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

Can a upper level, cooling minisplit cool a lower level bedroom in a PHIUS house?

Mike AbiEzzi | Posted in Mechanicals on

I’m building a 3,000 sqft PHIUS passive house. It’s basically a 2 story shoebox east to west. the main floor consists of a master bedroom (1/3 of the area in from  the west wall) and an open floor plan kitchen/great room (2/3 of the area in from the east wall) with an open stairway to the upper floor (along the north wall of the great room). The upper floor consists of the said open stairway across the north wall and a hall with three bedrooms along the south wall. 

The cooling load of the whole space is 5000 btu/hr. 

I’d like to place a mini-split sized for cooling at the north wall open staircase near the upper level floor (or will it need to be closer to the ceiling to be effective?). There’s a mitsubishi indoor unit that can split blowing directions and blow up and down at the same time. I’m fairly confident that it will be able to cool the upstairs rooms (with the doors open) and the kitchen/great room. What I’m not sure about is if it will be able to cool the main level master bedroom, even if the door is left open. 

I understand that passive homes can maintain consistent temps, so I’m hoping it will work, but I’m not sure it will in this case. Obviously adding another mini-split for the main bedroom would be way oversized, so that’s not an option. 


EDIT: Amazing community, thanks everyone for your replies!

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. Keith Gustafson | | #1

    How about a ducted minisplit?

    1. Mike AbiEzzi | | #9

      That would work, but would require a lot of ducting and would also hurt my plans of adding sound proofing to each room.

      1. Keith Gustafson | | #26

        Why a lot of ducting?

        to go from one source to two?

        I think there are sound proofing strategies for duct, the most obvious of which, for low use, would be a cover

        The idea here is that to go from zero AC or heat in the master to overcoming the small load does not require a giant duct, and if you are clever in placement of the unit, does not require 50 feet of it.

        And having the door open works very well, but negates sound deadening quite a bit I would think....

        1. Mike AbiEzzi | | #29

          This topic could probably warrant its own separate thread. If I have 6 zones and use a high static pressure horizontal ducted unit, I’ll have a trunk with branches for ducting supply. I’m not sure how returns would work.

          Then I would have to do something silly like zig zag the ducts and line them or create sound dampening boxes for each branch so sound doesn’t transfer between rooms. Even then I’m not sure how well that will dampen sound compared to no ducts at all. And how will those sound dampening modifications affect static pressure? And then what will the design of the returns be like?

          I admit I’m an amateur thinking about these designs, but it just seems like ducts would be unnecessarily complex and unscientific/risky to include sound dampening, all for several weeks of cooling a year.

          Don’t get me wrong, I would love to have zoned temperature control with one horizontal ducted unit paired with Airzones, but it would be expensive, sacrifice sound isolation and create maintenance (cleaning out ducts) and vents aren’t attractive.

          In an ideal world there would be a very low load multi-split outdoor unit with tiny recessed ceiling indoor units for each zone so I can have zoned temp control without any of the drawbacks of ducting. But such technology doesn’t exist. I’m actually going to try to run some straight conduit of some sort where bends in copper could be made at strategic points and allow any easy retrofit, just in case one day it does exist.

  2. Expert Member
    Dana Dorsett | | #2

    It will probably work the door open. To get a better idea, what is the design cooling load of the master bedroom?

    1. Mike AbiEzzi | | #5

      The design cooling load for the master bedroom is about 850 btu/hr.

  3. Deleted | | #3


  4. Jon R | | #4

    How well air circulates (for ventilation or heating/cooling) is a common question. Unfortunately, the best answers require computational fluid dynamics (CFD).

    A rough answer is that an open door with 1C delta-T will move about 200 CFM which is about 400 BTU/hr. A 2C delta-T (ie, moderate over-cooling of the open space) should work. Or a fan to increase air exchange.

    1. Mike AbiEzzi | | #6

      Hmm, interesting. So if there were a double wide french door, it could work?

      1. Jon R | | #10

        Probably - that would move twice as much air. And it's probably reasonable to ignore solar gain while sleeping.

        1. Mike AbiEzzi | | #16

          We actually won't have any solar gain in the summer for that area. Windows in that area will be 100% shaded during the summer.

  5. Joshua Van Tol | | #7

    Are you going to be heating with minisplits too? If so, you'll probably want one unit downstairs, primarily for heating, and one upstairs primarily for cooling. You don't need to run both of them all the time. I.e. run the upstairs one in the summer, and the downstairs in the winter. Run both when it's really hot or cold out. The smallest Hyperheat Mitusbishi would probably work, depending on your need for heating.

    1. Mike AbiEzzi | | #8

      Yep, that’s the plan, a dedicated heating mini-split downstairs in the great room.

      Heating load 20k btu/hr

  6. Trevor Lambert | | #11

    It might depend on how particular you are about the temperature. A 5kBTU/h cooling load is essentially nothing; probably half of that would be internal gains. Cooling the areas around it will lower the effect of all the gains on that room.

    I wouldn't even bother cooling a house like that.

    1. Mike AbiEzzi | | #12

      Yea, it’s really just for a few weeks a year that we’ll need it. Good point. I have a lot of trouble sleeping though in anything above 68°, so I’m motivated to make it comfortable year round. Although it’s only going to get hotter.

      Ok, I think you convinced me to not worry about. I forgot the amount of time that I’d care is pretty limited.

      1. Trevor Lambert | | #17

        If 68F is your limit, and you'll be sleeping in that room, then I would be worried. Is the load calculated at 68F? Usually cooling load is done at something like 75F.

        1. Mike AbiEzzi | | #18

          You're right. It's actually 77º. Hmm, I supposed I could have them calculate load at 68º to see how far off I am.

          1. Jon R | | #20

            Or 65ºF (for the open area) if you want the extra cooling to get the bedrooms to 68ºF.

  7. Jon R | | #13

    Keep in mind that load and temperature are only loosely related. One could have a 1K BTU/hr design load resulting in a 100F interior temperature.

    1. Mike AbiEzzi | | #15

      Can you elaborate? I'm having trouble deducing the risks based on your comment.

      1. Jon R | | #19

        Imagine a foam cooler sitting outside surrounded by 100F air. Eventually, with no cooling, the interior will be 100F. But it might take only a few BTU/hr of cooling to keep it at 68F.

        Summary - I wouldn't use "well insulated, low load house" alone to justify "no AC needed" (a possible interpretation of the last line in #11). But factors like moderate weather, thermal mass, duration, and tolerance can justify such a conclusion.

        With open doors and some over-cooling, your risk is low.

        1. Mike AbiEzzi | | #22

          Thanks for the explanation!

        2. Trevor Lambert | | #24

          A house with that low of a cooling load implies not only that the house is well insulated, but that the climate is not very warm. The small amount of cooling required to temper the house is typically available most nights, and having the well insulated house helps to buy you that time. With night time cooling, large swaths of sub-code houses in zones 5 and up do just fine with no active cooling.

          I understand the point you were trying to make, but I question whether the 1kBTU/h cooling load and 100F house scenario actually is possible. Given the fact that cooling loads include internal gains, for any typically occupied house a cooling load of 1kBTU/h actually means the house is losing heat, i.e. the ambient temperature is below 77F with no solar gain (or if there is solar gain, the outside temperature is sufficiently low to counteract it and then some). I think the hypothetical structure you suggest would be something like an underground storage shed near a hot spring.

          1. Mike AbiEzzi | | #25

            We can some pretty hot weeks in the summer. 90-100ºF+. And it's only going to get warmer with global warmin. If the house is warm to begin with, it's going to take some time to cool down passively if you want to go to bed at a reasonable time. According to WUFI passive, cooling is required to keep the home below 78ºF. A small cooling mini-split for a whole house is a pretty attractive proposition to keep a fairly large home comfortable year round.

  8. Deleted | | #14


  9. Expert Member
    Dana Dorsett | | #21

    Assuming the peak cooling load is 850 BTU/hr at some point during the day, with the bedroom mostly unoccupied, it won't have any problems at all staying cool enough to sleep if the door is left open during the day. In a high northern latitudes with late sunsets and solar gain from northwest facing window coming in after 9PM it there is some potential, but takes hours to raise the temperature by very much with only 850 BTU/hr of excess heat gain.

    A pair of sleeping humans adds about 500 BTU/hr to the load, but for most people and most houses that would only happen 6-10 hours after the peak cooling load from solar gains, etc.

    1. Mike AbiEzzi | | #23

      Great! If the main area gets warmer because we have the sliders open or something, we can always close the bedroom door to keep in the cool temps until later in the day we cool off the house again. Sounds totally managable.

  10. Keith Gustafson | | #27

    The problem with bedroom cooling is humidity. Two people breathing and sweating will drive up the humidity and you will want airflow to take care of that.

    Keeping the door open works great, but if you plan on staying up late to watch that hitchcock marathon in the living room, and the wife wants to sleep, prepare for complaints.

    1. Mike AbiEzzi | | #30

      Our climate is pretty dry (Colorado) and we’ll have an ERV, so I assume humidity wouldn’t be an issue for us.

  11. Robert Opaluch | | #28

    I built a passive solar home in Boulder in town. R-60 ceiling, R-20/25 walls, far from Passive House, with 4" tiled slab (thermal mass) on the first floor, carpeted bedrooms upstairs.
    (See section titled "Free air conditioning from the thermal mass", if interested)

    During the summer, yes it gets into the 90's outdoors. But IF the windows stayed closed during the afternoon, the downstairs stayed below 80F (upstairs with no slab thermal mass, into the 80's). During the night, one small open window upstairs allowed cool night air to float down the stairs and cool the thermal mass of the slab. Temps would drop to 68F in the morning, and rise 10F during the day. Interestingly, got the same temp range mid-winter (without auxiliary heating) on that first floor. Upstairs with no thermal mass, daily temperature variation was more like 15 degrees without heating or AC. Wish I had paid more attention and collected more data for the upstairs.

    My guess is that if you open a window in the downstairs bedroom or upstairs hallway overnight, and you have a slab or other source of significant thermal mass on your first floor, you'd get less variation than 68F-78F without using AC. With AC, larger SQFT, plus your higher PH insulation and air-sealing, your place should perform better, probably much better, than my old Passive Solar design, and be cooler during the summer.

    Yes, global warming may be changing that to higher temps, but you have AC to make it cooler and dryer during those hot sunny afternoons.

    1. Mike AbiEzzi | | #31

      Hi Robert. I currently have a condo in Boulder and even with windows open, the night isn’t cool enough to cool down the home within a reasonable amount of time in order to sleep. Even with the windows open all night, it might not get cooler than 75° by morning. Temps in the area seemed to have increased over the past decade. 100°F+ days are becoming more common in July.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |