• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 39 subscribers
  • Views 705 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.

TIDA-00650: Accuracy

A K
Intellectual 905 points
Part Number: TIDA-00650

We have manufactured TIDA-00650 boards using the provided design files. I have modified the lookup table and J1 connector for using a T-type thermocouple. Since the voltage output of T-type is within the same range as K-type, the filtering components are unchanged.   

I have tested up to 5 boards with two T-type TCs and experiencing large board-to-board differences in accuracy. The swing is anywhere from 0.15C to 1C with respect to a reference Agilent data logger. I have validated that the board accuracy remains the same by swapping TCs thus eliminating TC inconsistency.

I would appreciate any input on where I should focus my attention to understand the reason behind these large swings. Thanks!

  • 0
    TI__Intellectual 1510 points

    Hello AdtK,

    the variation is only from board to board, correct? when you look just at one board, than the variation is not that much, isn't it? have you looked at the ADCcode histogram? For example taking 1000 samples for a known input voltage and plot a histogram. How does this look like?

    The reason for a board to board variation can be minimized by calibration. Have you calibrated each board? Offset and Gain error calibration should be performed. Offset error can be calibrated also be chopping the inputs. gain error is most likely different in each board. the calibration has to be done for the TC and also for the RTD input.

    I hope this helps!

    Regards

    Alex

  • 0 in reply to AlexW
    Intellectual 905 points
    Thank you Alex! I am using the chopping method for offset calibration. I was expecting variation between boards, just not by a degree so thought I would verify. I will calibrate for gain error per board.