I am getting improper internal CJC from either SCXI-1303 or TBX-1303 under MAX.

[WTSIII]

I am running MAX under WinXP (and Win2000). When I configure two type K thermocouples on a TBX-1303 (or SCXI 1303) and use the built-in CJC, I get a low reading (-120 C). If I select user specified 25C, the reading is good and the TC's track the finger test. One odd thing is that the reading using the built-in CJC starts about correct, then shoots very negative. The one single time it seemed to work correctly, it started very positive (+120 or so), and then shot into range (25 C or so) for the remainder of the trace. Several hours later, the original problem returned with no change to the hardware or software configuration. What is going on? Why doesn't the internal CJC work? One other p
oint. If I read the CJC directly with a string from Labview, then the value comes in correctly. MAX just cannot read the value to use for compensation.

[RIO-Todd]

Are you sure that your thermocouples are good? It sounds like the junction of the thermocouple might be loose. Do you have other thermocouples to try? Also, make sure that you have the leads of the thermocouples connected correctly to the SCXI-1303 (positive and negative leads). Which SCXI module are you using with the SCXI-1303 terminal block?

Regards,

Todd D.
Applications Engineer
National Instruments

[WTSIII]

Toddd, Thanks for the response! I have 10 Omega thermocouple assemblies model labeled "P01L7 17K4FZ" at least 4 of the 10 behave the same way. I believe they are good because when I specify a cold junction of 25C, the thermocouples will track the temperature of my finger. The problem appears to be in MAX's use of the CJ temperature. I should say at this point, that the graph from MAX seems to show the first reading good, and the rest saturated low. (I can see a ramp from close to the correct value toward -117) The NI field engineer said that he thought the software install was bad, but I have uninstalled everything and reinstalled it, to no avail. If you read the thermocouple voltage and thermister voltage into labview and manually compensate it, the
n the TC reads correctly.