Originally Posted by
Mike W1
I've saved this post that'll tell you how good they are for out purposes.
Before I retired from Zeiss, at least 2-3 times a year I had to engineer systems that would measure the exact temperature of
machined parts prior to measuring their dimensions. (Engine blocks, heads, etc. in Big 3 auto plants.)
So I've done more than my share of study of temperature measurement methods. Let me say categorically that these
non-contact "laser" measurement systems are almost worthless. They measure infrared radiation from the target, and that
radiation varies wildly with what is called the "emissivity" of the surface.
The industrial suppliers of these things claimed that they could tell me a temperature within 2-3 percent of displayed
value, IF we sent a large number of samples to them so they could measure the emissivity and calibrate their sensors. Any
time a sensor required replacement, they had to go through the whole rigamarole again. Any time the surface finish
changed, ditto.
Without calibration, and given unknown surfaces, they wouldn't guarantee accuracy to better than 25% of the displayed
value! And that's industrial grade equipment costing thousands of $$. Bottom line, if you calibrate a Harbor Freight
instrument against an accurate contact thermometer, you may be within 5% for the exact object you calibrated for. Any
other object, you can be wildly off.
BTW the laser has nothing to do with the measurement process. It's just an aiming aid.
Best method is a low-mass thermocouple or RTD device, (thermistor). These are not expensive. Some under-$100
Volt-Ohm-Meters come with a thermocouple probe. I have had two of these, and they worked pretty well. I've never stuck
one into melted lead, but there are versions that are encapsulated in ceramic which I would.