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

Electric boiler for SIP building in Pacific Northwest

walkoffsingle | Posted in Mechanicals on

I have a workshop/garage constructed with SIPs panels and would like a little input regarding the viability of an electric boiler. The shop is used daily and I’m shooting for a minimum interior temperature of 60 F in the winter.

We live in a rural area where there is no access to natural gas so that leaves electric, propane, or heating oil. I’ve pored over the other postings in the archives regarding electric boilers and the consensus seems to be that they are prohibitively expensive when compared to other fuels.

We have an oil boiler in our home but there is only one person left servicing them in our area and he’s not well recommended so I’d like to avoid oil. I’ve completed a heat loss calculation using an online calculator (builditsolar) but I’m not completely confident in the results (garbage in-garbage out). The shop is is 60’x40′ with an18′ ceiling. The SIP walls are R-24, the ceiling has blown in insulation at R-49 and there is R-10 foam board under the entire slab as well as around it’s perimeter. The Pex tubing is already in place under the 3″ slab. There is 80 sqft of double glazed windows, 168 sqft of insulated garage door, and 21 sqft of insulated entry door.

The design loss calculation estimates 33,191BTU/hr and a UA of 738btu/hr-F. I’ve dabbled with some of the calculations myself regarding gas vs electric and the heat loss calculator will spit out a cost comparison but I’m not certain I’m comparing apples to apples.

We currently pay 6.5 cents KW/hr while propane is $2.69/gallon plus a delivery fee. Does an electric boiler make sense in this situation? Btw the shop currently has it’s own lightly used 200amp panel so the cost to hook up the boiler should be minimal.

Any help is greatly appreciated.

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. Kevin_in_Denver | | #2

    Some of the ductless minisplit heat pumps put out 2-4 times as much heat per dollar as an electric resistance boiler. They also run on electricity. The right models operate down to -15F, when electric resistance may kick in. With your energy costs, this type of heat pump will cost roughly less than half as much as propane.

    There is really only one heat pump that could use the pipes in the slab: Daikin Altherma. But it's too expensive. Ground source heat pumps are even worse cost-wise.

  2. GBA Editor
    Martin Holladay | | #3

    Electric resistance heat is expensive. Even if you decide that you want to heat with electric resistance, using a boiler doesn't make much sense. Electric baseboard units are much cheaper to install than a boiler.

    I agree with Kevin's suggestion; a ductless minisplit would make sense for your application.

  3. davidmeiland | | #4

    In this case, electric resistance heat is less expensive than propane. If you want to use the slab, install an Electro-boiler or Thermolec boiler or something similar--they are very common here (our fuel cost picture is a lot like yours, with more expensive electricity) and work fine. A ductless heat pump would be much cheaper to operate if you don't have to heat the slab. Is this a woodworking shop?

  4. walkoffsingle | | #5

    Can a ductless mini split evenly heat a shop space that large? Btw my thought was I would eventually add hot water solar to pre heat water.

  5. walkoffsingle | | #6

    It is multi use but a significant part of its use is as a woodworking shop. I take it that would influence your choice David? Dust issues with the air handler?

  6. Expert Member
    Dana Dorsett | | #7

    If you place the head correctly you can get reasonably even heat from a mini-split in a large open space. If there are multiple walls with a lot of window area you may need more than one ductless head, but if you can tolerate a 5F delta between the coldest & warmest corners convection will take care of it for you.

    Calculating the heating load is a critical first step to figuring this all out. The builditsolar heat calculator is ultra crude, and prone to error but is probably good enough for sizing a boiler. What were the indoor & outdoor design temperatures you used? Got a ZIP code? (For sanity checking the outside design temp.) What U-factor did you put in for the windows?

    A name brand 2-ton mini-split can deliver about 30KBTU/hr @ +5F, and significantly more at 25-30F. In the PNW east of the Cascades you'll average a COP of about 2.8-3, west of the Cascades it's more like 3.2-3.5, sometimes better. If you need all the gory details you could wade through this:

    For a shorter and less technical perspective this is a bit easier to digest, but not PNW-specific:

    With 6.5 cent electricity there's no good reason to be heating your house with an oil boiler. If you have ANY zone with reasonably open floor areas installing an appropriately sized mini-split to heat that area would cut your oil use enormously. The cost of the heat from the mini-split will be less than 1/4 that of $3.50 oil in an 86% burner:

    oil: 138,000BTU/gallon x 0.86 = 118,680 BTU/gallon or 8.43 gallons/MMBTU. At $3.50/gallon that's $29.50 /MMBTU

    mini-split: 3412 BTU/kwh x 3.0 = 10,236 BTU/kwh or 97.7kwh/MMBTU. At $0.065/kwh that's $6.35 /MMBTU

    Even heating with an electric boiler is cheaper than heating with $3 oil at your electric rates, but it's still well worth applying the cost of the electric boiler toward a ductless heat pumps.

    The sheer upfront cost of solar thermal makes that a fairly difficult project to rationalize. Using the same roof area for photovoltaics has a much better payoff, though at 6.5 cents/kwh delivered you'd need a significant subsidy to get the lifecycle cost of full-retail/no-subsidy $5/watt PV under your retail electricity cost. But PV WILL get cheaper, and soon. In Germany they're already paying half that (unsubsidiezed) or less using the same panels, inverters, and racking systems, and it's not because their labor costs are cheaper than in the US. Most of the cost of a US system is the soft-costs, not the hardware, and as rooftop solar ramps up to becoming a competitive commodity and code regulations & net metering provisions become better standardized $2/watt solar should be coming here too, at which point the lifecycle cost of PV will become competitive with 6.5 cent retail electricity. (Recent pricing in MA has hit the $4/watt range before subsidy.) But hydronic solar-thermal (even the DIY kind) can't compete in any way shape or form with 6.5 cent electricity used in a mini-split heat pump at a COP of 3.

  7. davidmeiland | | #8

    Neil, yes, my concern would be dust affecting the indoor unit. I haven't put a mini-split in my shop for this reason, although I also have yet to talk to any of the manufacturers to see what they say. Your shop is quite large so I think it's worth looking into one or more mini-splits.

    I agree with Dana that solar thermal is a non-starter. Solar PV is the best place to make an investment at this point. You can't sell extra hot water to the grid.

  8. wjrobinson | | #9

    I would pump $50 worth of electric heat thru the already installed floor at winters coldest less other months via optional controls of your choice. Get the rest of your heat with one mini split. Turn off the split when your dust creating work is in progress and when applying coatings. Install a good dust system our two.

  9. davidmeiland | | #10

    Wow, all the expense of installing a boiler/controls/piping, AND one or more ductless also? It would be cheaper to use an electric heater or two when the ductless shouldn't be running. Then of course there's also the response time of the slab to consider.

  10. Expert Member
    Dana Dorsett | | #11

    Mini-splits have filters, some can even accomodate HEPA filters:

    In a high dust environment you'll have to clean/change them more than the once-a-year-if-that schedule some people do in their home units.

    Radiant floors are good for combustion efficiency on fossil burners, and provide a real cushy-comfort-factor for the barefoot home environment (or even a socks-at-the-office environment), but there's not really much advantage in a shop building heated with an electric boiler. Cap the PEX to keep it clean, preserving the option, "just in case", but odds are it'll never be impressed into service. Air-to-water heat pumps or ground source heat pumps could use the PEX loops, but the upfront cost of those solutions are as daunting as any credible solar thermal solution would be.

    If electricity triples or quadruples in price and wood or pellets don't, there may be a rationale for better-class wood boilers or pellet boilers to heat the radiant floor. But that day is pretty far off- longer than the anticipated lifecyle of a ductless mini-split.

    AJ: At his estimated heat load and likely outside design temp he can already do the whole thing with one mini-split. In a tight high-R building with an insulated slab for thermal mass with active work under way inside you can turn off the mini-split for hours during the daylight and not lose much ground on interior temperature. The equipment, lights, and solar gain may already be more than carrying the load during the day- as long as the mini-split keeps it at the desired temp overnight there would be next to no load during the day- most days in the PNW are pretty temperate. Even "bright cloud" days can induce modest cooling loads in the middle of a Puget Sound winter in a building with R20+ walls and an R50 roof.

  11. walkoffsingle | | #12

    Dana, thanks for the links. Sounds like the argument for a mini split is pretty compelling. As an aside,attached to the north end of the building there will be a 1200 sqft SIPs mother-in-law apartment on a crawl space(complete with mother-in-law) and I was planning on going with a heat pump or mini-split system there. As for my calculator inputs I used 25 F as my outside temp (calc asked for coldest temp expected in normal year). I'm shooting for between 60 and 62 F as my inside temp. My U value on the windows was .55 (r 1.8). The interior of the space of the shop is open though I've toyed with the idea of enclosing the wood shop portion. I'm pretty fanatical about dust collection but could that be a problem for the head unit? I agree that there's no good reason to be heating our main house with oil and truth be told we haven't used the boiler in the last three years. I have quite a bit of dead standing timber on the property so we've been heating mostly with wood...but the house is a project for a later date.

  12. walkoffsingle | | #13

    Sorry didn't refresh my browser before my last post so I missed the answers regarding dust. AJ makes a good point about turning it off during dust producing times and given what Dana suggests about not losing much ground during the day time it probably makes sense to forgo the boiler due to the expense and just be vigilant about keeping it clean. I suppose when the shop is cold I could use the pex system as a wort chiller when I make home brew.

  13. walkoffsingle | | #14

    Would it be preferable to use separate mini split systems for the shop and apartment or a single large unit with multiple heads? The apartment consists of two small bedrooms and open living room and kitchen.

  14. davidmeiland | | #15

    You might find the cost of two separate units to be close to the cost of a multi-head, and the efficiency is better.

    What brands are you looking at?

