Experiencing increased channel noise levels on PXIe4497 with 1071 fan on HIGH

[rtaub]

I am using a PXIe-4497 to collect microphone data and when I switch the PXI chassis 1071e fan to HIGH setting, the noise level increases.  The chassis is outside of an isolated test chamber, so the noise isn't occuring acoustically.  Please see the FFT plot below of fan on AUTO (white trace) and fan on HIGH (red trace).  Has anyone had this issue?

[rtaub]

There was some noise getting the pass through into the chamber.  We added more sound insulation to block it out.  Here's what it looks like with the added insulation (FAn on HIGH, white trace is with added insulation):

But as you can see there are ~256Hz harmonics.  The data below is for MIC Y which is last channel the PXI-4492 (first module).  The difference between AUTO and HIGH settings is now:

Here is MIC Z which is first channel input on the next module PXI-4497:

Auto and HIGH fan settings are close to being similar for MIC Z>  Here are MICs X (white), Y(red), and Z (green) plotted together with fan on HIGH...X and Y are on PXI-4492 (last two inputs) and Z is on PXI-4497:

As you can see MIC Z doesn't have the 256Hz harmonics.  We swapped the Z with the Y channels to see if it was MIC or channel related.  Below is MIC Z in last input channel of PXI-4492 (where MIC Y was originally):

The noise follows the input channel, not the MIC.  Here is the other side of the swap...MIC Y in first input channel of PXI-4497:

No significant change now between Auto and HIGH fan settings.

This seems to be input module related....

What now?

[Roohcifer]

Hello,

Unfortunately, the images you posted in your previous reply are appearing as broken links.  I'm not sure why this happened, but I think I understand what you are seeing.  Still, could you summarize exactly which channels on which cards and which slots of the chassis exhibit this behavior?  (For example, ai0 of the 4497 in slot 2.)  Could you also answer these questions:

1.  Does this behavior replicate in all slots?  This may help us determine if it's an electrical issue or a fan vibration/proximity issue.

2.  Does this behavior replicate with no microphones or cables attached to the modules?  I want to do this to completely decouple the issue from any cable or microphone noise.

3.  What is the part number of your chassis?  There is a known issue regarding faulty fans on certain PXIe-1071 B-revision chassis (i.e., part number 195659B-01L).  You don't have to post your serial number here, but the entire list of affected chassis' serial numbers are listed in a spreadsheet in this KnowledgeBase article:  http://digital.ni.com/public.nsf/allkb/B9BB6E483EDB5353862578180081FA6A?OpenDocument

If your chassis is on the affected list, we can pursue replacing the fan (at no cost).  Else, we can troubleshoot accordingly depending on the answers to your other questions.

Regards,

[rtaub]

Hello David R:

Seems the pics I posted didn't show.  The attachment shows the issue.

We disconnected the mics and took data.  Noise still on MIC X and Y, and not on Z.  Noise is worse with FAN on HIGH. This is shown in the last two slide of attachment.

Chassis info:

PXIe-1071

P/N 195659B-01L

S/N 1697700

Regards,

Rob

[Roohcifer]

Hi Rob,

Thanks for the images.  Please do the following:

Unplug the mics AND cables from the devices such that there is only the card in the chassis.  Run FFTs on ALL channels (not just the three mic channels) to see if this noise is prevalent on other channels.  Do this for BOTH cards in ALL chassis slots (not just slots 2 and 3) to see if there is any variation in behavior.  I saw that you called in to report this to another engineer, so I will let him continue troubleshooting with you so we don't duplicate efforts.  He spoke with me about this since he saw this forum, so we'll discuss this issue together offline and keep communication going through your current service request.  Regardless, please perform the above experiment so that we can find out if the magnitude of the problem is slot-specific and in any way related to noise picked up by the cables.

Regards,