Found an expired 1984 patent for a creosote detector — thinking about building one

[mcjeepers]

Been reading through the monitoring threads here and went looking for prior art before starting anything. Turned up US4486743, "Creosote buildup detector and annunciator," from the early 80s — two temperature sensors with a heater to differentiate them. Plus a chimney fire detector and a couple of stove temp alarms from the same era. All long expired.

Somebody worked this out forty years ago, back when the sensing cost more than the stove, and we're still buying magnetic dials.

I've got a 3D printer and I'm thinking about building one for myself. K-type in the flue, ESP32, local only — I read the thread about services shutting down and I don't want a device that dies when someone's server does.

The thing I keep circling: what I actually want to know isn't the current number, it's cumulative hours below 250°F since my last sweep. That's the figure that predicts what's building up in the pipe, and nothing I've found logs it.

Two questions for people who've built or bought something:

1. Is that the number you'd want, or is live temp plus a high/low alarm enough on its own?
2. For anyone on double-wall pipe — how are you getting a probe into the flue? Cable routing seems to be the part nobody's solved.

 

[EatenByLimestone]

There are a ton of variables involved in creosote formation. One of the largest, and probably most easy to control, is the moisture content of the wood you’re burning. The less water in it, the less condensation that can occur on your flue.

 

[stoveliker]

My probe thermometer (bimetal coil, so no cables required) going in the flue (rather than magnetic on the outside) costs the same order of magnitude as a ESP32 + amplifier and K-type...
Just drill a hole and look at it from across the room. Granted not from another room.

I don't care how long I'm below 250, as one does not make creosote at the end of the burn, when the flue cools down also below 250.
So that cumulative time won't give you anything useful.

 

[Whitenuckler]

Been reading through the monitoring threads here and went looking for prior art before starting anything. Turned up US4486743, "Creosote buildup detector and annunciator," from the early 80s — two temperature sensors with a heater to differentiate them. Plus a chimney fire detector and a couple of stove temp alarms from the same era. All long expired.

Somebody worked this out forty years ago, back when the sensing cost more than the stove, and we're still buying magnetic dials.

I've got a 3D printer and I'm thinking about building one for myself. K-type in the flue, ESP32, local only — I read the thread about services shutting down and I don't want a device that dies when someone's server does.

The thing I keep circling: what I actually want to know isn't the current number, it's cumulative hours below 250°F since my last sweep. That's the figure that predicts what's building up in the pipe, and nothing I've found logs it.

Two questions for people who've built or bought something:

1. Is that the number you'd want, or is live temp plus a high/low alarm enough on its own?
2. For anyone on double-wall pipe — how are you getting a probe into the flue? Cable routing seems to be the part nobody's solved.

I wonder if you had a creosote deposit test area, and a probe to measure thickness of the deposit. I can't imagine a worse place to install something like that. If you look flue with a camera you might see what's going on.

 

[mcjeepers]

Thank you all very much for the prompt and thoughtful responses. The cumulative metric doesn't survive the end-of-burn objection and doesn't seem like something to pursue any further.

 

[stoveliker]

Yet a good temperature measurement in the flue is a good idea to make sure you burn safe and clean.
A probe thermometer is good for (the better) double wall stove pipes.
If you want something electronic, Auber has stuff. There's also another thread here where someone is making a (nicely designed, wood accents etc.) "alarm" light that works with a sensor that you can set to alarm. I forgot the details and don't have time to look it up, but I presume alarms both for too high temp (danger) and too low temp (dirty and delayed danger).

 

[begreen]

All good points brought up here: The dryness of wood, stage of the fire, stove operation, will influence the potential for creosote build up.

The other main factor is the design flue system from stove to chimney top. An insulated flue system that is mostly within the heated home envelope stands a much better chance of keeping clean than one that vents through an elbow then a tee of an outside chimney system. While the flue gases may be leaving the stove at say 300º, they may drop to below 200º by the time they exit an exterior chimney, especially when it's cold outside.

 

[gthomas785]

What if you compared the internal and external temperatures of the pipe? Maybe you could establish an R-value of the buildup layer and somehow estimate the thickness of it from that. Seems hard to do correctly though.

 

[stoveliker]

That would also need the flow rate (and heat capacity) of the gases in the flue, and the cooling power on the outside of the flue...

 

[gthomas785]

That would also need the flow rate (and heat capacity) of the gases in the flue, and the cooling power on the outside of the flue...

Perhaps simply the change in r value (or a proxy for it) could be found, after calibrating to a known clean state for a particular chimney setup.

I'm not saying it would be exact, but could still be useful. We all have heard the rule of thumb that the skin temp of single wall pipe is *approximately* half of the flue gas temp. Perhaps if one found that the skin temp was instead 1/3 of the flue gas temp, this would indicate significant creosote buildup in the pipe.