RE: ADS122U04: RTD calculation gives incorrect temperature

Hi User5877261,

I split the thread from the previous topic as the question changed from initial communication to actually measuring temperature with the ADS122U04 using TIDA-01526.  Can you tell me more about the RTD you are using and in particular how many wires?  Do you have R17 through R20 installed?  Can you tell me the actual conversion result value for each RTD measured?  I would prefer the 24-bit hex value prior to any conversion to temperature.

Best regards,

Bob B

Hi Bob B

    I'm glad to hear from you. I've mixed up the posts I posted! Again, I use the two-wire PT1000, R13 with 3900 ohms. R17 to R20 are installed. The register setting of my experiment today is register 0 (61H), register 1 (0ah), register 2 (03h), register 3 (a1h). My room temperature is about 20 ℃, and PT1000 is placed at room temperature. The ad value I receive is 447659. After the interpolation conversion in the program, a very unusual temperature value is obtained. I use a multimeter to measure the resistance of PT1000, which is 1089 ohms. Please help me analyze. What's the problem?
     

Thank you very much.

Hi User5877261,

With your original register settings, those settings were for PT100.  As you are using PT1000, you cannot use more than a gain of 2 at room temperature as you will exceed the input measurement range at a gain of 4 and 250uA of IDAC current.

For your second set of register settings you are using 100uA of IDAC current.  When using a 3900 Ohm reference resistor, the voltage output is only 0.39V and the minimum reference for the ADS122U04 is 0.75V.

So, pick a valid operating range and then let's see where we are with respect to the code results.  Make sure you have a valid reference voltage for the ADS122U04 and then verify that the output across the resistor agrees with the calculated value.  Then measure the voltage across the RTD and report back to me the measured voltages and the code being returned by the ADC.

Best regards,

Bob B

Hi Bob B

          Thank you for your reply. I just readjusted the register settings and recorded the relevant parameters,

This is the first group；

This is group two，

Thank you very much！

Hi,

I don't think the results are very far off.  By doubling the PGA gain group 2 is 1/2 the output code of group1 which is correct.  Could you take one more measurement for me and measure the actual resistance of the reference resistor to see how close it is to 3.9k?

Thanks,

Bob B 

Hi Bob B

My R13 resistance may have some deviation. I used a multimeter to measure 3880 ohms on the circuit board. But what I don't understand is that according to formula 17 (in zhcu422.pdf), the resistance value and ad value don't correspond to each other? (the position of the red circle should be equal to the sign)Thank you very much！

Hi Bob B

      In addition, according to the first set of data I tested above, ad value is 2264792. According to this data, I look up the table and interpolate to get the temperature. What I get is a negative number. When I look up the table and interpolate with the second set of data, the data is also wrong, because the interpolation function is the same, and the delivery parameters (2264792 and 1132448) are obviously different, which is what puzzles me.

   Thank you very much！

Hi User5877261,

You are correct about the equal sign and that is a mistake in the document.  I also made a mistake in my quick calculations. The code 2264792 would be appropriate for a gain of 1 and a gain of 2 should be approximately double that number or 4529584.

The voltage values agree with 250uA of IDAC current and 3900 Ohms Rref, but the output code results do not make much sense, as you have noted. It is interesting to note that the output code is approximately 1/2 of the expected, and the internal reference is approximately 2 times greater than the external reference.  According to the register information you sent, the register settings for register 1 is appropriate, but I'm wondering if the register has been written correctly.  The default register setting would use the internal reference which would explain why the codes are off from what they should be.  Can you verify that the registers have been written correctly by reading back the contents?

Best regards,

Bob B

Hi Bob B

      Sorry, we have a holiday, so reply late, your reply let me find the root cause of the problem, thank you so much! It's true that the value of register 1 is not written in, which causes the received value to have problems. I modified the delay after the write command to ensure that the read back data is consistent with the written data! After that, the data read back is normal. In addition, can you understand that the ad value interpolation table provided by the example program is a data table under the configuration of PGA = 2, Rref = 3900 ohm parameters. If the value of PGA is changed, then the table data can also be applied by multiplying or dividing by proportion?

 Thank you very much！