Hi, I am measuring temperature of a metallic box using K type thermocouple probe along with MAX6675IC to digitize its output.For this I am placing probe in the cavity of metallic box but here's the problem,As soon as probe is placed my temperature readings goes to 0C.I suspect it an issue of short circuit of probe with the metallic box.Suggest some way to insulate the probe but still measure the temperature or suggest any other new sensor which can measure the temperature of box.

 

your t/c supplier will show you

 

A bare, uninsulated (exposed wires) junction is the fastest responding thermocouple junction but a bare junction is subject to a ground loop if the object that the thermocouple touches is at a potential different than the 'ground' potential of the measuring circuit.

I suspect that the ground loop is enough to drive the indicator downscale which happens to be zero Deg C.

Any sort of electrical insulation that prevents the junction wires from contacting the metal box (or liquid in the box?) will prevent the ground loop, like electrical tape (assuming this a temporary test situation).

For a along term solution, an MgO thermocouple is an assembly with the thermocouple sensor inside of a metallic sheath,but which can be insulated from the metal sheath when it is manufactured as an "ungrounded" unit. The Magnesium oxide insulating powder insulates the T/C junction from touching the metal sheath.

Any insulation will add thermal lag time to the measurement of temperature changes; laws of physics.

 

K Type Thermocouple probe short circuited with metallic body
You need to show a picture of your T/C, and explain how you have attached the metal short to the t/C wires

 

use a mineral insulated thermocouple and you will not have a problem.
See: https://cascade.net/en/thermocouple/

 

The majority of thermocouple probe you buy is short circuited with metallic body !!! If you need a isolated, you need specify with your supplier.