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

Which ductless minisplit?

jackofalltrades777 | Posted in Mechanicals on

Zone 4 climate – 735 sqft – R25 walls – R40 roof – R5 windows – very tight ICF home – glazing is minimal with a low SHGC

A Manual J was done and it showed a 1 ton unit would heat/cool the home. I’m undecided if I should bump up to a 1.5 ton unit? Here are my options:

A – DuctlessAire Mini-Split – 1.5 ton – 21.0 SEER – $1,400
B – DuctlessAire Mini-Split – 1 ton – 21.5 SEER $1,060

Is it worth the extra $340 to up-size to the 1.5 ton unit?

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. GBA Editor
    Martin Holladay | | #1

    The reason to perform a Manual J calculation is to find out your heating and cooling load. Oversizing your equipment usually results in worse performance, not better performance.

    Trust your Manual J.

    For more information, see Saving Energy With Manual J and Manual D.

  2. jackofalltrades777 | | #2


    I remember reading on GBA that over-sizing a ductless minisplit is not bad.

    My other concern is that the extreme temps are rising. Yesterday it hit 105F out here and that was record breaking. The extreme heat waves are becoming more and more prevalent and the Manual J's don't reflect those extremes we are now getting.

  3. GBA Editor
    Martin Holladay | | #3

    Peter L.,
    If you are worried that global climate change will alter the outdoor design temperature during the next 15 years (the expected lifetime of the equipment you will be installing), then use the outdoor design temperature that you anticipate when you perform your Manual J calculation.

    A high-quality inverter-driven ductless minisplit is less sensitive to oversizing than some types of equipment, because it can function efficiently at partial loads. But the turn-down ratio isn't infinite, and oversizing can come back to bite you during the swing seasons, when the equipment needs to operate at partial loads all the time. (Remember, peak loads only occur a few hours a year.) You don't want the equipment to short-cycle.

  4. keithhoffman22 | | #4

    A few thoughts for you:
    - are you buying a heat pump unit? If so, be aware that the cooling and heating capacity of the units differs. It might settle the sizing question for you.

    - I believe if you are talking about an inverter unit, you are probably talking about a heat pump unit (does someone sell a/c only inverter driven units in the USA?)

    - I think if you are buying a non-inverter model (a/c only, lesser brand?), then I believe sizing issues are same as with whole house ac.

    You might find this article helpful for understanding why sizing in relationship to your inverter range is important.

  5. jackofalltrades777 | | #5


    Yes, it is a variable speed inverter unit with heat pump.

  6. Expert Member
    Dana Dorsett | | #6

    Variable speed is never infinitely variable speed. The minimum modulated output and your actual loads both matter.

    The HSPF / SEER test submittal sheets for mini-splits specify the minimum modulated output at +47F out/70F in, and at 95F out/ 80F in, as well as the maximum capacity under those conditions.

    Without knowing your load & +47F & +95F as well as your design loads at your 1% & 99% outside design temperatures AND the minimum modulated output of the mini-split under the standardized test conditions there's really no way to know which (if either) is going to be appropriate.

    I couldn't find submittal sheets on the DuctlessAire site, and they only state the nominal heating & cooling capacities, not the minimum or maximum. The maximum is sometimes the stated nominal, but for heating mode it must be able to deliver at least the stated nominal when it's +17F outside, +70F inside.

    That said, I'll go out on a limb and assert that NEITHER the 1 ton or the 1.5 ton is appropriate here.

    A 725' higher-R ICF house with low SHGC U-0.20 windows in US climate zone 4 I'd be shocked if your actual loads are even as high 9000 BTU/hr, and CERTAINLY not 18,000 BTU/hr unless you sleep with the windows open, or decide to bake bread and boil pasta all day when it's 105F outside. A code-min house that size might be nearing the max capacity of a 3/4 tonner, but not an ICF house with high performance windows.

    I've seen too many pro-forma Manual-Js done by HVAC contractors that used all code min assumptions for R values and windows, and Manual-J's that also assumed duct losses/gains even though the mechanicals were ductless, etc where they basically ignored the actual R & U & SHGC of the actual house design.

    Who did the Manual-J?

    What indoor & outdoor design temperatures were used?

    What where the ventilation & infiltration numbers used?

    What were the internal heat source assumptions on the cooling load?

    Finally, what were the heating & cooling loads?

  7. jackofalltrades777 | | #7


    Manual J was done by a local AC company. I will have to dig up the docs to get all the info

    It looks like I will go with the 1-ton Mitsubishi MUZ-GL-12NA-U1 which is a 23 SEER and 12.5 HSPF and COP at 47F = 3.84 and 3.1 @ 17F. It puts out 12,000BTU of heat @ 17F and 9,7000 BTU at 5F. Unit cuts out at -10F below zero


  8. Expert Member
    Dana Dorsett | | #8

    HVAC contractor Manual-Js are usually not worth much more than the paper they're printed on, but dig it out anyway, so we can see if there are any obvious errors. The natural tendency of HVAC contractors is to use conservative assumptions about everything from air tightness to R values, or even indoor & outdoor design temperatures, rather than risk a callback from an irate customer if it doesn't keep up with the load, and that approach usually oversizes by quite a bit.

    It's written in Manual-J instructions to be AGGRESSIVE in assumptions, giving load lowering credit wherever and when ever possible. Even then it sometimes oversizes, but is much closer to reality.

    At typical 725' code min house would have a design cooling load less than 12,000 BTU/hr even in Texas (which is on record as being hotter than hell), and about 11,000 BTU/hr @ 0F at the cold edge of climate zone 5 somewhere.

    An R25 ICF house with U0.20 low gain windows windows and heat recovery ventilation (or low exhaust only would likely come in under 7000 BTU/hr @ 0F for heat load and under 9000 BTU/hr of cooling even with 5 of your sweatiest friends dancing in the living room.

    At almost the same web-store price, point Fujitsu 9RLS3 can deliver 12,000 BTU/hr @ +17F outdoors, 70F in, and can deliver 12,000 BTU/hr @ 95F out, 80F in, and might be a better fit overall, despite the higher minimum modulation than the GL12NA:

    Depending on what your contractor's Manual-J looks like after you dig it out, the FH09NA might be even better, even though it is $150-200 more money than a GL12NA. It too can deliver 12,000 BTU/hr @ +17F and +95F, but has a slightly lower minimum modulation than the GL12NA and :

    For sanity checking outside design temp and average weather data purposes, where are you located?

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |