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

Poor Dehumidification With Minisplit

tvrgeek | Posted in General Questions on

My workshop, about 24 x 36 has a Mitsubishi mini-split.  As far as temp, it works great. ( which surprised me how good they have gotten) .  But, it does a terrible job of de-humidification.  I have had it looked at twice and it is running as it should. Coils are running about 5o degrees, but darn little moisture is removed.  I need to keep it in the 50 to 55% range. The summer it runs over 70 until I added a cheap de-humidifier which does pump out the water for as long as they last, about one year which really adds up!

Is this normal for a mini-split?  No, the “dehumidification mode does virtually nothing.

Central NC, Zone 4

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

    Poor dehumidification at less than rated load is common. My guess is that the excessively high CFM/ton makes the specs look better. An effective alternative is the Daikin Quaternity - but it's expensive. Maybe you can reduce the airflow. Or trick it into running at close to rated load.

  2. tvrgeek | | #2

    Thanks. Confirms what the contractor said. Reducing the airflow does not help. Found out the hard way. It is in my woodshop so before I upgraded my dust collector ( ClearView 1800! ) it loaded up. I now have the intake enclosed in a very large MERV18 plenum. Coils clean, full airflow, plenty cool and warm. I also used an old furnace filter for an ambient air circulator, MERV 18 filters.

    1. this_page_left_blank | | #3

      Reduced airflow because of clogged coils is a different scenario than planned reduced airflow. With clogged coils, there's minimal surface available for condensing. The way lower airflow works to remove more moisture is that it keeps the coils colder.

  3. walta100 | | #4

    I have to ask since you did not say.

    Did you try setting the unit in it dry mode?

    Did you try setting the fan to its lowers speed?

    Do you know where the moisture is coming from?

    Is your floor sealed with epoxy?

    Does your unit have a continuous variable speed blower motor on the indoor unit? (This is a feature on the high end models)

    Never having lived with a mini split what you are saying does make sense to me. The compressor is slowed so that it is only running fast enough to keep the temp of the room at the set point and never cycle off. This means the compressor almost never operates at full speed and the coils almost never get as cold as they could and the coils generally are not much below the dew point. So you could see why moisture removal could be a problem.

    You may want to try setting a few changes in the temp like letting it warm over night and then cool it 2° below your normal set point for an hour 3 times a day. This will force the unit to operate at max speed so it can get below the dew point and remove more moisture.


    1. tvrgeek | | #9

      Yes to all. Actually RustBullet on the floor. Slab is over foam and poly. Moisture is coming from being in NC and outside humidity being very high. Been bad recently as it has not been very warm, but rather humid. The vapor barrier 1n the walls is of questionable quality. (previous owner) Garage door seals only pretty well. Yea, the compressor does not run very fast or very often.

      I'll look at the programing. It has all kinds of features I don't use.

      Low speed fan is "reduced airflow" but it does not circulate as well in the overall shop.

      Ironic. Super efficient to cool, but I have to run a low efficiency de-humidifier. So, is the combination as good as just a plain old low efficiency unit? Bet not. Cost sure is not as it seems de-humidifiers last about a year. On my third one for the shop (cheap HD ones) but my crawlspace has a top line expensive commercial unit which has now failed for the second time in 3 years. So, a $1200 unit is no better than a $200!

      The other thing I learned is not to hang the outside unit on the wall. I wanted it clear of the ground but low frequency vibration. Going to make more advanced mounts, maybe using automotive hydraulic mounts.

  4. walta100 | | #5

    Does your dust extractor filter the air and return it to the shop or does it dump the air outdoors?


    1. tvrgeek | | #10

      Dust extractor is all internal. ClearView 1800. Cyclone, HEPA filters. So no, I am not sucking 1500 CFM into the shop. It does a great job by the way. Much steeper cone than the Jet/Griz Chinese OEM stuff. Almost nothing makes it to the filters even when cutting MDF.

  5. _JT | | #6

    I find dehumidify works well but leaves the house too cool. So I toggle between heat amd dehumidify and that works well. But I suspect the mitsubishi may have a temperature set in dehumidify which is limiting drying potential. Toggling between heat and dry may help.

  6. Expert Member
    PETER G ENGLE PE | | #7

    You should also consider moisture sources. In a wood shop there should be very little moisture addition to the air, unless you are curing wet lumber inside (which you probably should not). If you are running high humidity, you might have excessive air leaks. Sealing the air leaks will stop exterior humid air from coming inside. It will also save energy. Unfortunately, that will mean the minisplit runs with even less load, possibly doing even worse at dehumidification.

  7. joshdurston | | #8

    I have the same problem sometimes with my 1:1 Mitsubishi FH series. My fly at your own risk solution is to set the fan on low and progressively block the intake with stacks of magazines and books until my discharge air temp drops into the mid to upper 40f range.

    I think the Sensibo climate react feature might be able to issue mode changes and commands based on temperature and humidity (essentially a dry mode with a low limit).

    The multi zone units have dip-switches to set the target evaporator temperature in cooling. Unfortunately the Mitsubishi 1:1 units prioritize uselessly high SEER rating over good mixed sensible/latent performance. I wish there was a override to lower the max coil temp in cooling mode into the 40f’s. Honestly, I wish I got the 12/15 instead of the 9 because of the higher min mod and much better SHR and higher pints/hr.

  8. walta100 | | #11

    I am sorry to ask the dumb questions but from time to time the answer is yes I am doing something dumb so I can’t help but ask.

    If you are correct and the moisture is coming from the outdoor air, it seems you must have lots of air leaking in to carry the moisture. Consider having a blower door test before and after having air sealing work done.

    If you want to remove moisture from the air you must cool the air below the dew point. To do this you must limit the air flow. I like Josh’s plan and use a box fan to circulate the cool air. Do so understand you are playing with fire if you go too far. The coil could get below freezing ice would block all the air flow. If that happens the liquid refrigerant can’t absorb enough heat to become a vapor again. Should that liquid get back to the compressor the compressor would likely be damaged.

    When you want to kill vibration mass is your friend pore a concrete pad and bolt a stand to that is taller than any snow you are likely to accumulate.

    Have you stored any green wood in the shop?


  9. tvrgeek | | #12

    With a 9 foot garage door on one end, 12 inch through wall exhaust fan, several windows, yes I am leaking a bit. But as I suggested, the vapor barrier is very poor in the walls. It is just the kraft face with 1/2 plywood over it and a coat of latex paint. If I caulked it better and used a shelac based paint, then it woudl have a good VB, but emptying the shop woudl be monumental!

    We just about don't get snow, ( I did move to the Piedmont for a reason) but dirt grass clippings etc. so I wanted it higher. I did not want to build a big ugly block pier. Not being attached would of course stop transmission, but there are other ways too. Lowering the brackets and good isolation mounts will work.

    I checked, the dedicated portable dehumidifier coils run about 38 to 40 degrees. The lowest I have gotten the Mitsubishi is a bit over 50. Just enough to make a difference.

    Now, I am looking at getting a blower door and IR test for my house.

    1. willymo | | #17

      What model indoor/outdoor unit is this?

      Try running the split in dry mode, minimum fan speed, set the temp to 61°F (the minimum). You may have to turn it off periodically if it gets too cold. You have to force it into high compressor/low fan. It's the only way it will dehumidify.

  10. tvrgeek | | #13

    Hardest part: Finding a dehumidifier that lasts more than year. Sani-Dry in my crawl space has failed twice in three years. Four portables failed in my shop. ( loss of coolant, pump failure on two, humidistat on one. ) I notice the crawl-space units list great sounding warrantee, but will not honor them. "Professionally installed" OK, pay someone to plug it in and connect a drain hose? Give me a break, this is fraud. Almost as bad as Ridgid lifetime warrantee on cordless tools.

    1. Jon_R | | #14

      Mini-splits and dehumidifiers are basically the same technology - except the mini-split is exposed to much more extreme conditions. I'm curious why the former is expected to last 15+ years and the latter far less.

    2. willymo | | #16

      Look at the Santa Fe dehumidifiers. Small, efficient, mine has lasted 10 years in a VERY damp basement.

      Not Cheap!

  11. tvrgeek | | #15

    And I have had window units last 20. Cheap ones.

  12. tvrgeek | | #18

    Santa Fe seems to be the brand most suggested. 10 years is better. 20 is what I expect. The Sani-Dri in my crawl space has failed every year. If it lasts another year after they fix it this time, my warrantee will be up and I would put in the Santa Fe. Same company has 4 brands. Don't know if they are al the same or not.

    I only need 20 or 30 pints a day worst case by bucket testing, so they are all too big. Same with my crawl space. Don't need defrost cycles in either as both are in heated/cooled spaces.

  13. willymo | | #19

    The mini 70 is small! Better to get higher efficiency that maybe doesn't run all the time.

    1. tvrgeek | | #20

      Even that is twice as large as I need for either space ( crawl space and shop)
      Looking at economics. $1200 unit that lasts 5 years, or a $100 that lasts a year and used 20% more electricity. I am in my shop every day, so I can notice a failure before my tools rust, but not in my crawlspace, so I need remote monitoring and would like an alarm. Both last failures I found by accident. Now I am going to have an additional air quality test and mold inspection as the CS likely was above 65% for more than 2 weeks.

  14. tvrgeek | | #21

    MD33 might work for the shop. 33 pints. Under a grand

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |