Pellet Stove Controllers

I’ve developed a controller for a US Stove Model 5660. Is anyone interested?


It might be helpful if you were to explain more about what your controller does?
Does it control heat or something else and by what means and how does it do what it does?
Does it have a display or maybe can you talk to it and give commands?

Maybe you can list a set of features.

To save others from searching, explain what a 'US Stove Model 5660' is anyhow?
Is it coal, wood, oil, gas, electric or fission?

I'm not personally interested in such a device though.


All projects are interesting. Please post the code, schema and pictures on the Exhibition/Gallery section.

I fully understand the pellet stove equipment you are interfacing with the arduino. I am very interested in what you have done with the pellet stove controller. I'm running an Englander 55-SHP10 that looks like it is very similar from the control perspective to your 5660 with the addition of a secondary auger that runs continuously (whenever the combustion blower is on) to deliver pellets from the pulsed primary auger. I would be very interested to see what you have for controls, display programming etc. I'm currently interested in more finely controlling the primary auger pulse width among other things.


For anyone considering any such project: As much as I love my Arduino, and think I know it all....

Think fail-safe.

For the specific project here, for instance, I would add two things to the system:

An old fashioned thermostat which which would over-ride everything else and turn my stove OFF it the temperature went above, say, 75 degrees F.

A ditto to turn the stove ON if the tture went below 38 d. F.

Both could be wired in very simply, no micro-electronics involved, "right next to" the "stove on/off" circuits. Preferably by a sympathetic qualified electrician, 'cause I can promise you if you get a frozen-pipes-damage insurance claim, you won't find a sympathetic insurance company... and that pleases me, because I don't want my premiums going any higher.

Did you ever explain what your actually doing? I also have an Englander 55-SHP10 and have been searching for a better way to monitor the fuel level in the hopper. The default storage hopper only has a capacity of 40ibs so it requires refilling at least once or twice a day. If I forget to fill it and it goes out, it sometimes takes up to 24 hours to fully heat up the basement again on really cold days. It would be easy enough to set up an alert via SMS etc... but the biggest problem has been finding a level sensor for solid mass. The only other builds I've found online use an Ultrasonic sensor to activate the alert when the space is empty ( but I can never find any example sketches to build off of and calibrate.
Has anybody else figured out how to do this?

I wonder if an indirect approach would be sufficient?

If you can get access to the Archimedes screw which feeds pellets to the burner, then something to count revolutions, and send that SMS when "too many" revolutions have taken place. (You'd have to figure out how many revotions that would be... but that would be a simple exercise in trial and error.)

Alternatively, and even more indirectly, if you monitored the system which "runs" the stove, from how often it called for heat, and how "strongly" it called for heat.... I believe they "command" the stove to burn "hot" if it is cold out, or just "warm" is only a little heat is needed?.... then from that you could work out pellet consumption, couldn't you?

What does the surface of the pellets look like, a while after you have filled the bin, and after the burner has used up much of what you put in? Presumably like the sand in the upper chamber of an hour glass?

Couldn't you put something... a tennis ball?.. on top of the pellets? It would "float", wouldn't it? Maybe something heavier than a tennis ball. And have a string leading up from the tennis ball to a microswitch over the pellet hopper. And when the ball gets too low, the string goes taut, and operates the switch?

Pellets are flammable. (Duh.) Hoppers, I suspect, have covers to exclude sources of ignition, and to contain and limit any burning of the pellets in the hopper. Don't compromise those safety features with the ball- and- string.... i.e. by just leaving the cover off. Mount microswitch to underside of cover? Or something.

Yes, putting anything into the hopper is definitely risky, especially something set so close to the top auger. The stoves these days are set up to shut off before there's a fire danger, but you don't want to end up having to replace an auger or auger motor...
Its amazing that this particular use-case is such a tough nut to crack! I know I've seen forum posts about the same issue since 2010-2011!
I did contact the author of this post- as I would like to try his Ultrasonic sensor method, and he very kindly replied right away. Unfortunately, the Ultrasonic method has major problems reliably receiving sound waves in small, all metal, enclosed environments so he's trying to use a pressure sensor instead. My feeling is that it will ultimately come down to finding the right kind of pressure sensor. There has to be something people use to measure solid mass in agricultural settings right? I wonder how they measure things like animal feed or grain?

