Variation in thermocouples (+ - 2°C) and these are under the same conditions

[Jesús Hernández]

Hi, my english is not the best but I trust in your intuition and intelligence ability.

Actually Im connecting multiple thermocouples (J type) in a DAQ 9213. I managed to take all measures (7 thermocouples) and also the thermocouples are calibrated. My problem is that the 3 thermocuples give me a 20°c (60F) and the others 4 give me 22°c = 71.6°F, and all the thermocouples are under the same ambiental conditions.  The easy solution is to put an offset and compensate the error but it is not a good solution.

Can anyone tell me what is happening? I thinks it is wiring problem maybe the COM, but I have referenced all the negative signal (Low signal) to the COM,  also tried not referencing the negative signals to the common but the result is worse because the signals go crazy.

Please help me, please!

[johnsold]

Have you tried interchanging the thermocouples at the DAQ device to see if the difference is in the channels or in the thermocouples?

What is the construction of the thermocouples? Bare wire or sheathed? Do the thermocouple wires make electrical contact with the object being measured?  Do the two groups (20° and 22°) have anything in common with regard to what they are measuring, routing of the wires, or possible exposure to interfering electromagnetic fields?

I agree that wiring may be at least part of the problem.

You said that the thermocouples are calibrated.  Did you do the calibration or was it done by another party?  The standard error for Type J thermocouples is about 2°.

Lynn

[Henrik_Volkers]

In addition to Johnsold:

Same batch of TCs? 

[Julio9 · Accepted solution]

Hi Jesus Hernandez

Also remember that the thermocouples habe a postive and a negative side, so if you connect them wrong the temperature will be different and instead of increasing when something hotter si close it will decrease. 

Best Regards

Julio Mena

[Jesús Hernández]

Hi Julio :

Exactly, my mistake was think that the white wire was the negative (low) and the red wire was the positive (high) then. The error was not big because also I had the negatives referenced to each other, so I was making two big errors. Now the thermocouples are giving me an almost negligible variation.

I appreciate all your help.

Jesús Ignacio Hernández Cruz.