GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Transoms and Ductless Minisplits

WildBunchFarm | Posted in Green Building Techniques on


I am trying to heat and cool my second floor with one 12,000 btu ductless minisplit. I have heard that this is possible if I keep the bedroom doors open during the day to facilitate circulation of the hot or cool air.

However, does it get uncomfortable quickly in the night when the doors are closed. If so, I was thinking of putting in operable transoms over the doors. The only problem is they are kind of expensive (~$275 from Attached is my floor plan. I will be putting ceiling fans in each bedroom so maybe the transoms would be overkill?

Thanks for any advice you can give.

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. Expert Member
    Dana Dorsett | | #1

    How quickly the rooms cool off or whether it gets uncomfortable at all depends on the R-values/U-factors of the walls, windows, & ceilings, and how cold it is outside, and the design heat loads of the doored-off rooms relative to the partition wall area to the space that has the mini-split.

    Installing an operable transom is never a solution for temperature balancing- you would need a large opening at both the top and bottom of the wall to get appreciable convection going. Transoms are sometimes a partial solution for daylighting though.

  2. GBA Editor
    Martin Holladay | | #2

    Take the money that you would have spent on transoms, and use it to buy more air sealing tape (or slightly thicker insulation in your attic).

  3. WildBunchFarm | | #3

    Thanks for the responses. I probably won't do the transoms then. My walls are double studded 10.25 inches thick with dense packed cellulose. And the attic will have 20 inches of blown-in cellulose.

    I think I'll still do radiant heating in the bathrooms though. Or maybe a heat lamp is more cost-effective.

  4. Expert Member

    What sort of ventilation are you thinking of?

  5. user-4243359 | | #5

    If your bathroom floors will be tiled, in-floor electric resistance for back-up is hard to beat for comfort and flexibility; cost is reasonable too. Put it on a timer to welcome your feet on winter mornings.

  6. Expert Member
    Dana Dorsett | | #6

    With R30-ish whole-walls performance the design heat load of the bathroom is probably going to be roughly the heat emitted by and adult human and a couple of LED lights (unless the bathroom window is large or low performance.) Unless you want to keep the bathroom warmer than other rooms in the house it's unlikely to need supplemental heat of any sort.

    What is your 99% outside design temperature?

    Run a Manual-J or I=B=R estimate of the heat loads of the doored off rooms.

    A sleeping adult human puts out about 250 BTU/hr, which will come directly off the heat load.

    A 2x4 partition wall with empty cavities and 1/2" wallboard on each side are about U0.25 - U0.30. So at a 5F temperature difference (70F in the main space, 65F in the bedrooms) the walls will be delivering about 0.3 x 5F= 1.5 BTU/hr per square foot. A 15' x 9' partition wall is then delivering another 200 BTU/hr under those circumstances.

    A typical interior door runs about U0.75, and about 18-20 square feet, so that's another 60-75 BTU/hr.

    So, order of magnitude between the human occupant and the partition wall & door area, at a 5F difference there's 400-500 BTU/hr being delivered into the room. That may or may not cover the average overnight heat load of the bedroom in your location- it'll probably come down to window size & performance (usually does.)

    Even if not fully covered, the thermal mass of the room's contents as well as the thermal mass of cellulose in the exterior walls & ceiling means it can sustain a 1000 BTU/hr loss for hours without losing a LOT of ground.

  7. WildBunchFarm | | #7

    Haha thanks Ed. That's the problem we've been having in our 700 square feet house right now. It's heated with a ductless minisplit, but the bathroom can get cold because we have a continuous ventilation fan in there. I've tried to insulate the ducting for the ventilation in the attic as much as I can but you can still feel the cold air through the grate.

    @Malcolm - I'm probably going to go with the Zehnder system. I was quoted at $8700! I was thinking maybe I could skimp on some the ducting and only have supply vents in a few rooms. That might shave off $1000. I'm probably going to get skewered for trying to save money on something so important as fresh air. I've also requested a quote from UltimateAir.

    Do you have any other suggestions?

    Here's some updated pictures of my project. First picture is of the completed double stud wall. It took forever. Second picture is of the subcontractors doing the sheathing and taping seams.

  8. WildBunchFarm | | #8

    Thanks Dana for your response. Read it after my post. I may go with the heat lamp after all after your analysis. I did do a heat loss calculation per Martin's blog article. I'll go back and split that up by room.

  9. user-4243359 | | #9

    Very impressive Jimmy! I encourage you to keep us up to date on your project including post-construction results (performance, costs, etc) and lessons learned.

  10. STEPHEN SHEEHY | | #10

    Jimmy: Zehnder will design your HRV if you send the drawings to them. My experience was that their design was just right and avoided any significant material waste. Their contractor's discount was significant. I'd be reluctant to reduce the number of supply or exhaust vents just to save $1000.

    We heat with minisplits and also installed Ditra Heat under the bathroom floor tile. I never bother with it, but my wife loves it. Do it for family harmony.

  11. WildBunchFarm | | #11

    Thanks Ed and Stephen.

    @Stephen - I did get a Zehnder designed quote. Did you have to drill a bunch of holes through your joists to run the 3" tubes? I suppose that's not a problem with I-joists as long as I follow the recommended spacing of holes. Was it a pretty easy install?

    @Ed - Thanks. A lesson I just learned was that in a double stud wall or thick wall using exterior foamboard, you're supposed to frame the interior load bearing wall 3"-4" wider than the door rough opening if you are installing an "outie" door that swings inwards. This is to avoid the latch from smacking into the thick wall and to let the door swing open more fully.

    I didn't realize this until after I had framed out my double stud walls. So I went back and added two studs to the exterior door rough opening to make it 34" instead of 38". The interior rough opening is still 38". Thank goodness my other exterior doors are outswing French doors that open up onto a porch. These little details can be overwhelming.

  12. GBA Editor
    Martin Holladay | | #12

    Jimmy (and other interested GBA readers),
    For more information on rough openings for doors in double-stud walls, see Windows and Doors in Double-Stud Walls.

  13. WildBunchFarm | | #13

    Thanks Martin. That's the article that alerted me to the issue with the door. As an alternative for someone with a thick wall door problem, they can buy an extended latch strike plate up to 3" wider than normal. That may help in some cases when used in combination with a 6 9/16" door jamb. But those special order door jambs cost 2 to 4 times more than the standard exterior doors in stock. I find that door swing radius is usually not a problem compared to the latch striking the wall and leaving a mark. Doors are usually placed close to a wall anyways, where door opening radius is limited to begin with.

    In terms of ductless minisplits, how big of a difference is HSPF 9 compared to HSPF 12.5? I'm comparing LG minisplits to Mitsubishi minisplits and LG will be cheaper, but Mitsubishi has the better efficiency numbers. Mitsubishi also gets my vote in terms of finding replacement microfilters. I can't find those special 3M LG microfilters anywhere.

  14. STEPHEN SHEEHY | | #14

    Jimmy-the Zehnder HRV installation was easy. My HVAC guy was impressed at how easily everything went together. We didn't need to drill any holes, since the tubes were run inside a utility chase above the ceiling.

  15. Expert Member
    Dana Dorsett | | #15

    Run your heat load calculations at an outdoor temperature of +47F, and compare the minimum-modulated output of the mini-splits @ +47F as spelled out in the submittal pages. (It's a standard test temperature for the HSPF test protocol.)

    If the minimum output at +47F is substantially higher than heat load at +47F the unit won't actually meet it's HSPF numbers (whatever they happen to be) since it will be spending a lot of time cycling on/off at lower efficiency rather than modulating.

    If sized correctly for their loads, an HSPF 12.5 means that it will deliver 12,500 BTU per kwh on average over the season, and an HSPF of 9.0 means it will average 9000 BTU/kwh. That's about 39% more heat per kwh if you go for the HSPF 12.5 unit, but only if it's modulating rather than cycling. If it's so oversized for it's load that it's cycling most of the season rather than cycling it's as-used HSPF will be under 10, maybe even under 8.

    The -FH09NA Mitsubishi modulates down to 1600 BTU/hr out @ +47F, lower than any others that I'm aware of, which often makes it the best candidate for low-load houses.

    LG's -LS090HEV modulates down to only 3070 BTU/hr @ +47F, which may be fine, or not, depending on the load:

    Fujitsu's 9RLS3 also modulates down to 3100 BTU/hr @ +47F, but has an HSPF of 14.2 (14,200 BTU/kwh), nearly 50% more heat per kwh than the LG, and 14% more heat than the Mitsubishi:

  16. WildBunchFarm | | #16


    I have been reading a lot of your comments on GBA for months. You are hands down the heating and cooling expert.

    I hope you can answer another question for me.

    Is it considered short-cycling if my minisplit turns on and off every 10-15 minutes during cold winter days? What is the appropriate amount of modulating time? For my current house, I have the LG181HSV and it's most likely oversized for my needs. I know that some people complain of short-cycling minisplits that turn on and off every 3-5 minutes but is short-cycling every 10 minutes during February a big problem in terms of energy usage and longevity of the equipment? The rest of the year the minisplit seems to "modulate" for long periods of time.

    Thanks for any advice you can give.

  17. WildBunchFarm | | #17

    Does an erv help distribute heating/cooling in a house with just ductless minisplits in a few locations? I am probably going with a Zehnder system. I will also have ceiling fans in all bedrooms. Although the calculations work I'm still afraid of a few rooms being too hot or cold.

  18. GBA Editor
    Martin Holladay | | #18

    Q. "Does an ERV help distribute heating/cooling in a house with just ductless minisplits in a few locations?"

    A. No. During the winter, an ERV will introduce air at a lower temperature than the average indoor temperature, thereby slightly cooling the room where the air is introduced. During the summer, an ERV will introduce air at a hotter temperature than the average indoor temperature, thereby slightly warming the room where the air is introduced. ERVs don't help with space heating or cooling.

    Moreover, an ERV moves relatively low volumes of air -- not enough to balance temperatures between rooms.

    For more information on these issues, see these articles:

    Rules of Thumb for Ductless Minisplits

    Minisplit Heat Pumps and Zero-Net-Energy Homes

    Practical Design Advice for Zero-Net-Energy Homes

  19. Dana1 | | #19

    If the ERV supply registers are located near the mini-split heads and the exhaust registers are in rooms with no supply registers there is a modest redestibution of heating/cooling to to those other rooms, but it's very modest indeed:

    The specific heat of air by volume is about 0.018 BTU per cubic foot. A 20 cfm (=1200 cubic feet per hour) ventilation flow at a temperature difference of 5F only delivers (5F x 0.018 x 1200cf/h= ) 108 BTU/hr of heating or cooling.

    Putting that in perspective, a sleeping human emits 225-250 BTU/hr

  20. WildBunchFarm | | #20

    Just wanted to update everyone on my project. We finally finished the interior framing and stairs. Stairs with landings can get confusing with all of the adjustments you have to make for the different thicknesses of flooring at each level. But after I got the hang of it, it went pretty fast. This article really helped me out: Although, there is one thing in the article that tripped me up. Page 78 talks about adding and subtracting the tread and finished floor from the last step before the landing. I just added the finished floor height to the overall height measurement. I'm not sure why the article talks about messing with calculations for the last step. It should be treated no differently than any other step.

    Meanwhile, we hired a contractor to put on our 1x3 rain screen since we are going to use LP Smartside siding and we have 10.25" thick walls. He used thinner ring-shanked nails meant for cement fiberboard siding to prevent too much splitting. Although, when he ran out, he switched back to the regular 2 3/8" framing nails and there was no splitting there either. He just had to keep them a little more centered on the 1x3 furring strips. Interestingly, he has been building houses for 20+ years and has never installed a rain screen.

    I looked into Cor-A-Vent strips to do my first and last rows of the rain screen but it was too expensive. So I decided to wrap hardware cloth around the first and last rows in order to keep insects out. We also routed out 1/2" thick swaths 1/4" deep from the back of the furring strips to aid air movement and water drainage since most of the house was done with horizontal furring strips to accommodate our eventual vertical board and batten siding. It might have been overkill to do the routing but it didn't take too long to do once you have all of the pieces laid out on a table. We routed the furring strips about 16" on center.

    We are now starting to put in the windows. We decided to use a combination of Protecto LWM200 (liquid applied flashing) and Siga tape for flashing. The Protecto LWM200 ($40/gallon = 65 sq. foot) is so much cheaper than these newer vapor permeable tapes so we decided to apply that over our exterior window frames made from the furring strips (see picture). Plus, the Protecto LWM200 sticks pretty well to the tapes and felt paper. We also used plastic corner guards. The window will be sitting on a sloped sill that we ripped on the tablesaw. However, a sloped sill seems silly in our case since we are using flangeless Intus windows. We are going to tape the gap between the window and rough opening so where is the water going to go if it gets in? I guess I could cut some sponges up and staple them to the outside frame and then tape the window to the rough opening just like how some people put shims underneath the bottom flange of their windows to allow gaps for drainage. In any case, we're going to make a backdam out of backer rod and caulk. I'll post some pictures when we do that.

    We are going to install the windows flush with the rain screen. This may make the top of the window more susceptible to leaks, but we tucked the 1x3 over the top of the window under the felt paper (see picture). We will also make sure to tape that area really well and then install a drip cap.

    Lots of work. I've seen houses being built near me that started well after me and now they are pretty much finished. Sometimes makes me wonder if I'm doing too much.

  21. Expert Member
    Dana Dorsett | | #21

    For maximum effectiveness the furring needs to be mounted vertically so that stack effect convection purges the air in the cavity with outdoor air, and liquid water drains quickly. When the furring is configured as horizontal girts the gap provides the same excellent capillary break, but considerably limits the air exchanges, and rain that penetrates doesn't get a gravity assisted exit.

    With the depth of overhangs here the amount of rain getting in won't be huge though, and it will still dry better than with no air gap.

  22. WildBunchFarm | | #22

    I just came across this article by Hammer and Hand on how they recommend horizontal rain screens.

    Basically, they put on vertical furring strips first and then another layer of horizontal furring strips. I'm not sure that's necessary in my situation as I have already routed weep holes into the back of my horizontal furring strips just like vinyl siding has. The addition of an extra layer of furring strips would have cost me around $1200 with material and labor. Routing weep holes on the back of 370 pieces of 1x3s took about 2 hours.

  23. Expert Member
    Dana Dorsett | | #23

    The cross-hatched 1x furring is somewhat superior to vertical only, since it also reduces wind-driven infiltration drives better- air getting through the siding goes around the house, lowering the pressure difference between windward & leeward sides of the primary air barrier.

    In your location the cross-hatched approach isn't as beneficial as in the rainy Pacific Northwest.

  24. Jayraja | | #24

    Newbie here...

    I think the cross hatched horizontal strapping is the nailing base for vertical siding. I don't think it enhances the performance of the rain screen.

    Please correct me if I am mistaken. I have just started to read up on construction techniques to utilize in our forthcoming custom house in the PNW.

  25. Expert Member
    Dana Dorsett | | #25

    Having the nailer for the vertical siding was the express purpose, but it does indeed give modestly better moisture handling and modestly better thermal performance enhancement to the assembly. The cross-ventilation doesn't do too much in less windy places, but in coastal or windy ridge-top sites it can be meaningful (but still probably not worth paying extra for.)

    There has been a fair amount of research on pressure equalization techniques in rainscreen assemblies for larger buildings, and how different approaches affect both rain penetration of the siding as well as infiltration pressures at the sheathing layer. For house-sized buildings the additional rain penetration of the siding in a cross-ventilated system isn't very important, since it also leaves quicker. It gets more complicated in tall & wide buildings with bigger stack effect drives and bigger wind pressure differentials.

    Ideally in a house sized building it would always be vertical channels with both top & bottom venting, but if the primary channels are lateral, adding some amount of cross-ventilation helps the drying rate.

    In Jimmy Nguyen's case it doesn't much matter unless it's a high wind area, with lots of wind-driven wetting of the siding, (which it isn't.) Based on the depth of the roof overhangs and the height & symmetry of the trees on that site (an indication of modest peak & average wind speeds) it doesn't look like there's anything to worry about- the girt-furring approach won't suck in so much moisture on the windward side that it would require cross ventiliation.

    Unless your new PNW house is on the rainy-windy coast or in the Columbia Gorge or some other super-wind area don't sweat the cross ventilation. Stick with vertical-only furring unless driven to girts by vertical siding, and even then use your best judgement on how much direct wetting it will get before taking the cross-hatch approach. In zero overhang vertical siding situations that are wall-thickness sensitive, using furring-width squares of 1/4" XPS as spacers between the girts & weather resistance barrier at every furring fastener would be enough cross ventilation in all but the worst setups, saving a half-inch of wall thickness (but it's a lot more labor-intensive than crossed 1x furring.)

  26. WildBunchFarm | | #26

    I just did some research and found that my area gets 43.4" of rain annually, which is more than Seattle and Portland. Our rain just comes down with more volume whereas you all have it spread out over more days.

    I had a quick question about dense-packed cellulose in 10.25" thick walls at 24" o.c. My insulation contractor said that I would have to put up the drywall immediately after to prevent the cellulose from blowing out too much. I've read that some people suggest going with 5/8" drywall to prevent future drywall bowing.

    I've read others who do furring strips horizontal to the studs and then hang the drywall over that. I really don't want to do more framing! But if the consensus is to take this latter route, then I suppose I have to do this before the electrician roughs everything in since he'll have to put the electrical boxes flush with the furring strips.

    Picture update - got most of my Intus windows in last weekend. Major oversight on my part, but I didn't realize the windows opened up to the interior. I thought like most casement windows it would open towards the outside. Just keep that in mind if you are looking at any of these European windows. The installation was very easy and adjusting the operable windows to get the right seal was straightforward. They were very heavy though, which made having no flanges on the windows an advantage. You can install them from the interior and for the operable windows, you can easily remove the sash and install the jamb in the rough opening first. The put the sash in later.

  27. Expert Member
    Dana Dorsett | | #27

    The frequency of wetting and the drying intervals make a bigger difference than the absolute annual rainfall numbers. With deep overhangs the direct wetting isn't much.

    The rain in the northwest varies a lot even within fairly shot distances due to the effects of the mountains. Sequim WA gets less than 17" of annual rainfall (less than half of Seattles, despite closer proximity to the ocean) since it's in the rain shadow of Olympic range, while Forks WA on the Pacific side of the range gets about 120". Seattle's reputation for rain has more to do with the 155 annual days of measurable rainfall, not annual inches. That's a lot of rainy days with short average drying intervals, which is why deep roof overhangs to limit direct wetting and rainscreens to promote rapid drying make a real difference. (The Bengalis make great use of the "bungalow" concept of deep roof overhangs, necessary to manage the relentless rains during the monsoon season, which is both high frequency AND high volume.)

    At 24"'ll want to use fatter gypsum, but it should hold up just fine. For absolute flat-wall purists adding furring (or a layer of OSB) and shimming it out to laser-precise flatness might be necessary, but not the rest of us.

  28. WildBunchFarm | | #28

    Thanks Dana. That's a relief that I don't have to do additional furring strips. It looks like 5/8" sheetrock costs the same as 1/2" if I buy over . So it won't add to material cost.

    On to another question:
    I've gotten 4 quotes from HVAC installers and almost all of them are talking me out of using ductless minisplits. The consensus is that heating shouldn't be much of a problem in our climate (Zone 4), but they are all worried about cooling and humidity control. Since I am installing an ERV, would this be sufficient to remove humidity from each of our rooms? Also, in a supertight, superinsulated house, is the additional ducting from a ducted heat pump system going to cost me that much more over doing ductless minisplits. Moreover, the ducted system can come with a cheaper ERV system compared to the Zehnder.

    Some of the ducting will have to go through my attic, but I could always drop it down into my 20" of blown-in insulation after the inspectors have left. I designed my house to use ductless minisplits, but all these HVAC contractors are making me have doubts.

  29. GBA Editor
    Martin Holladay | | #29

    Q. "Since I am installing an ERV, would this be sufficient to remove humidity from each of our rooms?"

    A. No. For more information on this issue, see Misconceptions About HRVs and ERVs.

  30. Expert Member
    Dana Dorsett | | #30

    In a zone 4 climate you can probably do just fine with the Daikin Quaternity series, which can dehumidify to whatever humidity setpoint you like without sensible cooling when there is no sensible cooling load.

    They're not always a good option for zone 5 or colder though- it's not a cold-climate series.

    Making penetrations in the upper floor ceilng for ERV/HRV ducts is usually a mistake. Even if the ducts are under the insulation, the extra holes in the pressure boundary of the house are hard to keep air-tight over the long term, and any duct leakage would then drive air infltration. Ventilation ducts are small and can be run in soffits under the ceiling if need be.

  31. Expert Member
    MALCOLM TAYLOR | | #31

    You wrote: "I've seen houses being built near me that started well after me and now they are pretty much finished. Sometimes makes me wonder if I'm doing too much."

    The only answer to that when you are building spec houses is whether you end up making sufficient profit that you can sustain your business. Less projects per year with a higher margin per project can be a a good hedge against drops in real estate prices. A good business plan going in will help you know whether you are on track or not.
    I put a lot of my own labour into the houses I build. Two recessions mean I've sold several houses for what they cost to build, but never lost my shirt. That's worth something. The downside it it's hard to scale up. You don't end up making a fortune either.
    Good luck!

  32. dsmcn | | #32

    Jimmy, I'm responding to your question about stairs in post #20.

    Stairs are confusing even for professionals, and it is wise to think them through twice before cutting a stringer. I find it especially helpful to mark on the stringer material itself the rough and finish material on the upper and lower floors, and on the treads, before I cut it (I draw the tread finish material on just the first and last one or two, enough to check that all unit rises will be equal). If there are landings that have a finish floor of a different thickness than the treads, then a story pole will be helpful.

    For example, if your stringer lands on a slab that will be tiled, and you are using tread material that is 1 1/4" thick, and if you have cut the risers all to the same dimension, the first riser will be an inch higher than the second one. You may have a similar issue on the uppermost riser, and at landings. A mistake results in a significant tripping hazard, and is much more common than it should be.

    Similarly, if the finish material on the face of the riser is different from the thickness of the material where the stringer lands at the top, the uppermost tread may be narrower or wider than all the others. Not so much a hazard, but certainly embarrassing to the carpenter!

  33. WildBunchFarm | | #33

    Thanks Malcolm and David.

    @David - Yeah, I believe I did the stairs correctly. It took a lot of time. I found out that I could have gotten each stringer prefabricated but the cost would have been $400-$600 each. In the future, when my labor costs are higher, it would make sense to do the prefabricated stringers.

    @Malcolm - Is there a way you can share your email address with me? I have some questions and ideas to run by you.

    @everyone - Does it make sense to get a Zehnder ERV when there are cheaper ones out there made by companies such as Honeywell? The price difference is $9,000 (Zehnder) vs. $1500 (Honeywell).

  34. Expert Member
    MALCOLM TAYLOR | | #34

    If you are comfortable posting yours I will email you.

  35. WildBunchFarm | | #35

    are you able to email me here nguyen.jimmy17 @ g m a i l . c o m ?

    I wrote it out that way to confuse the bots.

  36. WildBunchFarm | | #36

    ERV update - I've decided to go with Renewaire ERV. I will be going with the EV200. The unit costs around $1,000. The Zehnder ERV costs around $10,000 (it does come with all of the ducting and diffusers and includes commissioning). The Ultimateaire Recoupaerator 200dx goes for around $2100.

    Some of the main reasons I went with the Renewaire is lower cost, no need for defrost mechanism, and the fact that they've been around for 33 years. Running an ERV continuously will end up costing me around $10-$12/month, which is more than I expected. But a supply of fresh air is important. I like that the Renewaire has no defrost requirement because that would have added to my energy consumption during the colder months. I also like the fact that it's easier to clean than the other ERVs.

    Solar panel update - I've decided to do the solar panel installation myself. It will save me $5,000-$6,000. My friend who is a solar installer in Arkansas said that it's relatively easy to do and he could supply me with the panels for $1.50/watt. Some of the estimates I've been getting in Virginia for solar installation is between $2.70-$2.85/watt. Although, it's important to note that some installers use the microinverters and some installers use string inverters. There's a big debate on which one is better, but I'm going to go with the string inverter because I believe the Tesla Powerwall will not be compatible with the solar panels with microinverters. This is not to say that it will eventually.

    I'll keep you posted on DIY solar panel installation. If more people feel comfortable doing it, especially homeowners, it may increase the use of solar panels. I am installing them on my porch roof that is 10 feet off the ground - so it is not quite as scary as doing it up high on two-story roof.

  37. user-2310254 | | #37

    Jimmy. How do you plan to handle indoor humidity? The Renewaire doesn't appear to address this issue. That is one reason I went with an Ultra aire ventilator/dehumidifier.

