Minisplit Questions (Wall Mount Vs. Ducted, Zoning)
![](https://secure.gravatar.com/avatar/ac6b00fa059dd6bb9f662d476d42146d?s=52&d=https%3A%2F%2Fimages.greenbuildingadvisor.com%2Fapp%2Favatars%2F200x200%2Fj.png&r=g)
Hello,
I am planning on installing a either a Midea dual zone wall unit minisplit or a Midea 24,000 BTU Ducted minisplit unit in an apartment that I have. The only items pre-existing are the cold air return and heat vents. If I go with the ducted option, I am running all new ductwork.
The apartment is 455 square feet. It has a kitchen/living room combo, 1 bath and 1 bed. Please look at the attached image for the room square foot measurements as well as the current vent locations/sizes. I am located in Northern Ohio. This place will be used primarily in the summer months (On a river) but I wanted it to be usable year round.
Here are my questions regarding the 2-Zone Minisplit option:
1. What I am planning on using is a Midea 27K BTU multi-zone compressor with 2 head units (One in the kitchen/living area and the other in the bedroom). I would either go with the high wall or console units. Is this the appropriate sizing? I know I have overkill on cooling, but if my calculations are correct, I will need around 22,000 BTU’s of heating.
2. Which is better? The highwall or console (Floor) style units?
3. Can the condensation drain line be ran out side? I am assuming condensation is only produced in the summer and there is no concern for this line freezing.
4. What is the optimal distance from the ceiling for highwall evaporators?
Here are my questions regarding a ducted minisplit:
1. The cold air returns located in the ceiling are panned inline with each other in a 2×6 joist space in the attic (14″ x 6″). I am hoping to run a single 14″ x 6″ rectangular duct to them through the bathroom closet. This duct would connect at the ceiling and run through the closet floor into the crawl space where the mini split is going to be located.
Is a 14″ x 6″ duct adequate for both cold air vents or do I need to go larger? Could I use just 1 wall space cavity instead (14 x 3.5″)?
2. Does the Kitchen/Living are need another hot air vent or is the current single vent in the room adequate?
3. Should I run 6″ or 8″ round pipe to hot air vents?
4. Should I use round hard pipe or is the round flex duct ok to use?
5. Do you think this 24,000 BTU ducted minisplit is the best option for my AC/Heat? Is it an appropriate size for my apartment?
I appreciate everyone’s thoughts and opinions! Thank you in advance!
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb1.png)
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb2.png)
Replies
The notion that a 455' apartment has a heat load of 22,000 BTU/hr implies you're leaving a window open, or that the place has no insulation and is located at 10,000' of altitude in Patagonia or something, not northern OH (US climate zone 5.) Even a barely insulated 455' space would come in WELL under 15,000 BTU/hr @ 0F unless it's an air-leaky 455' wind-tunnel.
What is the actual ZIP code (for outside design temp estimation, etc)?
2x4/R11, 2x6/R19, or something else (hopefully better)?
What's under it- a garage, slab on grade foundation, some other conditioned apartment, unheated basement/crawl or ???
What's above the ceiling- vented attic (if yes how much insulation), another apartment, or something else?
Looking at the layout I'd hazard that it would do just fine with a half-ton head in the bedroom, a 3/4 ton (or maybe a 1 ton if the slider is the world's air-leakiest single pane glass.) Midea doesn't offer a half ton head, so you'd be looking at pair of 3/4 ton heads, and they're overkill.
The 1.5 ton MCH2U-18PHH2 specifies 14,100BTU/hr @ +5F which for 455' of space would be a ratio of a whopping 31 BTU/hr per square foot. Almost any reasonably tight 2x4 framed place with R11s in the walls, storm windows, and R19 in the attic would come in under 20 BTU/hr per square foot at that temp, for a total heat load under 10,000 BTU/hr. You have only ~40 square feet of window in the bedroom, and a comparable amount of slider window in the kitchen/living area, and the east side is mostly partitioned from conditioned space, with the west wall from protected but unconditioned wall space. It's hard to imagine how you can get to 22,000 BTU/hr of heating load. Even if the glass is single-pane the 40' of window or slider only amounts to 2600' of load at a temperature difference of 65F (70F in, +5F out.) The exterior wall areas would be even less.
Very few locations in OH have 99% outside design temps lower than +5F, and no locations have design temps below 0F. (Yes, it gets colder than that, and if you happen to be there during a record setting polar vortex cold snap it wouldn't keep up, but designing for the 99.99% condition would be a mistake.)
http://articles.extension.org/sites/default/files/7.%20Outdoor_Design_Conditions_508.pdf
With heated space on 4 sides, unless the bathroom leaks a TON of air through the floor or exhaust vent it will be adequately heated by the occupant when occupied. If it runs too chilly when its below 10F a tiny radiant panel operated by an occupancy/vacancy sensor switch would cover the load.
A ducted solution in the attic would be a lousy idea, but if you go that route you'd be shooting yourself in the foot with flex. Go hard-piped with no sharp throated ells, and use duct mastic on every joint & seam, and caulk the register boots to the ceiling gypsum. You would also have duct sweating problems during the humid OH summers unless the ducts are fully inside conditioned space.
It's fine to run the condensate lines on the exterior, but make them accessible for cleaning/blowing the crud & slime out of 'em.
For wall mounting the installation manual (not available online) probably specifies a minimum distance from the ceiling, but whatever that is, for a heating dominated climate double it, or mount it near the floor.
Thank you very much for your reply. Regarding my BTU estimation, I was operating specifically off of the attached BTU map. This map indicates roughly a 20K-22.5K BTU requirement if their calculations are correct (45-50 BTU's per square foot). My zip code is 43558. Is there another formula for calculating BTUs?
All of the east wall is connected to heated space. The remainder of the walls are not. Both the north and south walls are exterior and the west wall is partially exterior and mostly connected to an unheated garage. The house is built over a crawl (If I used the ducted unit, that is where I would install it). The windows and patio door are new (No storm windows though). The insulation in the walls is R13 and the house has a brick exterior. The vented attic will have about 10" of blown in cellulose insulation.
The reason I was planning on going with the larger outdoor unit is for heating purposes. I am concerned that due to the nature of minisplits losing heating capability as the temperature decreases, that I may have an issue heating the apartment.
My understanding is with the newer units using inverter technology, is that you have more flexibility without oversizing. For example, the BTU range of the 12K wall unit is 4,000 BTU-13,000 BTU. Is this a correct understanding? I figure it is better to go on the large side than too small.
Do you expect issues if I went with the 27K outdoor unit (Rated at 4K-36K BTU), a 12K unit in the kitchen area and a 9K unit in the bedroom? I am open to other options as well.
My goal is do do this right once and be done. I am trying to do my homework wisely. Your input and expertise is greatly appreciated!
I've seen that map before, and it's complete & utter crap, INSANELY oversizing gas furnaces even for barely insulated 2x4 ranch houses by ~2x, and R13 2x4 single family houses by ~3x.
And apartments have a lot less exterior surface area to lose heat than single family ranchers. Your losses to the east wall are effecitvely zero, and the losses to the garage far less than to fully exterior walls.
Realistically your heat load is probably on the order of 5000 BTU/hr @ 0F when occupied. That means a compressor with a minimum output of 4000 BTU/hr would almost never modulate. The 1.5 tonner modulates down to 2500 BTU/hr, which means it will likely modulate whenever it's below freezing and dark out but would be cycling on sunny days even when it's freezing.
Let's run a quick & dirty rough calc. on that WAG:
Nearby Toledo's 99% temperature bin is +3F, code min interior temp is 68F, so you're looking at a design temperature difference of 65F.
Eyeballing it you have about 300 square feet of ~U0.1 wall: U0.1 x 300' x 65F= 1950 BTU/hr of wall losses.
Assuming the garage has doors on both ends that never close that would be another (worst case) ~250 square feet of exterior-ish wall for: U0.1 x 250 x 65F= 1650 BTU/hr of losses. If the garage has doors that are normally closed it'll be no more than 1000 BTU/hr.
With ~R38 of fluff in the attic and a 7% framing fraction of joists, with a vented attic you're at about U0.03 for the 455' of ceiling:
U0.03 x 455' x 65F= 887 BTU/hr of heat loss out the ceiling.
If the windows are new I'm going to assume that they are no worse than U0.5 double panes, and are likely better than that (but if they're new you can look up the published U-factors, or look at the sticker that came on them. You have about 80' of glass so that's going to come in around. U0.5 x 80' x 65F= 2600 BTU/hr of window losses.
Assuming a reasonably tight house and 20cfm (1800 cubic feet per hour) of ventilation/infiltration at a thermal mass of 0.018 BTU per degree per cubic foot that works out to 1800 x 0.018 x 65F= 2106 BTU/hr of ventilation & infiltration.
Add it all up:
1950 BTU/hr of exerior walls
1000 to 1650 BTU/hr of garage wall (I'll assume 1000)
887 BTU/hr of ceiling losses
2106 BTU/hr of ventilation
That comes to 5943 BTU/hr plus whatever you might be losing to the crawlspace, which may be next to nothing, but lets call it 500 BTU/hr for yuks
OK you MIGHT have a heat load as high as 6500 BTU/hr, but that's before subtracting ~250 BTU/hr for one sleeping human (now 6250 BTU/hr) and another 300 BTU hr for the refrigerator (now you're 5950 BTU/hr) and something for the standby losses of a water heater, other plug loads etc.
