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

which one of these Fujitsu wall units is the better choice

kieran973 | Posted in General Questions on

I’m talking to a Fujitsu installer this week. I plan to ask for two one-to-one hyper heat wall units: one for downstairs, one for upstairs. I’m pretty sure that the 9k btuh wall unit is the proper size for the upstairs load. For the downstairs load, I’m not sure whether to request the 12k or the 15k btuh unit. Originally, I thought the 12k would be best, but after taking another look at the 15k’s COP at 47F (which is basically the average heating season ambient temperature in my area), it seems that the 15k would actually be more efficient.

The short version of my question: given that the average heating season temp in my area is 46F, which is a more important indicator of a mini-split’s average heating season efficiency and energy use, the HSPF rating or the COP at 47F rating? 

The longer version of this question:

Climate Zone 5
99% design temp: 11F
Average heating season (September – May) temp: 46F
Manual J heat load at 11F: 31,600 btuh
Manual J heat load at 46F: 11,000 btuh

12,ooo btu Halcyon wall unit
Outdoor Unit #: AOUG12LZAS1
Indoor Unit #: ASUG12LZBS
HSPF: 14
Heating COP at 47F: 5.34 (at min 3,100 btuh) ; 4.64 (at rated 16,000 btuh)!/product/51106

15,000 btuh Halcyon wall unit
Outdoor Unit #: AOUG15LZAS1
Indoor Unit #: ASUG15LZBS
HSPF: 13.4
Heating COP at 47F: 6.06 (at min 3,100 btuh); 4.59 (at rated 18,000 btuh)!/product/51107

My 1,000 sq ft first floor is about 63.5% of the heat load while the 500 sq ft second floor is about 36.5% of the heat load. So with an 11,000 btuh whole house heat load at 46F, the first floor’s load would be around 7,000 btuh and the second floor’s 4,000 btuh. Assuming a perfectly linear drop in COP between the minimum and rated outputs (3,100 btuh to 18,000 btuh) of these two units, my math says that the 47F COP at 7,000 btuh for the 12k unit would be 5.12, while the 47F COP at 7,000 btuh for the 15k unit would be 5.67. So it would seem that despite being larger, and despite having a worse HSPF rating, the 15k wall unit is actually more efficient at my average heating season temperature.

The 15k has some other benefits as well: more cooling output (though not a huge factor since I only use AC around two months a year); an outstanding cooling COP; and more importantly, a slightly higher max heating output at 11F which would better match the 11F design temp load of the first floor – this first floor design temp load is 20,000 btuh, and the max output of the 15k at 11F is 21,250 btuh while the max output of the 12k at 11F is only 17,050. (The 9k more than meets the upstairs design temp load of 11,600 with an 11F max output of 15,700).

In other words, for my 31,600 btuh whole-house design temp load:
option one: the 12k + 9k would produce 17,050 + 15,700 = 32,750 btuh
option two: the 15k + 9k would produce 21,250 + 15,700 = 36,950 btuh

The first option is more closely suited to the whole-house heat load (only 3.6% oversized), but the second option ensures that the first-floor load of 20,000 btuh would be satisfied at 11F; it would also mean having a more efficient unit on the first floor than the first option, which would reduce total annual electricity consumption; and while oversized for the whole-house load, option two is not grossly oversized (16.9%).

Anyway, sorry to go on and on. Any thoughts would be appreciated. Thanks. 


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. kieran973 | | #1

    I went ahead and requested the 15k instead of the 12k. And actually, I can see an argument for choosing the 15k over the 9k as well - yes, two 15ks would be oversized, but the 15k and 9k both modulate down to the same 3,100 btu/h, but the 15k has a better COP at 47F. So I don't actually see what advantage the 9k has besides a few hundred off the price. Anyway, thanks everyone for your help with my questions so far.

    1. bfw577 | | #2

      Most mini split manufacturers will use the exact same condensers between two similiar sized models like the 9/12k and 18/24. Their outputs are then software/hardware controlled. The service manuals are usually identical as well. My 12k Midea service manual covers the 9k model as well. The compressor and all the specs are identical.

      My guess is the 15k unit is just a derated 18k unit that gets better efficiency at low speed.

      1. ianrking | | #5

        Yea I can confirm this to some degree. If you look at part manuals for Mitsubishi indoor wall units you’ll see that almost all parts, including the heat exchanger coils, for 9k and 12k units are the same. The same goes for the 15k and 18k models. The only part number that differs is a circuit board.

  2. Jon_R | | #3

    > 12k Heating COP at 47F: 5.34
    > 15k Heating COP at 47F: 6.06

    Yes, despite the demonization of over-sizing, it can perform better.

    >more important indicator of a mini-split’s average heating season efficiency and energy use

    A lot of decisions are being made with bad data. For example, a study in Vermont measured an actual HSPF of 7.96 on a unit rated at 13.3.

    Manufacturers could be required to publish a model of the performance using various conditions as inputs. This would allow more accurate calculation of actual efficiency. IMO, this would result in considerable energy savings at minimal cost. Might also clarify that many heat pumps are being produced with excessively high SHR. This requires a dehumidifier, which kills system efficiency.

  3. kieran973 | | #4

    Good data about actual real-world COP would be very helpful. A lot of the decisions I'm making about which units to go with are based on certain assumptions - ie: that the combined COP of the two Fujitsus I'm considering is around 5.5 at 47F, whereas two Mr. Cools have a combined COP of around 3.5 at 47F - which could actually be flat out wrong....

  4. kieran973 | | #6

    Just an update on this. I had the 15K and 9K btu Fujitsu Halcyon wall units installed about ten days ago. Total installed cost after all rebates was about $7,000. I ended up putting both units on the first floor (no units on the second floor) mainly because the upstairs bedrooms are so small and boxy that it seemed like the 9K btu unit wouldn't have performed that well upstairs (heat would have had to travel in S-shapes and U-shapes around multiple tight corners, through narrow doorways, etc). Instead, I put the 9K unit in the first-floor bedroom (11x16) so it can heat that bedroom and its bathroom (11x10) - this bedroom/bathroom is a mid-century addition that has three exterior walls, poor insulation and air-sealing, and its own almost-flat, one-level roof, which I also believe has little to no insulation. I have this 9K unit set to 68F. Meanwhile, the 15K unit is in our dining room and it serves the semi-openish dining room/kitchen/living room. This unit is set to 73F right now. I have to say that both units are both uglier and more effective than I thought they would be. Uglier: just bigger and boxier than I expected, especially the 15K unit. More effective: the 15K unit in particular is really impressive. With the fan set to high, and the temp on the wall unit set to 73F (I've been told its thermostat is on top of the wall unit), it's 70F on the old eye-level gas boiler thermostat in the dining room, and with the bedroom doors open upstairs, it's around 65F up there. This means that we've almost completely eliminated our need for turning on our gas boiler. Right now, I sometimes turn it on for 20-30 minutes or so at around midnight (just to get some supplemental heat into one of the back bedrooms where the door is closed and the temp has fallen to around 60ish F), but other than that, we haven't had to use it at all.

    Long story short, we're very happy with these Halcyon units. We got so many quotes for $25K-$35K multi-floor ducted systems. It's amazing to me that for $7K we've been able to add a heat pump solution that takes care of 95% of our heating needs.

    So thanks to everyone on this site who helped me think through a lot of these details.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |