Real-Time Measurement and Control

cancel
Showing results for 
Search instead for 
Did you mean: 

NI9215 Failure Modes

Solved!
Go to solution
I have been running a cRIO system at a customer site for over a year with no changes in the configuration made in software or hardware during that time.  Everything was fine.  All of a sudden, the customer reported seeing very noisy data combined with a DC offset on the NI9215 module.
 
Troubleshooting seemed to indicate these were two separate problems.  After checking for loose connections with no effect, I replaced the rather long, circuitous input cable to the NI9215 with a much shorter one and the noise was practically eliminated.  The DC offset remained, however.
 
The DC offset is roughly 5.9% over the full positive scale of the unit.  I wrote a small patch to deal with the offset, but, the next day, the offset went away and the patch was no longer required.
 
Although there have been no changes made to this test setup, there have been a number of large-scale changes to the building recently (Pumps added, others removed, new AC units put in, etc).  The 9215 is connected as per page 13 of the manual (single-ended).  The temperature of the room does not seem to be changing much, but I am also considering local temperature changes to the module as being the culprit (cracked solder joint + heat = intermittent failure, sort of thing).
 
I realize this is rather thin info to base guesses on, but does anyone have any ideas what could cause this problem?
 
Thanks in advance.
0 Kudos
Message 1 of 4
(2,898 Views)

Unplugnow,

As you know it is hard to pin point the source of the noise occuring on a line between a device and the PC. It sounds like you have taken the right steps to start to reduce the noise by using a shorter cable as well as assested some of the envirnomental variables. These new pieces of equipment in the facility could very well be causing the spradic noise that you are seeing. To futher reduce this noise I would recommmend, if possible, changing your wiring from reference single ended to floating differential (as shown on page 12 of the manual). Other tips to reduce noise can be found in the "Solving Noise Problems in Measurement Setups" section of the Field Wiring and Noise Considerations document linked below:

 Field Wiring and Noise Considerations

Hopefully this points you in the right direction to reduce your noise and let me know if you have any questions/comments.

Thanks!

Cheers,

Jonah
Applications Engineer
National Instruments

Jonah Paul
Marketing Manager, NI Software
0 Kudos
Message 2 of 4
(2,886 Views)

Ok, thanks.  The shorter cable is doing the trick for noise, so I am not going to change anything else there.  I am more concerned with the 5.9% DC offset that appeared all of a sudden, and then disappeared the next day.  It might be indicative of a 9215 module about to go bad, but I don't have any hard evidence for that guess - the modules seem pretty bullet-proof.  I have taken some voltage readings at various places throughout the system case this happens again.  I will re-post if any new information comes my way when I find out for sure what the problem is/was.

0 Kudos
Message 3 of 4
(2,880 Views)
Solution
Accepted by Unplugnow

It looks like the failures were a result of a connected PXI-6704, not the cRIO module itself. 

 

The error came back for the 4th time in about 2 years but this time did not clear.  We went to swap out the PXI-6704 board (input to cRIO module) with a known good one and saw some pretty spectacular board-level damage.  It is clear that a small area of the PXI board was on fire at one point, and there are other areas of probable heat/current damage. 

 

The damage appears to be centered around a blown diode, but we are not sure of the exact cause.  There does not appear to be any damage to the PXI chassis itself or the adjacent cards.  We are forced to assume the board just gave up the ghost.   It is amazing that the only indication that there was a problem of this magnitude during operations was a very rarely incorrect voltage output from the card.  Smiley Happy

 

I will post again if this DC offset error ever comes back once the card has been replaced.

 

0 Kudos
Message 4 of 4
(2,529 Views)