Are two measurements involved? If so what sensor type (TC type, RTD) , wiring, and transmitters involved.
The only way two sensor can read the same (within uncertainty limits) is if the sensor and transmitters have been calibrated in a lab traceable to a national standard.
I have seen this happening before in kilns and furnaces.
It can have many causes, so in order to be able to help you better, you have to give a more detailed description of what the problem is, which type of thermocouple, type of process, how you measure etc. What device you use to measure. Are you using thermocouple to 4-20mA transmitters etc.
Is the temperature difference always the same in xx °C? Or is it varying?
But.....running completely blind....these are the things to start (and what I would do):
- swap the thermocouples position mechanically
- swap the cables to the measuring device of the thermocouples
- use a calibrator and check of the measurements are ok
- also use the calibrator to check if the thermocouples are measuring ok.
It can also be that the thermcouple cable/extension cable that has been used to connect the thermocouple has been swapped polarity on both sides (I assume you are using the correct thermocouple cable for that specific thermocouple type).
I have seen people using normal cupper wire......and they also did not use the correct thermocouple connectors....
--> don't be that guy or girl!
When you are using transmitters in the head of the thermocouple that convert the measurement to 4-20mA (or voltage, but I would -always- use 4-20mA)
- check if the 4-20mA signal that comes out of the transmitter still corresponds to what it should be
- over time these transmitters can start to deviate and they must be re-calibrated
- transmitters must be set for the zero and the span.
- remember that some analog transmitters have potmeters for zero/span calibration. These potmeters can have problems over time.
Misreadings can also be because of a problem with the coldjunction measurement:
- maybe the coldjunction measurement of 1st thermocouple is near to the connection. And very far away for the 2nd.
- the place where the coldjunction measurement is done (often only 1 for all thermocouples) must have the same temperature as the point where the thermocouple wire goes over to cupper. Any temperature difference there and you will also see that in the temperature
Causes for misreadings can also be caused by the thermocouple itself:
- there can be a interconnection inside the thermocouple tube. The point where the metals connect is then not at the tip of the thermocouple, but somewhere else. Check with a gas flame burner over the thermocouple at specific points where the spot measurement is.
- when you are using thermocouples inside a furnace, it can be that they are put inside via a hole in the side of the furnace. Sometimes these holes are not well closed with fibre and there can be air moving over the thermocouple causing a misreading (often too low temperature measured)
- when you are using thermocouples at a too high temperature, the metallurgy of the thermocouple changes and therefore causes a misreading.
- when you are using thermocouples in furnaces that have atmospheres with for example lead in it (piezo ceramics production), it can be that, over time, some of the particles inside that atmosphere get inside the metals of the thermocouple. This causes a misreading
Ok, here you have some things to check. I hope that will set you on the right path of finding the problem.
Please give us more information if you need more help. And let us know when you found the problem what caused it and how you solved it. This can be of help to other guys and girls that experience the same problems.