You're really ARE in the neighborhood of 5000 BTU/hr for a whole-house load, if not actually under 5000 BTU/hr (though it might be) Measure it all up next time you're there and plug in the real numbers for wall area (be sure to subtract out the window & door areas), and make a note as to whether the garage itself is in fact insulated from the outdoors, which would cut those losses roughly half.
From a sanity-checking point of view 5000 BTU/hr for 455 BTU/hr comes in at a bit under 10 BTU/hr per square foot of conditioned space, which would still be a bit higher than most town-house condos I've run numbers on at similar delta-Ts. The wild cards are the floor losses and the losses to the garage, and possibly total air leakage if the place seems really drafty (but that can be fixed.)
That makes even the 1.5 tonner sub-optimally oversized for the likely load but it would still modulate quite a bit if the 1.5 ton compressor can really modulate down to 2500 BTU/hr @ 47F, which is pretty low compared to most multi-splits.
Dana, Thank you very much for all of your insight. I am leaning towards going with a 18K (1.5 ton) outdoor unit. Do you think I am still good to use a 12K wall unit in the kitchen and 9K in the bedroom? With the minisplits having inverter technology, do you lose efficiency or run into other issues by over sizing either the head units or compressors?
The 18K MCH2U-18PHH2 multi-split compressor can be hooked up to one 9K and one 12K wall unit, but it's pretty clear that you don't really need it, even the 9K is MEHSU-09PHH2 is oversized for an individual zone by more than 2x, an if better distributed has enough capacity to heat the whole house.
The floor plan really demands a head in each room if going with a mult-split, and the smallest head Midea makes is a 9K. If they made 6K units and the compressor modulated lower, a 6K head in the bedroom and another 6K head in the kitchen/living area.
Yes they modulate, but yes, oversizing by too much leads to lower efficiency and lower comfort. The minimum modulation of the 18K compressor is ~2500 BTU/hr no matter what heads are used, and 2500 BTU/hr is approximately the whole-house heat load at ~30F. Above that temperture the heads & compressor will be cycling on/off, cutting into efficeincy a bit, but a pair of 9k heads is best you're going to do efficiency/comfort-wise with their lineup.
That's why using a 27K compressor would be a mistake, since it's 4K minimum output is high enough that with your load levels it almost NEVER modulates, and only cycles on/off except on the coldest nights, which is bad for efficiency, bad for comfort. With the 18K compressor it will modulate at least most of the time during the coldest 10-12 weeks of winter, but cycle a lot during the shoulder seasons.
The MCH2U-18PHH2 with a pair of MEHSU-09PHH2 is your best option in their product line, since that combination would have the maximal range of outdoor temperatures at which it can still modulate.
FWIW: One of the guys in the office with a 1980s vintage ~1700' (plus basement) 2.5 story town-house (middle unit, not an end unit) in metro-west Boston area very recently ran a fuel-use based heat load calculation on it, coming in between 12,000-13,000 BTU/hr @ 0F outdoors. This is not some super-insulated green gem- it's sub code-min on just about everything except the recently replaced windows and possibly air tightness. That's ~7.5 BTU/hr per square foot of fully conditioned space (not counting the basement.)
Comparing fuel use based load numbers on gas bills prior to replacing the air-leaky single panes, and before fixing most of the gaps in the sub-code insulation plus retrofit air sealing it was still under 25,000 BTU/hr, and less than 15 BTU/hr per square foot of conditioned space.
It is heated with an 80,000 BTU/hr condensing furnace, which is ( /1700'= ) 47 BTU/hr per square foot, sized to perfection per that idiotic map in response #2!. Yet it is more than 3x oversized for the degraded-from-original-condition real load, and more than 6x oversized for the "after" picture.
OK, so it's not enough to heat the place at absolute zero, but it's good down to about -350F or so, which means he'll still be able to heat the place with decent margin if they move condo to Titan, the largest moon of Saturn! :-) (Seriously- how much margin do you need?)
To make it even worse, it's currently fitted with a 3 ton central air conditioner, which never runs long enough to make a dent in the summertime humidity. The real 1% load is probably on the order of 8000-9000 BTU/hr, given the shading factors and window orientation.
BTU per square foot rules of thumb tend to be pretty lousy, but the map in response #2 is one of the WORST versions ever invented, yet it seems to have proliferated on the web. Try a web image search on the terms:
furnace sizing map
That wretched disinformation map shows up in 4 of the first 5 on google images!