GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X 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

Tiny home heat pump (Mitsubishi or Fujitsu) for a mostly empty house

Kate9791 | Posted in Energy Efficiency and Durability on

I am in the process of renovating a 1-level 470 sq ft basement walkout condo in Kittery, ME. It has two floors above it. I’ve consulted a contractor, and don’t agree with their suggestion for a mini split. So, I’ve started doing research!

They suggest the Mitsubishi MSZ-FH15NA (they did not do a Manual-J, so I’m not sure what it’s based on.) Based on what I have found, there doesn’t seem to be a reason to go this big. I’m totally redoing the space with pretty energy efficient materials including 1″ insulation in the slab. It should be a pretty tight space with a low load. I’m not the type to turn the heat on very high…in fact, I travel about 3 weeks out of every month, so the heat will be kept around 55º.

I’m thinking more along the lines of either the Mitsubishi MSZ-FH09NA or the Fujitsu 9RLS3H. The bonus with the Mits is the ability to control it with my iPhone from when I’m gone (peace of mind since there is no other heat source.) But the Fujitsu seems to have a better warranty, I’ve had one in the past and loved it, and has better efficiency.

Hoping to get any input. Thank you!!!

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. user-1072251 | | #1

    The size you need should be determined by calculating the load so you're not guessing. The choice between the two brands is best determined by the support in your area for those brands. Assuming you are buying from an HVAC contractor, they should do a "Manual J" calculation and be able to show you the calculations and results.

  2. GBA Editor
    Martin Holladay | | #2

    At an outdoor temperature of 5°F, the Mitsubushi MSZ-FH15NA has a heating capacity of 18,000 Btu/h. At -13°F, the heating capacity drops to 11,160 Btu/h.

    I agree with you that at first glance, this unit seem oversized for your space. But the only way to know for sure is to perform a heat loss calculation. For more information, see these articles:

    How to Perform a Heat-Loss Calculation — Part 1

    How to Perform a Heat-Loss Calculation — Part 2

  3. Expert Member
    Dana Dorsett | | #3

    The FH15 or 15RLS3 are both ridiculous overkill for the heat load of most 470' walk-out basements.

    You can proabably do this one with the FH06NA and save a hundred or two in hardware cost below the FH09NA.

    Code requires that the heating system be ABLE to automatically heat the place to 68F at the 99th percentile temperature bin, whether you choose to set it to 68F or not. The 99% outside design temp for nearby Portsmouth/ Pease AFB is +9F, so you won't have to use some ridiculously lower number than that. Portland ME's 99% design temp is +2F- split the difference if you like, call it +5F "just to be sure" and you won't go wrong.

    The output capacity of thehalf ton FH06 @ +5F is 8700 BTU/hr, which would be a load/area ratio of 18.5 BTU/hr per square foot of conditioned space. That would be a ratio 50% higher than a typical code-min fully above-grade floor! Insulated basements are typically well under 10 BTU/hr per square foot, due to the limited amount of exposed above-grade surface area, and witntertime soil temps WAY above the 99% outside design temp once you're even 6" below grade or deeper.

    The Mitsubishi -FH09 and -FH06 both modulate down to 1700 BTU/hr @ +47F, whereas the Fujitsu -9RLS3 can only back off to 3100 BTU/hr @ +47F. For their greater modulating range at the low end makes the small Mitsubishis preferable to the small Fujitsus, given your very-low-loads.

  4. Expert Member
    Dana Dorsett | | #4

    As part of your renovation be sure to insulate the foundation walls properly if that hadn't already been taken care of.

    An IRC 2015 code-min foundation wall has to have the performance of R15 continuous insulation, which could be 3" of polyiso held to the foundation with 1x4 furring through screwed to the foundation with masonry screws, or an inch or more of polyiso trapped to the wall with a 2x4 studwall insulated with unfaced batts.

    In your location and intended low standby temp it's probably better from a mold & moisture point of view to bump that to 1.5" of polyiso if taking the studwall approach. Put an inch of EPS or XPS (but not polyiso) under the bottom plate of the studwall as a capillary break and to keep the wood above the subsoil temp in summer.

