GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X 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

Minisplits and moisture removal effectiveness

greenenvy | Posted in Mechanicals on

Considering that the better variable refrigerant flow mini split heat pumps, sized properly, are designed to operate largely without cycling off, it would seem that they would offer superior latent load performance (moisture removal) over the typical cycling split system. Is that true in the real world? I’m specifically interested in performance in high humidity environments like here in central Florida.


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

    In theory, your thesis makes sense. However, I'm not aware of any rigorous study that supports the thesis.

    There have been studies of air conditioner oversizing that failed to show that oversized air conditioners are less effective at moisture removal than right-sized units. The issue is complicated.

  2. Expert Member
    Dana Dorsett | | #2

    With continuous compressor operation you don't have the issue of condensation on the coils being re-evaporated and injected into the air the way oversized AC systems set up to run the air handler continuously while the compressor cycles on/off do.

    The sensible heat ratios of most mini-splits operating a at low speed is indeed quite a bit lower than most 1 & 2 stage split systems under low sensible load conditions, which is another way of saying they remove more moisture than the most common ducted systems. But that's not to say that they always remove ENOUGH moisture in those low sensible load conditions. A mini-split helps, but isn't necessarily the whole solution. Low sensible load houses will often need a separate dehumidifier to manage peak latent loads.

  3. charlie_sullivan | | #3

    " Low sensible load houses will often need a separate dehumidifier to manage peak latent loads."

    Just to expand a bit on why this is important---a well insulated house in a hot, humid climate will have the same latent load as a poorly insulated house in the same climate, but will have a lower sensible load. So moisture removal becomes a bigger fraction of the HVAC system's job as the house gets better insulated.

  4. greenenvy | | #4

    "The sensible heat ratios of most mini-splits operating a at low speed is indeed quite a bit lower than most 1 & 2 stage split systems under low sensible load conditions, which is another way of saying they remove more moisture than the most common ducted systems"

    Is broad terms, is there typically a crossover point in high sensible and high latent load conditions where the typical oversized split system will outperform the mini-split on the latent load, assuming the mini-split is sized properly? I'm trying to determine if there is a performance down side to the mini splits or, if done right, they are better across the board, with the exception of greater temperature imbalance depending on the distance from the mini-split.

    Thanks so much for your insights

  5. GBA Editor
    Martin Holladay | | #5

    I don't think that there is any performance downside to minisplits -- certainly not when it comes to moisture removal. If you are trying to heat and cool your entire house with just one or two ductless minisplits, of course you have to consider the problem of temperature differences from room to room. But that's another issue.

  6. Dana1 | | #6

    If undersized and running at max speed most the time the SHR would run on the high side, over 80%. If oversized and cycling all the time it's latent moisture handling won't be nearly as good as when it's modulating. It would have to be grossly mis-sized to fall much below the performance of a single-speed split system.

  7. Reid Baldwin | | #7

    I recall reading that some systems have a dehumidify mode in which the air flow rate is reduced but the compressor speed is unchanged such that the coils get colder. Is that effective? I can see that you would pull more moisture out of each ft^3 of air, but you would also remove more sensible heat from each ft^3 of air. Are these effects different enough in magnitude that the SHR changes markedly?

  8. Dana1 | | #8

    Most mini-splits have a dehumidify mode that operates at a fixed rate, but only the Daikin Quaternity series operates in dehumidify mode without sensible cooling, and the only mini-split with a dehumidistat control:

  9. Jon_R | | #9

    For the record, many mini-splits operating at less than 1/2 rated load (a common occurrence, even when perfectly sized) have little to no latent removal (humidity reduction), worse than a unit that cycles (turning off compressor and fan so that little re-evaporation occurs). Why - because the mini-split interior fan doesn't modulate down as far as the compressor, leading to excessive CFM/ton. Regarding dehumidify mode: in most cases, not a viable solution. But let's see some specs regarding part-load SHR, temperature control, etc.

    Plan on running a dehumidifier in any new house in a humid climate.

    Table 7:

  10. kjmass1 | | #10

    With just one of my 12K Fujitsu splits running 24/7 during peak humid summer months, I've yet to achieve the documented pints/hr. It keeps my 1800sf house at 72 no problem and certainly removes buckets of water and keeps humidity in the house under control, but take those numbers with a grain of salt. Dry mode might remove more but you lose temp control and it will just keep cooling.

  11. aunsafe2015 | | #11

    Just to add some anecdotal info to this thread, I have a Mitsubishi MUZ/MSZ-FH09 system, and it is so poor at moisture removal when operating at its lower capacities that I need a separate dehumidifier in the room.

    But once it ramps up to higher operating capacities on days that are 90+ degrees F, I can easily shut the dehumidifier off and the Mitsubishi keeps rel humidity in the room well below 50%.

  12. Expert Member
    Akos | | #12

    I just put a couple of books on top the mini split head, works great for reducing the air flow.

    Most modulating units have an coil temperature sensor and won't allow the coil to freeze up even if you restrict the airflow. Still more efficient than running a dedicated dehumidifier.

  13. aggierockdoc | | #13

    "Low sensible load houses will often need a separate dehumidifier to manage peak latent loads."

    I’m finding it hard to understand why this might be.

    My two premises are this:
    • A split AC system with a condenser outside and the evaporator inside can push both the latent heat of condensation and the condensate from the dehumidification process to the outside of the enclosure and process the interior air to give cooler, dryer air to the inside.
    • Yet a standard dehumidifier, where the process air passes first over the evaporator coil and then over the condenser, retains the latent heat of condensation within the enclosure.

    My conclusion is this:
    Having both a dehumidifier and AC running in the same enclosure means I have to spend more “energy-cash” in order to (1) run the AC to remove the sensible heat derived from both within the enclosure AND the dehumidifier (2) run the dehumidifier.
    My objective is to minimise energy usage and maximise system value:
    I’m happy to pay for value (ie good cost/benefit) in terms of achieving my goal ie keeping energy consumption down. (Energy costs are HUGE here – around 0.3 to 0.4 USD per kwh)

    My questions:
    What am I misunderstanding?
    What would you recommend in terms of an efficient HVAC configuration that delivers on low energy usage and high dehumidification capability with cooling a secondary nice to have?
    NB. I live in Costa Rica where dewpoints are at around 25C in the summer and around 18C in the winter. Ambient temperatures range from 22C to 35C in winter and 25C to 30C in summer. If I had a bumper sticker for all HVAC engineers here in Costa Rica, it would read “it’s the latent load, stupid”.

    1. Jon_R | | #14

      Designers of AC units seem to optimize for specs, not real-world efficiency or comfort. So they use the wrong CFM/ton at many operating points. So while it would be more efficient if the AC removed the moisture, it often doesn't and one has to use a dehumidifier to be comfortable.

      You want a unit where CFM/ton goes down with decreasing load (decreasing SHR). Most go way up at low loads. Adding a fixed air flow restriction is similarly flawed.

      A Daikin Quaternity will do a better job. Would be interesting to know if it would pay for itself in CR.

      > objective is to minimise energy usage

      Air sealing/testing and insulation is always important and usually cost effective. Solar AC is interesting.

      1. aggierockdoc | | #17

        Jon, thank you for responding. It's really helpful to hear an expert confirm my hunch that AC unit design is pretty much sub-optimally configured for the climate in which I live. It baffles me that the manufacturers kind of ignore the needs of the 2 billion people who live within 15 degrees of the equator.
        On a technical note, I'm particularly intrigued by your comment about needing a unit where air flow (ie fan speed) vs power consumption (ie compressor activity) should decrease with a decreasing load. I'm trying to get my around that.
        If in my scenario we assume there is no sensible load... the load would only change if I lowered the humidistat... so am I right to conclude you are saying something like...
        even though I want the device to decrease the SHR as I decrease the humidity set point, most (?all?) devices are too stupid to do that?"

  14. fcserei | | #15

    My solution was to use a regular thermostat to cycle on/off a Pioneer mini split in Dry mode around the thermostat set temp. I'm in Delaware. I had no problem to keep 75F, 45-55 deg humidity inside this summer. Electric bills were surprisingly low too.

    1. johns3km | | #16

      I do something similar- using a Sensibo it’ll toggle between dry mode and high speed cool for hot days.

    2. aggierockdoc | | #18

      At 24C/75F, I'd have to get my woolly socks out...

      In dry mode does your Pioneer just reduce the fan speed to provide the lower SHR... ? or does it slow the compressor as well?

      NB. My ideal set point would be to get humidity a little below 60RH with max daytime interior temps of 26C - 28C (and have ceiling fans keep us comfortable). Anything below that in terms of RH or temperature would be $ in the bin.

      As Jon R alluded to I'm planning on using 50 to 100mm foam batts all around the house enclosure, sealed up with a vapour barrier so that we can choose to let the outside in on good days, or keep the outside out on bad ones. Construction starts in 2 weeks!

      1. fcserei | | #20

        In Dry mode the fan speed is even lower than the low speed for the heat/cool modes. The compressor seems to be modulating also. The ducted mini split temp for the dry mode is set to 73F at the unit, slightly below the actual target temp I'm looking to get in the room.
        The best thing is that this way it is a set and forget thing, I don't have to fiddle with the mini split during the day like when I tried with the unit's original controls only.

    3. cldlhd | | #23

      I did something similar for my Mitsubishi multi splits in my bedroom. I have three wall units connected to the same compressor and the wall units were designed to have a range of 1700 to up to 12,000 BTUs cooling even though they're rated at 9,000 BTUs. Unfortunately the compressor they're attached to can only go down to 6,000 BTUs so often my well-insulated bedrooms will cool down quickly and then the fan will continually run and push moisture from the coil back in the room after the compressor outside shuts off. I bought these things called Cielo breeze on Amazon and set them to have the air conditioner come on when it hits more than 72° and have it shut off when it gets to 69 degrees. It turns the entire unit off including the fan which has prevented the humidity level from kicking back up was temperature is satisfied. We're basically in the same climate area, I'm just north of Philadelphia in Bucks county PA

  15. Deleted | | #19


  16. cldlhd | | #21

    I know this is an old thread and maybe it bit off topic because I'm talking about multi splits but unfortunately I learned after my contractor installed my Mitsubishi wall units in our bedrooms ( 2 bedrooms got msz-fh09na's and 1 got and 1 received a msz-fh06na) connected to a mxz-3c24nahz2-u1. Now according to everything I've looked up the wall units, if hooked up to their own outside compressor as a one-to-one mini split setup, are capable of going all the way down to 1700 BTUs in cooling mode but since the outside unit has a range of 6,000 to 23,600 BTU s that means the lowest my wall units can 'throttle' down to is 6,000 BTUs. So with well insulated rooms they cool too quickly, the compressor shuts off and the wall units fan keeps running distributing the moisture from the set coil back into the room giving high humidity. The work around I've come up with is some Cielo breez controllers from Amazon . I have the wall unit turned off but I program 'comfy' mode on the breeze to call for cooling (67° fan on auto) when it gets higher than 72°f then shutoff when it hits 69°. That way whenever the wall unit fan is running the compressor is and vice versa. That way the fan isn't pumping the coil moisture back into the room. Not sure how efficiency is affected though

    1. Jon_R | | #22

      There is a lack of data from multi-splits, so perhaps you can help.

      What is the bedroom design load? Other than humidity, was comfort acceptable without the Cielo controller?

      > lowest my wall units can 'throttle' down to is 6,000 BTUs

      If two bedrooms are calling for cooling and the compressor can modulate to 6,000 BTUs, doesn't this mean that each one can receive 3,000 BTUs? Or does the compressor then run at higher BTUs (15K or 18K)?

      > the wall units fan keeps running

      There is no fan mode that turns off the fan when there is no refrigerant flow? That would be a major design flaw in any AC.

      Very unfortunate that one has to use an add-on device to basically turn the unit into a single speed unit.

      1. cldlhd | | #24

        Honestly not sure what the bedroom design load is, it's a 70-year-old house and I'm not an expert. I think there is a mode within the Mitsubishis if you get into the programming where the fan will shut off without refrigerant flow but they're not like that from the factory. That's not a problem during the winter with heat as it helps distribute the heat around the room better but it's a problem in the summer with humidity. I don't have their kumo cloud app with that level of control so maybe there's a setting within it where you can do this. Yes it is kind of unfortunate to have to buy an ad on device luckily they weren't too expensive.
        Well I've operated with that theory If two of the bedroom wall units are asking for cooling then the 6,000 BTUs could be split between them get 3000 BTUs each, which still isn't 1700 BTUs which is what they're designed to be able to do down to if they have their own compressor rather than sharing one. Not really sure if they do split the minimum 6,000 BTU load, if so have an all free running with be able to get them down to 2,000 each But as soon as one unit is satisfied temperature wise it would stop calling for cooling and whatever it was using would go to the others.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |