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

Minisplit advice

okrobie | Posted in Mechanicals on

I have figured out the BTU of my house and would like to install a mini split system. I think I need a 6000 BTU unit for each of 6 rooms and I would like to put a 18,000 BTU compressor on each of the sets of 3, but I don’t see them marketed in those combinations. My house is about 30,000 BTU total. How do I find out if my plan is appropriate and how do I go about buying 6 – 6000 BTU ceiling cassettes and 2 – 18000 BTU compressors? Thank you.

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.

Replies

  1. GBA Editor
    Martin Holladay | | #1

    Jim,
    There are several issues here.

    Issue #1 is a basic design issue: if you install a ductless minisplit head in every room that needs heating and cooling, you'll end up with too much capacity for your house. The usual solutions are (a) to build a very tight, very well insulated thermal envelope for your house -- one that includes high-performance windows -- and then install just one or two ductless minisplit heads for the whole house, or (b) to switch to one or two ducted minisplit systems, so that one unit can serve several rooms.

    Issue #2 is getting down to brass tacks and actually choosing the model numbers you want to install. To do that, talk to your Mitsubishi or Fujitsu rep to find out which models are available.

  2. okrobie | | #2

    Hi Martin, thanks for your reply. I guess I have a lot to learn about mini splits. I'm used to my current forced air system where you need a register in each room. I'm trying to match my capacity with my estimated load. I will attach a rough drawing of my load estimates. According to my drawing, I need about 6000 BTU per room and possibly a little more for the kitchen. Would you please have a look and tell me what you think? Thanks, Jim

  3. GBA Editor
    Martin Holladay | | #3

    Jim,
    1. Where do you live? We need to know your climate zone.

    2. Is this an old house or a new house?

    3. What type of wall insulation does the house have? Attic or roof insulation? Floor or basement insulation?

    4. What type of windows does the house have? (Single, double, or triple-glazed?) Do you know the U-factor of the windows?

    5. Is the house built with attention to airtightness, or is it leaky?

  4. okrobie | | #4

    Hi Martin,
    1. I live in Jacksonville Florida. The climate here is jokingly called South Georgia

    2. It is an old house. Built in 1933

    3. As far as I know, there is no wall insulation. The attic is blown in and the floor is not insulated. No basement. It is built up on the foundation (not slab)

    4. We have the original 1933 windows. Single glazed

    5. I would say LEAKY ( a friend who just installed a mini split in his house was here yesterday and said it was very leaky)

  5. GBA Editor
    Martin Holladay | | #5

    Jim,
    You are prepared to pay about $18,000 to $24,000 on ductless minisplits. This is almost undoubtedly overkill.

    Moreover, the thermal envelope of your house is a disaster from an energy perspective. Your best course of action is to take at least half of your planned investment -- that would be $9,000 to $12,000 -- and use that money for air sealing and insulation. You may even be able to replace a few west-facing windows with new, low-SHGC windows for that kind of money.

    Once you've done that, you'll be able to install half as many ductless minisplit units as you are now planning. You'll be more comfortable, and your energy bills will be far lower.

  6. Expert Member
    Dana Dorsett | | #6

    What is the wall type & stackup (from the interior paint, to the exterior paint)?

    An uninsulated framed wall adds significantly to the cooling load, but an uninsulated masonry wall not so much. It's still worth insulating a masonry wall, but it's not the same energy-emergency that an uninsulated framed wall would be.

    Who did the cooling load calculations? They look high, even for an uninsulated framed house. Or was that just a best-guess apportionment based on the existing cooling equipment size and the relative room sizes? A so-so code-min house that size would have less than 1.5 tons of cooling load, maybe less than 3/4 tons. A 1930s antique with single pane windows, some attic insulation and no wall insulation might have a total house load of 2 tons, but that can probably be brought down to 1.5 tons (or less) cost-effectively, and the building improvements would do more for comfort than any cooling/heating system can.

