QuoteFrom that I got an idea that (mainly for MFMP and other replicators), should someone try to separate thermocouple further from too hot body by purpose. For example attach wide and thick enough (half?)ring of clay or something around body and inject thermocouple there half way. Temperature reading of the thermocouple would be lower, but would protect it better. I think reading then follows some pretty accurate heat transfer/heat dissipation formula and thus can be calibrated and compensated (at least) in post test calculations. That way replicators could have control readings to verify Optris readings. Clay was just one example, as long as material have some thermal resistance and can stand the heat.
Hi Argon. This is quite doable, and it is possible to construct accurate calibration curves for a 'cooler zone' thermocouple, something I have done often. The trick is to runs a 'slow and steady' calibration at a range of temperatures, comparing the cool zone indicated temperature with that of one placed inside an un-fuelled reactor. After several runs 'up and down' to amass plenty of data points for both a warming and a cooling system it is my preference to take the hot-zone temperature up to the point where that particular thermocouple is destroyed all the while taking data-points from the cool zone- it is a small price to pay for good calibration.
ETA - here's information pertinent to this - http://www.lookingforheat.com/…g-a-model-t-lenr-reactor/