Residential energy efficiency experts have begun to tackle the effects of decreasing a home’s heating and cooling needs on the indoor environment. Newly adopted building codes require homes to meet specific airtightness measurements and increased insulation levels. While few in the industry will argue against the need to build tight houses, such energy-saving measures can have unintended consequences. Many homes, especially in mixed-humid and hot-humid climates, experience interior moisture buildup that results in uncomfortable living conditions and, more importantly, indoor air quality (IAQ) issues.
Reasons good indoor air quality matters
Indoor air quality has been ranked among the top five environmental risks to public health by the Environmental Protection Agency. Even more alarming is that according to Underwriters Laboratories, six out of 10 homes are hazardous to the occupant’s health.
- We spend 90% of our time indoors.
- We breathe up to 3000 gal. of air/day.
- Adults take approximately 16 breaths/min.
- Newborn babies often breathe 30-60 times/min.
- Toddlers breathe 20-30 times/min.
- The average person breathes approximately 13 million cu. ft. of air during their lifetime.
There are four critical components to achieving good indoor air quality: building a home with a tight envelope, mechanical ventilation, high-efficiency filtration, and dedicated humidity control.
Build tight, ventilate right
Controlling the air that leaks in and out of the home as much as possible is considered the first step when the goal is to improve the home’s comfort and IAQ. Unless you know where the air is coming from and what it is bringing with it, you have little chance in controlling the quality of the air. Once the house is made tight, mechanical ventilation and moisture control methods need to be designed to match the upgraded building envelope, or the result could lead to an unhealthy living environment.
Bringing filtered outdoor air…
Weekly Newsletter
Get building science and energy efficiency advice, plus special offers, in your inbox.
This article is only available to GBA Prime Members
Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.
Start Free TrialAlready a member? Log in
37 Comments
Dehumidifier has been a massive comfort benefit for us.
An article on the impact of adding a dehumidifier during the Manual J/S/D process would be useful, there didn't seem to be a ton of info on that out there.
WIth newer mini-splits you have a very efficient Dry Mode which will cool the coils just enough to cause condensation. This is very low energy and removes large amounts of water from the air with some minimal cooling. Many new models also have cell phone options to set special settings with timers on the mini-split. This would be handy for folks who are fine at 84 or so degrees with low humidity. Very low energy use with a good 20 SEER or better mini-split which are common now and not expensive.
Vince,
What you’re saying in these comments is incorrect and the opposite of what Orr is quoted saying in the article. The newer higher SEER mini split heat pumps dehumidify less and less. “Dry” mode will further reduce the temperature in the room making it clammy in my experience. In a humid climate, all homes including new or leaky can benefit from whole home dehu.
Air conditioning has what called the sensible heat ratio -- SHR -- which is the ratio of sensible heat to total heat load. (Air conditioners remove two types of heat: sensible heat, which you "sense", by reducing the temperature of the air, and latent heat, which removes humidity without lowering the temperature of the air.)
A perfect dehumidifier would have a SHR of zero -- 100% of the cooling goes to dehumidification, there is no cooling of the air. Unfortunately that can't happen with just a cooling stage. The way that dehumidification works is that the air is run over a cold surface and cooled below its dew point, at which point the moisture condenses out. Unless you have a post heating stage that air is going to come out colder than it went in.
The colder your surface the lower the SHR, so if you want more dehumidification you set up the air conditioner so that the air is cold. However, there is a practical limit to how cold you can make it; if the condensing surface is below 32F the condensation will freeze and block the airflow.
To insert some sample numbers into the conversation, let's say your room air is at 72F and 50% RH. That air has a dewpoint of 52F, cooling it to any temperature above 52F will have a SHR of 100%, all of the cooling goes into making the air colder and there will be dehumidification. Cooling it to 45F gives a SHR of 74%, to 35F gives 67% and going all the way to 32F gives 66%. So even in the best case the majority of the cooling goes to making the air colder and not removing humidity. Going with a cold temperature and low air flow may work in situations where you need more than typical dehumidification and only a little cooling, but they don't work in places that need a lot of of dehumidification and no cooling -- like say a basement.
