DAQmx 9172/9477 modules lose PC connectivity

[DEppAtNGC]

After some time, the USB driver seems to lose contact with the 9172 modules. I don't think there is a power glitch becuase other instrumentation seems ok... but for no apparent reason, the 9172 modules (specifically the 9477, but also others) are no longer active/valid when our LV application tries to access them, and the Measurement & Automation Explorer shows them as inactive. When power is cycled on the 9172, the USB plug and play beeps occur and everything is live and operational again.

Any clues about why this is happening?

Here is the error message the application pops-up:

Measurements: Device cannot be accessed.  Possible causes:

Device is no longer present in the system.
Device is not powered.
Device is powered, but was temporarily without power.
Device is damaged.

Ensure the device is properly connected and powered.  Turn the computer off and on again.  If you suspect that the device is damaged, contact National Instruments at ni.com/support.

Device Specified: cDAQ1Mod5

Task Name: _unnamedTask<0>

[Jason_D]

DeppAtNGC,

Do you have this plugged into the back of your computer?  Does the computer go into standby or hibranate?  If for any reason the device loses power this will occur.  The last issue I had on this the guy plugged his chassis into a monitor USB connection and the monitor was powersaving and even though the computer was always active the device still lost connection.

If you can narrow down when exactly this happens it will help the troubleshooting process immensely.

[DEppAtNGC]

The 9272 is plugged directly into the PC using the supplied USB cable. The power is plugged into what is supposedly site filtered power for the lab - no other instrumentation / PC's exhibit any problems / reboots when this hang condition occurs, so if it's power related, it's because the 9172 is very sensitive.

I was originally thinking it may be a hibernate / standby issue, but another user told me that it occurred during active use of the system. I think I've only seen it after returning to work "the next day" / or similar. So I'm tentatively going to say NO, it is not related to a PC standby condition. Even if it was, it should come out of standby cleanly like the USB keyboard, mouse, and other peripherals do.

Is there any knowledge base issue related to the 9172? Is there a suggested monitor/re-initialization test I can perform short of cycling the 9172 power?

[Jason_D]

DEppAtNGC,

Here is a forum post with a similar probelm:

http://forums.ni.com/ni/board/message?board.id=250&message.id=46253&requireLogin=False

You might want to try the process located in this KnowledgeBase Aritcle.  If this works I would form it into a batch file.

http://digital.ni.com/public.nsf/allkb/1D120A90884C25AF862573A700602459

Does this problem happen on other computers too?

[DEppAtNGC]

HI Jason,

Thanks for noting this... From the article, it sounds similar to my issue... for some reason, the 9172 is losing connectivity. The Microsoft tool simply works around the problem rather than solves it. Another easier work-around is to cycle the power as we're doing now.

What I'm looking for is for NI to discover the root cause and correct it so we don't need a work-around. Again, the PC seems to stay connected with the mouse, keyboard, and the other instruments don't "reboot" or lock-up. Why does the 9172? It's helpful to understand that I'm not the only one experiencing this, meaning it's not something specific to our hardware or site.

Could you please escalate this problem within NI?

Thanks very much for your excellent help so far!!

Best Regards,

David

[Jason_D]

DEppAtNGC,

I can try to escalate this for you, but I will need some more specific information from you mainly:

What is the make and model of the computer with the problem?

Do other computers near you have similar problems?

Is there anything consistent about when this problem occurs?

Is there any more details or anything else you can think of that may help us to replicate the problem?

Replication of the problem has been the main hinderence to fixing this problem so far.  So your answers to the above questions will help significantly. 

I understand what you are saying about the mouse or keyboard not locking up and agree that this should not happen, but I think comparing a complex data acquisition instrument to a keyboard is a little unfair.

[DEppAtNGC]

We are using an HP Compaq dc7800p P/N KA594UT#ABA (Core2 Duo E8300 2.83GHz w/2GB RAM), with Windows XP Pro SP2.

We have not tried this with any other computer; but apparently the NI user base has, and it fails on the PC's they were using.

I know of nothing that seems to cause this issue to occur.

How many times has NI encountered reports of this issue?

[stilly32]

Hey David,

Just to be clear: do you only loose connection with the NI-9172 when you reboot or power cycle the computer? If this is the case, then we are working on it. As Jason mentioned, we were not able to reproduce this in house until very recently when a customer sent in a computer for us to test on. We are looking at the issue and will keep this and other threads updated. 

If you loose connection to the device in other cases, please clarify what cases (if you can). 

thanks, 

Andrew S.

Multifunction DAQ

Product Support Engineer

[DEppAtNGC]

Hi All,

This is Not related to a reboot or power cycle.  It happens randomly with everything powered up. The PC is not set to sleep; although the display may be going into screensaver mode. One claim is that the system was being actively used and it occurred when the PC had been idle for only about 5 minutes; although I did not personally see that issue. Read my previous posts carefully for more detail.

Thanks,

David

[lost_in_translation]

Hello Folks...

let me add my problem to the mix and see if maybe someone has some input or similarity?

I have a burn-in rack that's using TestStand 4.0 with NIDAQmx ver 8.6.1f0. It was working fine for a number of months and would burn-in my generators for the full 24hrs without a problem. Now, we're seeing an intermittent failure that causes the program to halt and wait for an input from the 9172...when I come around to check it, I have to kill all the threads in TS and restart to get the thing going again. I try starting up MAX to check on the 9172 and it's three components (9205, 9401 and 9477) and MAX freezes during startup. In ether condition, if I turn off the power or unplug the 9172 device, the programs will continue on. In Teststand,  I'll get the following error or one just like it but with a different .vi, associated with unplugging or resetting the 9172 after it's frozen.

Test Sequence Error:

DAQmx Create Channel (DI-Digital Input).vi:1

Device Specified: cDAQ Mod2

Task Name: _unnamedTask<107CD>

MEasurements: Device cannot be accessed. Possible casuses:

Device is no longer present in the system.

Device is not powered.

Device is powered, but was temporarily without power.

Device is damaged.

Ensure the devicce is properly connected and powered. Turn the computer off and on again. If you suspec that the device is damaged, contact National Instruments at ni.com/support

After closing (killing) the TestStand interface, and starting MAX but having to cycle the power on the 9172 device, MAX will not show the modules as being active. I'll try and perform a reset on the 9172 and I'll get a Error -88705 and I'll have to restart the computer to get the whole thing up and running again. the initial problem is very intermittent...this morning it ran for an hour, last night it ran for 15 hours...

here are the stats on the setup:

the computer is a rackmounted system built around the following system: http://www.logicsupply.com/products/sys7677

it has embedded XP on it and just the required drivers for the software.

the modules contained in the 9172 are the 9205, 9401 and 9477. they talk to 4 independant power generators via a support connector.

the 9172 is plugged into the CPU directly.

 I've replaced the computer with a spare, back-reved my teststand software, replace (one at a time) the 9172, 9401 and 9205...with no fix in sight. I'm getting ready to replace the 9477 now

anyone have any ideas? I'm quickly running out of components to swap and my production line has been down now since late last week...

thanks in advance.

-Scott Frushour