Thermocouple offset error w/ CJC

[dragel2014]

Hi everyone,

I am having an issue with my thermocouple reading in my PCI-6221 DAQ. I am setting up my SCB-68 accessory to use the onboard CJC compensation, and when I check my readings, it seems to be 5 or 6 degrees higher than what it suppose to be (checked with a multimeter). Here is my system info:

PCI-6221 M series DAQ

SCB-68 accessory with CJC enabled (in differential mode)

K-type thermocouple connected to AI4 and AI12 for differential reading w/ negative side shorted to AI GND

DAQmx 9.9

Labview 2010

I've checked my DAQmx parameters and things look all good. Any clue what this might be caused by? Is this a calibration issue? 

[Henrik_Volkers]

What do. you read, when you replace the TC with a short wire?
Should be room (CJC) temperatur...

[dragel2014]

It looks like it is measuring the same temperature as my thermocouple (which was at room temperature). Warming up the onboard sensor with my hand seems to have a response. Would this be an issue with calibration then? Shoud I put in a software offset?

[dragel2014]

With some more tinkling and reading ("How Can I Read the Value of the Cold Junction Compensation of the SCB-68 (SCB-100)?"), I found that my cjc thermometer seems to be working okay (reading 21C, which is what I expected), but shorting a port together and to AI GND didn't yield similar results and nor did shorting it to AI SENSE.

[Kelly_B]

Hi dragel24, 

Are you checking these readings in NI-MAX or in LabVIEW?  

Also, did you set your thermocouple to be K-Type in NI-MAX?

[dragel2014]

Yes, I was using NI-MAX, and I set it to K-type in there. Also set it to use built in CJC on the SCB-68 and configured the switches.

[Kaleb_A]

dragel2014,

May I ask why you are shorting the negative terminal to AI GND? This would not be a differential measurement anymore if you do this.

Also, here are some more information in this document: http://digital.ni.com/public.nsf/allkb/96B996BAD82A84E9862570690066B03F 

[dragel2014]

Anderson,

hmm... I have grounded the negative side to ground because it was to "prevent the differential voltage readings to drift out of the adc range". I have previously removed it, but didn't see any changes. On top of this offset, i also find my signal rather noisy. I know my thermocouple wires are a little long, but i have not see the same with another Agilent DAQ that I have.

[Kaleb_A]

What input range do you have your DAQ set on? The smallest voltage range for this device is ideal, +-0.2V, because the K-type thermocouple outputs 41 µV per °C voltage change. Any other voltage range and this 41 µV per °C would be undetectable. 

[dragel2014]

I have it configured to use K-type thermocouple inputs. Would changing this to voltage input and doing the calculations myself help? You would think that by setting it to read K-type it would automatically configure to use the right voltage range. I have set the expected input temperature range from 0 to 100C.