K Type Thermocouple Probe short Circuited with Metallic Body

[arham1026]

TL;DR Summary

I am Measuring the temperature of metallic box with K thermocouple probe. As soon as probe is place in cavity of metallic box, the temperature readings goes to 0C.

I measure the temperature of metallic box using K thermocouple probe along with MAX6675IC to digitize its output. As soon as probe is placed in the cavity, the temperature drops to zero. I suspect that probe is being shorted with metallic box.Suggest some way to insulate the probe or suggest some other new sensor(contact based )

 

[DaveE]

I used to glue small TCs onto electronics (transistor cases and such) with epoxy. A very thin coating was enough insulation, but didn't interfere with heat transfer too much. Not a "probe" just the wires, which is mostly all we ever used. A roll of TC wire is cheap enough that you can just cut the end off and make a new junction or a whole new probe. It wasn't unusual to see a power supply in temperature tests with lots of TC wires attached.

 

[Baluncore]

Welcome to PF.

One side of the thermocouple pair is grounded, so you should expect a short to result in an offset or noise, rather than zero. Test that condition by grounding the TC probe to T- and to ground.
I expect it is more likely that the TC pair is open circuit, when stretched. Check bit D2 in the serial data. "D2 is normally low and goes high when the thermocouple input is open". Does an open TC circuit return zero data?
Check that T- has a ground connection.