Why would a particular analog input channel cause other on that card to drift?

[RicK.]

I have 4 RTD probes and a torque sensor connected to an SCXI-1122. Whenever the torque sensor is included in the I/O device list the values from the RTD probes drift about 5 - 7 degrees C. If the torque sensor is not in the device list the temperatures are steady (only ~ .3 degree fluctuation. I have experienced this with my own VI and with the Collect N Scans example VI. Why would this occur? I have attached the MAX support file. The SCXI-1122 is connected through a PXI-6052E which is in a PXI-1010 chassis.

[Filipe A.]

Rick;

If amplitude of the analog values is way different, the instrumentation amplifier of the DAQ device might need an extra time to settle properly, and that might cause the values to drift of.

One thing you can try is to increase the interchannel delay in between channels (not the sample rate). That will give some extra time to the instrumentation amplifier to settle properly.

Hope this helps.
Filipe A.
Applications Engineer
National Instruments

[AVR]

Is it possible that the torque sensor is a grounded source and the RTDs floating sources? Unless all inputs are the same reference, floating or grounded, I have found that mixing them on the same device will cause interchannel interference (ie a ground loop). I have been told that it is not possible for this to occur; however, I have six different data systems in my lab (PC/PCI, PXI, DAQCard, MXI-3, embedded controllers, SCXI ... you can imagine the variations), and I have never been able to mix floating sources and grounded sources on the same input interface.