  5. Kate9791 | | #5

    Thanks for all the awesome info!

    Dana, after we waterproof the space, the walls will be insulated with 3" of R21 closed cell polyurethane. Plus 1" of rigid foam under the 3" inches of concrete in the floor and 1" of foam and about 4" of batt in the ceiling. Also putting in 5 triple pane Anderson Casement windows to replace the existing ones. The basement is below grade only on one side as the slope is pretty significant.

    I'm now leaning towards the Mitsubishi because there are way more dealers/contractors in the area and don't seem to be as many options for the Fujitsu. Plus the option of having the Kumo Cloud really makes a difference so we can check in on things when away.

    I have asked to see the load calculation the contractor did which brought him to quote the Mits FH15NA. It's my understanding that a Manual J calculation can only be done with speculation since the space is not finished yet. However, I don't think there will be a huge level of error since it's such a small space with a high amount of energy efficiency. That said, I still can't see how he thinks this unit is appropriate...

  6. Expert Member
    Dana Dorsett | | #6

    The 3" of closed cell spray foam works, but it's not nearly as green as doing it with 3-4" of polyiso (especially if you use reclaimed roofing polyiso) or a polyiso/fiber combination, due to the high environmental impact of the HFC blowing agents used, and the high polymer/R ratio.

    If you are plannning using 3" of foam between 2x4 studs in above grade walls it's a waste of good foam, since you would get comparable "whole wall" thermal performance out of 3.5" of open cell foam at a fraction of the cost. Going with 3" of closed cell foam on a fully above grade 2x4 wall comes in at about R 11.35 "whole-wall" after factoring in the thermal bridging of the studs, adding in the insulating value of the sheathing & wallboard, etc. . Going with 3.5" of R3.7/inch open cell foam comes in at about R10.35. That's a VERY expensive R1 you gain by going with closed cell. Installing a mere 1/2" of polyiso under the wallboard of the open cell wall brings the whole-wall performance to over R13, a significant improvement over the R11.35 performance of the closed cell insulated studwall, at much lower cost, and you'd be giving up less than 2 square feet of interior living space with that added wall thickness.

    With continuous R21 foam on an 8" poured concrete foundation wall, with a furring + 1/2" wallboard finished interior the U-factor of the wall is about U0.045 BTU per square foot per degree difference. Assuming a design temp of +5F outside and a code-min 68F inside that's a 63F difference, so the above-grade section of foundation walls are only losing 63F x U0.045= ~2.8 BTU/hr per square foot of exposed foundation wall. Below grade it will average something like half that much, worst-case.

    The bigger losses will be the windows and the fully above grade walk-out wall, but it takes quite a bit of exposure to add up to 8,700 BTU/hr.

    Seriously, build yourself a room-by-room heat loss I=B=R methods spreadsheet and see what it all adds up to. I'll hazard you'll come in less than 5000 BTU/hr before ventilation, and if you use heat recovery ventilation the ventilation heat losses are quite low.

    If you're digging out to insulate under a new slab, an inch of rigid is a bit on the parsimonious side to be commensurate with your R20+ walls. Go for 2" of 1.5lbs density "Type-II" EPS, which comes in at about R8.4, which should come in under $500 even if using virgin stock foam. Reclaimed roofing foam can be dirt-cheap. Either EPS or XPS is fine under the slab, but not polyiso (which takes on moisture over time if in contact with the soil.) For new foam use EPS, which is cheaper per R, and has a much lower environmental footprint. There are several reclaimers operating in MA, I'm not sure about Portsmouth/ Portland though.

  7. keiththarp | | #7

    Hi Dana,

    Thanks for the discussion, I am Kate's fiancé, we're working together to end up with the best all around solution for our project.
    We are leaning toward closed cell because we a sub grade on all four sides. 85% at one side sloping to about 60% 2/3 of the way toward the other side then dropping to @ 5% on the entry side with a bump out for the actual entrance that is just above grade. It is also a 100 year old property with a partial stone foundation and a history of mold and moisture. So with both of us being sensitive to mold and air quality issues we are reluctant to go with anything that won't play well with moisture, especially in places we can't see.

    There is a complete demo happening back to the stone walls, then there will be a poly vapor barrier installed against the walls. The previous finish was done on top of dirt floors. The 1" of rigid foam under the new 3" concrete pour was suggested by our consultant to create a thermal break, there will be a vapor barrier between it and the pour, and likely a vapor barrier on top of the floor prior to a floor install. There won't be any digging out, just removal of the existing layers of wood floor. Ceiling height is a concern so reducing the space we chew up under the floor is a concern.

    There will also be a drainage channel under the pour with a sump-pump.

    We are currently looking into Lunos for ventilation, a decentralized HRV system.

  8. Kate9791 | | #8

    Finally got the company to give me their heat loss calculations. I'm not sure what the numbers are based on. He didn't do any measurements or anything. Anyone want to comment? I erased the company's name as I'm not looking to call them out.

  9. Expert Member
    Dana Dorsett | | #9

    The outside design temp used is 8F and the interior is 70F, both are reasonable. That's just a cover page summary - they didn't include the sheets on how they got there, so there's no telling how far astray they might have been on U-factors.

    That said, the design heating load is 10,295 BTU/hr @ +8F, and the design cooling load is 4609 BTU/hr. Code min interior design temp is 68F, so from a code perspective you'd be looking at a heat load of 10,295 x (68-8)/(70-8)= 9,693 BTU/hr

    The calculated heating load is just a bit above the 8700 BTU/hr output of the Mitsubishi FH06 half ton, but within the 10,900 BTU/hr @ +5F heating capacity of the 3/4 ton FH09:

    That's only about a 1.1x oversizing factor, assuming the Manual-J was really correct, adjusted for a 68F interior temp.

    Unless Kittery has minimum oversizing rules (the way NY state does) the FH09 would do it, but there's no way you'd need anything bigger than the FH12, which delivers 13,600 BTU/hr @ +5F, which would be a 1.4x oversizing factor for the load using a 68F interior design temp:

    A Fujitsu 12RLS3H would have even more margin, and modulates down to a bit less than the FH12 Mitsubishi, but no way do you really need a 1.25 tonner. This would be the biggest mini-split to consider installing, unless going for a mini-ducted version (but if headroom is an issue you won't be going with mini-ducts).

    Given that the FH09 can modulate down to roughly half that of an FH12 or 12RLS3, and still covers the calculated load, I'd personally be inclined to go with an FH09, and presume that during Polar Vortex events it might need a bit of resistance heater to keep it at 70F. The calculated load still seems a bit high given that they gave it zero infiltration rate, but without the whole shebang to look at it's hard to say. If you're concerned, go with either of the cold climate 1-tonners, but not the 1.25.

    In cooling mode the FH15 can only modulate down to 6450 BTU/hr at +95F outside, which is WAY above your 1% outside design temp of 86F (per the Manual-J), and way above your calculated heat load at that temp, despite being calculated at a cooler inside temp. The thing will literally never module (ever!) in cooling mode. In heating mode it's minimum output at +47F is well over half your design temp heat load- it probably can't modulate any time the outdoor temp is above 25-30F or so, which is most of the heating season.

    The FH12 can modulate down to 3700 BTU/hr @ 47F, which is only slightly more than your heat load @ +47F, which is fine. It's min-mod cooling is 2500 BTU/hr at +95F out/80F in, still higher than ideal, but still not bad- it would modulate some during the cooling season, and would have beaucoup excess capacity for cooling it off quickly if you decided to just leave it off while you're away. The Fujitsu 12RLS3H would be fine too.

    If it's a fieldstone foundation closed cell polyurethane is indeed the right option. If it's quarried stone and reasonably flat you might still get there with rigid polyiso, but it's a judgement call.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |