VirtualBench

[RickRick]

We are collecting temperature data on 5 to 7 channels.   Occasionally one or more channels will become very noisy and the data collected will be either 18032, -17968 or a real (or close to real) temperature.    I know that 18032 indicates an open channel.  What is the significance of -17968?     If one channel has this problem, it obscures the chart so that the non-noisy channels cannot be easily viewed.

Any suggestions to fix the problem?

[Jeff_T.]

Rick,

It is difficult to say for sure what -17968 means without knowing what device you are using.  I have seen behavior similar to this before and it almost always has to do with a bad connection with the thermocouple.  If you are confident in your connections could you please tells us all the hardware you are using to take the data and what type of thermocouple you are using?  Can you also state how / if you are implementing cold junction compensation and what drivers you are using with the device?  Thanks Rick.

Best Regards,

Jeff Tipps
Applications Engineer
National Instruments

[RickRick]

Jeff,

 Thanks in advance - sorry for the late response.

We are using Virtual Bench with NI4350 for ISA (PXI/Compact PCI).  We must be using the board's internal cold junction compensation.  I could not find a product name or number on the juction board.

Currently we have 2 type K tcs, 2 type R tc's and 3 type S tc's hooked up to the system. 

Hopefully this will help you.

Regards

Rick

[Jeff_T.]

Rick,

Like you said earlier, open thermocouple detection will pull-up your connection to above 100mV, giving you a very large temperature.  Since thermocouples represent negative temperature with negative voltage then your channel must be getting coerced to some large negative voltage.  Besides a bad connection, I wonder if your device is damaged in some way.  Can you please tell me how long you have owned this 4350?  Has this device worked in the past and is this a recent issue?  Do you have another 4350 that you can swap out with this one to see if the problem persists?  Thanks Rick.

Regards,

Jeff Tipps
Applications Engineer
National Instruments

[RickRick]

The NI 4350 has been here for many years, in a dusty environment.   We also tend to have voltage brownouts, blackouts and likely spikes.

We are planning on replacing the entire system sometime this year (I'm slowly gearing up for this and am attempting to learn somewhat from our past mistakes...)

To cut to the chase then, it appears that we either have a wiring problem or a problem with the 4350.   

Unfortunately, we do not have another unit to swap in.

The problem has been with us off and on for a year or more I believe (no one pays a whole lot of attention to it as it is so problematic).

Thank You

Message Edited by RickRick on 01-08-2007 03:40 PM