ESP8266 Keeps Tabs On Wood Stove Temperature

Wood heat offers unique advantages compared to more modern heating systems, especially in remote areas. But it also comes with its own challenges, namely, keeping the fire going at the optimum temperature. If it’s too cold you risk buildup in the chimney, but if you’ve got it stoked up more than necessary, you’ll end up burning through your wood faster.

To keep the fire in that sweet spot, [Jay] decided to put an ESP8266 and a thermocouple to work. Now, this might seem like an easy enough job at first, but things are complicated by the fact that the flue temperature above the stove lags considerably behind the temperature inside the stove. There’s also the fact that the top of the chimney will end up being much colder than the bottom.

Mounting the thermocouple in the flue pipe.

In an effort to get a more complete view of what’s happening, [Jay] plans on putting at least two thermocouples in the chimney. But as getting on the roof in December isn’t his idea of fun, for now, he’s starting with the lower one that’s mounted right above the stove. He popped a hole in the pipe to screw in a standard K-type probe, and tapped it a few times with the welder to make sure it wasn’t going anywhere.

From there, the thermocople connects to a MAX6675 amplifier, and then to the WeMos D1 Mini development board that’s been flashed with ESPHome. [Jay] provides the configuration file that will get the flue temperature into Home Assistant, as well as set up notifications for various temperature events. The whole thing goes into a 3D printed box, and gets mounted behind the stove.

This project is a great example on how you can get some real-world data into Home Assistant quickly and easily. In the future, [Jay] not only wants to add that second thermocouple, but also look into manipulating the stove’s air controls with a linear actuator. Here’s hoping we get an update as his woodstove learns some new tricks.

9 thoughts on “ESP8266 Keeps Tabs On Wood Stove Temperature

  1. The next obvious step is to get an actuator to control the influx of, and implement a PID to deal with the natural delay in the system – although it gets tricky as the control function will probably change a lot depending on how much fuel is on the chamber. Nice project!

    1. I see three main reasons:
      – because it’s fun to do and learn new things
      – because it can optimize your life
      – because other people will take benefit of the work you shared on the Internet (given the fact that you publish it :).
      I like to heat my house with wood, I have access to “free” wood, and thus every time I heat my house with it I save something like 3€, compared to what my heat pump will draw on electricity power.
      To get that number I took benefit of things I have done earlier:
      – an apparatus to measure the power usage of my home
      – sensors to measure temperature from my chimney
      – retrieve the heated air from my chimney to drive it to the input of my HVAC
      – an apparatus to pilot my HVAC through modbus
      – some python script that automatically detect chimney temperature, and force my HVAC in ventilation mode in order to spread the heat all other my house. Once the fire is off, the HVAC returns to “heat mode”.
      I didn’t invent anything (maybe the python script), because for these jobs I’ve read many builds such as this one to make my system. Aside the (lot of) fun I had building this system, I know that I save some money every day I run my chimney because I have ways to measure and thus optimize my set up.
      Of course, every part of my project gives me benefits to optimize my home, my budget, and my life:
      – measuring the house power usage allows me to detect failures: for example: the water heater can draw more power if there’s a leak in the security system
      – having my HVAC communicating with modbus allows me to set the temperature when I come back from holidays
      – playing (prototyping) with all of these concept made my learn lots of things, some were “useless” for my work, other were priceless: learning modbus and industrial protocols, playing with my HVAC, and developping ways to measure power consumption gave me the opportunity to lead the project of making the supervision system of the datacenter I work in today.
      It was definitely worth to do and learn these things :)

      1. The function of having a dumb_me light system is my goal. Not for it to do the actual flu maintenance but to remind me it’s time to turn it at certain times I’m barely having knowledge (just got an uno and a couple boards to tinker with as practice). I thought about using some car sensors as indicators but ignorant on some of the coding and logistics. Having the fan system integration is fantastic idea though since most heat from stoves is wasted in the main room it’s set up in. Any advice or do and don’t you/y’all is welcome.

  2. Big box hardware stores sell sweep kits for under $60. Just do regular maintenance and inspection, because this isn’t going to replace the need to do regular maintenance and inspection.

  3. Nice, I wonder if there would be useful warning conditions that could be added in HA. That might actually contribute some safety? I’d be tempted to whack on a BME680 for some room data.
    Maybe blueooth monitoring for presence detection?

    But definitely would not want to try and automate with it in case I started a fire.

    Side note: those sterling engine (I think that’s how they work) fans on stoves have always held a spot in my heart since childhood.

    1. It’s “Stirling engine” (not “sterling”) and, no, those stovetop fans don’t use that — they are thermoelectric (Peltier/Seebeck). The more clever ones also have a bimetalic strip that bends the bottom (hot mating surface) away from the stovetop when it gets too hot.

  4. When I was a kid we heated our house with a wood-fired cookstove and (later) a wood-fired furnace. The cookstove had standard singlewall flue pipe, in which was stuck a thermometer with a dial and a needle indicating tempeature. It was very useful to keep the flue the right temperature, but a PITA when it came time to sweep the flue: we had to pull the thermometer out so the brush wouldn’t damage the thermometer probe. Despite the monitoring and the cleaning, we still had chimney fires. They were dramatic, but never any damage.

    The furnace had a double-wall heavily-insulated flue pipe that was a straight shot through the roof three stories up. No monitoring on that flue, and it was self-cleaning. Squeaky clean every time we checked, and we burnt about 3 full cords/year.

    The furnace had bang-bang control on the air supply by a standard household bimetal/mercury tilt switch thermostat. Shortly after LEDs were invented I wired one across the switch: LED ‘ON’ meant the house was warm and we probably had enough wood in the firebox. LED ‘OFF’ for more than a while meant “go feed the furnace”. I thought about making a proportional control for it, but figured that the bang-bang control scheme with its intermittent hot fire probably did a lot to keep that flue clean.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.