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
Energy Solutions

A Pioneer of Low-Energy Homes Since 1973

Bruce Brownell has been building foam-insulated homes in the Adirondacks for decades

Image 1 of 2
Bruce Brownell believes that the best way to insulate homes is with exterior rigid foam, following the PERSIST system.
Bruce Brownell believes that the best way to insulate homes is with exterior rigid foam, following the PERSIST system. A home built by Adirondack Alternate Energy in Northville, New York in 1986.
Image Credit: Adirondack Alternate Energy

Bruce Brownell, of Adirondack Alternate Energy, has been creating low-energy, largely passive-solar-heated, resilient homes in the Northeast for forty years — and he’s still going strong. Since 1973, Bruce has built more than 375 homes in 15 states, a third of them in very cold (over 8,500-degree-day) climates. Most require just a few hundred dollars of heat per year.

He told me that he’s done enough monitoring to know that even in very cold climates his houses will never drop below 47°F if the power and supplemental heat is shut off. The fact that these houses will never freeze makes them popular as vacation homes; they can be left closed up with no heat all winter without worry.

I’m surprised that Bruce isn’t better known. While a few of us hold him up on a pedestal as one of our leading low-energy pioneers, most of today’s low-energy designers and builders have never heard of him. I’ve pondered why that’s the case, and I think it must be that Bruce just rubs some people the wrong way.

Always a renegade

I’ve known Bruce since the early 1980s, having met him at various solar conferences. I can remember getting into arguments with him back then about some of his ideas. I recall, for example, him arguing that caulk is a bad idea, and he has always shunned heat-recovery ventilators.

Bruce’s strong opinions turned off a lot of people, I think, including editors of the periodicals we all read. So his houses haven’t received a lot of attention. But he keeps at it, and his track record is certainly impressive.

Bruce is still building much as he was in 1975, though with a few refinements over the years. And his houses seem to keep working — really well. He’s done informal monitoring of hundreds of these homes, and the New York State Energy Research and Development Authority (NYSERDA) has done more in-depth monitoring of a few of them.

Occasionally, weather events have tested his houses. Bruce told me recently that when a 1988 snowstorm knocked out power for three weeks, some of his homes served as refuges, with the owners’ friends or family moving in. The same thing happened with the January 1998 ice storm that knocked out power for up to six weeks in parts of the Adirondacks.

What makes Bruce’s homes perform so well?

His houses are all wrapped with four inches of polyisocyanurate insulation — using two layers with overlapping joints and all seams taped. All six sides (walls, roof, floor) are insulated with this system. Bruce claims this achieves about R-36; I suspect that it’s no more than R-30 — and probably a bit less that than. But because it’s a continuous layer of insulation, not thermally broken by wall studs or rafters, and because it’s fairly airtight, the performance seems to be very good.

A big part of the performance comes from passive solar design features (augmented by fans). Adirondack Alternate Energy houses are oriented with a long wall and much of the window area facing south. A small fan pulls air from the peak of the house down through an air-shaft and into a network of pipes buried in a bed of 70-100 tons of sand, providing thermal mass.

Heat from this thermal mass radiates upward into the house. Backup heat can be supplied by a wood stove, domestic water heater, boiler, ground-source heat pump, or air-source heat pump.

This house air is filtered using a moderately efficient (MERV-8) filter, which removes most dust and other particulates. Bruce doesn’t believe a heat-recovery ventilator is required, and while I don’t agree, it seems from anecdotal evidence he reports that his approach is keeping occupants healthy.

Keeping all the wood on the interior of the insulation allows it to dry out, and it sounds like there have been virtually no moisture problems over these four decades. Ice dams never occur, he told me.

What’s ahead?

While most of the rest of us, including the building science community, seem to shift their recommended building practices on a fairly regular basis, Bruce’s Adirondack Alternate Energy keeps at it with a system he’s tested for decades. Bruce is getting older, and I don’t know whether others in his company will carry on his vision of low-energy houses when he retires.

But his completed projects present a larger and larger collection of case studies of a simple system that seems to work well.

