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.

Thermal drift in the ADS1231

Part Number: ADS1231
Other Parts Discussed in Thread: , ADS1232
Hello Bob, I am doing tests with the ADS1231, I have based on the previous experience with the ADS1231REF.
I use a load cell of 50kg for a maximum of 30kg capacity, my circuit works well, it has a precision of 1gr, I have implemented the circuit calibration option which I do with a calibrated mass of 5kg.

After doing the calibration, I weigh my calibrated mass of 5kg and it gives me the exact weight. My problem is that when starting the circuit the next day there is a deviation of the weight that can
be 1gr or -1gr by weighing the same calibrated mass of 5kg. After 2 hours more or less, the deviation is corrected, I think you have to see the fact that the circuit is cold and the temperature of the
room.



  • Hi Manuel,

    Just so that I'm clear, are you using the ADS1231REF for these experiments, or do you have your own board?

    One consideration is when calibrating with with a 5kg weight, you are only calibrating 1/10th of the total range of the 50kg load cell. Gain and offset errors will be much more pronounced because of this calibration. Generally you would like to calibrate at or near the maximum weight to be applied.

    Most likely the issue you are seeing is related to some sort of thermal drift which could be related to either the load cell or the ADS1231. Discovering which one is causing the biggest issue may be challenging. Instead of using the 5kg weight as calibration, I would suggest using raw codes and do a no load and 5kg load measurement and track the raw codes. Compare the results for when the system is warm, and then compare to a cold system.

    Best regards,
    Bob B
  • Hello Bob, I work with my own board.

    I understand what you tell me that I'm only calibrating 1/10 of the capacity of the load cell. From now on I will do it with a much larger calibrated mass (20Kg or 30Kg), I hope better result.

    I'm going to look at the raw codes a bit difficult, but I'm going to try.

    Assuming that the thermal drift has to do with the ADS1231 integrated circuit, would you get better results with the ADS1232 integrated circuit?

    regards

    Manuel D.

  • Hi Manuel,

    It is difficult to speculate on what you are seeing and if a device change would improve the result. If the drift is load cell related, changing the ADC will not make any difference. If the drift is device warm up related, it may help to use the ADS1232 which has a self-offset calibration. If what you are seeing is related to a combination of system warm up and the calibration method then improving the calibration may help.

    One other thing that may be affecting a perceived offset may be related to the reference. If you have a lot of filtering on the reference or reference path that differs from the excitation, there could be a settling issue of the reference. I would be glad to review your schematic and layout to see if there may be something related with the design that could be causing this.

    Best regards,
    Bob B
  • Hello Bob, I have worked more in the software than in the hardware, my schemes are focused on the power and the ADC module, they are 95% extracted from the ADS1231REF. Do you want to do it privately?

  • Hi Manuel,

    You should be able to click on my picture icon and start a conversation with me privately. Just attach the information and send it to me.

    Best regards,
    Bob B
  • Hello Bob, thank you very much for your help.

    regards

    Manuel Díaz