How can I wire a thermocouple to monitor wood stove catalytic converter?

Question

Many modern wood stoves (wood space-heaters, actually) use a catalytic converter to dramatically increase efficiency and dramatically reduce pollution. It's important for users to monitor the temperature of the converter (also called a combustor), primarily because it must be brought up to an initial temperature of 500 degrees-F or so before engaging the bypass damper that forces the smoke to go through the combustor (as opposed to bypassing it and going straight out the chimney).

The temperature is normally monitored with a crude thermometer based on a coil spring and a probe that extends a few inches down into the stovetop just above the combustor. I've had considerable problems with these devices failing, and I also think it'd simply be fun to use a more modern approach. Thus, I'm considering the use of a thermocouple and solicit thoughts on the viability of this approach.

I've found a device called the "Type K" thermocouple, that seems like it could fit the bill. They can handle temperatures of at least 2500 degrees-F; the combustor normally operates in the 1000-2000 degrees-F range, and almost never above 2000 (furthermore, guarding against such excessive temperatures is another reason for monitoring temperature). They have a sensitivity of about 20 microvolts per degree-F, easily monitored.

https://assets.omega.com/pdf/tables_and_graphs/thermocouple-type-k-fahrenheit.pdf

I suppose the biggest concern is protecting the wiring between the thermcouple-proper and the measuring device (a cheap DVM, or perhaps an Arduino), as well as the measuring device itself. The latter problem can be solved by distance, as the wiring would be carrying no current to speak of, and the voltages are high enough (roughly 10-50 mv in the range of interest) that EM interference is unlikely to be of concern.

Which leaves the wiring. The Type K thermocouples seem to be geared towards kiln users. These have comparable temperatures to the wood combustor, but the exterior is well-insulated from the hot area, whereas just the opposite is the case for a wood heater. So it's not clear if the wiring can survive the stovetop temperatures, and if not, how it might be protected. Here's an image which shows the basic situation. The gloved finger points to the tiny hole where the probe would be inserted. I imagine the wiring extending to the rear (near the flue collar where the stovepipe attaches) to the block wall behind the stove (tan stucco, barely visible in the photo) The top of the stove and the flue pipe should not exceed 1000 degrees-F or so.

No doubt it'd work to suspend (in the air) the wiring between the themocouple and the stucco'd block wall that surrounds the stove, but it'd be awkward and unsightly.

So, is the some way of providing the interconnect that can easily survive the stovetop temperature ? Any other concerns about the efficacy of this approach ?

Answer 1

Thermocouple wiring must only be extended using the same material which the Thermocouple itself is made of.

Since a Thermocouple is formed simply by the junction of 2 dissimilar conductors (nickel alloys alumel & chromel for Type-K), using normal copper wire to extend your Thermocouple will actually create another 2 Thermocouples - one in series with each wire - which could have a significant impact on the accuracy of your readings.

So for your application, you need to find thermocouple extension wire with insulation which is rated to withstand the 2000F temperature - it'll probably be glass-fiber mesh since silicone and teflon don't go that high.
You'll also obviously need to make sure that the Thermocouple itself is made with insulation rated for that temperature!
If the connection between the Thermocouple wire and the extension is also going to be subjected to high temperatures then you'll probably need to use something like ceramic-body screw terminal blocks. In this case you'll want to make sure that the wires from each side extend all the way through so that you're only really relying on the terminal block to mechanically clamp the wires to each other (as opposed to having it become "part of the circuit").