/INTLDO pin

Thank you, what will happen if that resistor is not present? Is there a latent failure mode and are there warning signs?

Hi David Ridgway,

Let me look into this and get back to you in the next few days

Have you used the L01 in a design and did not include this resistor, is that why you are asking?

-Bryan

Bryan, short answer, yes. I have been chasing down some unusual behavior in a deployed design and during a schematic review found that this resistor is missing. This may not be related to the issue I am seeing, but I need to run it to ground.

Hi David Ridgway,

While I am waiting for additional info from our design team, would you like to describe the issue you are seeing? I would not expect this missing resistor to cause any significant issues (maybe if you are constantly powering up and powering down this pin), so we might be able to rule out this as a potential problem while we wait for feedback from the rest of my team

-Bryan

Bryan, I would also not expect this pin to be an issue. Currently, this pin is tied to the board return, but I could see how if there is a ground bounce, that might create an issue with the IC if current were to flow into this pin. This IC can get power cycled multiple times on a daily basis.

The issue I am seeing is that the gain of the signal chain is off by almost a factor of 10. The system has a SE sensor AC coupled into a DE opamp which sets the gain and filters, then through an analog MUX to a THS4551 with a gain of 1 to drive the ADS127L01. Output of the first two stages are within acceptable range, but the ADS127 is reporting a lower signal than expected. It is reporting that there is a signal, just that it is almost 10x lower than what is expected. Other systems with this configuration are working correctly.

I am also performing an error analysis of the ADC to see if there is something missing in our conversion process.

David

Hi David Ridgway,

Am I understanding correctly that you have multiple boards - all the same configuration - that are all working except for one? If so, have you tried an A-B-A swap to see if the issue follows the device or the board?

-Bryan

Yes, there are multiple boards (same config) and A-B-A testing has been done and the issue is isolated to one board.

If the issue is that the 10K resistor is needed to prevent damage to the ADS127 and this one board is the first to have a failure due to this resistor being absent, then the approach will need to be changed here as to how this is handled. Understanding what the failure mode of the ADS127 is without this resistor, is needed.

David

Hi David Ridgway,

Have you tried shorting the ADC inputs to see if the noise still meets datasheet specifications? This will give a good indication if the ADC modulator, filter, etc., are still functioning properly

The next step would be to apply a known voltage e.g. 1V, from a precision source directly to the ADC inputs (bypassing the FDA). Can the ADC convert this voltage correctly? If the voltage is way off, I would then probe the reference voltage to be sure it is accurate. Also check if you can pull data from the ADC DOUT pin instead of through your controller to make sure there are no transmission or conversion errors.

I am still waiting for some feedback from our design team, but again the missing resistor does not seem like it could cause this issue. As noted, the EVM does not have this resistor and we have not received any feedback that the boards don't work or give grossly inaccurate data as a result

-Bryan

Bryan, thank you for your suggestions, the ADC has now been ruled out and the issue was resolved.

However, now that it is known that the boards we are using are not per the datasheet recommendations for the ADS127, we still need to know what the failure mode and/or potential issues we could see with not having the resistor installed on this pin.

Thank you,

David

Thank you for the information regarding the resistor.

It was discovered that there was a schematic change that was not documented. The one board that had the error was built using an older schematic, bom and bare board; once this was discovered and the board properly modified, the ADC output was what was expected. Since this effort has started, two more boards were found to have been built to the older documentation.

David