Do Mini-Splits Really Have a Dehumidification Problem?
It’s time for another update of my ongoing personal heat pump case study. I wrote here in 2020 about how the inverter-driven mini-split heat pump serving most of my main floor is undersized for cooling and for heating. And I followed up with how well it performed in a heat wave in 2022 and in an arctic blast in 2023. Now it’s time to show you how it’s been doing with humidity control on hot summer days. Many people say that mini-split heat pumps have a dehumidification problem in cooling mode. But is it true?
The video above covers a lot of what’s here in the article and provides more context.
3 days of data
I’ve got three graphs of temperature and humidity data below. They’re from the HOBO data logger in my bedroom. In the next section I’ll discuss what the outdoor conditions were like during this period, but first, here’s the graph of the dry bulb temperature (also known as just temperature) in my bedroom.
As you can see, the temperature goes up and down throughout the period. It went as low as 73 °F and as high as 78 °F. Those two extremes, however, were outliers in the 72 hours covered here. Mostly, the temperature was between 75 °F and 77 °F. The average of all the temperature data is 75.6 °F. The red horizontal line at 75 °F is the indoor design temperature I used in the load calculation.
Next, you see below the graph of the relative humidity in my bedroom. Again, the red line is the design condition, or 50 percent relative humidity in this case. Unlike with the temperature, though, this parameter spends almost as much time below the red line as above. So, over this three-day period, we averaged 50.6 percent relative humidity. Only once did it go above 60 percent relative humidity.
And now let’s look at my favorite psychrometric variable for humidity, the dew point temperature. The design conditions of 75 °F and 50 percent relative humidity correspond to a dew point of 55 °F. That’s the red line. Our average dew point during those three days was 55.9 °F. Who says mini-splits have a dehumidification problem?
Based on the data from these three days, I think it’s safe to conclude that my inverter-driven mini-split heat pump doe NOT have a dehumidification problem. But there’s more to it than what I’ve shown you so far.
The corresponding outdoor conditions
Of course, one of the first questions you should ask is what were the outdoor conditions. That has a significant bearing how the heat pump and the house will perform. First, it didn’t rain any of those days and had been dry for a few days before as well. The screenshot below shows that it was mostly sunny. And we had typical summer weather for Atlanta. In fact, those three days were pretty close to being design days, with high temperatures of 91 °F, which is just below our summer design temperature of ~92 °F. The low temperatures were 70 °F each day.
But what about the outdoor humidity? According to Weather Underground, our dew point temperatures ranged a low of 63 °F to a high of 69 °F over that period.
- 4 July 64 – 69 °F
- 5 July 64 – 69 °F
- 6 July 63 – 67 °F
That’s definitely on the humid side, but it’s not nearly as bad as it gets in Florida or other hot, humid climates. Orlando, for example, had dew points between 70 °F and 78 °F during the same period.
The outdoor temperature matters because the hotter it is, the more the heat pump runs. And the more it runs at higher capacity, the more it dehumidifies. The outdoor humidity matters because our house is not an ultra-high vacuum chamber. Some of that humid air does find its way indoors.
And that brings me to the last part here…
Other factors that help
As you can see from the above, the indoor conditions were close to what I designed for. The outdoor conditions were also close to the design conditions, and that helped. But there were a few other factors that played a role here, too.
Aggressive sizing. I mentioned above that my heat pump is undersized for cooling. Actually, it does meet the ACCA Manual S requirement for equipment sizing…almost anyway. Manual S allows you to size an air conditioner or heat pump in cooling mode as low as 90 percent of the Manual J cooling load. The one I’ve been discussing here is slightly below that, at 88 percent. And that helps because it keeps the compressor cranking at above the minimum capacity for more hours of the day. A lot of people think you can’t oversize a mini-split because of its variable capacity, but they’re wrong. When you size to the lower end of the range, you get better dehumidification.
Airtightness. My house was built in 1961. Before 2012, it had a lot of air leakage. Since that time, our blower-door-measured airtightness has gone from about 17 air changes per hour at 50 pascals (ACH50) to 7 ACH50. That means we’re not fighting the outdoor humidity nearly as much.
Energy recovery ventilator. A final reason that our indoor humidity is under control is that we have a Zehnder Q600 energy recovery ventilator (ERV) for whole-house mechanical ventilation. It brings outdoor air into the house, but it reduces the amount of humidity that comes in with that outdoor air. Yes, it does still add humidity to the house, but it doesn’t add nearly as much as it would if we were bringing outdoor air straight into the house with moisture recovery.
What about ductless mini-splits?
My focus in this article has been on the performance of one of the three heat pumps in my house. That heat pump has two indoor units, both ducted. And it performs pretty much perfectly during regular summer weather. But what about ductless mini-splits? Do they have a dehumidification problem?
The photo above shows the relative humidity (91 percent) and dry bulb temperature (41.5 °F) of the supply air coming out of the unit. That’s an astounding 39 °F dew point! The temperature in the room was about 72 °F. So yeah, that thing makes a super cold coil and was giving me a 30 °F temperature drop at the time I took the photo.
Is it condensing water on that cold coil and dehumidifying the room? You bet! I know that for a fact because we had a little leakage problem in the condensate line that dumped a bunch of water onto the counter below. We got that fixed without too much difficulty, so now all that condensate is going where it’s supposed to go. So we don’t seem to have a dehumidification problem with our ductless units either.
No sweat!
The answer to the title question is yes. Well, at least that’s the case for my Mitsubishi heat pump on hut summer days. As they say about cars, though, your mileage may vary. The aggressive sizing, airtightness, and energy recovery ventilator all help. So does the moderate outdoor humidity.
The real takeaway here is that mini-splits can do well at controlling humidity in hot weather. Do things right and you shouldn’t need a dehumidifier for those conditions. That doesn’t mean you don’t need a dehumidifier, though. There are plenty of cooler, humid days when no air conditioner or heat pump will control humidity, especially in the fall, spring, and even winter. For those days, you’ll probably want a dehumidifier.
Allison A. Bailes III, PhD is a speaker, writer, building science consultant, and the founder of Energy Vanguard in Decatur, Georgia. He has a doctorate in physics and is the author of a bestselling book on building science. He also writes the Energy Vanguard Blog. For more updates, you can follow Allison on LinkedIn and subscribe to Energy Vanguard’s weekly newsletter and YouTube channel.
Related Articles
My Undersized Mini-Split Heat Pump in a Heat Wave
The Perfect Weather for a Dehumidifier
Make Dew Point Your Friend for Humidity
What Is a Mini-Split Heat Pump?
Comments are welcome and moderated. Your comment will appear below after approval. To control spam, we close comments after one year.
I’ve been wrestling with this for a few summers now. We have a new house (built in 2017), sealed tight and overinsulated with thick insulation on the walls and in the attic. Not passivehouse, but well above normal. Great blower door results (don’t remember the number but it was single digits.) Got a Manual J and purchased minisplits (two ductless downstairs and 1 ducted upstairs) that are variable speed. For the life of me I can’t get the indoor humidity below 60%. I think it has to do with the crawlspace connected to my basement. I think I need more sealing done down there. I have two dehumidifiers down there running constantly and it still is high.
Clay: With an airtight house and two dehumidifiers, you shouldn’t have a humidity problem. One potential problem could be the hygrometers you’re using. Some are quite inaccurate. Another could be mechanical ventilation. If you’re bringing in a lot of humidity through a mechanical ventilation system, that can lead to humidity problems. And maybe your dehumidifiers aren’t doing as much as they should. Have you looked at the output?
Any recommendations for a good robust hygrometer? I find the my Hobo Onset UX100 units a bit off the two from Extech units I have HD500 and RH390. Differ by 10-15% consistently. Could be drift, but the Hobo units aren’t that old.
I have had good luck with 5 different Sensor push Ht.w’s. They also offer a better spec version that has a RH accuracy of 1.5%. The app calculates dewpoint. The higher end models are made in the US.
I also have a Testo 605i that is equally accurate but more importantly long term stable. These are all bluetooth devices so can do long term logging. The Sensor push is easier to use.
Thanks for the response. Good points. I need to do some more investigating.
Is the crawlspace floor sealed? If unsealed that can be a source of a lot of moisture.
Lee: Yes, the crawl space is encapsulated. I’ve seen some leakage sites in the band joist, but the ground and foundation walls are covered pretty well.
Hi, Allison…I was curious about Clay’s new construction crawlspace and I wasn’t sure if you were responding for him or responding about your own crawlspace.
LeeH: Oops! Yeah, I thought you were talking about mine.
Are you monitoring the humidity in your Attic?
Lee: Yes, indeed. See this article I wrote about that in 2021:
Can an Exhaust Fan Control Humidity in a Spray Foam Attic?
https://www.energyvanguard.com/blog/can-an-exhaust-fan-control-humidity-in-a-spray-foam-attic/
I’ve been making some changes since then and will have more to say about that when I have more data.
Wow, 7 ACH50 still seems like a lot of leakage. Do you have thoughts on where those areas of leakage might be in the structure? I assume if they were easy to get to, you would have addressed them already. As I have posted, all of my new builds are less than .5 ACH50.
Thomas: Some of it is from the crawl space, which was encapsulated in 2015 but still has too much air leakage. The rest of it is mostly from the above grade walls on the main floor.
Interesting – is that leakage measured as living space to crawl, or crawl to outside? I.e., the crawl is part of the conditioned living space and blower door testing includes the crawl?
Allison:
Just a quick check here. Using the HOBO UX100-003, there is only dry bulb and RH measurements. Did you just calculate dew point temperature from the logged time series? Script or coded in Excel?
Adriano: True, that’s what it measures. But when you download the data through HOBOware, it gives you the option of having dew point calculated from the T and RH data. At least I assume that’s what they’re doing.
I’ll have to check this on the desktop application.
First of all I’d love to live where you are since the weather is so much nicer than TX… No offence but when you get some actual hot weather it’ll be interesting to see the humidity removal. Those temps are almost cool night time temps.
As for he humidity removal you don’t need to do any fancy testing to see how well they remove moisture. Just put a bucket under the drain. If I do that with the shop unit I can more than half way fill a 3 gallon bucket in about 45 minutes. They are exceptionally good at moisture removal and I can get the humidity down in the shop to the high 20s which makes it far more comfortable. For me 50% is not comfortable 25%-40% is ideal for me.
And for oversizing a minisplit yes there are reasons to. If you are going for peak efficiency(with a possible hit on dehumidification) you want to derate the unit to 80% as that’s generally the peak efficiency range of minisplits. And in climates like ours yes you definitely want more capacity when design temps are around 100* and it’s not uncommon to have a week in the summer of 110*-117* with temps still in the 90s at midnight.
You look at sizing etc differently when you live somewhere with extreme temperatures and then those temps can go well above “design” temps for long periods of time. And they do go well below design for long periods for a week or more every few years. And shoulder months are generally shoulder days to weeks here so they are generally ignored and you can have a few shoulder days then it’ll be 100* and then back to a couple more shoulder days. And it’s not a dry heat here.
That said this year our summer has been greatly extended as misery season usually takes over in june and for the first time in decades the high pressure is not sitting on top of us.
Interesting article. I live in central Florida and have had a 4T Trane (traditional style) for 20 years. Over the years, many projects to improve the efficiency of the house (attic insulation, wall insulation, replacement doors, window tinting, replaced entire duct system, new sealed can lights in ceiling). Typical summer days the AC would run continuously trying to keep up (without success). 2300 sq ft, cathedral ceilings, large volume of air for the (1) system. Replaced in 2024 with 5T Daikin inverter system. House stays cool but the indoor humidity is high. The system does run a lot; enough where humidity should be controlled. I cannot believe the system is oversized because the 4T was never able to keep up with the heat load.
Kent,
what do you consider “cool” indoor temp? (e.g., I think 78F is too chilly!),
and what is “high indoor humidity” when the temps are “cool”…
and when you say system runs a lot – the article did mention the undersized unit runs for “longer periods” before entering off-mode – compared to over-size that runs very short (blows out lots of very cold air) and then goes to off-mode but also for short periods… hence, as you prob know, is why it’s called “short-cycling”.
78* is very warm for most people. For me 74* and 40-50% humidity is still a little warm but sticky. I’d be Ok with 74 if the humidity was in the 20s without that lower humidity 72 is a better temp if you are stuck closer to 50%.
With a set point of 78* humidity is going to be a big problem except on the days when it’s 105-115* and the AC is running allot more.
RH in 20s? wow – that’s super low even for winter temps in the teens, 20, 30’s. R u certain? 74F and 40% is a really typical comfort target (except for those of us who prefer 78-82F indoors! lol – admittedly oddball!)
I’m probably swimming upstream here, but my personal experience with a Pioneer ductless mini-split is that the control is just awful. I have the wired wall-mounted controller, but honestly, it is not noticeably better than the built-in controller. The unit often overshoots the set temperature by several degrees in each direction. The air handler does not shut off when the compressor is not running, so it adds a lot of humidity during those periods, as the coil is way above the dewpoint for long periods of time. In my mind, it is a design flaw that is shared by many ductless mini-splits.
So really poor temperature control, and no active humidity control seem to be hallmarks of the industry. By direct comparison, I have an central HVAC system with a single-stage condenser that does a very good job with both temperature and humidity. It has a HE communicating furnace, because that was required with foam insulation of the roof deck, so that opened the door to an ION controller, which is able to do some humidity control tricks, even with the single-speed compressor. For example, it starts the compressor before it starts the indoor fam ensuring the evaporator is below the dew point. And then it can also reduce the speed of the blower in order to keep the coil cold. And it most certainly doesn’t run the fan after the compressor shuts off.
I live in the Houston area, so the humidity is real. It’s made worse by the foam insulation, as very little humidity escapes via the chimney effect that happens with conventionally insulated homes. The main system is intentionally undersized at 2.5 tons for 2000 square feet of 1980’s construction. And that rather basic system typically keeps the temperature within one degree of the set temperature, and the RH within 2 percent of 50 percent setting. It works very well for what it is. Better than perhaps it rightly should.
Meanwhile, by direct comparison, the highly-vaunted mini-split technology is just awful. The manufacturer glibly admits that plus or minus four degrees is expected, and they can’t seem to tell me how to make the fan stop with the compressor. It just seems stupid, because with variable speed air handler and compressor, this thing should be a humidity control champion. Your graphs seems to indicate that your system has some of the same tendencies, with fairly wild temperature swings and spikes in humidity roughly corresponding to the times that the compressor is off. You have compensated by smart design, but you still some big swings compared to my lowly single speed compressor central system. And if Mitsubishi can’t do it, then what chance does Pioneer have? I’m reasonably sure that the unit is made by Gree, so Pioneer probably doesn’t even know how it’s actually supposed to work.
I take this to be another industry blind spot, where the concept is more appealing than the reality.
Minisplits generally use a temp sensor in the head itself which leads to issues. One issue is that if you have an undersized unit in a large space and the temp it sees near the head is getting closer to the set temp the system will start slowing down and reducing capacity. So in a large space say like my shop the air on one end of the shop will be getting cooler and the unit will reduce capacity. Just because it doesn’t have the throw to get the air way out there and really mix up the air well. That’s why they are meant for smaller spaces. So the way to fix that is to use an external thermostat. (some brands can use the temp of the remote)
There are several manufacturers of minisplits. I have installed a number of them here and there for myself and others etc. Even as far back as some 3 zone LG units back when they were making them. For me Mitsubishi would be my last choice. There are several reasons that aren’t worth getting into but the biggest one is ROI is not there for them by a long shot but they heavily advertise to the building sector so they get the most attention. Also for parts some are better than other for parts availability I did commercial refrig at one time before being an electrician but after an ASE master tech.
I get what you’re saying about temp sensor in the air handler, but I have a separate wire controller, which is not at the “head unit”.
My mini-split is 2 tons. It has a LOT of air throw. In the room it’s in, it spins the ceiling fan and billows the drapes on the opposite wall. It is a relatively small room, but it’s hard to imagine a room so large it wouldn’t be able to keep the air stirred up.
I didn’t buy Mitsubishi because they don’t fail to support owner purchasing and/or installation. But I do wonder if they (or their dealers) understand their own products better than Pioneer understands theirs.
As to ROI, I think that most energy upgrades probably never ROI. For example, I have seen fierce defenders of tankless water heaters, but it’s a thing that can never ROI in most use cases. Every week, someone comes to my door trying to sell me windows that are supposed to save more energy than the total amount that I use. I can’t imagine that windows would ever ROI, unless it was because of low-e glass in a truly hostile sunlight environment. It’s hot here in Texas, but direct sunlight never hits my windows during the summer months, so low-e would just make it less inviting to hang out in a sunny spot during the winter months.
Mini-splits in particular are hard to justify because they are not DIY-friendly and professional installation is very expensive. I’ve done a few of them. The second one took me about four hours, so its hard to imagine how an installer can justify charging thousands of dollars. They do because they can get away with it. But if anyone really wanted to promote the use of mini-splits, installation cost is a major barrier to adoption. As ism, in my mind at least, the idea that the air handler must always be on an exterior wall, and won’t fit above a window or a door in any house that I have ever lived in. There’s also no significant thought given to replacing existing window units.
For example, I have an ancient 2-ton window A/C unit in my garage. I framed the back wall to accommodate it. It’s mounted about mid-wall, just above the level of my truck’s hood. It takes essentially no useful space away from the garage. I would happily replace it with a mini-split heat pump, but there is no form factor that can easily replace the ancient “Coolerator” unit. The high wall units require too much space above and below, so I’d have to sacrifice a lot of shelving. The low wall units intrude into the space, so in my case, my truck would no longer fit in the garage. And those air handlers are huge and don’t work well mid-wall. My 2 ton Pioneer probably has four times the surface area of the window A/C and sticks into the space almost a foot. I considered a ducted mini-spilt, but they are even more expensive to install, and the garage doesn’t have a great attic. And it seems that 2-ton cassettes are too big to fit within the confines of any ceiling joists around here. And then there is also the issue that mini-splits tend to have a very minimal filter system, which wouldn’t be ideal in a garage.
But hey, I didn’t mean to hi-jack the post with me dogging on mini-splits. I just noticed that Allison’s temperature graph looks a lot like mine, which I found all the more interesting in light of his experience and choice the premium brand versus my value brand.
Robert, good points – the lack of total space air movement is my biggest complaint of the small wall mount mini-splits. It would be easy (maybe) to have air movers/ventilators placed in strategic dead zones thru out the space to either move hot air to the mini split intake, or move cool air from the split to the dead zone. NOTE: it is more efficient to fill a space with positive pressure and displace that air (and its contaminants) than to try to suck it out like a vacuum (i.e., negatively pressurize the space you want to change). This means, likely better to fill the “dead zone” with the air coming from the mini split unit.
Interesting article which includes dew point explanation. I live in northern Maine and built a double stud home I built in 1982. I have been monitoring temperature, humidity and dew point using SensorPush system main floor and basement. Few summer weeks of humidity is controlled with ductless heat pump. Hot water heat-pump seems to be controlling humidity in the basement.
As a long-time industrial hygienist with specialty in built-environments and health, I’m interested in two other factors . When I am assessing I include surface/material temps and air movement in a space (not just ACH or some other gross measurement of ventilation for the building. This is because these two effect mold growth more than humid/vapor alone. While some “fear” the high humidity, and even reference values at which “mold grows” these almost always ignore those two factors, material temp and air movement flowing over the material at risk. Absorption, not just condensation plays an important role in mold growth. I often refer to this condition as “basement closet” mold growth – where there is no visible liquid water – no leaks, drips, condensation, puddling – but materials are covered in fine dusty growth. This happens when there is no air movement (trapped in “closet” or crawl, attic, the corner behind the dresser, etc) AND the materials are hygroscopic (and there are many !).
The air movement must also be considered when assessing occupant comfort complaints, particularly in offices, but sometimes in some rooms in the home.
What do you think these two factors are in your home? good, average, poor? obviously, limited to certain areas.
And when do you use these factors in your work and designs?
What model hygrometer are you using to measure the minisplit? Does it read fairly quickly?
Just a couple of additions to a really great topic for those of us in humid climates. 1) I oversize my handlers at least one step up over the condenser unit size just for humidity issues. 2) I never rely totally on HVAC system for humidity control due to shoulder months where that unit is not running very much. A dedicated whole house dehumidifier is standard for all of my builds. Lately, I have been running it parallel with the air handler, pulling from the return trunk and feeding back into the supply side with a damper. For most houses a 70 or 80 pint is sufficient. Target RH is 45%. This setup lets me use the same trunking and filters as the main unit.
Thomas, have you done any volume comparisons – how much are you pulling from the return compared to the entire return volume? 5%? 20% etc? Can’t be that much – and compared to whole house volume, you’re dehumidifying so little with that relatively small dehumidifier. Do you know how long it takes to lower the entire home 10% or 20% to get to your target of 45? thanks.
I must point out that I am not an HVAC contractor/expert. I am a builder of concrete hurricane resistant homes in a very humid coastal climate. I started 15 years ago building these and found that my blower door numbers were really low 1 ACH50 and less. I realized I needed to learn more about the consequences of building tighter homes. My HVAC sub began learning together. Many of my homes are 36,000 cu ft in volume, 2000sf conditioned, 2 ton condenser, 3 ton handler, 70 pt Aprilaire dehum, minimal crawl space plus attic total envelope conditioned, .5ACH50 with QFAMD makeup air set at around 90-120 CFM with high temp/humidity limits set. Humidity set at 45% year-round. It handles humidity during swing months and supplements AC if needed during summer. Per my owners, humidity never exceeds 50%. My personal home is set up the same and I live on water.
Many of the minisplits I have seen, including my mid-range American Standard unit, have an explicit dehumidify mode. Mine is in a Houston garage, so it is not very airtight at all, and it will not control humidity in the regular cooling mode. But once I engage dehumidify, it sucks the moisture out of the air like crazy. It can easily maintain a dew point below 50°F as measured with 3 different accurately calibrated psychrometers. It appears to do this by running the fan very slowly then modulating the compressor to keep the coil just above freezing. The fan never shuts off in this mode nor does the compressor but at night the compressor is turning very slowly.
I had a similar experience in a hotel in Miami a month ago. I had to use the dehumidify mode.
Allison, I have a ductless mini-split with 1 outside unit, and 2 inside units each of which can operate independently (although one room cannot operate in heating mode while the other room is operating in cooling mode ). You have a ducted heat pump at your house (with a fancy pants duct), that serves two (or more) rooms/zones, right? When you turn it on, can zone #1 set a different temp from zone #2? I assume that when the air handler is running, air is supplied thru all of the ducts to all of its zones. Thanks for letting me know. ~from a retired armchair BS fan!
PS You don’t have to post my question, but I’d love to learn that little tidbit.
James: My two ducted air handlers have completely separate ducts and each has its own thermostat. So, yes, I can set them at different temperatures. But they’re using the same compressor and refrigerant, so it’s not perfect. I discussed the issue of refrigerant flow to indoor units even when there’s no call for heating or cooling in this article:
6 Reasons to Do Single-Head Mini-Split Heat Pumps
https://www.energyvanguard.com/blog/6-reasons-to-do-single-head-mini-split-heat-pumps/
I love your new name for my pair-of-pants duct! From now on, I’m calling it the “fancy pants duct.”
I’m still have the issue of I see all that date but I see no data on the most pertinent piece of information. How much water was removed by the minisplit. Anyone saying their X unit isn’t controlling humidity and isn’t measuring how much water the unit is actually removing doesn’t have any data to stand on. Your system may be removing 10 gallons a day and you still have high humidity from whatever sources in your enclosure.
So without that data point of how much water the unit is actually producing it’s all an interesting exercise but not very useful. It’s like in school where in algebra you went through all that work to just reduce an equation into a smaller equation which is useful at wasting time but without all the numbers you are still left with a question and no usable answer.
Robert: The amount of water removed by an AC or heat pump in cooling mode is important only in the effect on the indoor humidity. Measuring it may be a fun exercise for those of us who enjoy the geeky side of HVAC, but it’s not what really matters. It’s the indoor humidity that matters, and that’s what I’ve shown here.
Allison, I’m not sure I agree with “not measuring water” – for determining appropriate size and capacity of a dehumidifier, it is crucial to know that volume – then, apply it to the room volume and the water in the air (i.e., measured in grains). That leads to choosing the type, speed, volume, capacity, efficiency of the dehumidifier you want to buy and install. Others, you could be running a 40pint dehumidifier all day all night and never reach your target RH in the space. And of course, oversizing the unit might simply waste money on initial cost and/or operating energy cost.
Scott: Robert’s comment was about measuring condensate from mini-splits, not dehumidifiers. But yeah, with dehumidifiers, measuring water is more important.
Allison – oh, oops… sorry! ok I agree – needing to know mini split volume of water is not a thing! lol. Except maybe to help demonstrate why it’s not doing what people think it should – lower dehumidity to their “comfort” %.
If you are going to blame the minisplit for not removing humidity you need evidence to support it. Having high humidity does not mean the minisplit is at fault. It may be dumping out lots of water and you have a more moisture infiltration and sources than you realize. Or it may not be dumping lots of water but if you don’t get data to trouble shoot the problem you will just be guessing and making assumptions.
If your humidity is high and it’s not dumping allot of water you may need to lower the fan speed which will lower the coil temp and allow water to condense faster.
If I were walking into a customers site with the complaint of high humidity one of the most basic things to check is the condensate drain and how much water is flowing through it. Because you will know right off the bat if it isn’t flowing much you have an issue there. It’s an extremely important factor to know in troubleshooting. It’s a common sense item to check in a diagnostic approach. If there is allot of moisture in the air we want to know why and see if the unit is actively dumping a good flow of water and if it is your problem is elsewhere.
I think the problem with mini-split dehumidification is real, but is not quite captured by the test. It’s not that they don’t pull humidity out when actively cooling — it’s a combination of application-specific factors:
They’re often installed without external controllers. This means they sense temperature at the head, and must keep the fan running to get an accurate temp read. When the compressor is off (in low-load conditions, especially if it’s a multi-zone system that doesn’t have a great turndown ratio), they will re-evaporate humidity into the air. With short several-minute cycles in a small room, they may not manage any net dehumidification, as the coils may never become water-loaded enough to drip off. To defeat this behavior, for Mitsubishi, you need a hilariously overpriced external controller and to literally cut a jumper on the control board to let the fan shut off. Even window units have figured this out, and will only run the fan for a few seconds every 10 minutes during the off cycle to sense temperature.
The other factor is actually caused by the turndown ratio. If you consider a Mitsubishi 6000 BTU/hr single-zone system, the minimum fan speed is 137 CFM, and the minimum capacity is 1700 BTU/hr. Even if it doesn’t ever stop cooling, at nearly 1000 CFM per ton, is this unit going to dehumidify well when its temperature setpoint is satisfied? I don’t think so!! This 6000 BTU indoor unit, incidentally, has the same dimensions, weight, and fan speeds as the 12,000 BTU unit… but is rated to remove 0.2 pints/hr (!) vs. the 12,000 BTU unit’s 1.9. You can see it right in the submittal that it doesn’t want to dehumidify!
So, there are 2 problems here:
* For units that cycle, unless you install an external controller and cut the jumper, it’s in “fan on” mode
* For units that manage to ramp down, they’re not measuring humidity and won’t care that they’ve dropped their latent capacity to 0 under low sensible load
Some unhelpful anecdata: My medium-sized bedroom has a Mitsubishi 6k unit on a multizone system. It drops the humidity great during initial pulldown, but if I let it cycle all night set to 73, I’ll wake up at 65-70% RH!
I would try to link sources (the Mitsubishi documentation) but I think pasting the giant URLs may have made my comment fail to publish 🙂
Thanks for saying it better than I did. I have wondered if it’s just that the places where mini-splits are typically engineered and originally intended to be used, might just happen to be places where humidity is not a driving factor. I have to believe that if the engineer was working in a space somewhere along the Gulf Coast, cooled by his own creation, that he would have made some revisions on the very first day.
It’s so obvious to me that the fan shouldn’t be running while the coil is above the dew point. And it would be so easy to monitor the coil temperature and then adjust compressor and fan speed in an intentional manner. Caleb’s observation about a jumper that has to be cut in order to get the unit to behave as it needs to seems to exemplify wrong thinking. That thinking might make great sense in an arid climate, where it makes sense to wring every single BTU out of the dry air passing over the evaporator. In that situation, it might seem like a missed opportunity NOT to keep air moving over the coil. And the easy way out – putting the temperature probe inside the air handler, might even seem sensible. If only all of the heat was sensible heat.
Admittedly, all of my mini-split experience is with “affordable” brands, probably all made by Gree, but if luxury brands like Mitsubishi fail to address such a fundamental weakness, then no wonder the copy companies are following suit.