GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter 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

How necessary is air conditioning in an ICF house with decent shade and roof overhangs?

JackieTreehorn | Posted in Mechanicals on

Oh man, where do I start… I’m building a daylight basement 2700 sq ft icf house climate 4c close to Portland Oregon which will be my full time job.  Since I will be available for a lot of the labor and details, I originally envisioned doing radiant floor since I’d be able to lay the pipe etc and save on some of the high labor costs associated with it.

Obviously the downside with radiant is no a/c.  I grew up in a 70’s daylight basement house about 800 feet away with 2×4 walls and who knows what for insulation.  It was heavily shaded and up until a few years ago, never had a/c until some trees needed to be removed.  In the summer hot streaks we opened windows at night and closed in morning and it was tolerable. So I went from there to a cookie cutter house with a/c and no shade so those are the only two comparables I have to go by. 

I’ve been researching some hrv’s and a couple have a nice summer bypass feature where it sees what the temp is outside and bypasses the core to bring in cooler fresh air.  When I saw that I thought maybe I could get away with radiant floor and use the hrv summer bypass to cool down at night. 

My manual j says house will be around 26k btu for heating and 21 for cooling.  I planned to do a ceiling fan in great room and one in master. The downstairs to the house I grew up in never got warm during heat waves and I have a feeling my house will obviously do better with more insulation and air sealing details so I kind of feel like I may only need one mini split upstairs.  House is rectangular shaped and open concept.

Obviously with icf I don’t want to mess up and wish I had done a/c because changing things down the road is going to be a bear.

I added pic of house site, the ol’ toyota is about where garage will be and house will sit to the right.  I’d guess maybe 3 hours a day in the summer I’m in full sun.  Trying to convey as much info as possible, thanks for reading!

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. Tom May | | #1

    A/C is only necessary if one feels they need it....lots of people live in this world without it.

  2. Andrew C | | #2

    Where do you start? With a nice building site! Looks beautiful.

    You started at the right place: asking questions in the design phase. Good job, you’re ahead of the game, even doing a Manual J.
    My personal experience with basements is that they tend to stay cool, but you will want to control humidity and you will want to mix air. We currently have a similar layout (two level walkout). We have a split forced air system and it never comes on in the summer, only the upstairs comes on and circulates air, so the basement air gets a bit stale without manual stirring.

    I think the general consensus on GBA is that if you have a well-insulated and air sealed house, it’s difficult to use radiant heat because you have to turn the water temperature down so much that the floors are cold to touch, so it defeats the purpose. And if you’re putting in mini-splits, you already have both heating and cooling systems, so the radiant won’t be used. If you insulated under the slab (and use a vapor barrier) and also insulate the basement walls, you should have a dry basement with no mildew or mold that is also easy to heat and cool. Make sure you put in a passive radon system while you are addressing sump pump lines, etc.

    I suspect that the experts are going to suggest having two different minisplits, or heads, for two different floors. Two mini-splits will be cheaper than one mini-split and radiant, and will give you heating and cooling.

    In my reading, HRVs are important for fresh air, but the energy that they carry just isn’t enough for any heating or cooling.

    My $0.02. Have fun.

    1. JackieTreehorn | | #9

      Thanks for the reply, I read the gba article about how a well insulated house and radiant wasn't a preferred method. I also thought it was in reference to something in the 12k btu and lower passivhaus stuff. The warm toes wasn't so much of a concern to me as was not having vents, and not listening to the vents blowing air (again can only compare to places I've lived).

      Was planning on r10 under slab, and possibly r15 if I can fit in budget (friend owns an insulation company so I should be able to source it at decent price). I also live in a high radon county so radon abatement is required.

      The only downside to the mini split is I think I'd have to go with ceiling cassette to make wife happy. Not sure how they compare to the regular wall mounted ones, but from an aesthetic standpoint they look nicer.

      1. Charlie Sullivan | | #16

        I agree that radiant floors can make sense for advantage of being quiet and, compared to panel radiators, allowing more flexible furniture placement.

        They also work with low temperature water, if your heat load is low, which makes them efficient to supply from a heat pump such as Chiltrix.

        Might be good to consider radiant ceiling panels instead. If you are lying on the couch, you get no direct radiation from a floor but you do get it from the ceiling, so you can get more comfort that way.

      2. Expert Member
        Dana Dorsett | | #17

        >"Was planning on r10 under slab, and possibly r15 if I can fit in budget (friend owns an insulation company so I should be able to source it at decent price)."

        Price notwithstanding, it's nicer to the world (and usually your wallet) to use Type-II (or denser) EPS under the slab rather than XPS. (Most ICFs are made with Type-II EPS) It's usually cheaper per rated R as XPS, but has a lower rated R/inch. At 2" it would be R8.4, AT 3" it's R12.6.

        XPS is the same polymer, but uses HFC blowing agents which are all extremely powerful greenhouse gases. While the mixtures of HFCs are proprietary information held by the vendor, the predominant component is always HFC134a, which runs about 1400x CO2 @ 100 years. EPS is blown with isopentane (~7x [email protected] 100 years), most of which leaves the foam and is recaptured while still at the factory, not vented to the atmosphere.

        Not prominantly featured in the marketing fluff, the blowing agents are responsible for it's higher R/inch, and most of it diffuses out over a few decades. The warrantees are typically only for 90% of rated performance (or R4.5/inch) @ 20 years, and at full depletion it's performance is the same as EPS of similar density, R4.2/inch. While there is an early-years performance benefit, that marginal benefit comes at a steep environmental price, albeit a modest financial price. The financial upcharge is usually about 20-30%, but the CO2e footprint (at the same rated R, not lifecycle average R) is about 800-900% higher than EPS, making XPS hands-down the least-green insulation material in common use in the US:

        Whenever possible (which is about 100% of the time) it's far greener and usually cheaper to design XPS out of the system completely.

        Portland Oregon's climate is dry enough (in absolute air humidity or dew point terms, not rainfall, not outdoor RELATIVE humidity) to use radiant cooling & heating with a reversible chiller (Arctic , Chiltrix, etc.). Even during the peak humdity July/August season the outdoor dew points are under 60F more than 90% of the time:

        With air that dry the latent cooling load is usually negative, and indoor humidity can usually be controlled but the ventilation rate on the HRV even without air conditioning, but a dew point of 60F would still have to be designed for to manage a radiant cooling approach, since a slab or panel temp less than 60F could sweat condensation on peak humidity days, an the chilled water lines would need to have air-tight insulation to avoid condensation.

        Bottom line, it's not nearly as simple to design for radiant cooling as air source heat pumps, which can also operate at high efficiency in Portland's temperate climate.

        A RIGHT SIZED modulating air source heat pump (ductless or ducted) would have much lower noise & wind-chill issues than the typical 3x oversized hot air furnace or old-school single/double stage heat pump. If your Manual-J numbers are correct, a 2 - 3 tonner with a vapor injection scr0ll compressor (sometimes called "hyper heating") would have adequate excess margin even for the typical cold snaps that drop below the 99% outside design temp by 10F or more. Even some of the mini-duct cassette type modulating mini-splits with big turn-down ratios would have adequate capacity, eg:!/product/26524!/product/25351!/product/30364

        (There are others. )

        With larger turn-down ratios the blower on the air handler cassette runs at a VERY quiet minimum speed almost constantly, while the compressor modulates the output up/down to match the heating load. (These are not yer grandma's R12 refrigerant heat pumps.)

        1. JackieTreehorn | | #21

          Holy cow, this is info overload, thanks for the long response. So it looks like eps is the way to go and I'll have to do 3" which gets me close to the r15 anyhow.

          So the cooling with radiant kind of scares me because I'd be relying on someone to design it correctly and with so many different setups etc it worries me something would go wrong.

          I somehow didn't see this post and replied to your other post further down, but I'm starting to see the bigger picture of doing non ducted or ducted mini. The system in my old house was just on, blow hot air, turn off and then two minutes later your face felt cold and ten minutes after that it'd kick on again. I'm really trying to avoid noise, and having vents on the floor. Sounds like some of the units you linked probably would check those requirements off the list.

          I had energy vanguard do the manual J, it does say the required total capacity at .80 shr is 2 ton (not sure what shr means) and the heating summary equipment load is 25739 btu. Ideally I'd like to have all my stuff figured out and squared away before meeting with an hvac consultant just so I know what I'm talking about but also to make it easier as I can just say I want model x etc. Time to start some research.

        2. Eric Habegger | | #22

          Dana, you are a fount of knowledge but I think you may be misapplying it in this case. I'm seeing a lot of people applying what's working for them in their world and trying to apply it to Mr. Treehorne's problem. I'm not sure the OP knows that almost all the heavy contributors on this forum operate from with East, South, and Midwest. Not Malcolm though. It just seems to be the cultural bias of this site.

          I don't go on every discussion about air conditioning a house that is situated in humid summer climates and tell them to use whole house fans. I know better. It won't work. But I'm seeing an intentional effort by most here to either ignore or deny that the OP doesn't live in that type of humid climate. He lives in a moderate climate that is not humid. He has shown an interest in not adding a heat pump to cool his house and to use the cool night air. It will also be difficult for him to install a mini split in his ICF walls or in the ceiling and maintain an intact ceiling air boundary.

          Why is everyone ignoring the obvious? A whole house fan will work where he lives and it will work harmoniously with the heating system he is using for winter. I think you guys are selling Jackie short. Just because you guys have a "hammer" (heat pumps) for the summer conditions where you live does not mean the OPs problem is a "nail" for the far less humid conditions where he lives.

          Give the guy a break. He's got a plan and I think with some little adjustments it will work AND be cheaper and more energy efficient than what you guys are proposing. I'd like to hear the big Labowski weigh in.

          1. Tom May | | #23

            Good reply Eric. Just because a house is in a certain "zone" doesn't necessarily mean all the conditions associated with that zone apply. Think of someone who lives at the base of Mt.Washington and one who lives at the top only a mile away. Totally different conditions. Same can happen to a house that is exposed as opposed to being surrounded by woods a block away or one that is on the coast as opposed to a house a mile or so inland. Many factors drive a design and there is no "one fits all" solution.

          2. JackieTreehorn | | #26

            Well I am starting to lean towards the mini ducted, or ceiling cassette. The pro's for radiant for me are, no noise, no floor grates in the way, and the possibility of reduced labor by diy the tube install. In my old house which was cookie cutter 2x6 walls without any shade, we still manage to not run the heat or a/c for half of September and most of October, and again in April and May. The house was drafty and had builder grade windows so I know in comparison to how my new house will be it seems like I should be able to pull off the same amount of no heat/ac.
            It's overwhelming a bit trying to make a decision especially with having walls that won't be that easy to change my mind with.

          3. Expert Member
            Dana Dorsett | | #27

            >"Dana, you are a fount of knowledge but I think you may be misapplying it in this case. I'm seeing a lot of people applying what's working for them in their world and trying to apply it to Mr. Treehorne's problem. I'm not sure the OP knows that almost all the heavy contributors on this forum operate from with East, South, and Midwest. Not Malcolm though. It just seems to be the cultural bias of this site."

            I've spend about a third of my life in the PNW, advise friends & relatives in the region regularly on HVAC, building science and construction issues, and have worked and lived in Portland as well as other Oregon locations (mostly Willammette valley and coastal areas), as well as WA (on both sides of the Cascades.) What's "...working for..." ME is math, experience, and attention to weather data.

            Making me out to be a New England biased regional player wouldn't be a correct characterization. I've corrected others multiple times (including the honorable Martin Holladay :-) ) on mis-perceptions of humidity levels in WA/OR/BC. Dew points are an expression of absolute humidity, and even though it's easier to design radiant cooling for the drier (absolute humidity) of the Pacific Northwest than where I currently live, it's still not a trivial exercise.

            I have multiple relatives in western WA already heating & cooling with right sized mini-splits. I am currently trying to convince a sibling WA to get rid of the propane tank & ludicrously oversized propane hot air furnace and go with a ducted mini-split (keeping the wood-burner is fine in his location) which would be an EASY retrofit on his existing ducts. Summers seem to keep getting warmer in that area and electricity leveraged with any heat pump is dirt cheap compared to propane. But even a 1.5 ton Mr. Cool DIY (or the identical Senville & Pioneer DIY models, all manufactured by Midea) would be better than coming home to a sun-baked house and cranking on a noisy oversized window-shaker in the bedroom, which is his current cooling strategy.

          4. Eric Habegger | | #28

            Dana, my apologies if I lumped you in with individuals that are unaware of the generally non-humid conditions in summer of the Pacific Northwest. I can only relate to the OPs climate as just a slightly different version of my own. I live in Northern California about 150 miles north of SF. It's a slightly less extreme version of the Pacific Northwest with less total rainfall and hotter temps in summer. It has gotten up to 108F once this summer.

            I only use a whole house fan for cooling. You don't have as much controllability on indoor temps without a heat pump system. But for a person with a certain temperament like mine it kind of makes me feel connected to the outside world. My house is in no way completely optimized for passive cooling but it still works well, though not perfectly. By far the biggest improvements that make it work were these:
            1. Tightening up the 80 year old house to 2.25 ACH50.
            2.Putting a perforated radiant barrier below the roof deck.
            3.Building a wrap around porch on the south side of the house.

            It's not perfect by any means. When there are consecutive 100F + days it can eventually get up to about 82F indoors just before I turn the WHF on. I can easily live with those occasions because they are relatively rare and because the living room and both bedrooms have ceiling fans. I like the feeling of being not too cosseted in an isolating totally conditioned room.

            Like I said, the house wasn't designed from the ground up for this. It still has can fixtures in the ceiling with LED lights. They are relatively airtight but they take up space away from the cellulose insulation. Also, I don't have adequate cellulose on the perimeter because the rafter just meet in a bird's mouth and do not have anything like a raised truss. Finally, the roof shingles are not a low emissivity type. Everything consider I'm amazed at how well the WHF does. My biggest regret is I undersized the fan by about half so when the temp outside goes about 62F my house only get down to 69F. So even on the worst days the temp rise in the house is only about 12F.

            So you can see why I think a whole house fan would be ideal for the OP in his climate with new construction using radiant heat in winter. He doesn't need a heat pump for heating. And he can do all the things to optimize low heat gain during the day that I wasn't able to do since it is new construction. And of course there is the fact that the high temps where he lives will be significantly less than mine are. With the thermal stabilization given by the concrete and foam in the ICF walls that house would be just made for a WHF. That's why I'm so disappointed that no one here recognized that he had a really good idea from the get-go. It just needed refining.

  3. Eric Habegger | | #3

    I think I have to agree that HRVs won't be adequate because they just don't move enough air when there is a narrow delta between indoor and outdoor temps. A moderately sized whole house fan, which is the normal method for doing what you are trying to do, has cfm of around 2000 to 4000 cfm for your size house. That versus a HRV of maybe 100-200 cfm.

    I think you should just trust your instincts about cooling using cool night air and stick with a WHF. It will be much cheaper and much more efficient to have a dedicated WHF or two than minisplits or a big central heat pump. Cheaper in upfront cost AND energy costs.

    If you do go the WHF route then you obviously require a vented roof with adequate venting for soffits and ridge. This is really the way to go with new construction in non humid climates, like yours is. As always, air seal the ceiling/ attic transition and bulk up on cellulose in the attic. Preferably use raised heel trusses so you are well insulated at the perimeter of the attic.

    1. Eric Habegger | | #4

      Something I might add. It's wise to keep HRV and WHF switches together. I often tend to forget to turn off the HRV after I've turned on the WHF but grouping them together tends to avoid that. Of course, provided you go the WHF route.

    2. JackieTreehorn | | #10

      The whole house fan hadn't quite crossed my mind. Now I have a whole new thing to research. The cfm didn't even cross my mind either but makes sense, I just read summer bypass and thought it was a neat idea. I specified raised heel too...