  38. WildBunchFarm | | #38

    @Steve - I didn't really think of that. Do you think it's essential in my Zone 4 climate? I was hoping the minisplits could do most of the dehumidification. I'll do some more research on this issue. Thanks for alerting me to it.

  39. WildBunchFarm | | #39

    @Steve - Here is something from their website -

    "If RenewAire's ERVs recover moisture, can I still get rid of excess winter humidity in my home?

    RenewAire's ERVs are effective at controlling excess winter humidity in tight houses because they don't recover all of the moisture from the exhaust air. The RenewAire energy exchange core is designed to recover less moisture than heat, so plenty of excess moisture is exhausted out of the house. And because cold weather air is very dry, the long-term effect is making the house drier. Excess winter humidity is one of the top problems people solve with RenewAire. Studies show that enthalpic-transfer (transfers both heat and humidity) air-to-air exchange ventilators like RenewAire, when operated at the ventilation rates recommended to control poor indoor air quality, have plenty of moisture-removal capability in cold climates."

  40. user-2310254 | | #40

    Maybe Dana will chime in. Some mini splits seem to do a good job at warm weather dehumidification. If the Renewaire isn't introducing too much moisture, it may be okay. Which mini split are you considering?

  41. GBA Editor
    Martin Holladay | | #41

    Steve and Jimmy,
    Most homes don't need a dehumidifier. Operating of a ventilation system is almost always adequate to address high indoor humidity levels in winter, and operation of an air conditioner is almost always adequate to address high indoor humidity levels during the summer.

    If a house needs a dehumidifier, it's usually an indication that something is wrong with the house (or the occupants' lifestyle). There are a few exceptions -- for example, for houses in the Houston area.

    For more information on this issue, see All About Dehumidifiers.

  42. user-2310254 | | #42

    Hi Martin. Jimmy has a tight house in a southern climate. As the article notes, "If you live in an energy-efficient house in a humid location like the Southeastern U.S., you may need a dehumidifier."

    My situation is similar, which is why my energy rater recommended a dehumidifier/ventilator to provide the 80 cfm of fresh air his design required. At the moment, my outdoor humidity is 63%. Tonight, it will be around 80%.

    That said, my unit normally just ventilates. The dehumidifier only kicks in when the indoor level raises above the controller's set point. (I usually keep the house between 40% and 50% during the cooling season.)

    In Jimmy's case, do you think his minisplit will run enough to keep the house comfortable? The RenewAire unit retails for around $1K. An Ultra Aire would be two to three times more. If he needs more dehumidification, would his best course be to buy an inexpensive Energy Star rated dehumidifier?

    Just curious about the best way to approach this issue if someone lives in a tight house in a warm/humid climate.

  43. GBA Editor
    Martin Holladay | | #43

    My usual advice is to design a house that addresses moisture-entry issues, and to pay attention to potential problems surrounding moisture-generating activities inside the house. I don't advise installing a dehumidifier, which can be expensive to buy and is always expensive to run, in the absence of evidence that there is a problem.

    If a problem develops, a dehumidifier may be unavoidable. If it is, it's always best to start with a $250 stand-alone dehumidifier rather than more expensive equipment.

  44. user-2310254 | | #44

    That makes sense, Martin.

    Jimmy, it seems your ERV approach makes the most sense from a cost/benefit standpoint.

  45. Dana1 | | #45

    In winter dehumification can usually be achieved by ventilation alone, even in the southeastern US. In summer, a tight air conditioned house in that region will usually be adequately dehumified by the air conditioning, but not on all days, unless the AC is operated with a dehumidstat control, not just a thermostat.

    ERVs aren't anywhere near as efficient at returning moisture as they are at returning sensible temperature, and thus can still control wintertime humidity via ventilation. ERVs also allow higher summertime ventilation rates than HRVs for the same amount of latent load. But whether that is "worth it" depends on just how much summertime ventilation you think you really need.

  46. WildBunchFarm | | #46

    Well it looks like I'll just stick with the Renewaire. Thanks for the information everyone. I haven't been doing a good job of monitoring the relative humidity in my current house. I just looked at my monitor and it showed 70% relative indoor humidity, which is high. I turned on my LG minisplit's duhumidstat and now just have after 30 minutes it has dropped to 58% relative humidity.

    In pricing out minisplits, LG has always come out the cheapest by several thousand dollars. One contractor quoted me for Daikin minisplits and that was over ten thousand dollars more. Some of that can be attributed to higher labor pricing on their end, but that's still quite a difference. So far my one LG minisplit in my current house has been performing pretty well. I might just stick with them for this new house.

    I also found out that some minisplit manufacturers like LG make cassettes with one additional port that you can duct into a nearby room. I am going to use two cassettes in order to prevent some key areas from having cold/hot spots.

  47. WildBunchFarm | | #47

    Question for the community: my minisplit installer is going to use 1 - five ton condenser to run 8 indoor minisplit units. He says that it will be sufficient for the load and it will save me $3,000 by not having to do 2 - three ton units. I'm a little worried, but he seems very confident. His heat load calculation for my house came out to 40,210 btus and 27,409 btus for cooling. My ownn heat load calculation following Martin's blog ( came out to 42,000 btus for heat load. So at least we are close in numbers. The condensing unit can put out 41,000 btus at 17 degrees outdoor temperature. That's cutting it pretty close, but I am told that all of my units won't be running at max capacity all at once. For instance, my two minisplits in the basement probably won't need to be run at 70 degrees during the wintertime because it's not "living space."

    Does anyone have any advice on whether or not this is okay?

  48. WildBunchFarm | | #48

    Here are some specs on the unit. In case anyone was wondering, the whole package is going to cost $17,500, which gets me:

    1 - 12,000 btu for in-law suite
    1 - 7,000 btu for in-law suite bedroom
    1 - 7,000 btu for mud room
    1 - 12,000 btu for 1st Floor Main House
    1 - 12,000 btu cassette + ducting into one bedroom
    1 - 7,000 btu cassette for master bedroom + ducting into master bathroom

    2 - 9,000 btu units

  49. GBA Editor
    Martin Holladay | | #49

    This thread is four months old, and you're up to comment #48. Few readers will have the stamina to re-read all 48 comments.

    Can you remind us of your climate zone and the number of square feet you are trying to heat and cool?

  50. WildBunchFarm | | #50

    Hi Martin,

    It's Zone 4 and 2800 square feet with a 1600 square feet basement.

  51. Expert Member
    Dana Dorsett | | #51

    41K of heat load for a 2800' house (+1600' basement) would be about right if it were a code-min house in a location with a 99% outside design temp. That's a substantially high estimate for YOUR house @ +17F (probably 2x.) I'd hazard that your true heat load @ +17F ie lsss than 25,000 BTU/hr, and it could easily be under 20,000 BTU/hr, given the description:

    "My walls are double studded 10.25 inches thick with dense packed cellulose. And the attic will have 20 inches of blown-in cellulose."

    Who did the load calculations?

    As point of reference, the fuel-use measured heat load @0F of my antique 1.5 story sub-code 2x4 framed ~2400' house + 1500' basement (basement walls are insulated, but not the slab) is about 40,000 BTU/hr. (An agressive Manual-J says 40,723 BTU/hr, which plenty close enough.) My house at +17F would runs ~33,000 BTU/hr, despite having nearly 3x the wall losses of your house, and more than 1.5x window losses (all clear glass double pane or single + clear storms). If my house had your wall & roof U-factors and code-min windows it would be in the 18,000-20,000 BTU/hr @ +17F range. And this is a house with a much LESS efficient shape than yours to boot! (My house has a 14 corner footprint, and an unfavorable exterior-wall to floor area ratio compared to your comparatively elegant fooprint.)

    According the short-sheet submittal sheet provided ( )the minimum rated heat output @ 47F of the 5 ton LG is 17,940 BTU/hr which is MANY times your actual heat load @ +47F, and is a large fraction of your design heat load. It's not a very good fit- ridiculously oversized. The only advantage is the ability to micro-zone the hell out of the place, but it's not going to run very efficiently with most zones cycling on/off most o the time (even at +17F) rather than modulating. This really calls for a careful but aggressive room by room load calculation.

    Many of LG's individual mini-splits modulated down to ~1000 BTU/hr, but I'm not sufficiently familiar with they're mini-duct cassettes and the whole multi-split lineup to make good recommendations.

  52. WildBunchFarm | | #52

    Thanks Dana. I guess it's a relief that it's not undersized. Now I have to downsize some of those indoor units and possibly the outdoor unit.

  53. Expert Member
    Dana Dorsett | | #53

    Jimmy, seriously, do a real room by room load analysis on this house, or pay an energy nerd/engineer to run those numbers using aggressive rather than conservative assumptions on air leakage & R-values. Even half-ton heads are going to be extreme overkill for most rooms in a house like yours.

    Run the load numbers at both +17F and at +47F, which will give an indication of the outdoor temperature at which the multi-split compressor (and not just the zone heads/cassettes) begins to cycling on/off rather than modulating. Most Fujitsu & Mitsubishi multi-split compressors will still modulate down to 6000-7000 BTU/hr or so, independent of size, well below the nearly 18K of that 5 ton LG. Having some individual heads cycling a bit isn't nearly the hit that cycling the compressor would be. If your load at +47F is say, 5000 BTU/hr and the minimum modulation of the compressor is 18,000 BTU/hr it's going to take real toll on as-used efficiency compared to compressor that' can back off to 7,000 BTU/hr.

  54. WildBunchFarm | | #54

    Hi Dana,

    Can I pay you to do it? Or would you recommend anyone? I guess they don't have to live in Virginia if I can communicate over the phone or email.


  55. WildBunchFarm | | #55

    How do I determine what the condenser unit will modulate down to? The spec sheets give a range of 17,000-68,000. So if the BTU requirements in the house goes under 17,000 - the condenser will just shut off correct. I called their tech support and there wasn't much support given - if you know what I'm saying. Should I be looking for some specific word in the engineering manual?

    I know the indoor evaporator units can modulate down to lower btus, but I'm not sure if the condenser units can. I could downsize to the LG LMU480HV, but anything less than that and I can't run more than 4 minisplits off of one unit

  56. Expert Member
    Dana Dorsett | | #56

    I'm not currently in the business of running Manual-Js as a service. You want a competent professional engineer or RESNET/ HERS rater willing to calculate the U-factors on your non-standard wall / roof / other assemblies, and use the published U-factors for you windows actually used, not a pro-forma Manual-J on a house that shape & size with all code-min assumptions, including code-max air leakage, since your house is WAY better than code min. A quick directory search on the RESNET site ( ) comes up with:

    Building Performance Solutions, LLC
    1934 Old Gallows Rd. Suite 350
    Vienna, VA 22182
    Phone: 877.831.5061
    Email: [email protected]

    21750 Red Rum Dr, Ste 117
    Ashburn, VA 20147
    Phone: 703-797-2051
    Email: [email protected]
    Contact: Lee O'Neal

    Patuxent Environmental Group
    3975 Fair Ridge Drive, Suite T15S
    Fairfax, VA 22033
    Email: [email protected]
    Contact: Sandeep Kainth

    1431 W. Main Street
    Richmond, VA 23220
    Phone: 804.225.9843
    Fax: 804.562.4159
    Email: [email protected]
    Contact: Sean Shanley, Director of Residential Services

    There are others in neighboring states, and still others fairly far away that may be willing and able to run the numbers for you. You'll need the architectural drawings, the window & door U-factor and size data, etc to be able do it all remotely.

    The 17,940 BTU/hr (almost 18K) minimum of that 5 ton LG is it's capacity at +47F outdoors, 70F indoors. The minimum & maximum output changes with temperature (and humidity), so in fact it will be able to modulate quite lower than that when it's well below freezing, but probably not below 10,000 BTU/hr even at +17F. Your binned hourly wintertime mean temperature is probably about +40F give or take, where it's minimum output capacity is still probably pretty close to 17,000 BTU/hr. But your mean load at that temp is going to be quite a bit less than 17,000. The thing has about a 4:1 turn down ratio, and probably puts out at least 55,000-60,000 BTU/hr max @ +17F, which means it would modulate down to ~14K-15K at your design temp.

    That's probably lower than your actual heat load- it'll modulate some, but it's not a heluva lot lower than your heat load at that temp. I would expect it to be in a modulating range only a few hours per week, and only during the coldest weeks of the winter.

  57. WildBunchFarm | | #57

    Came here to post and I saw that I was mentioned in Martin Holladay's Musing of an Energy Nerd blog post. Finally famous! Although, the picture of the gentleman applying the capillary break over the footing wasn't me, it was my dad and he's 70 years old.

    I wanted to provide some updates on the house build. Taking advice from Dana Dorsett, I reduced the size of our minisplit condenser unit and some of the indoor heads. I did not get a room by room Manual J because the cost would have been around $750. I'm over-budget already and simply can't afford it. In any case, I downsized from a 5 ton condenser to a 4 ton condenser. The 4 ton was the smallest we could go to be able to power 8 indoor units. The alternative would have been to go with two smaller outdoor units, but that would have cost more money. By switching from a 5 ton unit to a 4 ton unit - we ended up saving $2,000.

    The total cost of 8 LG indoor units - (4 - 7,000 btu units, 2 - 7,000 btu cassettes, 2 - 9,000 btu units) = $15,500.

    One thing that I have discovered is that minisplits may not be that good at controlling indoor humidity, unless you run it in dry/dehumidify mode constantly. These posts are a bit outdated, but I don't think much has changed ( One of the posters said that right-sizing his minisplit solved the problem, but others seemed to think that this is a problem with all minisplits. Any thoughts? In my house, if I'm not in dry mode, the humidity can get up to 70%. In dry mode it stays between 50-60% depending on outdoor humidity.

    Siding updates: I went with LP Smartside lap siding for my siding. I am using it in a board and batten style, which will void their 5/50 warranty. It's a risk, but I think that with my rain screen and the borate treated LP Smartside I will be okay. I am also leaving a 3/16" gap between the vertical boards, which will provide some extra space for moisture to escape. Areas of our house, including the soffit, will receive 1x6 tongue and groove cedar STK stained with Sikken Cetol SRD. The total cost of materials for siding ~3400 sqft including Benjamin Aura exterior paint was around $11,000. This does not include my labor.

    For our front porch, I decided to go with a basket weave decking design. It wasn't the main reason I did it, but having angled boards does resist lateral forces better than decking placed perpendicular to the joists ( This will save me in the off chance 20 of my guests decide to run from one side of my porch to the other and do a jump stop. Kind of like when we used to try to tip over our school bus by all running to one side when the driver was making a sharp turn.

  58. GBA Editor
    Martin Holladay | | #58

    Thanks for the update! I have corrected the caption on the photo of your father; sorry for the mix-up.

    The photo of your famous father, taken (I assume) by famous photographer Jimmy Nguyen, can be seen here: Capillary Breaks Above Footings.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |