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

Heating a Superinsulated, Cold-Climate House

mpsterner | Posted in Energy Efficiency and Durability on

We’ve been designing a passive-inspired “Pretty Good House” for the better part of a year or so. We’re now hoping that we’ll be ready to build by summer–getting our ducks in a row and such. The goal has been to design an energy-efficient home and garage with a target budget of 400-450k (which is proving challenging), styled in the New England Colonial Farmhouse style.

I need help finding the right heating system.

I’ve read all (I think) of the articles on GBA on this topic, however, given my situation, I am still interested in more feedback…

Heating a Superinsulated House in a Cold Climate

Heating a Tight, Well-Insulated House

The house is 40’x26′ (exterior) with a breezeway and 28’x32′ insulated, occasionally heated garage.
– Full basement
– Full two stories
– 1975 interior square feet + basement.
– 12″ R44 Double stud wall based on Building Science Corporation design https://www.buildingscience.com/documents/enclosures-that-work/high-r-value-wall-assemblies/high-r-value-double-stud-wall-construction
– R80 in attic truss (blown cellulose)

I am wondering the following:

How to Heat
Mini-splits: According to most articles I see, mini-splits are the way to go for a house like ours. However, given my location I question if this is the best heating system for us. Additionally, I have called HVAC contractors around the area and while familiar with mini-splits, they’re really against the idea of this being my primary heat source.

Electric boiler: I like that this could be offset with photo-voltaic, however, I am concerned about the COP and cost of this for heating. What are the pros and cons of a hydronic system heating with an electric boiler? How will the cost of running this through a winter compare with propane?

Gas boiler: This is the gold standard in my area. Pretty much what everybody installs (both homeowners and contractors). I am not crazy about signing up for propane bills forever, however, it is comfortable and proven. I have been receiving estimates in the 20k-25k range though for the whole system, and that seems insanely high to me.

Other options: Are there other options that I am not considering?

I have many other questions with regards to systems, but for this thread, I am going to limit it to heating systems. Please let me know your thoughts and questions.

I’ve attached our house design. No, that is not an archaic fireplace–it is flue-chase enclosed within the thermal envelope and then cladded with brick veneer for the classic colonial farmhouse look.

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.

Replies

  1. user-723121 | | #1

    Michael; What about solar access, windows, high SHGC and the like. Any opportunities for off peak electricity where you intend to build. Sounds like good R-values for Zone 7A, you will need a well insulated foundation and sub slab insulation. I would shoot for 1 ACH50 or less for optimum performance.

    Get the details and Design Temperature heat loss figured out then look at the different heating options available.

  2. mpsterner | | #2

    Hi Doug, Regrettably, I omitted some incredibly important information from the original question. Thanks for pointing that out.

    The designer specified the heat loss at 22,744 btu/hr, including the basement. The basement is listed as R21 and slab as R19, but that may include the concrete (though that wouldn't be much).

    The air change figure that I have is CFM50/ssf 0.05, which I don't know how to get into the air changes per hour metric you mentioned. Can you help with that?

    The house is oriented south with roughly 8% south glazing. As you can see from the design, the main floor is pretty much one large room. We don't have any significant added thermal mass though. The windows are These windows are: whole window Ufactor of 0.15 BTU/hr-ft^2°F and a center glass SHGC of 0.56.

    You can do off-peak electric here but I don't know what the rates are. I can look into this.

    1. user-723121 | | #3

      Michael; CFM 50 to ACH50 is CFM50 x 60 divided by the house volume (I think). You have some great windows specified from my perspective. As to the design temperature heat loss. I have built 6 double wall homes with similar specifications and slightly larger on a square footage basis. Your windows are much better than I had available at the time. This house will perform at around 1 BTU/sf/hdd and probably better than that with the quality windows, good foundation insulation and 1 ACH50 or less. So your 3,000 sf house or thereabout is going to have an annual energy consumption for heating of 25,500,000 Btu's @ 8,500 heating degree days. Your hourly heat loss per degree F is 125 Btu's and at a -20F design temperature your design temperature heat loss is 10,625 Btu's.

      Your house on the natural gas grid would heat for under $200.00 per year. 60 cfm continuous ventilation will be about 12% of your heating load with a 75% efficient HRV.

      1. Trevor_Lambert | | #9

        I think you might want to check your math on the hourly heat loss. My house has this one beat in every single known category:
        -smaller square footage (all of it above grade though, so maybe balances out)
        -higher R value walls, ceiling, slab
        -higher design temp
        -equal or better windows (one number doesn't tell the whole story, because each window u-factor will be different depending on size / operability - most likely he has been told the best number, probably that giant picture window on the south side)

        And yet my modeled heat loss per hour is, by the most generous estimate, 11,500 BTU/h.

        The only unknown is glazing ratio. I'm guessing you used 8% for the entire house, but what he said was "8% south glazing". So is that 8% of just the south side, or is that 8% of the whole house is glazing on the south side? Judging by the picture, I'd guess the latter. The amount of glazing on the main south wall is almost certainly over 1/4 of the whole wall.

        1. mpsterner | | #13

          Hi Doug and Trevor,
          Our house volume of conditioned space - 27,384 cubic feet.

          I confirmed that this figure - .05 CFM50/ssf is for "surface square feet."

          The architect said this:
          Currently the most up to date air tightness standard set by PHIUS is .06 CFM50/ft2. Though passive house North America doesn't use ACH50, your calulated number (set at .05 CFM50) is .68 ACH50. The old original passive house standard was .6 ACH50. The CFM50sq.ft. is a stronger and more accurate way of measuring airtightness. (and you would need your surface square footage - 6237.5 Sq. Ft. - to calculate the ACH50 number.)

          So, with all that said. It sounds like this is based on achieving .68 ACH50.

          Doug, can you help me understand the calculation using our heat loss calc of 22,744 btu/hr and our annual heating degree days for my actual location (at 70° inside temp) of 10,958 HDD. The heat loss calcs were based on a 86° design temperature difference (70° to - 16°).

          As for solar glazing, the south side has 8% glazing (of the total living area not including the basement). The window ratings provided are for all windows except for the large center window–I haven't gotten numbers on that particular window. They're all "tilt and turn" windows.

          Let me know what you think. Thank you.

          1. Trevor_Lambert | | #20

            Michael,

            The only way for the different size and style of windows (fixed vs operable) to have the same u-factor, is for the frame and glazing edge to have the same u-factor as the center of glass. This is not possible. The edge of glazing is always higher than center of glass, and so usually is the frame. For example, on my house, two identical size windows and glazing, one fixed is u-factor 0.8 (metric) and the other operable is 0.86. Comparing the fixed one to another fixed, but larger window, the factor is 0.75. Having said all this, most energy modeling probably doesn't drill down to this level of detail.

            Just out of curiosity, what is the window brand?

    2. mpsterner | | #12

      Doug - I got the peak electricity information...
      Peak - Normal rate 12.5 cents/kw
      Off-Peak - 10 cents/kw
      Thermal storage - 8.9 cents/kw

  3. Expert Member
    Akos | | #4

    Michael,

    I really like the heating system from this build:

    https://www.greenbuildingadvisor.com/article/flatrock-passive-firing-up-the-heating-system

    The problem is in your area, you need AC in the summer, so if you are installing a mini split systems for that, it doesn't make much sense to have another backup heat setup.

    If you are worried about the performance of the mini split in the winter, an option would be to add in some electric underfloor heat in the main area and electric panel heaters upstairs as backup.

  4. GBA Editor
    Martin Holladay | | #5

    Michael,
    Q. "What are the pros and cons of a hydronic system heating with an electric boiler?"

    A. The fuel cost of all forms of electric resistance heat are the same, because these systems are 100% efficient. If you choose hydronic distribution, you get the worst of both worlds -- the high fuel cost of electric resistance heat, along with the very high installation cost of hydronic distribution. If you want electric resistance heat, install baseboard electric units or wall-mounted panel heaters -- much less expensive to install, and the same efficiency.

    My vote: install ductless minisplits or ducted minisplits -- you'll get summer cooling and winter heating, without separate equipment, and your heating fuel cost will be half the cost of electric-resistance heating.

    And your architect has drawn the most expensive duct termination for a kitchen range hood that I have ever seen.

    1. mpsterner | | #14

      Hi Martin,
      Thank you! My concern is that mini-splits are just not adequate for the number of days we have well below 0°F and the "even-ness" of the heat (especially in upstairs bedrooms). So then, I start thinking hydronic due to the even distribution, however, I'd prefer to not sign up for a lifetime of propane use–hence the consideration of an electric boiler that can be offset with PV.

      Point taken though that I am spending the $$$$ on the hydronic and then paying higher fuel costs of electric resistance heat.

      I just don't want to install a heating system (looking at you mini-splits) that I later find actually isn't really adequate or that the heat is uneven.

      Air conditioning sounds nice, but with Lake Superior right by us, it never really gets hot. Though it does get humid...

      1. mpsterner | | #17

        Martin - I forgot to respond on the duct termination. Are you referring to the chimney? That is actually a chase for a woodstove pipe, not the vent hood. But yes, I know. My wife is very, very stuck on having an exterior brick clad fireplace look as she is dearly attached to the colonial farmhouse look.

        It does have the additional benefit of preventing intrusion into the bedroom space above for boxing in the chimney flue.

      2. Trevor_Lambert | | #19

        My house is similar to yours. Very open main floor, more traditional second floor with bedrooms, bathroom and a hallway. Similar levels of insulation and air tightness. What I have found is that the second floor demands very little heat. We have a 600W radiant heater in each bedroom, but the frequency and duration they come on is low. The thermostats I have are modulating in 20% increments, and I've never seen one of these heaters go above 60% (360W), and very rarely above 40% (240W), and keep in mind this is with no other heating on the second floor. When we get some decent sun during the day, it's not uncommon for the second floor heat demand to actually be zero over a 24 hour span. Over the last week, we've seen some lows around 0F, highs typically around 20F. The amount of heat delivered to the second floor was in the ball park of 1-2kWh per day (less than 10% of the first floor heat). I could look up the exact numbers from my energy monitor when I get home, but I don't have remote access to it.

        So what I'm suggesting is that even if you find the mini split doesn't end up as even as you'd like, a very minor amount of electric heat in each room might easily make up the difference.

        I'm currently considering a mini-split with one head on each floor, and if I didn't have the desire for cooling upstairs, I'm pretty sure I wouldn't even bother with a head upstairs.

  5. Peter Yost | | #6

    Hi Michael -

    "The air change figure that I have is CFM50/ssf 0.05, which I don't know how to get into the air changes per hour metric you mentioned. Can you help with that?"

    I don't know what your air change figure means at all; can you double check this and let us know? The units/expression seem wacky.

    Peter

    1. Trevor_Lambert | | #10

      I'm guessing CFM50/ssf means airflow at 50Pa per surface-square foot. Ball park surface area on this house would be around 5000ft^2. So 4400*0.05 = 250cfm. Square footage floor area is about 3000, volume ~24,000, ACH50 =~0.6.

      1. mpsterner | | #15

        Yes, I've confirmed that this is exactly what it is.
        .05 CFM50/ssf (surface square foot). Surface square footage is 6237.5 Sq. Ft.

        So, .05 x 6237 = 311.85 x 60 (min) = 18711 / 27,384 cu ft volume = .68 ACH 50

        Thank you.

  6. natesc | | #7

    A wood stove is a great compliment to a minisplit in an efficient house.

  7. Expert Member
    Dana Dorsett | | #8

    What is your 99% outside design temperature?

    Even at -10F cold climate ductless is still running at a COP north of 1.5. Cold climate heat pumps have capacity tables as low as -13F to -15F, with a very few specifying an output at -20F or so.

    Though not specified as a cold climate unit, Fujitsu's mini-ducted units have the right type of compressors and decent capacity at -5F (the lowest temp in the capacity tables), but not the pan heaters to keep defrost ice from building up during extended cold snaps. The 1.5 ton unit is good for more than 18,000 BTU/hr @ -5F, and though capacity isn't specified, probably more than 15,000 BTU/hr @ -15F. They also have a signal output to call on auxiliary heat to allow it to be integrated into a "backup heat" system. The lack of a pan heater and associated control hasn't stopped people from using them in zone 7, but some amount of monitoring of the drain port & pan ice is prudent.

    Fujitsu's cold climate multi-zone units are compatible with the same mini-duct cassettes- they won't modulate as much when married to a multi-split, but in a high-R solar tempered house that won't matter as much.

    1. mpsterner | | #16

      Hi Dana, Thank you for your response.

      How can I determine what my 99% outside design temperature is? I am in 54891 zip code. It can be -20° for weeks, however, that is definitely not the norm. Most of the winter is spent (this is based off nothing but my experience) around 5°-25°.

      Understand the specs of the units you mention, my biggest concern is failure to operate. We can certainly install electric resistance backup heat and maybe that is enough to have a redundant and failsafe system... It is this complete lack of confidence in this solution that all HVAC people around here have in this option. They simply say "that doesn't work around here. Maybe further south, but I would never do that." That sentiment has gotten in my head!

      1. Expert Member
        Dana Dorsett | | #21

        >"How can I determine what my 99% outside design temperature is? I am in 54891 zip code."

        Based on location I'd hazard that the 99th percentile temperature bin in 54891 is within a degree of Ashland's -16F design temp:

        https://articles.extension.org/sites/default/files/7.%20Outdoor_Design_Conditions_508.pdf

        So Fujitsu's capacity numbers at -15F on their cold climate goods are pretty relevant.

        >" It can be -20° for weeks..."

        I doubt it. Not since the last ice age.

        It might get DOWN to -20F or cooler every night for weeks, but it doesn't STAY -20F or cooler for weeks. It will get to some temperature above that on most days, even during a Polar Vortex disturbance cold snap. A day that even averaged -20F would log 90 heating degree-days (base 70F) for the day. The two coldest days logged at the Ashland Kennedy Memorial weather station so far this year were 86.7 HDD70 and 85.4 HDD70 on the 30th and 31st of January 2019, according to degreedays.net datasets. So even on those days it had to have gotten considerably above -20F at some point during the day to hit mid-80s HDD70 .

        At those temperatures it would be good to avoid Mitsubishi, since the temperature sensors in the outdoor unit are't calibrate well enough to operate fully consistently at temperatures as "warm" as -18F. When the controls determine the sensor is out of range it turns the unit off until it hits -13F. While there are field reports of Mitsubishis still going strong deep into the -20s, the spec says it can turn off at -18F. Fujitsu cold climate units have a fully specified output capacity down to -15F, but keep going even at -30F. There are some Gree units fully specified down to -22F (wet-bulb), but the support network in the US for the Chinese vendors isn't as well developed as Fujitsu/Mitsubishi/Daikin/LG . (Home Depot may sell them, but I'd be reluctant to count on a box store for quality technical support on that sort of product.)

        At max speed -15F the 1.25 ton Fujitsu 15RLS3H is good for 16,000 BTU/hr, the 12RLS3H is good for 11,500 BTU/hr, 11,000 BTU/hr for the 9RLS3H. The 3 ton / 4 head AOU36RLXFZH multi-split delivers over 22,000 BTU/hr at -15F. A pair of any of the single zone units would be a decent amount of heat for a high-R house, and the 4 zone multi wouldn't be terrible either. The COP at max speed is in the 1.4-1.8 range for those units (depending on which unit), which is still substantially more efficient than an electric boiler. At your mid-winter average temp of about +17F (see https://weatherspark.com/m/12230/1/Average-Weather-in-January-in-Ashland-Wisconsin-United-States#Sections-Temperature ) the COPs are about 2 or a bit higher, using half the power of an electric boiler.

        1. mpsterner | | #25

          Hi Dana,
          Thank you for all of this.

          You're absolutely right, it doesn't stay -20. That was a bit of an exaggeration, though we can absolutely have many days, sometimes more than a week where it gets to -20 every day. Key phrase, "gets to," not "stays at." So, point taken. Thank you.

          Based on what you're saying, I am gathering that you feel very confident in the Fujitsu route, yes? Additionally, you're saying that these Fujitsu units are still functioning reliably and efficiently at -15°F? If so, we're really only talking about a handful of days each winter where the temperature will get below that, and even then, only during the nighttime hours. Even on the coldest of days it'll typically get up to -5 to -10.

          When it gets really cold, do they just stop working and just as easily start working again once it warms up above -15 or so? Or what exactly happens at the coldest temps? Can damage be done to the units themselves?

          Given a desire for redundancy and safety in the event that we had a really cold spell... How would you suggest planning backup heat? How to size the number, location and sizing of electric resistance baseboards? Do they need to be tied to the other system, or do I just set them to a lower temperature and they'd only ever kick in if things got uncomfortable?

          When it comes to being prudent and managing these units during super cold temps, my main concern is when we're away. We could be away traveling during a period of extreme cold and no one would be there to monitor things.

  8. Expert Member
    RICHARD EVANS | | #11

    Michael,

    What a beautiful home and layout!

    I too suspect your heat loss figures are higher than they likely are given your level of insulation.

    I would opt for a pair of Fujitsu ductless mini splits: one for the lower level and one for the upper level (one or both for heat and one for AC, upstairs). People are using these things in Alaska. You will be fine... (I would avoid Mitsubishi in your climate because of the massive drop off in output in really cold temps.)

    I would give up on trying to convince the mini-split installers that the heat pump will serve as your primary heat source. They will NEVER understand... (I know this from experience.)

    In my case, I visited our local wood stove company and picked up a catalogue of wood stoves for free. I brought it home and told our HVAC contractor that we were installing a 300,000 btu wood stove and that our mini split would serve as our back-up in the shoulder seasons. (Don't tell them that the stove is anything less than a 300,000 btu stove as they may think it is too small and try to up-sell you on a supplemental propane furnace.) It worked! They happily installed our single 12k Fujitsu unit and I couldn't be happier with it (It modulates in a way that no gas furnace ever could. It is truly sublime.)

    I later returned my wood stove catalogue and thanked the dealer for their time.

    1. mpsterner | | #18

      Hi Rick, Thank you!

      Are you saying that the insulation is low for the estimated 22,744 btu/hr heat loss calc?

      I hear you on not worrying about what the contractors say, however, I have my own concerns on it being effective all the way through the winter and more, comfortable throughout the house. Will the heat be even enough or on a dark, cloudy cold day are we actually going to be quite cold in spots?

      Since no one around here does this, it is hard for me to understand what the experience of living in a house with mini-split heat will be like.

      1. Expert Member
        Dana Dorsett | | #22

        A pair of 15RLF3H can deliver 32,000 BTU/hr @ -15F, so you'd have margin for colder temps assuming the heat can be reasonably distributed. A single head for the floor with the open floor plan would probably work just fine, assuming the design load is less than 16,000 BTU/hr for that floor.

        The floor with the doored off bedrooms might call for the 1.5 ton 2 zone AOU18RLXFWH cold climate compressor serving an ARU12RLF 1 ton mini-duct cassette zone, a combination which can deliver 17,900 BTU/hr @ -15F according to NEEP data.

        Together that would be 33,900 BTU/hr @ -15F, about a 1.5x margin on your calculated design load, which should be able to get you through those cold snaps that dive below -20F.

  9. user-723121 | | #23

    Michael; The 99% dry bulb design temperature for Ashland, WI is -21F according to my 1985 ASHRAE Fundamentals Handbook. I always used 65F as the indoor temperature when determining design temperature heat loss. I believe it was assumed people, appliances and lights would supply the other 5F.

    Using my numbers would give us an 86F Delta T. Your surface square footage is 6,237.5 (I come up with a bit lower number). Let's use you wall U-value as a straight calculation of .o23. x 6,237.5 for an hourly heat loss of 143.46 Btu's. We know windows will lose more, the ceiling will lose less and heat loss through the foundation is very complicated and best left to a sophisticated heat loss program or Dana.

    So, in very simple terms your DTHL is 86 x 143.46 or 12,338 Btu's. My original estimate was just a guess based on square footages and past experience but I knew the number provided by your consultant was way off. You can break your proposed home into elements, walls, windows, ceiling, foundation, infiltration and do the manual calculation. Area X U-value for hourly heat loss per element and add them up for a total hourly heat loss. Infiltration and ventilation air will likely be less than 20% of your total heat load if you meet .6 ACH50.

    Heating Degree Days for Ashland, WI
    https://w2.weather.gov/climate/xmacis.php?wfo=dlh
    http://degreedays.wi.gov/

    NOAA shows a mean of 8,647 hdd and Degree Day WI is 9033 hdd

    1. mpsterner | | #24

      Hey Doug,
      Thank you, again!

      Just to be clear, are you saying that the designers calculated heat loss of 22,744 btu/hr is wrong and that it is very likely a lot lower?

      Obviously, if this is what you're saying then I've got to determine best next steps for getting a more accurate heat loss calc, because that is a major difference!

      1. user-723121 | | #30

        Michael: I believe your design temperature heat loss is lower than 22,744 Btu. You can do a manual calculation (U value times area) for your homes elements, walls, ceilings, doors, windows, foundation and infiltration. For simplicity I would add 20% to the sum total of the other elements for infiltration and ventilation air reheating. Assign to each element an hourly heat loss per degree F and add them up. You now have the hourly heat loss in Btu's per degree F for your home. Use any design temperature you like along with any indoor temperature you want to use to determine the design temperature heat loss.

    2. Expert Member
      Dana Dorsett | | #26

      >" The 99% dry bulb design temperature for Ashland, WI is -21F according to my 1985 ASHRAE Fundamentals Handbook."

      Doug, are you sure that isn't the 99.6th percentile temperature bin, not the 99th?

      The ACCA and EnergyStar data sets indicate -16F for Ashland:

      https://www.energystar.gov/ia/partners/bldrs_lenders_raters/downloads/County%20Level%20Design%20Temperature%20Reference%20Guide%20-%202015-06-24.pdf

      https://articles.extension.org/sites/default/files/7.%20Outdoor_Design_Conditions_508.pdf

      The 2009 ASHRAE short list puts even the 99.6% design temperature for Duluth MN at -12F, whereas the ACCA puts the 99% design temp for Duluth at -16F, just like Ashland:

      https://www.captiveaire.com/catalogcontent/fans/sup_mpu/doc/winter_summer_design_temps_us.pdf

      I'm pretty skeptical about a -21F 99% design temp at that location, to say the least!

      >"I always used 65F as the indoor temperature when determining design temperature heat loss."

      That would not be legitimate in a Manual-J. A code-minimum indoor design temp is 68F, and the calculation is based on the difference in indoor & outdoor temperatures, not the heating/cooling balance point. The heating/cooling balance point is useful in determining the heating cooling equipment sizing, but that is distinct from the actual heat LOAD on the house. In an IRC 2018 code min house the balance point is usually about 60F, with internal heat sources making up the difference, but in high-R houses it can hit the mid-50s. Professional Manual-J tools often estimate the balance point.

      The thermal mass of foot-thick cellulose walls would also come into play in a more sophisticated load analysis than Manual-J too, all tending to lower the peak load. But if the heating system is designed for ~23KBTU/hr @ -16F and the actual peak loads are closer to half that it's not a disaster, and won't change the equipment choices much. It will still need an upstairs zone and a downstairs zone.

      Michael Sterner writes:

      >"Additionally, you're saying that these Fujitsu units are still functioning reliably and efficiently at -15°F? "

      In brief, yes I'm confident that a cold climate Fujitsu will perform reliably and effciently (compared to an electric boiler anyway) at -15F, and 3-4x more efficiently than an electric boiler during the shoulder seasons.

      The whole point of publishing extended temperature capacity tables is to put in writing what the actual capacity is and is not at the conditions in the table to make it possible to size it correctly for the load. People are using cold-climate Fujitsu mini-splits as the primary heat source in locations that hit -30F, despite the lack of a published capacity at that temperature. Unless the installer screws up on refrigerant charge levels it will pretty much hit those marks at -15F, and they're pretty reliable units putting out lots of heat even at temperatures well below -15F.

      A guy in Quebec who used to post here under the handle Jin Kazama heats his not-so-superinsulated house with four 3/4 ton Fujitsu and reports they kept his house comfortably warm at -30F outdoors, though the output air was pretty tepid when it was that cold outside. I know a couple of people in VT heating with cold climate Fujitsus in locations where it drops below -15F at least a couple of nights most years.

      The Northeast Energy Efficiency Partnerships maintains a list of cold climate heat pump models and capacities, including the capacity 70F indoors at the coldest temperature in the manufacturers' extended capacity tables. It used to be downloadable in spreadsheet format updated monthly, the most recent version published early January this year, but they're reformatting how they do it (the spreadsheet was pretty cumbersome, with hundreds of models and combinations listed), but are saying the information will become available in the new format in April 2019. Watch this space:

      https://neep.org/initiatives/high-efficiency-products/emerging-technologies/ashp/cold-climate-air-source-heat-pump

      They also have other useful info sheets on installing & using cold climate heat pumps here:

      https://neep.org/initiatives/integrated-advanced-efficiency-solutions/air-source-heat-pumps/air-source-heat-pump#Consumer%20Guide

      1. mpsterner | | #27

        Dana, thank you so much for this extensive reply.

        Seeing that Duluth's 99% is -16 gives me more confidence too. Duluth spends a significant amount more time colder than where I am. Even Ashland is colder since I am up on the peninsula that goes into Lake Superior.

        I am digging in deep to the resources you provided and my confidence in these units in our area is growing. One thing is for sure... I am going to have to either convince an HVAC person or find someone from out of the area to do the installation. Not having the right support is another concern of mind, so I'd much rather convince a local HVAC contractor to do this...

        From what you've seen in this thread thus far, does it seem to you that the heat loss calculation of 22,744 btu/hr is wrong and do I need to look into different options to get more accurate? The designer I am working with seems very competent and knowledgeable, but he did acknowledge that this is a relatively simple heat loss calc–not an extensive Manual J or similar.

  10. mpsterner | | #28

    Hey everyone,
    I appreciate all of the extensive replies. This has helped to continue my exploration of mini-splits. I have come up with an additional question that, I think, has major implications for the project...

    How would I heat the garage? The garage is going to be a heated workshop, not heated 100% of the time, but likely a great deal of the winter to 40-45° or so.

    The garage was not included in the heat loss calc and is likely too far away and divided from any potential mini-split location that it would require its own source of heat.

    With that, does that start to steer anyone towards a different heating system?

    Ideas and suggestions? Thanks.

    1. Expert Member
      Akos | | #31

      +1 on the a dedicated mini-split. For a garage, I would oversize the unit so you get much faster response when you need to get it hot (I have a 12kbtu unit in a 450sqft somewhat insulated garage just for this reason).

    2. Expert Member
      Dana Dorsett | | #32

      >"...likely a great deal of the winter to 40-45° or so."

      The Fujitsu RLS3H series all have a "minimum heat" mode to maintain a (non programmable) indoor temperature of 50F. At 50F indoors a -9RLS3H will have more capacity than it does at 70F indoors, but it won't bring a building stagnating at 0F or colder up to 50F very quickly. You may need some sort BTU-overkill solution for bringing it up to temp in a reasonable amount of time. Whether it's "worth it" to install a mini-split in the garage / workshop depends a lot on how much of the time it's being actively heated.

      Electric or propane fired overhead radiant heaters directed downward can make a workshop more comfortable even when it's still cold in doors, and can be a good solution for intermittent use applications.

  11. josh_in_mn | | #29

    For the garage, why not just another separate mini-split? Or perhaps just a simple resistance heater for those times you plan to work there. What level of insulation and air sealing do you contemplate?

  12. RMaglad | | #33

    Doug, I'm over in Ontario, Ottawa, and we see similar, maybe slightly warmer temps than you. Just coming off a brutal january and february, with many days not exceeding -22C (-8F), and dropping down to lows of -28C (-18F). I too have a 22.5Kbtu/hr design heat loss. I have 2 mitsu 6K units (1 in master bedroom, 1 in basement rec), a 9K mitsu in living room, and a 9K fujitsu mini-duct in the kids bedrooms/bathrooms/basement bedroom area. single level home, with basement, very similar levels of insulation. My theoretical capacity is about 33K btuhr, understanding that the fujitsu miniduct is not rated down to -25/-15F (my design conditions), however it does still put out heat at that temp.

    January 17 to February 17 is used 2,434 KWhr of electricity. My bill was $343CAD. This includes 2 electric heated floor pads which, during that time consumed about 450KWhr. Basic electric hot water heater, with 2 adults who like long showers, and 2 kids who regularly have baths. We run 1 load through the dishwasher per day, and regular laundry, and what i would consider above average kitchen use (all electric).

    The house never struggled to maintain set points. Solar gain helps when the sun is shining, but we had a few very rare super cold cloudy days, and even at that i maintained an interior temp of 71 in the living areas, and 67 through the rest of the house.

    1. mpsterner | | #34

      Hey Ryan, thanks for the details! It is really good to hear from someone in a similar climate, similar house, etc. with a good experience on these units. Considering the intensity of January/February, that is pretty reasonable for electricity when it includes your heating. Is that about as high as you've seen it? Where are you using the electric floor pads? In bathrooms? Sounds like you live similar to us–with kids and not skimping heavily on electrical use.

      I got a quote back from a local HVAC company and they suggested 1 outdoor unit: http://www.fujitsugeneral.com/us/products/multi/2-3-4rooms/aou36rlxfzh.html rated for 36,400 and then using 2 indoor heads – 1 on the main floor at the north door and 1 on the upstairs landing outside of the bedrooms. The outdoor unit would be underneath the north porch.

      Does anyone have any thoughts on this solution?

      1. Expert Member
        Dana Dorsett | | #35

        >"Does anyone have any thoughts on this solution?"

        Of course!

        A pair of heads on a single 3 ton multisplit is a second rate, somewhat marginal solution. See the last paragraph of response #21.

        The "rated" 36,400 BTU/hr is the modulation level at which it was tested for efficiency at +47F, which isn't particularly relevant to you. The more important number is it's maximum capacity at -15F for the -36RLXFZH is 22,110 BTU/hr, which is slightly LESS than your calculated 22,744 BTU/hr load.

        You would have more capacity and some capacity in reserve with a PAIR of separate 15RLS3H, even though the "rated" capacity of (2x 18,000=)36,000 BTU/hr is slightly less than the 3 tonner's 36,400 BTU/hr. When it's -15F outside each of the 15RLS3s are good for 16,000 BTU/hr , or 32,000 BTU/hr for the pair. That would be 1.4x the calculated load, which means you would have margin to spare well into the -20s, even though they don't publish the capacity at those temperatures.

        It's usually cheaper to install a pair of single-zone units than a pair of heads on a multi-splits, and the HSPF efficiency is also higher. It also means there are fewer single points of failure- a compressor could crap out completely and you'd still have the other one running, and the remaining 15RLS3 delivers over 70% of the calculated design load- enough for freeze protection even at design condition. The only down side to a pair of minisplits is the visual noise of having two outdoor units instead of one.

        1. mpsterner | | #37

          Very, very good feedback. Thank you. I will request that they adjust the quote based on this.

          One more question... How should I heat the basement if I do mini-splits? The basement is going to be an unfinished space for now, sort of a kids play space, but I'd like it to not be freezing cold and will someday finish the space for a guest room and family room...

          Thanks Doug! I can see that there are some issues with the heat loss calcs. I've gotten in touch with a certified passive house consultant to help me do accurate heat loss calcs.

          1. Expert Member
            Dana Dorsett | | #38

            >"One more question... How should I heat the basement if I do mini-splits? "

            What is the calculated heat load of the basement?

      2. user-723121 | | #36

        Michael; I just looked at your heat loss page again and this is what I come up with for the hourly heat loss per degree F. I went with an R-44 overall wall R-value, for your climate (9000 hdd) that is the minimum to be considered as superinsulation.

        Btu/hr
        58 Walls
        52 Windows
        13 Ceiling
        12 Glass doors
        9 Foundation wall above grade, assuming 12" above grade
        18 Foundation wall below grade
        9 Basement slab
        20 Infiltration
        16 Ventilation
        207 Btu/hr/F 84F Delta T 68F Set point, -16F Design temp.
        17,405 Btu Design temperature heat loss

  13. mpsterner | | #39

    Dana - I wasn't able to comment on your nested comment... Sorry.

    According to the basic heat loss I have, the 22,000 btu/hr at -16° includes the entire house, including the basement. I don't have a separate basement heat loss calc.

    1. Expert Member
      Dana Dorsett | | #40

      Try entering in the necessary information into a loadcalc.net (online freebie tool_ to ball park it. I suspect the load is really pretty low, but without getting at least some kind of handle on what the load might be it's hard to venture a reasonably well targeted solution.

      If both the walls and slab are insulated and the window & door area is low, with no joist insulation in basement ceiling it might track within 5F of the upstairs temperature even without actively heating it. If there's a heat pump water heater down there it'll lower the temp a degree or two, if there's a big refrigerator or freezer down there it might raise the temperature a degree or two. We can make some adjustments on it for those kinds of factors once we have a basic load number to work with.

      1. mpsterner | | #41

        I actually see that the HVAC contracted that quoted had included a manual J in their quote email.

        Their manual J does have some estimated numbers (it is showing the whole house at 34k btu/hr), so I think it is a little off, but it shows the basement at 6,051 btu/hr at -16°

        Currently, under-slab insulation is R16.7, basement walls are R20. I was intending an electric water heater. There will also be a freezer in the basement, yes.

        1. Expert Member
          Dana Dorsett | | #42

          >"Their manual J does have some estimated numbers (it is showing the whole house at 34k btu/hr), so I think it is a little off..."

          Dya think? (Like more than 1.5x beyond what other calculations are telling us, maybe?)

          This is pretty typical of HVAC company Manual-Js. They had to have modeled it with all code-minimum inputs or big air leakage/ventilation to come up with a number that big.

          >"...but it shows the basement at 6,051 btu/hr at -16° "

          That's not big enough to rationalize another mini-split, especially if they had their thumb on the scale. 6051 BTU/hr is about 1800 watts. A crummy electric baseboard heater that side is pretty cheap, and will probably never be operated at a 100% duty cycle. The standby loss of a 50 gallon electric water heater and plumbing is going to be on the order of 50-100 watts, call it 1700 watts, a conscious adult human would put out about 100 watts at idle, 75 watts while sleeping etc.

          If it's not going to be set up as a finished guest room space for awhile it's safe to defer the heating solution, and you'll be able to determine empirically about how much it's going to take when the time for the build out of guest quarters or playroom etc happens. A 1500 watt oil-filled radiator type space heater on a plug-in line voltage thermostat and a datalogger tracking duty cycle against heating degree days for week or three in winter doesn't take a large research equipment budget.

  14. JoshPriem | | #43

    Hello Michael!

    I'm looking at heating options for my house in northern WI. After reading this entire thread I am curious with what option you went with and how it worked out for you.

    Thanks!

  15. PAUL KUENN | | #44

    I'd love to hear as well! PK here in Appleton, WI but building up on Washington Island next year for a customer and I'll install solar PV with it.

  16. mpsterner | | #45

    Hi Josh and Paul,
    The house has gone through many, many changes since this post and conversation, still with many of the same attributes as before but it is largely a different design altogether.

    I certainly don't expect you to dive in at the level this other post covers but it runs the entire spectrum of options we were looking at among many challenges in getting an accurate load calculation and system design (even from a mechanical design firm), etc.

    To summarize:
    - We actually didn't start building until this spring so the heating system is just now being installed
    - It was really difficult to get the mechanical design agency to produce accurate load calcs and then design a system that was as good as other GBA experts said it could be. We finally got there, only with a lot of help from GBA.
    - We went with a 30k Mitsubishi heat pump with an air handler 40MBAAQ24XA3 ducted throughout the house and used zone dampers to zone the spaces.
    - We did a hybrid ERV system where we have dedicated exhaust and shared supply with the air handler

    The verdict is out still as it is just about to be up and running. I will let you know in a few months time. We're just about to do our blower door test which will help us understand if we achieved what we needed to achieve for this system to work well. We need to hit 1.1 ACH50.

    Lastly, I know that this is posted all over GBA but keep your designs relatively simple. Maintaining air barrier continuity with the design we have (while not super complicated) was very challenging. Especially since I was teaching the crew about these things while we were building. It was difficult to keep the knowledge flow caught up with how fast they were able to build a house.

    Best of luck!

  17. musielski | | #46

    Michael, I'm interested to here how this works out. Keep us posted. I started building a cabin in Hayward. It's been a labor of love. Current I'm heating it with electric baseboard & a wall unit. It's been down to at least - 26 F, and even 1/2 into March it's still into -teen's at night. Being in the middle of know where, I still have a smart meter so I can really track my usage.

    1. mpsterner | | #47

      Hi Dan,
      It has worked out well so far. We did a different house design than this original but it is basically the same size and similar construction. We did 5" of exterior Rockwool and were able to achieve a very airtight assembly. The house was easily heated with two 1500w space heaters most of the time during construction. Now that we have our full HVAC system set up, it keeps up easily even during the cold spells. We went with a Mitsubishi 30k heat pump with an indoor air handler that is ducted throughout the house. I was worried about distribution so we ducted the whole thing, however, after being in a super insulated and airtight building, you notice how "even" the heat is everywhere. It seemed like every room in the entire house was the same temperature, even when we were just heating with 2 space heaters.

      Any way, we're not quite done with the house yet and haven't moved in so that will be the real test! Seems to be working very well though.

      1. musielski | | #56

        Great to hear this Michael! You are absolutely correct...it's amazing how even heat is in an air tight how. Great Job!

  18. PAUL KUENN | | #48

    Great to hear Michael, Congrats!

  19. ErikOlson | | #49

    Thanks for the detailed follow up on your house.

    Looking back, would you still choose the ducted air handler? Or would you consider multiple smaller individual mini splits?

    1. mpsterner | | #52

      Hi Erik,
      I'd like to say "I'll get back to you on that soon" as my response. It was anywhere between -10 and 20 for most of January. It really seemed that we had a cold month. During that time, our heat system wasn't even hooked up yet and I heated the entire house to 60 with 2 1500w space heaters and at times to 75 (for drywall) with 3 1500w space heaters. It always seemed that the heat was (almost oddly) distributed throughout the house. This leads me to believe that a single split on the main floor in a central location and a single upstairs, perhaps with short distribution to each of the 3 rooms would have been easily doable and comfortable.

      All of this will become more clear with time once I am not working in the house and just living. Then I'll detect any subtle differences easily.

      I was a bit worried about cold and warm spots and that's why we went with a ducted system but my guess is we would've been fine without. I think you'd definitely be fine in a house plan that is very open. Ours is sort of "in-between."

      At the end of the day, we had a ton of help from GBA and one of the options that was proposed was going with two splits. The other option was the one that the people designing the system wanted to go with too so I just went with them.

      I will say, in my experience, the feedback you can get on GBA is often better than professional consulting firms, even in big cities. You really got to find someone that knows high performance homes.

      Sorry for a long answer. How's that for a novel...

      1. ErikOlson | | #54

        Thanks for the feedback, Michael. The layout I am planning is both open and symmetrical, so I am leaning towards two matching mini-splits. The bedrooms will have through the wall jumper ducts pulling from near the splits, with a common low-power exhaust blower for airflow when the doors are closed.

        I'm in MN so we've had very similar weather. Definitely agree with the quality of help on GBA. It really feels like this is the milieu at the leading edge of building tech.

  20. 88cch | | #50

    As an HVAC contractor, I would agree that any of the prior suggestions can heat the house so I will mention a few other concerns like comfort, redundancy, service/support, aesthetics, and future replacement. Comfort (I'm talking heating as your primary need) is subjective and it has a lot to do with the building envelope, but if our customer requests "the most comfortable heating system", its likely not going to be ductless. Zone-able radiant floors, radiant panels/BB (any source but not convectors), have the potential to deliver the most comfort, albeit at high install costs (except resistive types). You will need a separate AC system if desired so that adds to the cost. If your goal is cheap, efficient heat, ductless will suffice. If its comfort, ductless or any forced air is probably not a 1st choice. If IAQ is a desire, then you might want to look at ducted systems (HP or otherwise) that will allow integrated filtering as a compromise. A 70 degree room can be either comfortable or not, depending on the design. I question whether "just install 1 ductless head" on each floor is truly comfortable, or just tolerant because nobody admits it. Redundancy- do you want/need back-up heat? What if you lose power? Will a generator power the ductless or resistance back-up? It will surely power a propane boiler or furnace. Service- What if the ductless fails on a 0 degree day? 2 units is better than 1, always. What type of HVAC service do you think you get at the outside unit on a 0 degree day...quick and dirty? Likely. A propane boiler might be expensive to install and operate but you have a good chance of getting it fixed on X-mas eve if peace of mind matters. If the HVAC contractors your dealing with are reluctant to install ductless, what is the chance of quality service in the future? Future service on wall units can be ugly- moving big, expensive furniture, fingerprints on walls, drips from water leaks, dropped tools, etc. Again, radiant (or even forced air distribution) might be old-school, but it tends to be bulletproof in the living space. Aesthetics- its your house/choice but do you want to look at wall mounted units? That look doesn't fly in our higher-end market - mock up a unit on the wall to be sure your partner is on-board with the plan. Future replacement- some time in the future (5, 10, 15 years?) your HVAC guy will tell you its time to replace the outdoor unit, or maybe 1 indoor unit because the replacement part/labor costs more than a new wall unit. Will the replacements available then be compatible with the remaining system? Existing refrigerant pipe size and wiring going to work then? Refrigerant type? Maybe, maybe not? Innovation is fast, parts are very expensive out of warranty, HVAC talent is sketchy, and there are no guarantee manufacturers support all their equipment...ask me how I know this. That old-fashioned, un-sexy, less-efficient propane boiler might look expensive now, but in 10 years, it might be the cheapest and safest choice.

    1. Expert Member
      Akos | | #51

      Rich,

      Propane heat is dead in any application except for off grid. This is not because of some noble environmental reason but simple economics.

      In my area, it costs about 3x to heat with propane as it does with heat pump off $0.15 electricity. Even resistance heat works out to about the same cost as propane.

      Well insulated house like the OP has might loose maybe two degrees for each day the power is off in the dead of winter. Heat backup is not needed. The place is currently heated with TWO 1500W space heaters. Any budget generator from a box store will easily keep the house warm in case there is a long power outage.

      Most ducted mini splits can be had with a strip heater. In case the unit fails, the strip heat can carry the space if needed. Even if that fails, we are back to the needing to plug in the two space heaters.

      Radiant is also dead in any well insulated house that need cooling. Besides the fact that the floor will never be hot enough to be felt in any well insulated house, it simply costs too much to install for ZERO benefit. I have floor heat at home, and I can tell you about the only time I can even just feel it running is during a cold snap when I stand by the windows where there are extra loops. If I could go back in time, I would have never installed it.

      For proper comfort you will be installing ducts for cooling, it makes no sense not to use those same ducts for heating. Equipment cost between a heat pump and an AC only unit is not that much.

      1. 88cch | | #55

        Akos,
        I don't disagree with most* of what your saying, I'm certainly no fan of propane or oil and I have ductless wall units in my house as cooling/some heating use. They are ugly though. If you are building a house as was described, then yes, any properly sized HP (ducted or not) and 1.5kw back up heaters are fine, but they are not without risk with service, maintenance, comfort and future replacement as I mentioned. Yes, boilers, radiation, radiant floors, separate AC systems, pumps, might be " ZERO benefit" in your words for a high performance home, and certainly more expensive, but if I was in an area with sketchy HVAC services (which I'm afraid to say is most areas outside major cities), I would consider the more serviceable system for peace of mind and take the hit on efficiency, knowing what I do about my industry and heat pumps. The apparent belief that "mini-splits" are the solution for every HVAC application is not the reality of what we see everyday as an older, stable, service focused HVAC contractor. We seem to find a lot of messes made by HVAC contractors (or the homeowners/ builders that coax/bully weak HVAC contractors to do stupid things for short money). We are currently proposing a $30K replacement (including a $5K crane, but not the required painting, patching, carpentry to result) on 6 year old equipment, by a mid-level brand you would recognize, that no longer supports the units. Stay with the top brands on mini-splits.
        * in any home, high performing or not, radiant floors are very expensive, aren't always warm, could be considered a waste of money, but are more comfortable than any system that moves air- they just are- my opinion.
        Sorry to jump on this thread 3 years late, must be an HVAC busy thing...

    2. mpsterner | | #53

      Oofta, there's a lot here and I struggle with how much to dive in. It should be first noted that this post was reactivated by a comment, however, it is on our original design from 2019 and not the one we ended up building. There is another thread here where we explored heating systems again and settled on the approach. I don't expect you to read it but it's just for reference if others are interested.

      https://www.greenbuildingadvisor.com/question/heat-pump-hvac-systems-in-zone7-new-house-and-what-heat-system-for-backup#comment-210080?utm_medium=email&utm_source=notification&utm_campaign=comment_notification&utm_content=view

      My goal was to balance comfort (real comfort, as in, my wife says she's cold at 70° in our old farmhouse) with efficiency, cost, longevity, safety, indoor air quality, etc. I was not willing to risk comfort, hence going with a fully ducted solution. I was so concerned about potential comfort that I installed hydronic tubes in the basement slab. It is clear now that I won't be needing them.

      We have multiple layers of redundancy built in and I would expect that if we lost power on Christmas, we might just have a quieter time and enjoy it just fine! That said, one area that we completely ignored on recommendations is we do have a wood stove (Stuv) and ducted the air handler to be able to elegantly circulate the air. We had to be very careful on sizing it and get proper combustion air prepared. If you have a raging fire you need make up air if its a small fire the combustion air is enough. We also have the resistance electric heat accessory for our air handler so that can provide backup. Lastly, we can always plug in our 15oow heaters.

      Only time will tell on how this system holds up but they have a pretty solid track record on being long lasting. We don't have wall units since we're ducted but I would've been fine with them if they were positioned logically. It would've saved a lot of cost I am sure.

      I can tell you that this house is wildly more comfortable than the house that we lived in previously for 12 years with its hydronic floors. The floor was always warm but there were too many drafts and it pretty much always felt cold. I haven't tested it but practically ever surface in the new house seems to be the same temperature. I would also be concerned about the mass of water in a hydronic system in this house that can't cool down quickly if things are getting too warm.

      Unfortunately, in this case, the heat pump and ducting was more expensive than a top of the line hydronic system... This is probably my greatest disappointment. We live in a rural area and have very few contractor options and they're extremely busy so suffice it to say, we paid what we had to pay. From my perspective, the indoor air quality and A/C that comes along for the ride is far and away more beneficial.

      1. nickdefabrizio | | #57

        I am glad this worked out for you. I live in Zone 5 and heat most of my average home with (2) 12K Fujitsu low temp wall mounted mini splits mounted in the living room.. They work well; although our bedrooms remain several degrees cooler (which I like). .

        For those who are considering back up heating systems, for mini splits, let me suggest two things:

        First, if your utility company offers off peak systems/rates, look into them. Our rural Electric Co-op offers a system called ETS (Electronic Thermal Storage) from Steffes running on a separate off peak meter at 6 cents kw. It is essentially a box of bricks that heats-up during off peak hours, and then radiates this heat during on peak hours. This can be paired with a well insulated HW heater running off peak as well. The off peak hours are quite generous during winter times-including one hour during the day and all night and weekend...It could be a good back up to mini splits in that it can be configured as a hydronic boiler, air handler heat strips or a point heater. Best of all, the Co-op would allow us to hook up a car charger to also run off peak at 6 cents kw (!) and would allow a heat pump to also be run off this meter (which would in effect use the house to store heat) to combine efficiency and low cost. PV can off set this use as well (although the economics at 6 Cents kw take longer.)....... I am thinking of putting it in solely for the savings I would see for car charging...

        Second, check out ceiling mounted electric cove heaters. I recently returned from my brother in law's place high in the Wyoming mountains-one of the coldest climate zones in the lower 48. He heats his (average insulated home) with these ceiling mounted cove heaters. He keeps the thermostats low or off in rooms not being used and they seem to respond quickly if you want more heat. He claims that in practice they are significantly more efficient than electric baseboard because of superior control and comfort at lower output levels (he is an electrical engineer so he understands they have the same 1:1 COP input to output ratio but claims actual use affords better results). I was skeptical at first but I was surprised at how comfortable they were when the thermostat was at 55F . They are popular out there in his neck of the woods...and they are cheap and easy to install.

      2. musielski | | #58

        Having owned 3 homes outside Chicago that were 120+ years old, even at 70 F Deg you can feel cold. The "cold wall" effect and low humidity (20%) can create this easily. The Hayward cabin (Zone 7) can be heated with two 1000W electric baseboards. At .098/KWH I can't justify propane.

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |