GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Smart Thermostat for Resistive Heating

canada_deck | Posted in General Questions on

I’m excited to be moving on to the next stage of my energy efficient shed project and I am working on the detailed drawings and choosing specific materials.

This will be an 8’*12′ shed with 6″ of insulation in the floor, walls, and ceiling.  One insulated door and no windows.  I’m located in a climate similar to Seattle but a little colder.

I want to keep the shed from ever getting below freezing but I don’t want to use any more electricity than I need to.  I’m planning to heat the shed with a few heat lamps in ceiling mounted light fixtures.  I don’t want to rely on anything with moving parts (no fans) and I don’t want to limit my ability to store stuff on the walls so heat lamps seem like a good safe option.

I don’t want this to be hokey (nothing that relies on being plugged into a receptacle.)  I want to have a proper wall mounted hard-wired thermostat.  Ideally, it would be a thermostat that has WiFi and some advanced features so that I can log how many hours it is in heating mode each month to calculate how much I am spending without needing to install a separate energy meter.  It would also be nice to control it remotely.

Unfortunately, I can’t find anything like this.
So far I have found:
– Some themostats that can accurately turn on close to 35F/1.7C but they aren’t set up for WiFi or can’t directly control a baseboard heater (they are low voltage thermostats)
– Some thermostats that have the features I want but the lowest setpoint is 41F.  In my area, the difference between 41 and 35 is significant.  I haven’t run all the math but I wouldn’t be surprised if it almost doubles the number of days when the heat kicks on.
– Good anti-freeze thermostats that are designed as a plug-in module and don’t have WiFi.  You plug them into a receptacle and then plug a space heater into them.
– I have also noted that some thermostats for switching on baseboards have a minimum load.  E.g. they would not work if you were running a single 100 watt bulb.  I’d like to find something that is reliable with as little as 300 watts of load.

Has anyone seen anything that checks all my boxes?

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. Expert Member
    Akos | | #1

    I use a Mysa unit for a pump house. It has almost everything you want except it only goes down to 5C.

    Getting too close to freezing might be an issue if you have any air leaks, uneven temperature or stratification in the place. I would say the 1.7C target is a bit borderline.

  2. paul_wiedefeld | | #2

    Vancouver has averaged 290 HDD_40 and 109 HDD_35 over the past three years. If you need about 1000 btus/300 watts on design day, that's 164kwh vs. 62 kwh per year, or about $10. Leaving a light bulb on for Nov, Dec and two on for Jan and Feb would probably work for $50/year or so. Is this worth chasing?

  3. Expert Member
    BILL WICHERS | | #3

    The usual thermostats used for this kind of thing are known as "line voltage thermostats". That's what you want to look for. I usually just use simple ones that have no electronics in them at all, so "old school" style thermostats :-) I'm not aware of any smart ones. Smart thermostats are usually intended for full HVAC systems, freeze control is usually considered a pretty basic application, so the thermostats used for it tend to be very basic too.

    Another potential option for you is to get an old-school incandescent night light and put it under the thermostat. This will fool the thermostat into thinking it's slightly warmer than it really is, so it will let you trick the thermostat into holding a lower actual setpoint temperature. This might let you use a regular thermostat for the lower temperatures you want to control.

    I would make two suggestions though: if you want to CONTROL the heat source, don't use light bulbs -- use a small electric resistance heater, like a small electric baseboard or ceiling/wall mount unit. Light bulbs have very low resistance when they first turn on, which results in relatively high initial inrush currents when they turn on. This inrush current will wear out the thermostat contacts more quickly, and is also why light bulbs tend to burn out faster when switched on an off frequently. This is also why larger switches have a "tungsten" rating, which allows for the inrush. Electric resistance HEATERS don't have this issue, so they're much better suited to a temperature control application like yours.

    If you just want simple with a light bulb, my recommendation would be to just leave them on all the time. As an example, if you have a space needing 200 watts to maintain your target temperature, you can just leave 200 watts worth of incandescent lighting on continuously to maintain the temperature set point. If you have 400 watts worth of lights, you're thermostate will just switch them on and off with a 50% duty cycle, so you'll still average 200 watts over time -- there is no energy savings. Without a thermostat, your lights staying one will just keep your space some fairly fixed number of degrees warmer than whatever the outdoor air temperature is, so you don't have control, but you do have some extra warmth to help with condensation control.


  4. canada_deck | | #4

    Thanks for the responses. These are very helpful.

    A few things:
    I actually don't have a good sense of the design load. I am guessing (hoping) that approx 200-300 watts of heat will be sufficient given my location and the design. The inside space is 8'*12' by 8' tall. Walls, floor and ceiling will be 2*6 on 24" centers with Rockwool batts. Sheathed on both sides with plywood or OSB. There will be a poly vapor barrier on the inside side of the wall. I expect it to be essentially air-tight. There will be a single well-sealed 36"*80" metal exterior door. I would like to be able to keep it above freezing when the temperature outside gets as low as -10C (14 F).

    Point taken on 1.7C potentially being too close to the freezing point.

    Re: "Is it worth chasing." That's a fair point as well. I am partly doing this out of curiosity. When I complete the project, I want to try building an energy model of the space to see how closely I can match the model with the real results. I'm also going to be spending a fair amount of time and money building this shed so I don't mind spending an extra $100 for something that will save $20/year and be trouble-free

    Re the inrush current. That's very interesting and I did not think of that. Perhaps a ceiling mounted infrared panel would be a better option. Something like this:

    1. Expert Member
      BILL WICHERS | | #5

      Yes, one or more radiant panels is a better idea than light bulbs. And purpose-built heater is going to be better as a heater, and also safer to operate. Think of light bulbs as light sources than generate heat as a side effect, where a heater is intended to make heat as it's primary function.

      The easiest way to do your energy modeling is to use a simple on/off mechanical thermostate and two temperature sensors, one logging the temperature inside the shed, the other logging the outdoor temperature. Set the thermostat to whatever target temperature you want. Watch the temperature graphs over time. You will see that when the outdoor temperature drops below the target temperature, the target temperature will be held as a reasonably flat line by the thermostat (you might even see on/off cycling around the set point). At some point though, you'll see the indoor temperature fall below the target -- that's when things become interesting.

      When the indoor temperature falls below the setpoint, you know that the amount of heat output you have is no longer sufficient to overcome the thermal loss of the structure, so the indoor temperature falls. You have, at this point, lost control of the indoor temperature (what I like to call "losing the battle"). The thermostat will just be "on" at this point, keeping the heat source running constantly, since it can't produce enough heat to get up above the set point where it would normally cycle off. You will now see the indoor temperature 'track' the outdoor temperature's changes, with a relatively constant delta (difference between indoor and outdoor temperatures). That temperature delta will change very slightly with very extreme low outdoor temperatures, but for normal ranges, you'll see a pretty much fixed delta.

      The way it will look around the transition seasons is that during the day, you'll see the indoor temperature rise above the setpoint, since the sun will heat up the structure and the thermostat has no way to cool things off. As the outdoor temperature drops in the evening, you'll see the indoor temperature track the outdoor temperature, eventually reaching the thermostat's set point, at which point you'll see the indoor temperature become a flat line, or possibly an on/off cycling squiggly line "/\/\/\/\" around the setpoint. Eventually the outdoor temperature may get cold enough that the thermostat loses control the way I've described, at which point the indoor temperature will fall and again track the outdoor temperature until eventually the outdoor temperature rises enough that the thermostat will start cycling and that indoor temperature will be a flat line again.

      I've attached a pic from the temperature monitor in my dog's doghouse. The doghouse has a heater in it that is controlled by a simple electronic thermostat I designed (with chew-proof features), that implements simple "bang-bang" control -- it turns on when it reaches the low temperature setpoint, and it turns off when it reaches the high temperature setpoint. Because of this type of control system, it ALWAYS cycles on/off around a target, but never stabilizes at the target. This allows us to see some interesting things in the temperature graph though:
      1- As outdoor temperatures fall, you see the slope of the on- part of the on/off cycles changing, and flatten out. This is because it takes longer amounts of 'on' time to get up to the high temperature setpoint. You see then long 'on' times raising the temperature, with short 'off' times as the temperature falls after the heater shuts off. When things warm up, the slopes change with steep (short) 'on' times followed by relatively long periods of 'off' time while things cool down. Eventually (although I didn't have a good pic to show for this), the outdoor temperature gets too low for the heater to maintain control, and then you start to see the temperature fall below the low-temperature set point and start to 'track' the outdoor temperature as the heater is just on all the time, unable to raise the temperature.


Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |