C
I have calibrated 7 k type thermocouples that were used in an autoclave at 350°C for 12 months
Each thmcpl was around 6°C high. I check with brand new thmcpls and my kit was okay.
I did some research and found this:
“The reasons for the instability in Type K thermocouples are due to some inherent properties in the chromel/alumel material.
One problem that occurs with this thermocouple is an effect called short-range ordering. It occurs in a temperature
range of about 500°F to 1020°F (260°C to 549°C) when nickel and chromium atoms in the chromel leg tend to form an ordered crystalline
structure. The ordering produces a different metallurgical structure and if a temperature gradient exists, an erroneous
EMF is produced”
It doesn’t say over what period of time this may occur
So my question is: is it normal for the k types to drift that much over such a short period of time if used at 350°C
Great site.



Each thmcpl was around 6°C high. I check with brand new thmcpls and my kit was okay.
I did some research and found this:
“The reasons for the instability in Type K thermocouples are due to some inherent properties in the chromel/alumel material.
One problem that occurs with this thermocouple is an effect called short-range ordering. It occurs in a temperature
range of about 500°F to 1020°F (260°C to 549°C) when nickel and chromium atoms in the chromel leg tend to form an ordered crystalline
structure. The ordering produces a different metallurgical structure and if a temperature gradient exists, an erroneous
EMF is produced”
It doesn’t say over what period of time this may occur
So my question is: is it normal for the k types to drift that much over such a short period of time if used at 350°C
Great site.


Hi, thanks for the advice. Interestingly enough this has come back to haunt me with a new project manager on the job. He has asked how we can dictate when any of the large number k types has drifted enough to produce erroneous results. My answer is to shorten the calibration period but the tests that are run in the associated autoclaves run for two years. I also said we need to find a pattern in the failure rates ie age of thermocouples. Another suggestion I made wast to use one r type amongst the k's in each autoclave. This could high light the amount of drift in the k's themselves. However I know r types are very expensive. I'm at a loss to suggest anything else now 