Must be a trick spot welder to weld those hot junctions.
So - next challenge for Rob 'the miracle worker' Wilson - can you come up with a suitable rig (chrome plated of course) so we can all make our own?
Dave -- You don't
need to weld TC wires together (though I agree that it
is better and more accurate if you do), you
can just coat them thermally conductive grease, twist them together, and squash them in a press for
intimate contact. Probably half the TC's currently measuring experiments in R&D labs around the world are assembled this way.
Hopefully I have not lost it (a paper -- not my mind), but I have instructions for making a calibrating an
optical pyrometer that we used to measure hypergolic plumes a few years back. Other than gaining access to qualified high temperature test standards, it is a simple device that only costs a few $$$ to build (under $15 as I recall from 2005). As my unit was built using customer's parts on time they paid for, it ended up staying with the customer after my work was done.