The power indicators on the chassis along with the controller go out and system locks off.

[GC_-_LF]

We have a number of PXI-1036 chassis' with PXI-8102 controllers, that intermittently fail. The power indicators on the chassis along with the controller go out and system locks off. There is no control of the mouse or keyboard and the monitor screen is black.   Unable to power cycle to reset system.  Only by removing the PXI-8102 and pressing the reset switch (SW2), will the system return to normal operation.   These systems are at customer locations and pulling cards to reset is not something we want them to do.

Is this a problem with the controller or chassis?  

Is there an easier way to reset the system?

[Danny_Kaps]

Hello PJZ,

I think we are going to need more information to pin down the source of this issue. My first thought is that if this is happening reliably on multiple dispersed chassis/controllers, it is unlikely to be the same hardware issue in every case, and as such might be more related to what the devices are doing when they fail.  Here are some questions to start:

1. What defines "intermittently"? How often is this issue occuring, and what are the controllers doing when they fail?

2. Is this a new issue or has it been occuring since initial setup? If the former, did anything in the setup change before the first incidence of this failure?

3. You mention that the systems are dispersed at customer locations, how many are there total, and are they all having the same issue? It is possible that this may be related to the power coming into the chassis, so if only a subset of the deployed systems are experiencing this failure, can we isolate some common factor to only those setups?

4. Are these controllers running the same Windows OS they were shipped with, or was it changed to something else?

If we can somehow reliably reproduce this failure state, that would be ideal for troubleshooting. That being said, this seems like an issue that would require some dedicated digging by an Applications Engineer; if you have a service contract I would recommend opening a service request for more convenient follow-up (http://www.ni.com/support/).

Best,

Daniel 

[GC_-_LF]

Hi Daniel,

There may be a number of months of normal operation before a failure occures.  The test systems, for the most part, were not in use, but powered up.  These test systems have only been deployed for around a year, and there have not been any setup changes made since the initial installation.  There are over 10 test systems at various locations world wide, they are all identical, with the same HW and supplied OS. There have only been 2 systems that exhibted this failure, to date.  It does appear that the input power is linked to this failure, since we were able to link it  to severe weather and power outages.  We also noticed that even though there may be a relationship to "bad" power, it did not adversely affect other test systems on the same power line.  It appears that some NI PXI-8102 controllers are more susceptible than others, to flucuations with input power.   We have witness this failure at our location, with one of the returned failed systems,  after a power outage due to a storm in the area.  Other than that we have not attempted to simulate this problem.  We are in the process of arranging an NI Applications Engineer to look into this further.   Pulling the controller card out of the chassis, to reset this failure is not something we want our customers to do.

Thanks,

Pat