I think using the glow plugs as thermocouples for cylinder combustion temps is interesting.
I am quite intrigued by the concept, although I will have to think about it a bit to determine exactly where the "hot" and "cold" junctions really are, as I have to consider the glow plug itself canceling out it's internal Seebeck effects as it's base screw insertion into the block is apt to be within a few degrees of it's top terminal, albeit the top connector of the glow plug is a dissimilar metal than the engine block.
Note that thermocouples make use of the Seebeck effect to measure temperature differentials:
DuckDuckGo. Privacy, Simplified.
duckduckgo.com
So far, I see this as a roundabout way of measuring block temperature where the electrical connection is made between glow plug and the block ( hot junction ), which is between two cold junctions ( engine block to copper test lead ) and another thermocouple formed at the connection between glow plug power connector and the other copper test lead.
I believe this will work, as far as giving a reading as the engine heats up, but I see a problem with the other unintentionally formed thermocouples formed at all the junctions of dissimilar metals of unknown alloy making such a mess of extraneous Seebeck inputs as to render this approach untenable, when one could just use an off-the-shelf type K thermocouple with known Seebeck characteristics and thermal environment of the "cold junction".
Just about any joining of dissimilar metals makes a thermocouple...but the ones you buy ( e.g. type K ) use known stable alloys with thoroughly researched calibration curves that relate temperature difference to the voltage difference they develop. Thermocouples are made from many different alloys...for our uses, type K is commonly used.
But I find it intriguing nevertheless.
I have a tiny soldering iron for microcircuit assembly whose tip heater temperature is controlled by this method. They appear to be using two dissimilar metals bonded together as a heating element, then about once a second, the current is switched off for a few milliseconds and the millivolt Seebeck voltage is measured, compared to setpoint, then latched. If Seebeck voltage was below setpoint, another shot of current ( approx 1 second ) is delivered, if above setpoint, no shot, waits a second then tests again. The iron runs from a 19 volt DC "brick" same as those that power laptop PC's.
