Welding thermouples directly to barrel

While researching thermocouples, I saw some references to welding the thermocouple directly to the surface to get a more direct measure of temperature. While this would be a neat way to do it, I do not understand how this would work. I would think that the signal would be shorted out.

Does anyone have any information on how this works electrically?

... i think they 'make' a TC by welding a single wire of different material to the tube and measure directly the thermo-potential between wire and tube ...

I think the voltage you get depends on the two metals making the junction. Any pair of metals gives you a voltage such that if you get v1 for a given temperature between metals A and B, v2 between B and C then you will get v1+v2 between A and C.

So if you insert a third metal in between so that you have two junctions in series you get the same voltage provided both junctions are at the same temperature, regardless of what the third metal is.

---

[www.hydraraptor.blogspot.com]

I've been using a bit of a hack, where I cut my T-C thermocouple into several lengths so I get multiple thermocouples, and then just pinch the two different thermocouple wires between the nozzle and heater barrel, without welding them.

It's not as solid as a welded thermocouple, but it seems to be just as accurate (gives me the average temperature of the barrel and nozzle, in theory), and it has the huge advantage of not working if either of the thermocouple wires come loose from the barrel/nozzle. In that case, I get a temperature reading out of bounds, and an automatic shutdown is triggered.

Without that sort of safeguard, a loose thermocouple can cause a fair amount of damage, like this:

[builders.reprap.org]

Wade