device return 23 oC.

    Thanks all,

  • 0 in reply to loc nguyen
    TI__Guru** 113765 points

    Hi Ioc,

    Normally this measurement is made using a ratiometric measurement where the excitation of the reference and the RTD is the same.  It appears that the schematic is setup to use a ratiometric measurement, but in your configuration you are not doing so.  Is R12 connected in your circuit?  And if so, what is the value?  If R12 is not installed, then there is no current path for the IDAC current source to AVSS.

    I would highly recommend reading A Basic Guide to RTD measurements.

    Also, you could review the application circuit for RTD measurements in the ADS1220 datasheet.  One thing to also consider is the 50 and/or 60 Hz rejection only works for for 20sps.  The filter will not work for 90sps.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    TI__Guru** 113765 points

    Hi Ioc,

    I missed the comment that SP1 is shorted. Using the internal reference with low gain will include drift of both the reference and the IDAC source.  I would suggest making the measurement ratiometric to cancel this drift in the conversion result.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 80 points

    Hi Bob B,

    you would suggest making the measurement ratiometric to cancel this drift in the conversion result.

    => It means i need use External reference (REFP0, REFN0)  and Current sources on, 500 µA.

    Can i use analog supply (AVDD – AVSS) used as reference ? I will use REF3025.

    Thanks,

    Best reagards,



  • 0 in reply to Bob Benjamin
    Prodigy 80 points

    Hi Bob B,

    I tried with configuration: External reference (REFP0, REFN0)  and Current sources on, 500 µA. Rref=3.9kohm. when i tried with room temperature adc =3887, 5-10 oC chamber adc=5410.

    Some thing wrong ?

    thanks,

    Best regards,

  • 0 in reply to loc nguyen
    TI__Guru** 113765 points

    Hi Ioc,

    Making a ratiometric measurement involves using a reference voltage source that is established from the same source that is exciting the RTD.  You have not stated if you are using PT100 or PT1000.  I will assume that the RTD is PT100, and that at room temperature the expected resistance value is around 110 Ohms.

    If you are using the IDAC current of 500uA and a 3.9k Ohm resistor for the reference, you can calculate the resistance value for the RTD as a ratio of the code returned using the formula from A Basic Guide to RTD Measurements for the 4-wire RTD as shown in equation 68:

    RTD = (RREF * Output Code) / (GAIN * 2^23)

    So for the ADS1220, using the REF0 as the reference source, the RREF as 3900 Ohms and a GAIN of 2, I would expect for room temperature to see a code value closer to 473.2k.

    So something is considerably off with respect to the code you are seeing returned as it is closer to 0 then to the expected value.  Make sure that your wiring is correct and that the current is flowing properly in your circuit.  I would expect an approximate voltage drop across RREF (R12) to be 1.95V.

    Can you send me the actual register configurations you are using for each register?

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 80 points

    Hi Bob B,

    Sorry with my old test. I used filter,  room temperature adc =388712 => 3887, 5-10 oC chamber adc=541013 => 5410.

    New test:

    I tried again with resistor 100 ohm (instead of RTD), RREF 3.9 kohm => measure: VREF 1.94V, voltage drop on resistor 100ohm (instead of RTD) 50mV. 

    write_reg(REG_0, 0x03), write_reg(REG_0,0x05) write_reg(REG_2, 0x55), write_reg(REG_3, 0x10). 

    ADC count =780134. Theory ADC count = 430185. => It is very strange.

    Thanks,

    Best regards

  • 0 in reply to loc nguyen
    Prodigy 80 points

    Hi Bob B,

    I tested with room temperature, ADC count =780134, with 5 - 10 oC chamber, adc count = 766713 ( 20 minutes in chamber).

    Thanks,

    Best regrads,

  • 0 in reply to loc nguyen
    TI__Guru** 113765 points

    Hi Ioc,

    When in the temperature chamber you must also consider the potential drift of the reference resistor and the offset of the ADC.  However this will not account for 20k codes unless the resistor drift is very large.

    One thing I need to verify.  In your register listing you show:

    write_reg(REG_0, 0x03), write_reg(REG_0,0x05) write_reg(REG_2, 0x55), write_reg(REG_3, 0x10). 

    You are either writing register 0 twice or you meant that the second write_reg function should say REG_1.  If you are intending on writing register 1 with this value it would be incorrect.  You do not want to turn on the burnout current sources (BCS) for the conversion measurement.  So register 1 should not have bit 0 as '1'.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 80 points

    Hi Bob B,

    Sorry, I am wrong when typing : write_reg(REG_0, 0x03), write_reg(REG_1,0x05) write_reg(REG_2, 0x55), write_reg(REG_3, 0x10). 

    However this will not account for 20k codes unless the resistor drift is very large.

    => I don't know why the resistor drift is very large, it is very strange.

    Thanks,

    Best regards,

  • 0 in reply to loc nguyen
    TI__Guru** 113765 points

    Hi Ioc,

    As I mentioned in my previous post, you should not use setting 0x05 for register 1.  This turns on the 10uA burnout current sources which will add significant error to the RTD measurement.  Change the register 1 setting to 0x04 instead and you should see a significant improvement.

    Best regards,

    Bob B