    Hard-coat low-E storm windows are much much cheaper than replacement windows, and while they won't cut the heat gains as much as a high performance low SHGC replacement window, it'll still cut it significantly from clear-glass single panes. Even re-glazing the single panes with hard coat low-E glass makes a difference, but that's probably as expensive as low-E storms, that would perform better, and would not tighten up the window air leakage the way a decent storm window would. Larson low-E storm windows are available through the box-store chains, and are all custom orders. (It's worth the up-charge for going to their "Silver" or "Gold" series for the better hardware and higher air-tightness.) They might look at you funny ordering that sort of item at a box store in south Georgia location but it'll make a difference.

    Air sealing the house makes a large difference in comfort too, since the latent loads in your climate are pretty high. Are the existing ducts in the attic, above the attic insulation? Is there a crawl space under the floor, or is it masonry-on-dirt or something?

  7. okrobie | | #7

    Hi Martin, as you might have guessed, I'm in the very preliminary phase of this project. The numbers you are posting would rule out the project entirely. Naturally one of my first concerns is the financial feasibility. The input you are making is valuable. Thank you. Most of the work will be DIY. I have a neighbor who spent 30 years as a maritime engineer and he just installed a mini split all on his own without help. He even knows where to get a vacuum pump (and how to use it) Anyway he promised to help me, so the labor part should be reasonable. I'll address some of the other issues when I answer Dana in the next post.

  8. okrobie | | #8

    Hi Dana, thanks for your thoughtful post. The house is brick construction so it's not as bad as it could be. The load estimate was done by a seat of the pants method using a square foot to BTU converter I found on some website. The current evaporator is 3 ton (36000 BTU) and when my estimate numbers came out similar to the actual existing system, I was happy. Until I started getting feedback from Martin, yourself and others who were so concerned about insulation and leaks, I hadn't given it any thought at all. Floridians are commonly disinterested in energy conservation. I have never heard of the various types of glass you mentioned, so I have more research cut out for me.
    There are ducts in the attic but they are covered up by blown in insulation. The house is above ground on the foundation with crawl space underneath. The floor has never been insulated. I've wanted to insulate the floor but there is a lot of old wiring that needs to be replaced with Romex and I want to do that before foaming the floor. I'll follow up more soon.

  9. Expert Member
    Dana Dorsett | | #9

    Insulating the floor isn't a huge priority in your climate zone, but a crawlspace can be a better place to route cooling ducts than in a hot attic. Even with the ducts under some of the insulation it's adding at least a half-ton of cooling load up there. If the ducts leak much it can be a ton or more of additional load.

    Rather than insulating the floor, installing a ground vapor retarder and insulating & air sealing the crawlspace walls brings the crawlspace fully inside the pressure & insulation boundary of the house, in which case duct leakage won't drive infiltration rates nearly as much. A single 1.5 ton mini-duct cassette like the Fujitsu 18RLF or Mitsubishi SUZ/SEZ KD18 mounted in the crawlspace would handle the likely load, and you could then seal up the ceilings for better air tightness, lowering the latent load.

    http://www.fujitsugeneral.com/PDF_06/Submittals/18RLFCD%20Submittal.pdf

    http://www.fujitsugeneral.com/duct.htm

    http://www.mylinkdrive.com/uploads/documents/1249/document/SEZ-KD18NA_SUZ-KA18NA_Submittal.pdf

    http://usa.mylinkdrive.com/uploads/documents/4297/document/17_8_SEZ_Ducted_Heat_Pump_Systems.pdf

    While these don't have the soaring SEER numbers of ductless mini-splits, the output can be divided more appropriately for the doored-off room loads. With a sealed-insulated crawlspace in a 1-story house you can even use the crawlspace as the return plenum, and would only need supply ducts. While the ducts adds something to the cost and cut into efficiency it's nothing like the cost and comfort issues of a ridiculously oversized ductless solution that would be cycling on/off all the time to meet the minimum load requirements of the compressor.

