Thermal gradients in electrical paths of thermocouple conductors

[erichp]

I recently inherited a problem related to temperature measurement.

Thermcouples (d- and k- type) are being used in a high-pressure, high-temperature environment, with the measuring junction suspended in an electrically-insulating solid. The leads of conductors A and B extend out of the solid and, each is held in contact with a separate tungsten carbide cylinder. Around each cylinder, there is a press-fitted steel ring. From that steel ring, wires lead to the instrument where voltages are compared.

I realize this setup is non-ideal for several reasons, but the one I'm most concerned about is this: Each cylinder system is liquid cooled, and it's possible that the temperature at a given point on cylinder A is different than the temperature at the corresponding point on cylinder B. I'm concerned that the difference in temperatures along the electrical paths could alter voltages and lead to inaccurate temperature measurements.

Has anybody dealt with anything similar? Do you have suggestions that can help me get accurate temperature measurements?

Please forgive my ignorance on the topic - if more information is required, please let me know.

 

[Yuri B.]

I think the wires going separately gather unwanted signals - the "noise" .