Babar,

Thermal calibration is usually done poorly. I worked at a bozo-company a while back that did their temperature testing by placing their product in the temperature chamber for a quick 15 to 30 minutes which is enough time to take only the air to target temperature. As a result, their product temperature specifications weren't worth the paper they were printed on.

Also, the inaccuracy of the calibration process would worsen your results if you took two dat samples and attempted to curve fit your tables on that.

I've seen considerable differences from measureing temperature by various devices in the lab. I would say you more than likely would only be adding chaos.

I would agree that a sanity test should be made.

aka J