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

Minisplit Condenser Efficiency

rossn1 | Posted in Energy Efficiency and Durability on

I am looking for some guidance on mini-split condenser efficiency. I have calculated cooling loads for my home (accounting for elevation derating), for the heads shown below. 3 of the 5 heads are upstairs and will be used regularly during cooling season, 2 of the 5 heads (movie, breezeway) are downstairs (partially below grade), and will be used intermittently, at most. I am only using the system for cooling.

The DSB I ran (shown below) indicates a 48K BTU condenser will run the system fine – 125%. Considering the heads that will run regularly – 88%.

One installer, who I would like to work with, was recommending I go with a 60k condenser (100%) and actually included it in his bid.

Am I understanding correctly that, in my situation, the system will perform more efficiently with the 48K condenser?

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. Expert Member
    Akos | | #1

    What zone are you in? If these will be used mostly for cooling, you have to get the system matched pretty close to your house loads otherwise you'll get pretty low efficiency and moisture removal. The outdoor unit should always be sized for the expected load from the house. Having additional indoor head capacity is not an issue for efficiency.

    The issue you have to watch is the minum number of heads operating matching up with the minimum modulation on the outdoor unit. The problem with the larger system is the min modulation on the 60k unit is 2x of the 48k. So if you will be running only a few heads often the outdoor unit will cycle a lot (generally bad for efficiency and humidity control).

    Looking through your plan a 12k head, even de-rated is silly size for a bedroom, unless that bedroom is about 1000sqft.

    Wall mount/ceiling mini splits are generally more maintenance. For something that will running all the time, a regular air handler or slim ducted unit with a proper intake filter is a much better choice. This allows you size the duct to each room according to the load and about the only maintenance you need to do is regular filter changes. A wall mount is fine for something that won't be used often like the media room.

  2. rossn1 | | #2

    Thanks for your thoughts. I'm near Denver (zone 5), dry, and about 3.5 months of cooling each year.

    The master suite includes the bathroom adjoining without a wall, and is about 700SF with cathedral ceilings. With derating, the head puts out 9500 BTUs.

    That is really good to understand about the minimum modulation... I will look into that further. In my remodel, only 1 of the 3 upstairs units could possibly be ducted (sealed roof assemblies), so that is pretty much out, but good to know.

  3. Expert Member
    Akos | | #3

    In zone 5, I would look at using a hyper heat unit, chances are you can heat your entire house with it and remove the gas burner. Since humidity won't be an issue, getting the sizing spot on is not as critical. Even if you have hydronic floor heat, chances are that operating the heat pump for base heat and only using the hydronics to keep the floor warm will be cheaper to run. Definitely the case if your fuel is oil or propane.

    I'm in the land of houses with air handlers in the basement and ducting on the interior of the house. Usually inside walls and floors with some bulkheads around rooms. You don't need an attic for ducts. A well designed system will have much better comfort than a wall mount in every room type of layout.

    Wall mounts do work well in larger open spaces. Make sure to plan on how to access it for the occasional blower wheel cleaning and unclogging the drain (~every 5 years), so 14' in the air is not the best.

  4. rossn1 | | #4

    The house is radiant heat, and the comfort from that isn't the warm floor, rather the mode of heating, so I would plan on continuing to use it a my primary heat source since it was a large investment for comfort and only 1 of the 4 bedrooms will have a mini split head. A hyperheat wouldn't be able to provide the heat needed, anyhow, it is an older home. However, all this does raise the question about if the extra investment in the hyper heat model is valid, in the event natural gas prices rise significantly, so thanks for the thought. No chance of ducting in this house. The upstairs is generally large and open.

  5. Expert Member
    Akos | | #5

    Gas will likely not change in price any time soon, I think your original plan for AC only makes the most financial sense.

    Since you do have hydronic floor heat and your climate is pretty dry, one option would be to look at hydronic cooling. As long as you can provide enough cooling with the existing emitter area with water that is above summer time dewpoint, you can cool the place with something like a LG Mult VS with a low temperature hydro kit with the occasional wall mount where you want extra cooling.

  6. rossn1 | | #6

    Yeah - interesting call for sure. I looked into this a few years back, when I was doing the radiant design (it's actually radiant ceiling, and not radiant floor), and it seemed like the equipment to do so was getting pretty spendy (of course, the cooling is now running more than I thought I would spend - around $17k) and also having to manage that dewpoint. One bad day could lead to a bit of a drywall/paint mess. Of course, this would give me cooling everywhere and it is good you re-surfaced this.

    I'm not clear -- is the LG Mult VS a special unit, and how does it plug into the existing hydronic heat design?

    1. Expert Member
      Akos | | #8

      The Multi V S is their light commercial single phase VRF unit. The hydro kit is basically an indoor unit with a refrigerant to water heat exchanger in it. Connects up to the branch box like any indoor head.

      If you can run the system with a fixed temperature that is always above your expected dewpoint, it should be simple enough to setup up. I would still put a largish wall/celling mount somewhere in case you need more cooling. As with anything, the devil is in the details. It would be an interesting project.

  7. Jon_R | | #7

    > will perform more efficiently?

    Generically, could go either way - some compressors are so inefficient (in energy use or latent removal) at low loads that cycling is better than modulating - they modulate low for marketing reasons. The right way is to have all the part-load efficiency values (including when cycling) and do a bin analysis with your loads/conditions. AFAIK, no manufacturer provides these values.

    Looking at what MXZ-8C data there is, it appears that any partial load operation is significantly less efficient - so match compressor rating to Manual J load. Even better, slightly under-size and live with the occasional few degrees warmer than desired in some rooms.

    1. rossn1 | | #11

      Mitsubishi doesn't allow you to exceed 130% with connected heads

  8. jameshowison | | #9

    Consider one-to-one compressors outdoors for the less frequently used units? Also ensure that the heads in the rarely used section can be on but not blowing. Many of the Mitsu units have an issue that their fan doesn't stop and there is cold refrigerant cycling past the head (because other areas are calling), or the heads have to have a resister clipped to allow this (what a bizarre hardware design decision). Basically the system is using the piping as a buffer tank to store the currently unneeded refrigerant (this is even the case with junction boxes for the refrigerant). Defeats the point of "separate zones" (also makes the on-board temp sensors basically useless and confusing). Maybe this is less of a problem with non-Mitsu equipment?

    I'm told that (and it makes sense that) those issues go away with one-to-one compressors as well as better understood minimum capacity and efficiency. Can be better for routing lines too.

  9. rossn1 | | #10

    Unfortunately, there's only one spot to put a single condenser, and that is 6' up on a wall. Due to distances
    (1 upstair unit at 101ft from the condenser), a branch box is needed for some of the distance to work out, anyhow.

    My only other option would be downsizing the two downstairs 6k units (that is closer to the actual BTU need, though I have other reasons for going to the 9k units) and going with a 3.5T unit; however only the hyperheat model is available, which will add to the pricetag. Additionally, the media room space is somewhat joined with the adjacent bedroom because there are a pair of Lunos Spot HRV's, so 9K is probably a better option, putting me back to the 4T unit.

    1. Expert Member
      Akos | | #14

      Be careful where you mount the outdoor unit. Although these are very quiet, the compressors still vibrate. Best is to only wall mount onto masonry. If this is not an option, go for ground mount.

      If you must mount onto light weight construction, try to do it far from the bedrooms, preferably near a corner of the house or where an interior wall intersects the exterior wall. You definitely don't want it in the middle of a wall as it will become a giant drum skin. Even that is not a guarantee that you won't hear the vibration from the compressor. Best is to stick to masonry or ground mount.

  10. walta100 | | #12

    Be sure you understand the systems with multiple heads are not able to reduce their capacity as much as the single head units do. If you know the system is over sized single heads have a better chance of modulating at a low speed where they are most efficient instead of cycling on and off where they are least efficient. Also having separate systems gives you redundancy when something does fail you still will have the other system working.

    I do not like the idea of over sizing the heads. I am not sure the flow of refrigerant gets stopped when a head is not operating.

    As small as the mini outdoor units are it is hard to believe there is only one spot with room. Note the units are often stacked vertically.


    1. Jon_R | | #13

      > modulating at a low speed where they are most efficient instead of cycling on and off

      The MXZ-8 seems particularly bad in terms of part-load performance compared to single head units. Using data here (MXZ-8C48NAHZ2-U1, 98F/72F), I see a 6.7% decrease in efficiency when operating at 75% vs 100%. A typical degradation coefficient (loss due to cycling) is less than this at perhaps 6%. Not clear why they bother modulating at all.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |