Readings will shift or return NaN after some random amount of time.

[mgrigg]

I put together a system comprising of a SCXI chassis and several analog input modules to monitor mainly temperatures and a few pressures. The system ran fine for a couple of weeks non-stop with no problem. As of late, the system now give erroneous readings after a random time (from 20 minutes to several hours). All voltage readings shift or report NaN. Stopping the program and restarting will correct the readings for a random period of time. Labview 6i, Win2000

[RIO-Todd]

Greetings MGrigg,

I have some questions for you that will help me clarify your issue a bit more.

1. What SCXI modules do you have? Do you have an SCXI-1200 module?

--I have seen issues with the SCXI-1200 parallel port module returning a NaN. The NaN on the SCXI-1200 is most likely because the parallel port timed out during the transfer. Make sure you have the parallel port of the PC configured correctly (see the link below).

Parallel Port DAQ Device Troubleshooting
http://www.ni.com/support/daq/comprob/2000/partrb2k.htm

2. Are all of your signals floating?

--If you are acquiring from floating signals and aren't using bias resistors with the SCXI modules, the readings could start correctly and then eventually float to +/- the limits of the hardware.
They shouldn't report NaN though. SCXI modules will either have jumpers in them that will let you configure the module for a floating or grounded signal or the SCXI terminal blocks will have resistor packs that will allow you to add ground referencing for floating signals. Check out the user manuals for the modules/terminal blocks for more information on this.

Those are a couple of questions/suggestions that I have for now. Let me know if this does/doesn't help or if you have any additional questions. Have a great day.

Regards,

Todd D.
NI Applications Engineer

[mgrigg]

Thanks for your reply Todd!

The SCXI modules I am working with are the 1102B (with TC2095), 1102s and 1124s. The TC2095 is set for floating thermocouples & open TC detection. The signals on the 1102 are currently not connected. The inclusion / exclusion of the 1124's does not seem to have any effect on the problem being experienced.

The SCXI chassis is cabled to a 6034E DAQ card.

The code is set to stop on any error but no errors occur (i.e. see standard buffered acquisition routine).

I am now in the process of replacing each major system component one at a time to see if I can isolate the source of failure.

Any further suggestions would be greaty appreciated!

Mike G.

[RIO-Todd]

Hey Mike,

What module is the 6034E device cabled to? It should be cabled to one of the 1102/1102B modules and not to one of the 1124.

Are you using virtual channels to convert the thermocouple voltage readings to a temperature or are you reading the voltage directly from LabVIEW and converting to temperature within your program? I'm wandering if you have decreased the voltage range of the analog input to provide you with the maximum gain.

Creating virtual channels in Measurement and Automation Explorer (MAX) takes care of setting the gain for you. Within the data neighborhood folder of MAX, you can create thermocouple virtual channels and set an input temperature range for your thermocouples and the gain of the SCXI module and 6034E will be s
et to maximimize your resolution. If you aren't using virtual channels I would suggest to create one and see if you run into the same problem.

I hope this helps. Let me know if you have additional questions.

Regards,

Todd

[mgrigg]

Hi Todd!

The 6034 is cabled to the first slot which contains an 1102B module.

I am reading the voltage directly in labview and converting it to temperature. I use the INPUT LIMITs node of the AI CONFIG to set the gains. Correct?

I think I will try the virtual channel angle, at least to become familiar with it (old dog, new tricks).

Since we have last communicated, I have swapped the SCXI chassis and it modules with another almost identical unit (1 more 1102 and 1124) and the program has been successfuly running for 1 week now without experiencing the problem I encountered. I will now set the suspect chassis and modules up with another computer and start pulling modules one by one to see if the problem follows a specific module.


Thanks for your input, I will try your suggestions!

mike.

[Tom.]

Hello,

How was this issue resolved? Im seeing a similar issue with the same hardware.

Cheers

Tom