Thermocouple readings are opposite to whats expected

[Nadim_Virani]

Hi all,

We are working on a project utilizing type K thermocouples and the NI- USB 6210 DAQ and when we view the readings in lab view, the readings seem to be opposite from what they should be. For example, when we place the thermocouples over a heater, the readings show that the temperature is actually decreasing, while a seperate infrared thermometer is showing that the temperature has actually increased. 

Any help on this issue would be greatly appreciated. Thank you for your time.

Nadim Virani

Aerospace Engineering Student

California Polytechnic State University, San Luis Obispo

[AdnanZ]

Have you wired the thermocouple correctly? What values do you get if you use a DMM to measure the raw voltage (if that's possible to do)?

[Nadim_Virani]

Thanks for the response, it turned out to be a wiring issue.  However, now we are experiencing a lot of noise resulting in a 1.5 degree C error in each of our thermocouple readings.  Any advice on how to reduce thermocouple noise would be appreciated. Thanks again in advance for any help.

[AdnanZ]

Well, I woud suggest having a look at Table 1 (Analog Input Connections) at the following page:

Field Wiring and Noise Considerations for Analog Signals

I suspect that you are using DIFF measurement, so I would recommend using bias resistors when connecting the wires (which is mentioned in the table). Try it out and let us know.

[acanivell]

Similar problem here, but in my case the offset is much higher, around 10C

I tried to play around with different pullup and bias resistor values, but nothing changed much. I am using:

- SCXI 1102 plus 1303 screw terminal

The resistors I use are the 10Mohm resistors recommended in the manual of 1303 terminal.

Any ideas or things I could try?

[Dan_K]

Hello acanivell,

Have you seen the problem when using a different thermocouple on the same channel? Are you sure that you are specifying the correct type during configuration? Either of these could be the source of the issue.

Another thing you could do is feed in a known test voltage and perform a reading on this. This could help to see if the issue is located in the channel, the thermocouple or the configuration.

[acanivell]

Hello Dan

After a  lot of connections and disconnections I finally made it work:

- As I was using different colouring standards for TC, there was a mismatch of material.

- For the resistors, I finally removed the pull up and set the bias resistor to 10ohm. I do not see however why it does not work with the pull up, because the 1303 manual shows the 10M-pullup & 10-bias configuration as suitable for floating measurements ...

In any case, after these two actions, readings are fine. 

Thanks for the suggestions