As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
Wiring a thermocouple to watch a catalytic combustor means three parts that all have to match: a type K or type N probe sitting in the exhaust path just past the combustor, extension wire of the same alloy pair running the whole distance to the instrument, and a readout with a thermocouple input and cold junction compensation. The probe generates its own signal and needs no power. What fails in practice is almost never the probe, it is a copper splice, a reversed connector or an insulation that melts against hot steel.
What a thermocouple actually sends down the wire
A thermocouple is a junction between two dissimilar metals. Heat the junction and a small voltage appears along the wires, roughly proportional to the temperature difference between the hot junction and the point where the two alloys end. For a type K pair that sensitivity sits around forty microvolts per degree Celsius, which means the entire useful signal from a combustor running hot is a few tens of millivolts.
Two consequences follow immediately. First, there is no such thing as a thermocouple reading on its own: the instrument must know the temperature of its own terminals and add it back, which is what cold junction compensation does. Second, a thermocouple circuit measures a difference, so every place where the alloy pair ends and something else begins becomes part of the measurement whether you intended it or not. That single fact explains most of the failures described below.
Where the probe has to sit to see the combustor
A catalytic combustor only does its job above a light off temperature, and above a further limit it is being damaged rather than used. Both numbers belong to your stove, and the manual that came with it is the only honest source for them. What the wiring question turns on is placement, because the same combustor can produce three very different readings depending on where the sensor is.
A magnetic thermometer on the stove top reads the temperature of the steel, filtered through the mass of the plate and delayed by minutes. A probe in the firebox reads flame and air mixing, which swings wildly. A probe in the exhaust stream immediately downstream of the combustor reads the gas that has just passed through it, which is the quantity the light off and overfire limits are actually defined against. If the stove has a factory probe port, that port is the answer. If it does not, the decision to make one is a decision about a certified appliance, and a stove shop is the right place to take it.
Mounting hardware matters more than it sounds. A probe pushed through a hole and taped will move, and a probe that moves gives a reading that changes when the blower runs. A compression fitting or a thermowell holds the junction at a fixed insertion depth and seals the opening so flue gas does not leak into the room, which is both a measurement requirement and a safety one.
Wire connectors and the junction you create by accident
The signal chain is short and every link can corrupt it in a way that still produces a plausible number on the screen. That is the dangerous part: a wrong temperature does not look wrong.
| Link in the chain | What it has to be | What a mistake looks like on the display |
|---|---|---|
| Probe junction | Type K or N, sheathed, rated above the gas temperature expected | Reading pinned at the top of scale, or an open circuit message once the sheath fails |
| Extension wire | Thermocouple grade wire of the same alloy pair, not copper | A reading that drifts as the room behind the stove warms up |
| Connectors | Thermocouple grade in the matching alloys, correct polarity | A reading that falls as the stove heats, or sits near room temperature |
| Insulation | Fibreglass or ceramic fibre near the stove, never plastic | Melted insulation, a short that reads the wrong point on the wire |
| Routing | Away from hot steel and clear of mains cable runs | A noisy reading that jumps when a pump or fan starts |
| Readout | Thermocouple input set to the same type, with cold junction compensation | A good probe and a reading tens of degrees out, consistently |
Polarity deserves its own note because the colour convention surprises people. Under the North American code the negative conductor is red, which is the opposite of nearly every other wiring convention in a house. Reversing a thermocouple connector does not usually produce an obvious error at room temperature; it produces a reading that moves the wrong way once the probe gets hot, which is how it ends up discovered during the first real fire.
What the readout has to be able to do
Four capabilities decide whether an instrument is useful here: a thermocouple input of the right type, cold junction compensation, a range that covers flue gas temperatures, and a settable alarm so the number does something when nobody is looking at it. Logging is the fifth and it changes the value of the whole exercise, because a combustor question is about a curve over an hour, not a value at a moment.
No general indoor air monitor does any of this, and it is worth being blunt about that because the search results mix the two categories. What indoor instruments do show is the shape of the problem you are solving. The Autopilot CO2 monitor and controller states a remote sensor probe on a fifteen foot data cable precisely so the sensing element can sit where the measurement is and the display can sit where a person stands, which is the same layout a stove probe needs. Its maximum sensor temperature is not stated in the listing bullets, and in any case it senses carbon dioxide rather than temperature, so the parallel is one of arrangement rather than function. Treat it as a diagram of the wiring problem, not as a candidate instrument.
Room side instruments that answer a different question
A combustor probe tells you whether the catalyst is working. It cannot tell you whether the stove is spilling into the room, and that is the measurement with safety consequences. Two separate channels cover it.
Carbon monoxide is the first. A stove with a blocked flue, a cold chimney or a negative pressure problem in the house puts combustion products into the living space, and the human nose does not detect the gas that matters. A fixed alarm to the relevant standard is the baseline, and a portable instrument is useful while commissioning or fault finding: the NOYAFA JMS19 four gas detector states detection of hydrogen sulphide, carbon monoxide, lower explosive limit and oxygen with light, vibration and sound alerts and a response within half a second, which is the behaviour you want from something clipped to a belt while you work on an appliance.
Particulate is the second. Loading a stove, opening the door against a weak draught, or running a combustor below light off all put smoke into the room, and the effect shows up as a particle count long before anything smells wrong. A counter such as the Eujgoov particle counter states six size channels from 0.3 to 10 micrometres with particulate mass concentration and storage for thousands of records over a universal serial bus connection, which is enough to compare a reload against a steady burn. Choosing an instrument for that job is covered in choosing a particulate monitor, and a stove also quietly consumes room air, which a carbon dioxide and climate monitor will show as a rising trace in a sealed room.
Putting the catalyst temperature to work
Install the probe where the manufacturer intends, wire it with matching extension wire and connectors end to end, keep the insulation rated for what it touches, set the readout to the right type and then do one verification before trusting any of it: bring the probe to a known temperature, boiling water at a known altitude or an ice bath, and confirm the display agrees. That single check catches reversed polarity, the wrong type setting and a copper splice all at once, and it costs ten minutes. The same discipline applied to other sensors is described in calibrating against a known reference.
Then use it as a curve rather than a number. Record a full burn cycle, note where light off happens on your stove with your wood, and you have a reference to compare future burns against. A simple temperature and humidity logger in the room alongside it gives the context, since the same load behaves differently in a cold house with a cold chimney. The probe tells you about the catalyst, the room instruments tell you about the room, and running both is the only arrangement that answers the whole question.
Frequently Asked Questions
Which thermocouple type suits a catalytic combustor?
Type K is the usual choice because its range comfortably covers combustor temperatures and probes and wire are easy to obtain. Type N is more stable at sustained high temperature and type J has too low a ceiling for a flue. Whichever you pick, the probe, the extension wire, the connectors and the readout setting all have to be the same type.
Can I extend the probe with ordinary copper wire?
No. Joining copper to a thermocouple alloy creates a second thermocouple junction at the splice, and the instrument then reads the difference between the hot probe and that splice rather than between the probe and its own compensated terminals. Use extension wire of the same alloy pair all the way to the meter.
Where should the probe sit on a catalytic stove?
In the exhaust stream where the gas leaves the combustor, which is where the stove manufacturer places their own probe port if one exists. A thermometer on the stove top reads steel temperature and lags the combustor badly. The manual for your stove is the authority on the port, the depth and whether drilling anything voids the certification.
Does a thermocouple need its own power supply?
No. It generates its own small voltage from the temperature difference across the junction, in the region of tens of microvolts per degree. What needs power is the instrument, because converting that voltage into a temperature requires an amplifier and a measurement of the terminal temperature.
Can an indoor air quality monitor log the combustor temperature?
No. Indoor monitors read air temperature with a thermistor or a packaged humidity and temperature sensor, with a ceiling far below flue temperatures and no thermocouple input. They are useful around a stove for a completely different reason: carbon monoxide and particulate in the room.