Voltage offset between attenuator and DAQ in SCC-68?

[Cranky]

I'm troubleshooting a system that uses a PCIe-6221 and an SCC-68 without much success. There is a constant voltage offset on the analog inputs that doesn't show up with a DMM in the breakout box.

In one case, there is an A10 module in slot 1 (10x attenuator) that shows a ~4VDC offset in LabVIEW (MAX or an executable), which becomes a ~40VDC offset in our program. The measurement is set for differential. Changing to RSE or NRSE increases the offset to ~6VDC. Using a DMM on the differential inputs shows 0 VDC. Unplugging the connector on the A10 module does not change the offset. Removing the A10 module does not change the offset.

In another case, ai2/ai10 shows a 2.2VDC offset in the test panel in differential mode and 10.7VDC in RSE. Measuring between ai2/ai10  with a DMM shows 0VDC. Measuring between DGND and ai2/ai10 shows 15.5VDC on both.

In addition to checking voltages at all of the ground references, I've tried replacing the 6221, the cable, and the SCC-68 and the problem is still present. There is a 24VDC power supply connected to some of the automation components in the enclosure and disconnecting it removes the offset. The 24VDC supply is not directly connected to the DAQ system but does power a hardware "watchdog timer" relay. Removing the relay does not remove the offset.

I'm still learning how to troubleshoot measurement systems, so I've probably overlooked something simple. Anyone have suggestions on where else I should be looking?

Thanks!

[Cranky]

Disregard. It seems that the test bench was set up to connect the ground of the DAQ to the ground of the 24VDC supply through the shield of one of the sensor cables. Apparently there are several internal connections between the shield and ground on both ends of the cable. The 24VDC components were added after it was built, so I'm not sure why it was set up that way.