GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Audio Play Icon Headphones Icon Plus Icon Minus Icon Check Icon Print Icon Picture icon Single Arrow Icon Double Arrow Icon Hamburger Icon TV Icon Close Icon Sorted Hamburger/Search Icon

Community and Q&A

How much are mini-splits really affected by altitude?

Dave H | Posted in General Questions on

Hello –

I’m sizing a heat pump mini-split for my 20×20 garage, which has been pretty well air sealed and insulated with closed cell foam. I’ve had a series of installers out and the recommendations were all over the place, ranging from 12k to 18k BTU, and from $3k to $12k.

Two of the companies actually bothered to do a load calculation, and I tried it online as well – but those numbers had a broad range, too.

Putting them all together, I would guess that I need roughly 6k Btuh cooling and 22k of heating (although, one of the results had the cooling at 4k and another had the heating at 49k!). I am using these numbers because the installer who really took the time to get into the load calculation produced those numbers, which fell between the other two results.

Now, I can look at different models (like the Mitsubishi & Fujitsu) and choose a size. One thing I don’t understand, however, is how altitude plays into this. I know that mini-splits have reduced capacity at altitude, but how much would one typically deduct from the ‘perfect world’ capacity of a minisplit to account for altitude?

Is there a formula for this? I’m at around 5350 feet. thank you!

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. User avatar GBA Editor
    Martin Holladay | | #1

    Dave,
    I'm going to invite GBA readers to comment on your questions. In the meantime, here are links to two relevant resources.

    Selecting Heating, Cooling Units for High-Altitude Homes.

    High-Altitude HVAC (This PowerPoint presentation includes de-rating tables.)

  2. D Dorsett | | #2

    The short answer is that it varies, and you would have to consult with the manuturer. The derating will not be the same for each model or vendor.

  3. Bill Dietze | | #3

    A good rule of thumb is to decrease the mini split output in proportion to the absolute air pressure.

  4. Dave H | | #4

    Fujitsu responded right away with a very clear document. I have attached it.

    I also asked Mitsubishi for info.

  5. Dave H | | #5

    Things are getting clearer.

    The installer who seems most trustworthy and took the most time to analyze the load is recommending at Fujitsu 18RLB. In the specifications for that unit, it says 'Min/Max Heating BtuH 7,000~20,000'. If I compensate for altitude, roughly, that's about 16,600 of actual heating at my location.

    That same HVAC pro estimate a max load of 20,503 BtuH for my space. Doesn't this mean that the unit he's proposing is a bit underpowered on the heating side?

    I also noticed that for the construction quality, he entered 'semi-loose'. This is a tough call because I sprayfoamed and airsealed the garage like crazy, filled all the bays with closed cell, and installed the windows extremely tightly. There is a garage door in there, but I've put in a new vinyl seal on the exterior and a big brush seal inside, plus upgraded to a well insulted/sealing door. I have a feeling this would be considered slightly more on the tight construction side then semi-loose, but I'm not really sure.

    Each of the HVAC pros who saw the garage all said it was the tightest garage they'd ever seen. I sort of get a kick out of the airsealing. It's like a game so I'm spend tons of time caukling and sealing. But, there is a lot of thermal bridging since I wasn't able to mitigate that with my skill level.

    Obviously there are a lot of moving parts to this - but with so many professionals giving different information, it's tough to know what is 'best'.

    - Dave

  6. Nate G | | #6

    Does this altitude de-rating reduce the efficiency, or does the electrical consumption drop commensurately?

  7. Dave H | | #7

    I'm not certain nor do I completely understand all of this, but reading the text on the PDF they sent it seems like the it's the 'capacity' that is derated, so I think it's derating the BtuH of cooling/heating that the unit can produce.

  8. Charlie Sullivan | | #8

    Dave, none of these calculations are going to be able to reflect your actual air leakage, and without a lot of detailed calculations, they aren't going to reflect your actual thermal bridging. So there's some risk that you'll end up with the sizing off somewhat. The traditional solution is to oversize it to be sure. But you could also consider:
    1) You can probably tolerate having the garage temperature a little lower than you'd want your living room to be in the middle of the night on a cold winter night. That means you can get away with less heat being supplied, and the heat pump will actually be able to put out a little more heat at 60 F than it does at 72 F.
    2) If you want more certainty, you could heat it with electric space heaters for a month or so, measure their electric consumption, and compare with the temperature record in order to back out a precise relationship between temperature and heating load, and use that to calculate what you need, even if your experiment didn't include January.

    The high altitude affects the heat pump performance by changing the effectiveness of the heat exchangers. The internals of the heat pump don't know what the atmospheric pressure is, but the temperatures they see at either end are worse (lower on the cold side, higher on the hot side), and so it produces less heat and operates less efficiently.

  9. Bill Dietze | | #9

    Attached is some 2012 Mitsubishi data on high altitude performance.

  10. Stephen E | | #10

    Why spend 3k to 12k on a mini split in garage? This isn't living space. A gas garage heater would cost 10 times to 40 times less and work a lot a better. Even a electric version. Mini splits aren't made for on demand heating that a garage environment demands. Infrared is perfect for a garage. Heat the vehicle right before you take it out or work on it. Shut off the rest of the day. So not only is the initial cost many times cheaper, but the cost of running the equipment is cheaper (fuel & maintenance) when put on a timing system. Want AC put a $100 window AC and be done with it. Sometimes the easy solutions are the best.

  11. Dave H | | #11

    Well, you've provided some pretty dogmatic information without having any idea what I'm planning to do in my remodeled garage.

    How do you know it's not living space? I am remodeling the garage to convert it into an office/studio, and will be running a business that includes a video studio and and the temperature control and noise level are very important. We need to be able to do long video shots and keep the temp stable with low noise levels, year round.

    Your $100 window unit suggestion is a no go but thanks..

  12. Stephen E | | #12

    You said a garage which are used for vehicles. If you mentioned that is going to be an office I would recommend a different solution. Don't be offended. Sometimes clarifying yourself helps.

  13. User avatar
    Dana Dorsett | | #13

    By all means DO share what the intended use and occupancy patterns of this space will be.

    It's one thing to pay for a high end heat pump for heating & cooling a recording studio, another for heating & cooling a bedroom or living room, and still another for intermittent use as a pottery studio or glass blowing shop, and still another if it's going to be for cultivating cannabis (legal under some circumstances in some high altitude US locations.) The "one size fits all" or "one type fits all" approach to HVAC equipment selection isn't always going to end up with optimal solutions. A mini-split solution might be a reasonable choice, but it also might not be the most optimal.

    A blower door test might help in zooming in a bit closer on the infiltration contributions to the loads.

  14. Charlie Sullivan | | #14

    It sounds like my comment about it being OK if the heat didn't keep up on a cold night should be revoked too, now that I know what the space is used for. You still might decide to accept some risk there, but it would take some time to catch up, so mornings could be cold as well.

    If the garage door is not in regular use, there could be ways to temporarily seal it for the winter, making it a lot more air tight and perhaps also better insulated, while still allowing it to be used if and when needed.

    If noise levels are a concern, it might be that an oversized system running at part capacity would be quieter than a right-sized system running at full capacity. It would also catch up faster if you had to turn off the heat during a recording session.

  15. Dave H | | #15

    Well, the noise is pretty easy to deal with in this context. A high sort of 'whoosh' sound like a minisplit would make is fairly easy to deal with if you have appropriate mics, etc. It's that low, rumble, kind of sound that would come from an in-wall unit that really gets tricky. A 'hum' like that is easily EQ'd out but it's worth avoiding so the mini-split is a winner.

    In regard to sizing, the garage door (which has been pretty well sealed and is insulated) will very rarely be opened. And the unit can run continuously while recording. In fact, it's almost better if the cycles are long, so that there is less 'on and on' noise in the background if any of that noise manages to leak through.

    It seems to me that the risk here is that if I choose that particular unit, there is a possibility that in some circumstances it will have trouble keeping the room fully heated. However, that's likely to be quite a rare event (i.e. maybe a few days a year, and likely only at night). If that's correct, it's not a big deal as I can bring in an electric radiator from the house.

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |