Why is there cross-talk between channels on my USB6229 when they are enabled?

[BCM-CERL]

I have a USB6229 DAQ that reads a voltage fine when only 1 channel is active.  If I activate any other channels, no matter if they are near to the first channel or not, it affects the original measurement, even if nothing is plugged in to the additional channels.

Please help.

[Bueller]

Hi,

What you are experiencing is something known as ghosting. If you do not wish to see this effect, tie the inputs of the other channels to ground. If the analog inputs are set for differential mode (DIFF), tie both the positive (ACH+) and negative (ACH-) terminals of your open inputs to analog input ground (AIGND). If the analog inputs are in non-referenced single ended (NRSE) or referenced single ended mode (RSE), tie your open inputs to AIGND.

For more troubleshooting information, check the following KnowledgeBase

Regards,

Faris A

[bcmasters81]

Hello,

Grounding the terminals as recommended does not appear to help.  After changing the order of how the channels are read, putting a 1 ms delay in between reading the channels, the "ghosting" appears to be minimized.

The new issue now is that one of the channels has a perpetual DC offset, irrespective of what is connected to it (even a short).  The two channels in question have high voltage probes and compensators on them.  Even when the probes ares grounded to themselves, or each other, one channel still has a sizable offset, which, when amplified (these are 1000x attenuation probes, so I scale the incoming signal by a factor of 1000), usually amounts to 80V.  That means the "signal" is around 80mV DC, which doesn't seem to come from anywhere outside the DAQ.  Again, even without anything connected to this channel, this offset exists.

Any thoughts?

[BCM-CERL]

I have also noticed that some of the other devices connected to the DAQ will cause the apparent voltage on the high voltage probe to shift.  We have a pressure gauge that sends an analog signal to the DAQ, which will raise the voltage on the probes.

Help!

[RT4CY]

I have a couple questions for you.

-  What are you using to read your incoming signal (LabVIEW, Test Panels, etc.)? 
-  What channel are you seeing this behavior?
-  Have you seen this behavior in the past, or is this a new occurrence?

One thing you could try is performing a Reset and Self Calibration.  Open up Measurement and Automation Explorer (MAX), and select your device under DAQmx Devices.  Click on Reset, followed by Self Calibration shown below.

[bcmasters81]

Hello,

In answer to your questions:

-  What are you using to read your incoming signal (LabVIEW, Test Panels, etc.)?

LabView 8.5

-  What channel are you seeing this behavior?

Analog channels a4 and a7 for now, a1-a3 also have issues occasionally

-  Have you seen this behavior in the past, or is this a new occurrence?

This is a new occurrence since starting to use a4 and a7

One thing you could try is performing a Reset and Self Calibration.  Open up Measurement and Automation Explorer (MAX), and select your device under DAQmx Devices.  Click on Reset, followed by Self Calibration shown below.

Thanks,  I will try this immediately.

[BCM-CERL]

I tried resetting the DAQ as well as recalibrating it.... no change.

Thanks for the tip though.

[RT4CY]

What is your input terminal configuration (differential, RSE, etc.)?  If you are using differential, make sure you use bias resistors to provide a path to ground for bias currents.  This Developer Zone article provides more information about this, as well as several measures you can take to reduce noise from your system.

[ChrisR.]

---

@BCM-CERL wrote:
The two channels in question have high voltage probes and compensators on them.

---

Can you give a model number or schematic of these probes and compensators? If they have long time constants internally, you could still be experiencing a ghosting problem. Does the problem go away if you measure only one channel at a time (i.e., don't scan multiple channels)?

Chris

[PsJ]

I have same problem with my USB6211.  But I belive it's due to the very high impedance of my source, a photodetector.  If the impedance is at same order of magnitude as the impedance between to channels, you would see signal cross talk. 

Any comments on that?

Regards

PsJ