  4. Jon R | | #5

    > Obviously the downside with radiant is no a/c.

    Generically, radiant is fine for AC, you just need to rely exclusively (vs some of the time) on a dehumidifier for latent removal. The typical response to this is "it's horribly inefficient". But it completely depends on how much moisture needs to be removed. In most cases, it's a small fraction (perhaps 20%) of the total load. With say 80% highly efficient radiant (very low delta-T) sensible cooling and 20% inefficient (heat is added back into the space by the dehumidifier) latent, I get a quite reasonable net COP of 3.85 (assume sensible COP=5 and a 2.89L/kWh dehumidifier).

    Perhaps someone cares to flush out the numbers for your climate.

    1. Eric Habegger | | #6

      I think you may be confusing a rainy climate with a humid summer climate which so many midwestern, eastern, and southerners are familiar with. Unless I'm badly mistaken, and have been for a majority of my life, where he lives near Portland is a perfect climate for a WHF. I live in northern California where we get 30 to 40 inches of rain in winter. But it's dry, dry, dry in summer. Also, even if it rains in summer in Portland I don't think that equates to the humid and sticky conditions that you may be familiar with. Maybe Portland, Maine...

      1. Jon R | | #7

        I think you missed the last line.

        1. Eric Habegger | | #8

          OK, you're right that Western Oregon has much higher humidity that Eastern Oregon. I just came from a google search that said the average August humidity in Portland Oregon is 60%. Hopefully that helps determine if he needs dehumidification. At least in summer.

          In Portland "Maine" the average daily humidity is 79%. I think that level of humidity would be much worse than 60% and Maine might not be suitable for a WHF. I can definitely see that near 80% RH would see too much moisture being brought in doors in Maine and much of the rest of the Eastern half of the country. I've been acustomed to think that in much of the west as long as the night time temps go to 65 F and below in summer that WHFs generally work well. Not so in the East. I'm open to being wrong about that though.

          1. Charlie Sullivan | | #14

            The number to look at is dew point, not relative humidity. Weatherspark says that Oregon gets stays below 55 F dew point 60% of the time, even at the worse summer peak. That's pretty great. Portland Maine, is above 60 F dew point 60% of the time at the seasonal peak. that's much worse.

    2. JackieTreehorn | | #11

      I was reading this article and it's one that got me head scratching about whether I needed ac.

      In the article GBA Technical Director Peter Yost refers to a guy named Robert Bean. He's quoted saying “For marine climates (Pacific Northwest) and high-performance buildings with sensible cooling flux under 5 Btu/hr/ft2, mechanical cooling is not necessary. Use dedicated de-humidification and elevated air speeds (ceiling fans), and ventilate for IAQ."

      My manual J works out to be ~7.5 btu/hr/ft2, but I'm not as near to marine climate in the 4c as I am to climate zone 5. If I drive 20 minutes to the east I cross the border between the 4c&5 so not sure how that impacts the de-humidification requirements...

      1. Jon R | | #12

        Also note that the need for dehumidification isn't just about outdoor conditions - interior moisture sources play a significant role (eg, 1 pint/hr). In some dry climates/conditions, you want an over-sized HRV to remove such moisture at lower cost than a dehumidifier.

        Similar for cooling. It might be 82F outside, but solar or other gains might bring the room temp to 100F - making fans and thermal mass ineffective.

        1. Charlie Sullivan | | #15

          When outdoor temperatures are mild but cool, the heat produced as a byproduct of dehumidification is actually enough of a bonus that, combined with moisture removal, it can be a better deal than trying to run an HRV at high enough airflow plus running a heater, with a high enough efficiency HRV, depending on the heater type.

  5. Trevor Lambert | | #13

    Is there a reason you're building with ICF? You posted on a green building forum, so I'll point out that ICF is just about the least green building method. Both concrete and foam have massive carbon footprints. Without extra insulation, it's also not particularly high performing from an energy efficiency standpoint. It's expensive relative to other wall structures.

    Most of the benefits bandied about regarding in floor heating are straight up myths. They cost a lot, and perform similarly to other means of heating. I'd go with a ducted minisplit system. It will be at least no more expensive than the radiant system, and provide you with cooling if you need it. This is from someone who recently built a home with in floor heat, and replaced it within two years with ductless minisplits.

    1. JackieTreehorn | | #19

      Mainly for the diy aspect of it. My full time job will be building the house and after helping on a half dozen icf homes, it's something I can tackle 75% without help. That was also the main reason I was considering hydronic is so I could take up some of the labor.

  6. Expert Member
    Dana Dorsett | | #18

    >"Most of the benefits bandied about regarding in floor heating are straight up myths. They cost a lot, and perform similarly to other means of heating. I'd go with a ducted minisplit system."


    The number of hours per year a house with a ~10 BTU per square foot design heat load would be providing that extra-cozy bare-foot comfort that would be more than barely-noticeable is fewer than 150, and those would usually be the pre-dawn hours most people spend in bed.

    1. JackieTreehorn | | #20

      So to pivot from hydronic, there's an energy credit my state does for keeping all the ducting within the conditioned space. "All heating and cooling system components installed inside the conditioned space. This includes all equipment and distribution system components such as forced air ducts"

      So how does one do that with a ducted mini system? I attached couple pics of floor plan. Blue line would be a location that I could do a ducted mini(red being duct runs), but I have zero idea how far the ducts can be etc. The great room is 10', master is 9', so that little hallway between the master and great room could have a drop ceiling in it to hide the unit. Same goes for downstairs above the closet. If I lose some height in closet, not a huge deal.

      My other question would be if using an hrv, and say it returns 80% of the heat it removes, if those hrv vents dump into office and master closet, bonus etc, those rooms would still probably stay at an ok temp?

      I'm learning, albeit slowly...

      1. Expert Member
        Akos | | #24

        You keep the ducts inside the envelope exactly as you suggest. A drop celing bellow the actual ceiling for the unit and some bulkheads around the ducts that you can't run inside an interior wall/floor.

        Well insulated and sealed basements in my much colder climate need very little heating and not much cooling, there is no need for a dedicated unit just for the basement. You can heat/cool your entire house with a single ducted unit.

        The basement is generally the best spot to install the HVAC with ducts feeding the upper floor. The best spot to put the unit is near one of the outside walls, run a large trunk in a bulkhead either at the bottom or top of the wall the entire side of the house. From this trunk you put takeoffs for ducts that run inside the floor joists to the registers where they are needed.

        With a taller room, putting a largish return somewhere high up helps a lot with cooling performance.

        Make sure to put a well sealed and large filter (2" or 4") on the return duct. This will keep your ducts pristine for many years to come.

        1. JackieTreehorn | | #25

          So I do have one positive for the ducting, I'm using web floor trusses so should have plenty of room for the horizontal stuff in between basement and main floor. The basement also has a small mechanical room where I hope to put water heater and hrv. I could technically run one duct through center of house but is there a way to duct into the main floor that doesn't involve grills on the floor? My last house had a grill that just fired right up into the bottom of the sofa, so trying to avoid issues like that.

          1. Expert Member
            Akos | | #31

            You can run ducts up inside interior partition walls. Also another common solution here is to run them behind built in cabinetry, for example a kitchen pantry cabinet or built in shelf storage.

            You are probably in the land of flex duct, but there is such thing as 3x10 and 3x14 sheet metal duct which fits perfectly between wall studs. You also always bump up the walls a bit to fit larger ducts. I've done this with framing the space with 2x4s on flat on both sides, easy way to run an 8" trunk.

            Ducts inside the conditioned envelope don't need to be insulated, so they take up much less space.

  7. Peter Sperry | | #29

    Whether you opt for it at the beginning or not, why not stub a couple of 2 or 3 inch PVC pipes thru the ICF wall in likely locations for future mini splits. Costs you nothing but a hour of time and some spray foam. When we did our ICF stem walls I put a six inch and one four inch on each wall for when I change my mind down the road.

    1. JackieTreehorn | | #30

      I still have that on my list of things to do before pour. Can't hurt to have some sleeves for future.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |