USB 6211 causing Voltage shift

[Cablejon]

We are using a pCIE 6321 card for Data Acquisition with a BNC 2090A. The BNC 2090A has a 100kΩ Bias resistor installed and we are using NRSE mode single ended. I have two problems one is we have a very clean signal in RSE mode and a much noisier signal in NRSE. Our programmer has set the program for NRSE so we try to use that mode. second problem is we have started using a USB 6211 device to output a wave form signal to a servo control card for a hydraulic system. this signal is just an opposite signal to the feed back load cell feed into a summing node in an op amp circuit. the load cell signal is also feed into our Data Acquisition system using the pCIE 6321 card. both the USB 6211 and the pCIE 6321 are on the same computer but we get a voltage shift when the USB device powers on. I can eliminate the voltage shift if I separate out the power from the computer to the USB device and externally power it and make it a floating device. I am not sure this is a long term solution to this problem and I don't understand why I get a ground loop from 2 devices that are powered by the same PC or Computer. any help or ideas would be very welcomed.

[aputman]

I have several USB 6215 and during startup of the PC, the analog outputs have an offset of about 3 volts.  Is this similar to what you notice?  I too would like to know why this happens. 

[Cablejon]

we are seeing a 20-30mv shift on all of our BNC 2090A channels. the common ling is through the servo card that we build. the servo card has a connection from the feedback load cell and the wave form generated by the BNC 6211 device. the load cell output is also feed to the BNC 2090A and from there into the pCIE 6321 card on the same computer.

[bill_gilbert]

There is a phenomenon called noise rectification that can produce problems like these. Noise from something like a motor drive can come back into the computer or DAQ hardware an get rectified across a diode junction somewhere, showing up as a DC offset. Not saying this is your problem, but I have seen it a couple of times over the years. In my case, solved by installing an analog signal isolator module, breaking a ground loop. Neither device showed an offset when separated, when connected it appeared like magic.

When there are multiple devices and interfaces connected to one computer, galvanic isolation is beautiful.

[aputman]

I misspoke.  The offset from the USB device in my case is not 3 volts.  The signal drives a power supply and the supply output is 3 volts during the PC startup.  The analog offset is probably only about 250 mV.  As soon as my software is started and takes control of the supply programming, the offset disappears.  I've not figured out what causes it.

[bill_gilbert]

Aputman, sounds like a Local/Remote mode problem to me. The supply can only be controlled from one source at a time, the front panel controls on the supply, or the remote computer. The controls are probably set fro 3v in local mode, and you see that until the computer takes control through whatever port you are using.

that's my guess.

BTW you should start a separate thread for your problem.

BG

[aputman]

I was posting to see if this was related to the OP's problem.