With a dehumidifier, all of the heat that is removed (plus the heat from the electricity of operation) is returned to the air as sensible heat after the moisture is extracted. So the air exiting the dehumidifier is hotter than the air coming in. And drier, of course. That may or may not be what you want either.
Conceptually, the Daikin Quaternity is the Goldilocks solution, there is a separate heating coil after the cooling coil and they are independently controlled. So sensible heat is decoupled from latent heat, you can get as much dehumidification as you want at any temperature. Unfortunately I don't have any experience with them and people who do have experience seem to be vanishingly rare.
"A perfect dehumidifier would have a SHR of zero -- 100% of the cooling goes to dehumidification, there is no cooling of the air."
I see where you are going with this comment, but in reality this is not true. It is impossible to condense water with out cooling air. Theoretically if incoming air is very high relative humidity like 99% (which is impossible) you could drive very low SHR number but it would never be zero, because the dehumidifier has to cool air below the dew point in order to condense water. So that very statement implies some sort of sensible heat transfer. "Cool air below dew point". In reality the performance of the dehumidifier is largely dependent on the conditions it is operating in. The efficiency on the other hand is a function of operating conditions and the design of the dehumidifier itself.
I wish you'd read the following sentence: "Unfortunately that can't happen with just a cooling stage."
And then I go on to show why you can't get a SHR below 65%. If you need a SHR below that, you have to add heat to the air to cancel out some of the sensible cooling. That's what a dehumidifier does. However, dehumidifiers are blunt instruments, they add all of the removed heat -- plus the heat from the electricity used -- into the exhaust air, so the exhaust air is warmer than the incoming air.
Ideally what you want is a system that measures both humidity and temperature and only adds back exhaust heat when it is necessary, otherwise it gets dumped outside. This is how the Daikin systems are advertised to work, I just don't know if they actually work that way.
"very efficient Dry Mode which will cool the coils just enough to cause condensation" This statement sounds exactly like a perfectly dialed in ultra efficient air conditioner. This mini-split system dials in that coil temperature so that most of the compressor energy is going to cooling air. This also allows the maximum capacity to be extracted from the compressor, because of elevated coil temps. This is exactly the opposite of how a dehumidifier operates. I have always said, dehumidifiers are just really really inefficient air conditioners. So conversely any ultra efficient air conditioner is a very very inefficient dehumidifier, it is just psychometrics, that efficiency has to come from somewhere.
Dehumidifiers are highly efficient space heaters.
All of the heat they extract from the air is returned to the interior as sensible heat, as well as all of the heat from the energy consumed. This is typically on the order of 3-4 times the amount of electricity used so you're looking at a COP that compares favorably to good heat pumps.
My point is, why run a space heater when you're cooling? It's more efficient just to run the air conditioner colder and do more dehumidification.
I feel that in spring and fall the dehumidifier should be perfect for where I live. Add a little bit of heat to the house and dehumidify. Adding cool air (with A/C) to remove humidity would not make sense when temperatures are between 13-18C.
My experience with 'dry mode' has been underwhelming. We had a test hut with an oversized Mitsubishi wall cassette mini-split (not my call on sizing). Interior RHs were running in the 70% range in the summer. Very airtight, no ventilation, so outdoor air leakage/high dewpoint outside air was not the issue.
We tried turning on dry mode on 8/8/2017--per the temperature graph, you can see that interior temperature dropped from about 72 F to 67 F--a pretty significant subcooling, possibly pushing things out of the comfort range for many folks. Outdoor temperature is shown in light orange in our graphs.
Interior dewpoint did decrease, from maybe ~66 F to ~55 F, under the recommended 60 F dewpoint target. However, interior RH barely budged from 70% with the change in interior temperature.
The remainder of the graph shows something going wrong with controls after a power spike or something... interior temperatures dropped into the low 60s. Returned things back to normal operations around 8/22, and back to the same dewpoint/RH as before.
Thank you for sharing, Kohta. This information needs to be more visible in the industry.
Vince - do you mind sharing some makes/models that work well for this?
There's been periodic discussion on this site about how much humidity the Daikin Quaternity line can remove without sensible cooling... Ex: https://www.greenbuildingadvisor.com/question/personal-experiences-with-daikin-quaternity-ductless-heat-pumps
Would be great to know if there are others.
M
Mitsubishi, Midea, Senville, Darkin,, Mr Cool DIY and others have Dry Mode and work well. These are also high SEER, some over 25 to 33 SEER.
The Mr Cool works very well in a relative's home for last few years, and is easy to install, no fancy tools needed. Dry Mode did not seem to cost much in the utility bill.
These can dehumidify much more space than traditional older units.
Home should be tight and have good insulation to get the most/acceptable benefit.
It takes less energy to move heat than make heat as they say with Heat Pumps, and reduce dangerous heat-trapping excess carbon emissions and save money.
Vince,
From the data we are seeing from the manufacturers, typically, the higher the SEER, the higher the SHR (F). In the chart, you will see that a 33 SEER system has a SHF of .96. This means that 96% of the BTUs are focused on cooling and only 4% are focused on removing water. Dry Mode is typically a timer setting that is just overcooling to dehumidify, which can cause comfort issues and even lead to condensation problems.
That makes perfect sense. With a heat pump the amount of energy required depends on the temperature delta. You can make a heat pump more efficient by reducing the delta. Since you can't make the outdoor air colder, the way to do that is to make the indoor side warmer. If you look at my post #7, SHR goes up as evaporator temp goes up. "Dry" mode is just pushing the evaporator temperature back down to give acceptable dehumidification. It's really just as simple as reducing the air flow with the same refrigerant flow.
I believe the future is controls that have both a thermostat and a humidistat, so the SHR can be tailored to give acceptable dehumidification at the best possible SEER. Although I keep predicting this and it keeps not happening. Chiltrix has introduced this with their "Psychrologix" controllers but they're pretty far out of the mainstream: https://www.chiltrix.com/documents/Chiltrix-Psychrologix-TS.pdf
Note that a dehumidifier is not the solution, you're basically taking all the energy savings from the high-SEER system and giving it back. Dehumidifiers only make sense when the sensible heat load is too low for the latent load.
"Dehumidifiers only make sense when the sensible heat load is too low for the latent load." Totally agree, which in most green grass climates will be in the spring and fall - overnight time. If we look at ASHRAE peak moisture load times, we see they are very different from what we use in Manual J for peak sensible load times. For example in Birmingham, AL peak design load temp is 94 degrees, and the ASHRAE Humidity Design load is 83.
The whole reason we are tightening homes is to reduce the amount of run time on the largest energy consumption system in the home, but by doing so, we are also having to rethink how we control the humidity.
No matter how energy efficient a home is, if it is not healthy and comfortable, it won't matter.
"I believe the future is controls that have both a thermostat and a humidistat, so the SHR can be tailored to give acceptable dehumidification at the best possible SEER. Although I keep predicting this and it keeps not happening. "
It isn't happening because controls alone can't make the system operate in the conditions it needs to. It is purely a psychometrics problem, when there is a high sensible load an AC system can be tailored to meet the needs. However when sensible demand is low that is the problem, because that latent to sensible ratio is very much slanted to latent. In theory a 4 ton A/C system would need to operate from 4,000 BTU's/hr 100% latent to 48,000 BTU's/hr 100% sensible. Then your control argument could be valid, that type of system is impossible to design.
Your next statement
"Note that a dehumidifier is not the solution, you're basically taking all the energy savings from the high-SEER system and giving it back. Dehumidifiers only make sense when the sensible heat load is too low for the latent load."
This is only a valid argument if you fix all the variables in favor of a high SEER system. In reality, high SEER systems have overcooling specifically to address humidity control. So what you should be comparing is running a high SEER system in overcool mode vs. running a standard system and dehu with a 5 or 6 degree higher set point. In reality that is the actual comparison that needs to be done. In this case that 5 or 6 degrees of extra cooling far outweighs the energy consumption of the Dehu. A better way to think about it is, having a 2 stage system, but instead of low and high, you have a latent stage and sensible stage. A dehu does an excellent job at controlling excess latent, let the AC do an excellent job at controlling sensible heat with a corresponding latent heat benefit.
I can only say for sure on the Mr Cool unit that: in Dry Mode runs till turned off, removes lots of humidity quickly, fan runs in very slow mode and does do some very modest cooling. If you use the included wifi it will have many other options for timing and other things from the phone App. Humidity unlike heat equilibrates in the house somewhat quickly if it is open. This unit is a 21 SEER and is very effective at dehumidifying the home at very low cost. Also unlike the old dehumidifiers, the heat generated is dissipated outside. These are not super cheap but generally over time should pay back the cost and provide comfort. Like many new mini-splits this Mr Cool does very efficent heating a cooling. Great to see this discussion of lower cost heat and cooling with newer products. Thanks.
“[Deleted]”
Great conversation. I'm not sure why I don't see a reply button on Nicole Krueger's comment above, so placing it here.
I think using some figures showing the latent vs sensible loads over the calendar year (and showing them in different locations) would really bring some clarity to these discussions. Those could be combined with Santa_Fe_Dehu's suggestion that the important comparison is Approach 1: A/C plus dehu vs Approach 2: A/C in over-cooling mode.
I think it would be really great to see the energy consumption of each approach over the calendar year, as well as the temperature maintained in the house.
If I understand Santa_Fe_Dehu's comments right we should see that, in the shoulder seasons, with Approach 2 the temperature has to be driven down 5 or 6° to achieve the needed latent removal. And we should see that the energy usage for that is high.
With Approach 1, in the shoulder seasons we should see the temperature driven up a little (due to the dehu converting from latent to sensible), but then the A/C taking that sensible heat out (to maintain the desired temperature). The question is then to compare the energy used by the dehu plus the energy used by the A/C to compensate for the higher temp with the over-cooling energy use in Approach 1. Sounds like Santa_Fe_Dehu is arguing that Approach 2 is lower energy usage than Approach 1.
Of course, probably more important is the question of whether the upfront cost of the dehu is worth the two outcomes in Approach 1 (no dehu): either tolerate the higher humidity, or tolerate the over-cooling. Even if Approach 2 (dehu plus A/C) were actually more costly in running energy it delivers more comfort (in the shoulder seasons).
Seems to me that a vizualization of this over the calendar year is what is needed to help people see these trade offs well.
James,
Great summary but comfort and health should also be added to the mix which obviously can’t be graphed but may be the most important aspect of this.
I think you are right that the comfort and health concerns are key. But since they relate to the humidity levels (and are a problem in the periods where the A/C can't handle the latent load, without over-cooling) they can be reflected on the graph.
So, I think one could also graph expected relative humidity levels, given the latent removal capacity and expected runtime of the A/C only solution. Then it would be clearer that the problem is in the shoulder seasons.
One could even do this hour by hour, because there are times of the day where the A/C is not at design conditions for temperature and therefore doesn't have the moisture removal capability.
And over-cooling is a relationship problem (ie it's a comfort issue that is much worse for some residents of the house!)
Agreed. This is the big issue at night. I don’t see how this problem can be solved in a comfortable manner without a dehu or electric reheat which is only on certain systems particularly not on Mitsu. Unfortunately you can’t “energy efficiency” you’re way to dehumidify. Dry mode on these are awful in my opinion. I have 4 1:1 ductless in an old house which I’m trying to tightened up.
James,
Santa_Fe_Dehu, i.e. Ryan Marks, is making the opposite conclusion so far as I can tell. Approach 2 is more efficient is their claim, no?
DC on the other hand is arguing for approach 1...maybe? Or really he is just arguing for different systems than are currently available?
I'd say approach 1 and approach 2 aren't even comparable because only approach 1 achieves comfort. And the whole point of heating and cooling is comfort.
It's hard to predict the energy usage of approach 1. You've got the interplay of two systems: one the AC, cools the air to satisfy a thermostat, and as a side effect it dehumidifies. The other dehumidifies the air to satisfy a humidistat, and as a side effect it heats the air. Together they will insure that both the humidistat and thermostat are satisfied, which is how comfort is achieved. But I don't see how to model their interaction.
My intuition -- and I'll admit I haven't run the numbers -- is that it's possible to achieve the same level of comfort with a single system with humidity and temperature sensors, and to do it more efficiently, mainly because you're not heating and cooling the same air repeatedly before extracting the heat and dumping it outside. In the humidfier/AC setup the humidifier cools the air, then heats it, then the AC cools it. You've got two compressors running. You could accomplish the same end with one compressor and one heat/cool cycle.
"single system with humidity and temperature sensors,"
You may be right that such a system could be made more efficient (I have no idea). It sounds to me like your system would essentially be a dehumidifier that rejects just the right amount of heat to match the sensible load.
The two systems don't strike me as fundamentally that different, your system just eliminates a step (namely a sensible load transfer via interior air, vs having it all happen within one refrigerant cycle). I imagine there are a lot of factors to determine how efficiency such a system would operate in real world.
With a separate dehumidifier and AC, each evaporator could be optimized for their respective tasks. With a single evaporator, the coil would have to overcool (at times) and then would compensate by dumping some heat back in from the refrigerant via the interior condenser.
OK, either my eyes were crossed last night, or James' post has been edited. It seems improper to edit a post such that subsequent comments no longer make sense without noting it.
(Edit): Oh and by the way, the entire post didn't get properly edited so it is not internally consistent.
I don't recall editing the post, although I was grading papers at the time so much is a blur. Regret the confusion as this is a very useful discussion!
And FWIW, I think that the argument is that approach 2 (A/C plus dehu) is more efficient (as well as delivering comfort that approach 1 (A/C only) can't deliver. But even if the A/C + dehu was less energy efficient it delivers comfort that the A/C alone cannot (as DC contrarian points out!)
James, I may be off my rocker. I just remember thinking you had it flipped, and i re read it a few times to be sure. And now it looks to have reverted. Sorry, I'm either going crazy or the editors are doing some weird things.
In any case, I agree that a dehumidifier and an AC, rather than the super cool mode, is what Ryan was arguing in favor of. Cheers.
This comment is in reply to Ryan's comment, #14, because we can't nest any deeper. I had written:
"I believe the future is controls that have both a thermostat and a humidistat, so the SHR can be tailored to give acceptable dehumidification at the best possible SEER. Although I keep predicting this and it keeps not happening. "
He replied:
"It isn't happening because controls alone can't make the system operate in the conditions it needs to. It is purely a psychometrics problem, when there is a high sensible load an AC system can be tailored to meet the needs. However when sensible demand is low that is the problem, because that latent to sensible ratio is very much slanted to latent. In theory a 4 ton A/C system would need to operate from 4,000 BTU's/hr 100% latent to 48,000 BTU's/hr 100% sensible. Then your control argument could be valid, that type of system is impossible to design."
We should be careful using words like "impossible." So here's what such a system would look like. First, some basics. In a heat pump, gaseous refrigerant is compressed into a liquid on one side, called the condenser, which gives off heat. On the other side liquid refrigerant is allowed to expand, which absorbs heat, that side is called the evaporator. In a cooling-only system the condenser is outside and the evaporator is inside. In a heating/cooling system there are coils inside and outside and either one can be the evaporator or the condenser, depending on which mode you're in, and there is plumbing to switch the way refrigerant flows.
In a system like the Daikin -- I'm going to say "like" because I don't have direct first-hand experience with the actual Daikin -- there are three coils, two inside and one inside, and all of them can act as both condenser and evaporator. So you have the four following operating modes:
Mode 1: Both inside coils cooling, outside coil shedding heat.
Mode 2: One inside cooling, one inside heating, outside shedding heat.
Mode 3: One inside cooling, one inside heating, outside absorbing heat.
Mode 4: Both inside heating, outside absorbing heat.
Your interior can need either heating or cooling, and can either need dehumidification or not. (That's kind of a simplification). So there are the following operation modes:
Mode A: Need cooling, don't need dehumidification
Mode B: Need cooling and also need dehumidification
Mode C: Need heating and also need dehumidification
Mode D: Need heating but don't need dehumidification
Three of those modes are pretty simple to implement: Mode A goes with Mode 1, Mode C goes with Mode 3 (or possibly Mode 2 if you only need a little heat), Mode D goes with Mode 4. For efficiency, the evaporator temperature in Mode A should be kept above the dew point to prevent dehumidification. The tricky mode is Mode B and there are two sub-modes:
Mode B1: The SHR needed is above 65% and can be provided entirely by operating in Mode 1
Mode B2: The SHR needed is below 65%. Mode 2 provides this.
Now, this can't be done with a simple on/off switch like a thermostat or a humidistat. You need to measure the temperature and humidity, compare to setpoints for each, and figure out what mode you need to be in. But it's certainly not impossible. It just doesn't exist yet.
This sounds, conceptually, much like a dehumidifier paired with a high SEER cooling system. But if made to be one package, perhaps has appeal in that sense. Or is the claim it would be more efficient in some way?
This podcast had some related interesting insights into design of dehumidification, including relatively new designs. Can't say I understood it all!
https://positiveenergy.pro/building-science-podcast/2021/11/4/keep-it-dry-
"According to Building Science Corporation, most air-conditioning equipment will not have the capability of removing the moisture load without overcooling the space."
Most will not but some will? Which will?
“Most VRF, VRV, and ductless systems are programmed to prioritize efficiency over humidity removal."
Is Daikin Quaternity the only one that doesn't?
As I point out in post #9, you can increase the efficiency rating of a system by raising the evaporator temperature, which makes it remove less humidity. And rated efficiency sells systems.
Even if that weren't the case, almost all air conditioning equipment has just one sensor, the thermostat. So even if the system is installed with an eye toward dehumidification, the installer has to make a guess about what settings will do an acceptable job removing moisture. And that setting will have to do under all conditions.
I think soon we will see systems with both temperature and humidity sensors, and both temperature and humidity setpoints. That will give a much better balance of comfort and efficiency.
I'd like a ducted AC unit with an integrated counterflow heat exchanger (essentially an HRV-core) using a variable system controlled bypass. If the desired SHR is above the usual ~0.65 minimum, the heat exchanger is bypassed completely and the SHR is controlled by coil temperature/airflow alone. If the desired SHR is below that minimum, the heat exchanger is put into function, pre-cooling the intake air and re-heating the exhaust air, thereby extending the range of achievable SHR. With a high-performance counterflow HRV-core with efficiency above 90% that could be a decent range extension.
This exists in various forms and typologies already, of course, but I haven't yet encountered it in a neat package for residential in my part of the world.
Even better would have it come in the form of a “dumb-box” merely consisting of the heat exchanger, variable bypass and a coil meant for water. This would get integrated with external fans and an air-to-water monobloc heat pump. The system could do heating too with the heat exchanger completely bypassed. Made compact enough, zoning could be achieved by having multiple of them. Unfortunately, the air-to-water heat pumps still seem to stop a bit short of the ideal minimum water temperature for proper dehumidification.
Yes! Making one of those for myself is on my to-do list. My thought is to buy a small ECM-based HRV system and an air-to-water heat exchanger (a fan-coil minus the fan) to run chilled water through. Variable bypass on air-to-air exchanger would be ideal, but I might instead have a separate fan coil and vary the fan speeds on the two to control the SHR.
IMHO, most of the AC units that claim a SEER rating higher than 20 are basically cheating, by only giving you sensible cooling. There is almost a direct relationship between the claimed SEER and SHR on the Mitsubishi units. One feature that is on multisplits that I wish would come to 1:1 is a coil target temperature. The multisplits have dip switches to select a target evap temp appropriate for your building (at the expense of "efficiency").
The 1:1 prioritize lower "efficiency", but I believe this is a false economy since I now have to run a less efficiency dehumidifier and a the AC.
I have a Mitsubishi FH09 and lack of dehum without doing some really hacky things is my biggest complaint.
The only way this will change is if the energy efficiency ratings somehow penalize them for lack of dehum. There needs to be a giant (*)next to the SEER 30+ ratings that states you get next to no dehum. If I could do it again I'd get a bigger unit with a more appropriate SHR for Southern Ontario.
Log in or become a member to post a comment.
Sign up Log in