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

Mini Split Sizing for cooling and heating

yacpro13 | Posted in Mechanicals on

Hi everyone,
I’m new here. I am in the process of planning renovations for a newly acquired home. It’s unbelievable how many times I ended up on this site in my quest for information regarding energy efficiency and general best practices, so I thought I should finally join.

As part of the renovations we are planning on installing a mini split heat pump. The primary reason for installing the unit is for conditioning during the summer months; obviously the house already has a heating system (electrical baseboards) but no ac.

That being said, I am very much aware of the much greater efficiency of a heat pump vs resistance heating, and so if I will be installing one I would want to use it for heating as well as much as possible.

The house is near Montreal, so definitely a lot of cooling in the summer and a lot of heating in the winter. Cool Calc assumes the following design temps:

Summer Indoor Design Temperature: 73F
Summer Outdoor Design Temperature: 83F
Winter Indoor Design Temperature: 70F
Winter Outdoor Design Temperature: -7F

Here are the house specs (see attachment for a 3D pdf plan of the top floor).
-Built in 1987
-One floor and basement
-Main floor is open concept and has 2 bedrooms. Total area approx 900 sqft.
-8 ft ceiling
-Main floor windows to be replaced as part of renovations

So far, every single contractor I’ve talked to is proposing a 12000 BTU unit. However, some say that I will be able to use the unit for the majority of my heating while others straight out say there is no way and I will still have to mostly rely on resistance heating.

My top contenders right now are the Fujitsu 12RLFW1 and 12RLS3.

Your comments would be very much appreciated.

PS: By the way, manual J calculations seem to never be done around here, or at least not for project this small/simple. I did attempt to get a rough estimate using Cool Calc, but I am not confident in my input and therefore in the results.

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. Expert Member
    Dana Dorsett | | #1

    In your climate for space heating you should be looking at the "H" version, the 12RLS3H, or the Mitsubishi -FH15NA or FH12NA (with the optional pan heater installed). It gets cold enough long enough that defrost ice can damage outdoor units that don't have defrost ice protection designed-in.

    What did CoolCalc come up with for design heating & cooling loads?

    A typical tight 2x6 framed 900' house with decent insulation (including foundation insulation) and code-min low-E windows would come in with heating load in the range of 12,000-15,000 BTU/hr @ -7F (with numerous exceptions in both directions.) A Fujitsu 12RLS3H would be able to deliver the lion's share of the heat load (if not the whole thing) at -7F, but you are likely to need some amount of resistance heating in rooms doored-off. (I couldn't see the attached floor plan.O)

    The wild card in the calculation will be air leakage & ventilation rates, and it's possible to see 10,000 BTU/hr of just the infiltration + ventilation using default settings in some tools. If you're meticulous about air sealing and use heat recovery ventilation at modest rates that can drop a full order of magnitude.

  2. yacpro13 | | #2

    Hi Dana,
    thanks for the lightning quick response - I've actually read a few of your posts on similar topics in other threads.

    A system that can cool properly, and delivery the lion's share in heating is exactly what I am looking for. If distribution becomes an issue (because of the single source), I figure I can always install a WhisperGreen to take air from the open area into both bedrooms.

    Since you couldn't view the pdf, I am attached a screenshot with annotations.

    I did see the H and FH models. I had also found the RLS3 technical specs here:
    https://portal.fujitsugeneral.com/files/catalog/files/DTR_AS115E_01--ASU9-15RLS3_H1.pdf

    and saw that it can deliver 15000 BTU/hr at -7F exterior, 70F interior. I thought the main advantage of the extra low temperature units like the H was that it could deliver around 12000 BTU/hr at -15F, which is quite a bit below my design temp anyway...

  3. yacpro13 | | #3

    I just noticed I had to click the 'attach' button.....

  4. Steve Knapp CZ 3A Georgia | | #4

    With that simple floor plan, perhaps you could use mini split cassette to distribute the conditioned air.

  5. Expert Member
    Dana Dorsett | | #5

    The main difference of the -H versions is the automatic pan heater, which only adds ~$100 to the raw price. eg:

    https://www.ecomfort.com/Fujitsu-12RLS3/p65508.html

    https://www.ecomfort.com/Fujitsu-12RLS3H/p65511.html

    The RLS3 is pretty much identical to the RLS3H in all other respects.

    Heating a doored off space by moving 100cfm of air only a few degrees warmer with a WhisperGreen requires an insanely low heat load for that room. The specifc heat of dry air by volume is about 0.018 BTU per cubic foot per degree-F. A 100cfm fan is 6000 cubic feet per hour, and at a 5F temperature difference that's only 6000 x 5F x 0.018= 540 BTU/hr of heating.

    That is the heat output equivalent of a couple of sleeping humans + one sleeping cat.

    Regarding the Manual-J, the 6000 BTU/hr of infiltration can be cut to under 1000 BTU/hr if you get serious about air sealing an install heat recovery ventilation, which reduces that 24,000 BTU/hr heating load down to 19,000 BTU/hr. The 6000 BTU/hr number (=100 BTU/minute) at a 77F temperature difference implies 100/77F/0.018= 72cfm of natural air leakage- the equivalent of a bath fan running 24/7, which for a 900' house with 8' ceilings is about 0.6 air changes per hour of infiltration, which would be pretty high. At a code-max 3ACH/50 you'd have less than half that for natural infiltration, and if you got serious about air sealing it would be less than 1/4 that. Assume you can get that down to ~1000 BTU/hr, maybe less.

    The 3000 BTU/hr of loss through the floor being nearly 2x that of the losses for the ceiling implies there is a ventilated crawlspace with maybe R13 between the floor joists, or an unconditioned basement with no foundation insulation (?). Insulating the foundation with R15 continuous insulation ( 3" of rigid polyiso, or 1.5" of continous polyiso + a 2x4/R11 studwall) would reduce those losses by an order of magnitude (and deliver higher mid-winter barefoot comfort), reducing the total load to the 15-16K range.

    Working backwards from the area & heat loss, it looks like the walls are ~U0.068, which would be 2x6 studs 16" on center with R19 fluff in the cavities(?)

    Similarly, the window U-factor comes out to about U0.47, which would be typical for clear-glass double-pane windows (or single panes + clear storm windows), and no low-E(?). A minimalist low-E window (or re-glazing the storms with hard-coat low-E glass) would drop the windows to U0.35 or lower, which would knock the ~7600 BTU/hr of window losses down to ~5600 BTU/hr or less. In combination with air-sealing + foundation insulation that would bring the whole-house load down to the 12-15,000 BTU/hr range.

  6. yacpro13 | | #6

    Steve, I did take a look at the cassettes, but in my area they are quite a bit more expensive, not to mention they are less efficient, and I don't find them aesthetically more pleasing than a wall unit.

    Dana, thanks for the detailed insight. I get that adding a circulation fan doesn't make much sense with such a small temperature delta and such low air volume. Noted.

    The construction details are my best guess - the house this unit is for has been purchased but I am only getting the keys in 6 weeks. I don't know for sure the construction details at this time, though since I am replacing the windows, I will be able to input the precise data for those shortly.

    I find the Cool Calc design temperature (-7F) for heating a little aggressive. According to the nearest climate station, the average of the daily lows for this winter is around 15F, while the monthly one time lows are around -5F.

    Given that info I have narrowed down my options to 3 units:
    Fujitsu 12RLS3
    Mitsubishi MSZ- FH12
    Mitsubishi MSZ- GE15

    I looked up the heating capacity table as a function of indoor/outdoor temperature. I've placed all 3 tables in a single document (attached). Is it just me or the 12RLS3 completely blows the competition out of the water when it comes to heating? The Fujitsu's heating capacity is literally at least twice that of the Mitsubishi models.

    Am I reading this correctly?

  7. yacpro13 | | #7

    Can anyone confirm this? Does the RLS3 output 16600BTU @70/5F while the competition only outputs 6000-9000 BTU?

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |