cDAQ9188 zeros out all inputs

[ahhah]

My cDAQ9188 zeros out all inputs after running for a hour or so.  Sometimes after an hour.  Sometimes after 4 hours. The zero values then return to their normal values within 3 seconds (my log is recordeing every 3 seconds).  

There are 5 active modules. 2 X NI 9213 Thermocouple module (not using 1).  1X Analog In NI9205. 2XRelay9485. 1XDigital Output9477.

I'm not certain if the  digital outputs or relays return to 0 but the thermocouples and analog values do.

Any ideas? 

This is vexing since I've set a number of alarms to fire when the 0 appears.  I could implement some sort of averaging or code to reject the 0's but I'd rather resolve the hardware problem.

[USBarnes]

ahhah,

What kind of environment are your devices in? Is there anything that happens during these times that may correlate to these zero values? Is it possible for you to try your code in another 9188 chassis to determine if it is chassis specific or software specific? Also, are you using the onboard clocks on the devices or an external clock? Lastly, is the 9188 connected directly to your computer or is it connected to a USB hub/switch? Sorry for all the questions, but I'm just trying to gather a little more information. Thanks!

[ahhah]

1.  Environment?  Temp / humid controlled lab?

2.  Tracking temps and pressures and switching valves show no correlation to the zeroing.  The zeroing may not happen for 36 hours and then might happen after 2 hours monitoring the same stuff.

3.  Not easy to switch out the 9188.  We did switch out the temp module to no avail (same issue)..

4.  I'm not sure I know the difference between onboard and external clock.  The devices are timed with LV but perhaps the temps are being pulsed? 

5.  Ethernet.

[USBarnes]

ahhah,

I just realized that I should have phrased a couple of questions better. My apologies. When you set up your analog input task, is it set to Sample Clock? You'd probably know if you were using an external sample clock source. As for my question regarding the ethernet connection between your 9188 and your computer, is the 9188 directly connected to your computer through the ethernet cable, or is it routed through a hub or on a local network? Maybe we are seeing intermittent Ethernet connection issues. As you said, averaging a few values would smooth out the zeros. Do you have a way of monitoring the inputs of the cDAQ modules to see if it is the sensors that fall to zeros or the 9188? Could it be an intermittent connection issue between the sensors and the device? You might also try monitoring just one card at a time to see if the behavior persists.