Thermocouple Operation, Ice-Point Correction???

[DC712001]

Trying to do this on-the-cheap (out of necessity) and though I have the EMF/Temperature-conversion chart, and have read literature on roughly how it works, how do you go about correcting/calibrating for the Ice-Point junction?

Do I attach the thermocouple to an ice-cube, take a Milli volt-reading, then add, (or is it subtract) that voltage from what I would later read from the T/C when attached to the hot load/object in the furnace?

Am I completely off base? If so, how do you use the T/C or make it work?

Edited October 15, 2008 by DerekC

[Walking Dog]

Derek -

Thermal voltages appear at the junction of any two dissimilar metals. This means that if you make one thermocouple, say iron wire with constantan wire, you will get a thermal voltage at this junction, plus two more where the iron wire and the constantan wire are attached to your millivoltmeter. The reading will then be the sum of all three voltages.

The fix is to make two junctions, say iron-constatntan and then constantan-iron, bringing out two iron leads to the meter. The two junctions formed at the meter are the same type each (probably iron-copper), and at the same temperature, and produce equal but canceling contributions to the measurement.

The two measuring junctions are then used, one at the unknown temperature, the other in ice water at 0 degrees C. The measured voltage relates to the temperature DIFFERENCE between the two measuring junctions (one of which is at the known temperature of ice water).

You can avoid the mess of ice water by putting your reference junction at just room temperature, independently measuring the room temperature, and adjusting the measured temperature difference accordingly.

Omega Engineering has an excellent free reference book on thermocouples and other temperature measuring devices. Browse their web site at omega.com.

[DC712001]

Thanks WD.

[Ice Czar]

good advice ^


if for some strange reason you wanted to DIY calibrate really tight
DIY Triple Point Calibration + - Pro/Forums

of course the further your calibration point is from what your target temperature range the more error you'll have. Calibration at the zinc freezing point (787 F) would be ideal but not nearly as easy at either ice water, the triple point of water or boiling water.

there is alot more info to be had with the search query
"cold junction compensation"

Edited October 17, 2008 by Ice Czar