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.

ADS1246: PGA set to 1 but readings are 4 times to high

Part Number: ADS1246
Other Parts Discussed in Thread: REF5020

we are using ADS1246 for some years now without any problems but recently we encountered many devices which were overdriven.

We checked analog (-2.8V,+2.8V) and digital (3.3V) supply voltages as well as external reference voltage (2.046V, REF5020), which were al ok. Also analog input voltage is as expected. What we found is that the readout of the ADC is exactly 4 times higher than expected. Our conclusion was that maybe PGA is not set correctly. Register 03 is set to 0x02 which should set 20 SPS and PGA=1. We checked SPI interface and the correct bits (0x43, 0x00, 0x02) are sent to the chip. When we reduce the analog input by a factor of 4 the device works properly.

Any ideas?

Thanks for your help.

  • Hi Frank,

    The differential supply range on AVDD on the ADS1246 is specified at 5.25V, so operating the device at 5.6V is not allowed. This is actually outside the absolute max operating conditions of the ADC, so you could be damaging the device by applying these voltages. If that is the case, unexpected behavior can occur, which could be different from lot to lot, device to device.

    I know you said you confirmed that the correct bits are sent to the ADC during the WREG command. Have you read back your register settings after WREG to confirm that they were loaded correctly? 

    In the post you linked to, the engineer had purchased parts from a 3rd-party reseller. As such, we could not guarantee their performance, and it was up to the engineer to take his issue to the reseller. If you would like to provide the topside markings on your devices, we can try to track down where they came from. Please provide a picture of the device(s) in question.

    -Bryan

  • Hi Brian,

    thanks for your support.

    I do not know how I missed the supply voltage to high, even i found 5.5V in the data sheet but still to high. Will prepare a board with +-2.5V supply and unused chip...

    Will also modify sw to read back the WREG, checked only SPI bus so far.

    The devices are marked:

    ADS1246

    U7K G4

    C6XD

  • Hi Frank,

    Let me know if the updates make any difference, or if you are able to successfully read back the registers.

    Can you send me an image of the topside of your devices? Also, would you mind sharing where you purchased them from (and when)?

    -Bryan

  • Have to check with purchasing department where the chips are from...

  • Thanks Frank, I will dig into this and see if I can trace the lineage of this device. In the meantime, if you can find out where they are from, or if you have any update on the technical performance of your board after the changes, please let me know

    -Bryan

  • Had to get my hands on new LDOs to get the differential supply range below the recommended 5.25V, now we have got +-2.5V. Also changed firmware to read FSC register. We made 4 new boards with unused ADS1246 and the outcome is: ADC still overdriven by a factor of 4 and FSCs contain: 0xFFFFC0.

    As the next step we will try to write 0x400000 to the FSC an see if we can get correct readings, although the factory calibration will be gone.

  • Thanks for the update Frank.

    As an update from our side: the ADS1246 topside markings indicate that your device was not manufactured by TI. Therefore, we cannot provide any further support on these products. Instead, please communicate with whomever sold these devices to you.

    I hope they are able to resolve these issues for you, or at least provide you with replacements.

    -Bryan