As for your indirect suggestion, the ultimate answer would be to eliminate the need for multiple refillings altogether, by adding more capacity to the hopper. They sell expansions that simply sit on top of the existing pellet hopper, but you can only add so much weight before its no longer practical. Larger industrial pellet furnaces will have a large standalone hopper next to the unit with a long diagonal feeder auger leading directly to the burner. Such a setup could be controlled by a small motor and would only need to be filled once. I tried building one last winter with a medium sized bin and an hour glass shape on the bottom leading to a piece of PVC pipe. I ultimately had to give up when I couldn't find an archimedes screw shape that could be long and thing enough. I considered using a leaf blower as a pneumatic conveyance system but it would require way too much energy to move such a small amount at a time. If I did use a large standalone hopper however, I could use the pneumatic method to blow the pellets into the hoppers location in the basement.

Here's another cunning plan. (Baldrick)

Somewhere in the hopper: A transparent vertical tube.

Inside the tube, a number of light sensors, in a vertical array.

In an appropriate upper (inside) corner of the hopper, a light which can be turned on and off under computer control.

From time to time, turn light on, see which light sensors can see it, which are "buried" under the remaining pellets.

Is it not possible to have a plate at the bottom of the hopper with a load cell arrangement so you are then sensing the weight of the contents of the hopper?

roundhouselabs wrote:

There has to be something people use to measure solid mass in agricultural settings right? I wonder how they measure things like animal feed or grain?

Yes, using load cells placed at the base of each structural support stand. Often used for liquids and grain, as ultrasonics is not always a reliable method.

You could also think about placing a load cell under the hopper at one end, with the other end on a pivot point, but that will take a bit of structural work with possibly needing to lift the whole thing off the ground by a small amount.

Using a load cell will provide a long term reliable method of weight determination.


If you want a continuous readout of the level in the hopper... not so easy.

But do you need that?

Surely just a "getting low" signal, and a "you are now very low" signal would suffice?


Mount inside the hopper, on the wall, something a little like a weak mousetrap. Mount it against the wall, i.e. 90 degrees different to a mousetrap's ordinary orientation. Put a broad paddle on the "arm". When the arm is "closed", i.e. how it would be to catch a mouse, it closes a switch.

As you fill hopper, pull arm back against the spring, let pellets hold it held back. When pellet level falls, arm is released, switch closed.

A bit Heath Robinson? Perhaps. But Does The Job, with bits you can SEE, and know if they are working, or why not? And do-able. And inexpensive.

Hi Jim

I'm also interested on your Project, could you share it?


Also interested! Is it publicly available?

Hello. I have a US Stove 5770 and have been thinking of building my own controls for it. Is this available? Would be nice as a starting point.

I have previous Arduino experience with a mpguino and never had any problems with it over many years. But I'd have to learn more about the programming.

Interesting post thread !! I too have been working a project similar to this one idea. I have a Hammond PB105 pellet boiler and it works great (for the most part). It provides our domestic hot water in the winter time. (summertime we use electric) however there are those few days when either the boiler has a fault or the owner forgets to load pellets and it goes down in the middle of the night and we’re faced with a cold shower in the morning ! NOT GOOD !

I’ve setup a Particle Photon with 2 MAX6675 Thermocouple modules to monitor boiler output water temp and the outside temperature. The photon works great and sends data to ThinkgSpeak for a realtime graph and also to a Google spreadsheet via IFTTT. I’ve also added logic to the control scheme to send me a txt (via IFTTT) if the temperature on the boiler drops below 100f for more than 10 minutes. I’ve recently added Blynk into the mix, so I have a nice graph of the temperatures on my phone. This all works well but now I want more !

I want to expand my IO capacity so I’m moving to the Arduino Mega so I’ll have a bunch more IO. I’ll be adding a proximity sensor on the augerfeed mechanism to count pulses as the boiler feeds pellets, this should give me a good measure on consumption rate. Also want to try the SR04 ultrasonic sensor in the hopper to sense pellet level. Good insights in this thread about potential issues with that !

The Mega will do the majority of the monitoring and will be interfaced back to the Photon which will be my “communications manager”. It will sit there and read (over serial) the data from the Mega and then publish to the web. I think it’s a fairly decent design, at least in theory !!