    If local code requires that the crawl space be ventable or drainable (sometimes the case in flood plane areas) it's fine to build insulated and weatherstripped removable hatch covers for the vents to meet that requirement, and keep it air tight until you actually NEED to vent or drain. If the crawlspace is currently vented and has no ground vapor retarder, it's adding a lot of unnecessary moisture to the house. A 6 mil polyethylene or EPDM (roofing membrane) ground vapor retarder with the seams lapped by by 12" and taped/caulked sealed to the crawlspace walls with 2" of rigid EPS or closed cell spray foam is pretty air & vapor tight. If using the crawlspace as the return plenum you'd have to use intumescent paint or other firebarrier on any foam, or use fire-rated Thermax polyiso as the insulation.

    Are the walls double-wythe or triple-wythe brick with plaster applied directly to the inner wythe, or is there furring & air space with plaster & lath, or some other type of brick wall? How much roof overhang is there on the rake & eaves?

    It's not the easiest to use, but this online cooling load calculator is probably more accurate than most:

    http://www.borstengineeringconstruction.com/Cooling_Load_Analysis_Calculator.html

    You'll have to figure out quite a few of the parameters based on actual construction details, and take a WAG on some others (particularly the air leakage).

    Clear single-pane glass has an SGHC of about 0.81 passing through 81% of all incident radiation (and about 89% of the visible spectrum radiation.) A soft-coat low-E coating would cut that by nearly half but they aren't rugged enough for putting on exposed window surfaces. While re-glazing with hard-coat low-E cuts it by only ~5-7%, adding a hard coat low-E storm window to an existing clear window cuts the effective SGHC by more than 10% delivering ~0.65-ish performance, while reducing the window's air leakage. It's primarily the west facing windows that drive peak cooling loads, since the solar gain occurs later in the day when the outdoor temperatures are higher and the house has been sun-soaked all day long, with the brick starting to radiate inward. If you have west facing windows driving peak loads skyward in some rooms, heat rejecting window films applied to an exterior storm window can cut the effective SGHC to 0.4 or even less, depending on how much daylighting you're willing to give up, but the second pane gives you that option. IRC 2012 code-max replacement windows would have an SGHC of 0.25, and would have multiple low-E coatings.

  10. okrobie | | #10

    Hi Dana, thanks for all the information. If I can't find a way to put a practical Mini Split in, I don't think I will do anything on the mechanical side so I won't need a plenum or ducts in the crawlspace. It's laughable to call it a crawlspace because when I go into it I'm literally on my belly with about an inch or two clearance for my butt. I don't see putting anymore in there. I've got to do the electrical work I mentioned and I hope I don't have to go down there any more after that. I definitely don't want any city inspectors going in there, so I don't know what code requirements are for this type of crawlspace. I like your window suggestions especially for the West side of the house. I don't want to do window replacement for architectural reasons, but installing new panes is definitely workable. I have to remove a few of them anyway to replace sash cords. Many of the West windows had ugly looking film on them and my wife made me remove it. If I went to a dealer and asked for a bid on storm windows here in my region, they would laugh me out of the place. There is a lot of energy management ignorance in these parts.:) I looked at that estimator page, but I don't have a clue how to determine some of those parameters. I'll give your ideas more consideration and get back. I do appreciate your expertise and your willingness to share your knowledge. Thanks.

  11. Dana1 | | #11

    Whether or not you use the crawlspace for ducts, instead of insulating the floor, it's worth putting down a ground vapor retarder and insulating at the crawlspace walls. That is effectively "earth-coupling" the house thermally to the very favorable subsoil temperatures in your area (low 70s F), while isolating it from both ground moisture and outdoor humidity (from infiltration/ventilation).

    When you insulate between floor joists of an air conditioned house with a vented crawlspace in your area the joist edges can be colder than the crawlspace air's dew point in summer, leading to mold & rot issues.

    One way to get around that is to use closed cell foam and encapsulate the joists, not just insulate the floor. But with a 15-18" crawl space you would use a lot less closed cell foam by insulating the crawlspace walls instead, and gain the benefit of thermally coupling to optimal temperature soil, lowering both the heating & cooling loads.

    It's less risky, since it brings all the structural wood inside the fully conditioned space where it is warmer (=drier) in winter, and drier in summer since it is dried by the air conditioning, and not exposed to outdoor humidity.

    BTW: Which side is north in that floor plan sketch?

    And, what is the material stackup of the walls, just in case there is a reasonable & safe way to retrofit-insulate/air-seal without gutting the place.

  12. okrobie | | #12

    I'm originally from "Up North" (Pennsylvania, Milwaukee, Tulsa and Chicago) so locals still consider me a Yankee, even though I have lived in Florida since 1980. I have been asking around about crawlspace treatment and the consensus so far is to leave it alone. Thanks for the warning about leaving the joists un-treated that is really valuable information. The sad truth is that we get frequent flooding in our crawlspace and the open access points are necessary to allow the water to evaporate. You might be aware of a way to work around that issue, but I don't know of a way. The top of the drawing is North. The walls are plaster with lath underneath. The lath is supported by 2X4's so there is about 3.5 inches of air between the lath and the brick. I think I have addressed all of your points, so I will wrap up this post for now. Thanks, this has been a great learning experience. Jim

  13. okrobie | | #13

    When we get heavy rain, the water in the crawlspace reaches3 to 6 inches. That happens at least twice a year and my neighbors have the same experience. The funny thing is that my subdivision is officially called "Murray Hill Heights", elevation 15 feet above sea level.

  14. Expert Member
    Dana Dorsett | | #14

    With a flood prone crawlspace it's still worth putting down an EPDM ground vapor retarder and insulating & air sealing the crawlspace walls. Build removable weatherstripped hatches for draining & drying after the flood. You may have to put some washed smooth 3/4" stone screenings as ballast on the vapor retarder to keep it in place during floods. Alternatively 1.5-2" rat-slab with a 6-mil poly vapor barrier underneath it should work.

    You might be pleasantly suprised to get 0" of water in there after the torrent, and if you dig a sump and install a pump you can probably all but guarantee that. But if it takes on water popping out the vent hatches would allow it to drain & dry as rapidly as it did before, or even quicker since it won't be seeping groundwater for days. Once the visible water is gone the air conditioning would handle the rest, once you close the hatches.

    The alternative is to put ~2" of closed cell foam on all of the wood, encapsulating the floor joists.

    If it were possible to get rigid foam goods in there (doesn't sound likely) a continuous couple inches of foil faced polyisocyanurate foam on the underside of the joists with the seams taped with foil tape would be greener, and would be to current code, leaving the joist bays empty. Just be sure to air seal at the ends of the joist bays too.

    With 3.5" of air between the plaster and the brick it's probably going to be possible to insulate this house without gutting it, provided there is sufficient roof overhang that there isn't a lot of direct wetting on the exterior on a regular basis. You'd need to use good masonry sealers to limit liquid water uptake rates on the exterior of the brick. A "half-pound pour" of slow rise open cell polyurethane foam is one potential solution, and probably cheaper (and definitely more air tight) than a non-expanding injection foam solution, either of which can be done with a single 1" hole per stud bay (on either the exterior or interior.) Both are fairly vapor-permeable but air tight, and the studs could still be dried by the air conditioning, despite the lack of a ventilation cavity behind the brick. If you don't have at least a foot of roof overhang you have to approach this cautiously...

    A 2lb pour of closed cell foam would be more expensive, but could also work. Even if the exterior edges of the studs deteriorated over time, 3.5" of 2lb foam is structural, very adhesive and is fairly vapor tight. You'd essentially have glued the back sides of the lath to the brick with a 3.5" thick vapor barrier- no studs required! But it's not the greenest of materials- 4x the polymer of a half-pound pour, and high global-warming potential HFC blowing agents.

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |