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

Sizing Minisplit Capacity for and Small Rooms

nick_vk | Posted in General Questions on

If I have understood what I’ve read here, now that variable-speed compressors are the norm, the critical spec to avoid oversizing is the minimum BTU output.  So if I am needing to cool only 2oo square feet, an LG 9000 BTU, 27 SEER model that modulates down to 1000 BTUs/hr will probably perform better in terms of comfort/avoiding short cycling than a Carrier/Midea 9000K BTU, 40+SEER model that only modulates down to 5000 BTUs/hr,  and may even outperform a Mitsubishi 6k BTU, 33 SEER model that modulates down to 1500 BTUs/hr.

Am I thinking about this correctly?  I’m not ready to ask for design help yet (I need to redo my Manual J now that I’m just about done with envelope improvements), just trying to understand general principles.

I’m in Tucson, climate zone 2B.

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. Jon_R | | #1

    Roughly matching size to load is a good thing, but you are hung up on minimum modulation level and this has little to do with anything. So much so that some well designed units that could modulate lower avoid the "specs game" and deliberately start cycling. Because overall, this provides BETTER performance.

    Despite many such claims, less cycling does not necessarily imply better efficiency, comfort or longevity.

    For the right answer, check the specs (although some aren't available).

  2. Expert Member
    Dana Dorsett | | #2

    The cold-climate LGs with the more sophisticated compressors (eg LAN090HYV1 /LAU090HYV1 ) operate highly efficiently at their minimum modulation levels, whereas some other models with the same minimums have pretty lousy low-speed efficiency. The Mitsubishi FH09 and FH06 are both still very efficient at minimum speed.

    Note the difference in heating efficiency COP at minimum modulation @ +47F for these two SEER 27-ish 3/4 ton LGs:!/product/31912!/product/25817

    The LAU/LAN 090HYV3 only musters a COP of 1.5, compared the the LAU/LAN 090HYV1's 4.23, almost 3x the heat per kwh at that minimum output during temperate weather.

    The -HYV3 has a modest efficiency edge at max speed in cooling mode though.

    From simply a comfort point of view a minimum modulation rate that is 3-5x oversized for the design load is almost always going to be less comfortable, even if it is still managing to operate efficiently while cycling at a low duty cycle.

    1. Jon_R | | #3

      Note that both have horrible (1.0 and 1.5) COPs at 95F, 1023 Btu/h.

      Cycle fast enough and temperature variations are not an issue. Any unit at low speed should be quiet enough that variation in noise level isn't either. And if they aren't blowing on anyone, variations in airflow shouldn't be noticable either. Latent removal is typically non-existent at very low loads, so don't expect a difference with cycling there either.

  3. Expert Member
    Akos | | #4

    I would take all NEEP data with a slight grain of salt, it isn't always complete or proper. For a mini split COP when it is putting out 1000BTU doesn't matter much, the energy use is so small that in a house it is barely noticeable.

    The thing that matters for comfort is air flow. Most LG 9k wall mounts don't run much bellow 200CFM on low which is WAY too much in a room. Make sure whichever one you pick can go down to closer to 100CFM.

    1. Jon_R | | #5

      > 200CFM on low which is WAY too much in a room

      Do you have some data supporting this? Say in the case where the airflow is directed across the ceiling. ASHRAE information seems to indicate that 1 CFM/sqft is fine.

      Latent removal is a separate issue, but may not be a concern for Nick.

      1. Expert Member
        Akos | | #6

        This comes from painful personal experience of having to deal with a 9k head in a bedroom.

        The only way I got any level of comfort while the AC was running is by choking the intake down significantly. It is hard to get an exact flow number on a wall mount, but based on face velocity and outlet area, I'm down to about 1/2 the rated flow of the head. Even then I would call bearable.

        This is a high mounted unit blowing towards the ceiling and away from the bed.

        1. Jon_R | | #8

          You weren't comfortable because of humidity or because there was too much air movement? For the former, most situations will need about 400 CFM/ton. So 67 CFM when running at 2000 btu/hr. Agreed, most mini-splits can't do this (a big failure, even when perfectly sized).

          If the latter, I'm interested in the difference between your experience and ASHRAE guidelines (which seem to indicate that 200 CFM into a 200 sq ft room is fine).

          1. Expert Member
            Akos | | #9

            Too much airflow. No matter which way you adjust the vanes, you could always feel air flow. This is great for heat but uncomfortable for cooling.

            A properly designed HVAC system should never be felt.

            Most bedrooms have very little cooling load, trying to horseshoe in a wall/floor/ceiling mount in there and have decent comfort is impossible.

        2. ianrking | | #10

          @Akos. I'm having a similar issue with my units. How did you restrict the airflow?

  4. nick_vk | | #7

    Thanks everyone. Working on my manual J now, once that's done I'll post again.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |