• State Verified Answer
  • Replies 4 replies
  • Subscribers 893 subscribers
  • Views 362 views
  • Users 0 members are here
Options
Related

Sigma Delta ADC Differential input disconect detection

Juan

Hi all,

I am using R7FS1JA783A01CFM microcontroller with 4 sigma delta differential adc channels from SD00-03 (4 channels)  with 4 termocouples. Each Termocouple has the next design in order to have 3V in the entry of the positive side to get the signal to saturation in case the sensor has been broken.


However, the value is the same as if the termocouple was without voltage or with 0 degrees. The desing seems to be okey but the signal value is not. I have been checking the manual and I get to I get to the paragraph "33.3.11 Disconnection Detection Assist Function" however the solution seems similar to what i proposed but using a single ended entry.

Why I am not seing the result in saturation if I have 3V in the entry of the disconected sensor?

  • 0

    10M is a very large resistor.  I didn't look it up but I am sure the data sheet for the ADC will ask for a lower input impedance.  Try a much smaller resistor, like 10K as a test to see if this is indeed the problem.  Then you can investigate the optimum size.  Even if the ADC can a fraction of a microamp, you could easily lose that much current to leakage paths.

  • 0

    10M is a very large resistor.  I didn't look it up but I am sure the data sheet for the ADC will ask for a lower input impedance.  Try a much smaller resistor, like 10K as a test to see if this is indeed the problem.  Then you can investigate the optimum size.  Even if the ADC can a fraction of a microamp, you could easily lose that much current to leakage paths.

  • 0

    First of all:

    Thanks for your response & .
    In the case of Mark´s proposal, I changed the resistor from 10M to 0,5K impedance but It didn´t solve my problem.

    However, I have been making different tests and it seems the problem is due to the use of GAIN AMPLIFIERS.

    Currently, in the S1JA application we are using 4 channels (SD0:3). We want to use a maximum value of 25 mV. Each channel has a gain of 32U.   --> GTotal(32) =  G1(4)*G2(8) to take advantage of all the values to have more resolution.

    When the sensor is disconnected I have 3V en P and 1V in N

            - When SDADCCHn amplifications are (4,8),  we don´t see the signal in saturation mode. SDADCn ~= 10u --> BAD

            - When SDADCCHn amplifications are (1,1),  we see the signal in saturation mode.  SDADCn = 32767.      --> GOOD

    I don´t know why this might be hapenning or what should be the steps to follow...

    • For the moment I will try to stop the adc, change the amplifiers and check if we are in disconnection mode. Then go back to previous gain configuration and start the adc.
  • +1

    When calibrating the SDADC24, the differential input voltage must be in the range shown :- 

     

     

    With 3V in CHnP and 0V in CHnN, the differential input voltage will be outside the range allowed for calibration.

     

    Calibration must be performed when using differential input mode, the first time after a reset (or stored calibration values loaded) :- 

     

    That´s why when we use calibration in the ssp it is not working properlly. If we deactivate it, it works without problems.
    -------------------------------------------------------------------
    The response was given by the Renesas Customer Hub.