  15. Expert Member
    Dana Dorsett | | #16

    Run room-by-room heat load calculations on the apartment. You really need a heat load over 5000BTU/hr to rationalize a separate ductless head. Doored-off very low heat load rooms can be temperature balanced with the rest with radiant cove heaters (about the same cost but twice the comfort of electric baseboard) or oil-filled electric radiators. I suspect the heat load of the 1200' apartment would be worthy of 3/4 tons of ductless, maybe even 1-ton, but probably not more, if you don't have ridiculous amounts of window area.

    To get the most out of the ductless it's generally better to 'set and forget' rather than use setback strategies due to the much higher part-load efficiency compared to running full-out on a recovery ramp (any savings from the lower heat loss at lower temp get eaten up in lower efficiency during the recovery.) When used in combination with resistance heating in doored-off rooms remote from the ductless head it's better to set the mini-split to a temp at least 3F higher than the resistance heater in the adjacent room so that a large fraction of the heat to the remote room is still from the high efficiency ductless. Using radiant cove heaters and controlling them with both setback thermostats and an occupancy sensor cutout can keep the use of resistance heating to a minimum while still providing comfort. During the recovery ramp of the radiant cove zone comfort is much higher than with baseboards or even panel radiators, since it heats up the occupants directly, not the air. ( Don't oversize radiant cove heaters by a huge factor though- you don't want it to feel like you're in a broiler when it comes on. :-) )

    Whether it's a multi-split for the whole building or a pair of mini-splits comes down to ease and cost of installation. The outdoor units can have a high wintertime standby loss if they're cycling, since they effectively resistance-heat the compressor during standby mode to keep it from having startup issues, but that's only at temps much cooler than 25F.

    Do clear-glass U0.55 windows even meet code? In most of the PNW U0.35 is code-max, and in a building with an R50 roof and R24 SIP walls there's a good rationale for holding the line at U0.28 except where the higher solar gain would be useful & desirable. If there is any kind of low-E coating on any surface you're probably looking at no more than U 0.35, which is enough to make a difference in your heat load calculation.

  16. GBA Editor
    Martin Holladay | | #17

    Wort chiller? I always let my wort cool down at its own pace -- no hurry. Just sip a bottle from the last batch as it cools.

  17. walkoffsingle | | #18

    I was wrong on my u value for the windows. They are rated at .30.

  18. Expert Member
    Dana Dorsett | | #19

    80 square feet of window over-esimated by (0.55-0.30=) U0.25 at a delta-T of (60F - 25F=) 35F means you had overestimated the load by (80' x 35F x U0.25= ) 700 BTU/hr, bringing the total down to 32.5K from 33.2K, a difference, but not a huge difference (~2%). If the rest was done reasonably true to the real whole assembly R values you're still only looking at about 2 ton's of ductless- 2.5 tons if you want to run it a bit warmer. With any substantial 24/7 plug loads (say a refrigerator, a chest freezer, a water heater and a Tivo) and a coupla overnight live-humans you'd be able to stay above 65F @ 25F outdoors with only 2 tons of ductless.

    The leakage factors on SIPs tend to be pretty low, and the weatherstripping of your garage door may be the biggest single parasitic infiltration if you did a good job on all of the SIP seams, etc.

  19. walkoffsingle | | #20

    The garage door is something I've been spending a lot of time thinking about. It's definitely the Achilles's heal of my shop. The r values given by the manufacturers seem dubious. I'd love to hear other people's clever ideas about sealing around an "insulated" garage door.

  20. walkoffsingle | | #21

    @David, I really haven't gotten as far as looking at brands for a mini split since I was fixated on using my "wort chiller" to heat the floor. I think all of ya'll (you's guys for those on the east coast) have me convinced that's the way to go so I'm open to hearing about brands for my application. This is one project I'm unlikely to try and DIY but I'd like to go into it with as much info as possible so I can sniff out the corner cutters.

  21. user-444644 | | #22

    Neil, I'm gonna try to run with the big dogs here and agree that a mini split is the way to go- AND stick my neck out to add that if you just want to play with your hydronic setup, consider using a plain 'ol electric tank water heater for floor warming/supplemental heat during the "cold months" here in the PNW.

  22. Kevin_in_Denver | | #23


    I agree that the airtightness of the garage door is the achilles heel. I'm currently trying to solve the problem with a pair of homemade carriage doors, as described in another thread:

    I'm installing the doors this week, and will try to make another video.

  23. Expert Member
    Dana Dorsett | | #24

    There are many mini-split manufacturers, and many installers in the PNW, but you want to go with one that has good tech support and a number of installers in YOUR area. (In my experience specifying mini-splits for some relatives in Kitsap county WA, in western WA Mitsubishi seems to own more than half the market, with Fujitsu the next largest vendor but still rare in some locations, not sure who's #3.) Both distributor and installer support are important, but if you stick with the big-3 (Mitsubishi, Daikin, Fujitsu) the quality of the unit design & manufacturing QA tends to be pretty good.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |