Signal Conditioning

Showing results for 
Search instead for 
Did you mean: 

thermocouple gives wrong reading


I'm using thermocouple type T. I set up board PXI6221 board with Labview 7.1.

I can read 0C temperature +/-1.5C in ice water bath. But when I take the same ice water bath at distance of 15m, the temperature is around 10C. Does cable length gives 10C difference?

The Lavbiew DAQ for thermocouple have CJC "constant" set at 25. When do I change this?

Should I set CJC as "constant" and fill in room temperature reading?


0 Kudos
Message 1 of 6
Are you not using any signal conditioning Modules in between thermocouple and DAQ card??
Consider SCXI 1102 series with appropriate terminal block
also for using thermocouples at a distant location, you have to use proper compensating cables to connect
0 Kudos
Message 2 of 6
Hi.  I'm using SCB-68 box supplied by NI.  I connected 8 thermocouples to it.  First thermocouple goes to AI0 and AI8, second to AI1 and AI9, etc. The SCB-68 manual says to flip dip switch S5 in the box, but this does not work. The first thermocouple is way wrong, So I set default factory settings.
 I looked into SCXI 1102, but I'm afraid I dont have the time or the money to buy it. So I have to use what I have.
I just used copper cable #24 to connect thermocouple from rig to scb-68 box 15 m (~60ft) away.
Can I adjust CJC in labview so I read temperature to match room temperature?
I noticed that if I do that, the higher end of temperature is read lower than actual temperature. But I dont expect to reach temperature anyway.  
Sincerely and thanks.
0 Kudos
Message 3 of 6
You mentioned that you were using copper cables to extend the wiring. I suspect that may have something to do with it. Thermocouple create very small voltages based on temperature by using a junction of dissimilar metals. You may have created 2 more junctions when you connected the thermocouple leads up to the copper wiring. Ideally you should use type T thermocouple extension cable.
Also have you set up your scaling for a T type curve and are you using lower voltage input ranges on your DAQ card? I would expect the signal to be 100mV or less.
You could also be picking up induced electrical noise from your long run of copper cable.
"It’s the questions that drive us.”
0 Kudos
Message 4 of 6
In your post you mentioned that you are having some trouble getting a thermocouple to read accurate readings when connected directly to a Data Acquisition (DAQ) board without any signal conditioning.

Just as a quick refresher of how Thermocouples work, I would like to encourage you to check out this tutorial:
Taking Thermocouple Temperature Measurements

Essentially you will see that a Thermocouple measurement will be a very small signal (max of about 20 mV), requires a cold junction temperature, and because of the signal size it is subject to a lot of noise leakage.  Other factors such as an incorrect CJC value could cause problems with the actual accuracy of the device.

Essentially in your situation, the first thing that I would do is recommend that you have some sort of signal conditioning for your thermocouple (such as devchander mentioned in your forum post).  If you cannot afford to get extra signal conditioning, then make sure that you have the temperature range for your task set as accurately as possible for the expected values.  That will allow the maximum possible resolution to be used.  The CJC temp is given in Degrees C, so make sure that it matches the temperature of your terminal block as closely as possible (typically room temperature).  Finally, the noise on the longer leads could lead to the increased problem with the temperature signal.  The only way to really get rid of noise would be to use signal conditioning or oversample and average the measurements.

Considering the information that you gave in your second post, I think that AnalogKidtoDigitalMan is right about copper wire posing a serious problem with your measurement.  Definitely review the above tutorial to best understand how thermocouples work with measurement systems.

I hope some of the tips help you find a better solution than what you have now.  If you have any other specific questions about this then feel free to let me know.
0 Kudos
Message 5 of 6

Sorry to post here but haven't found out how to post new question -


I have a thermocouple system set up in rev 5 using a SCXI1102 with 1303 terminal blocks and 120 thermocouple type E.  This runs fine.

I also have another PC with rev 8.5.1 that I am upgrading the rev 5.1 program.  When I run this new PC I get three different levels of temp.

The only difference between the set ups is from the cable out of the SCXI box and the PC's.  I am using the internal CJC temp.


The three different leves of temp are -

The first group reads temp fine, just as the old setup.

The second group is about 3°C higher

The thrid group is about 7°C higher.


Why the differnet temp levels??

0 Kudos
Message 6 of 6