While I haven’t always agreed with Bruce, I admire his tenacity. He is indeed a visionary.

Alex is founder of BuildingGreen, Inc. and executive editor of Environmental Building News. In 2012 he founded the Resilient Design Institute. To keep up with Alex’s latest articles and musings, you can sign up for his Twitter feed.


  1. jinmtvt | | #1

    Alex, did he mention to you
    Alex, did he mention to you if he simply followed the PERSIST method or came up with it by his own ??

    Do you know what he is using under the insulation for air/vapor barrier ??

    R-36 ?? he must have a magical polyiso source we can't access ... if using a PERSIST
    layout and not sealing at gypsum level, the remaining of the structure adds almost no insualtion value, + the temperature affect on iso insuation ... seems a bit low for roof,
    but the air sealing must be very good and we all know that this plays a very large role.

    A few hundere dollars per year in 8500HDD is a very goot feat,
    something not many houses can attain in the north east usa/canada regions...

    Is there a collection of pictures of the built homes somewhere ??

    When you say he uses a MERV 8 filter for his homes, do you mean he is not having anything for the make up air except for leaks ??

  2. mfredericks | | #2

    Other examples
    Thanks Alex, this guy sounds interesting. I've always thought the PERSIST approach seems to be a great building method in any climate and I keep wondering why there are so few examples of builders who choose it for new builds. Does anyone know of developers/builders who often use PERSIST in new home construction, outside of Alaska?

    I'm also curious along with Jin about his air/vapor control layer(s) and the seeming lack of make up air supplied to what must be very tight homes.

  3. itserich | | #3

    rigid foam
    Here is some information on a home based on Bruce Brownell.

    And Bruce Brownell's site.

    As a novice upgrading my home, I am glad to see some appreciation for rigid foam.

  4. jinmtvt | | #4

    Alex: i read "urethane"
    Alex: i read "urethane" boards on both links, but you stated polyiso in your blog .. ??
    maybe he has used both ... anyhow, just pointing out !

  5. jinmtvt | | #5

    A key

    A key component of our energy efficiency is our sealed envelope which is many times tighter than a typical new homes and will blower door test at 0.2 air changes at 50 pascals. We have no interior combustion- no gas appliances allowed, airtight wood stoves are vented outside and a fan moves all interior house air 3 times an hour through high capacity MERV 8 filters. Most importantly, we are totally sealed from moisture with no dampness and not needing any humidifiers or dehumidifiers.

    what is code requirement in usa for air exchange of 0.2ACH house ??
    I see no mention of other exchange system with the exterior ...

  6. user-1119494 | | #6

    R36 or R36-equivalent?
    I've calculated our CA standard R13 2x4 stud walls at about R8 when accounting for bridging losses. Since polyiso runs about R6/inch, 4 inches=R24, and (13/8)*24=(approx) R36.....and it is far far tighter than our rotten CA walls.

    So, OK, I'll buy the R36 equivalence, even if it is not technically R36 insulation.

  7. watercop | | #7

    I suspect the annual energy
    I suspect the annual energy cost difference of heating a home whose walls PERFORM at R30 vs PERFORM at R36 is fairly trivial. Stopping air leaks is far more important, and it sounds like Bruce's methods do just that.

  8. itserich | | #8

    Both Thermax (sold at Menards) and Tuff R (sold at Home Depot, but appears to have been replaced by Thermasheath) mentioned on the links are polyisocyanurate.

    In the house I am working on I am using expanded polystyrene because it is cheaper. Cutting and fitting the pieces is tedious but I think it is a one time job with few potential performance problems as long as the foam is tightly fitted.

    It is confusing why the Brownell pages state R 36, someone should inquire.

  9. wjrobinson | | #9

    Bruce is a great guy. I have
    Bruce is a great guy. I have been in his designs. Alex, Bruce no longer uses sand. His plans now call for the air ducts to pass through 12" of concrete. The homes are very even in temperature from concrete lowest floor level to top of the interior.

    I posted about Bruce here at GGA for years now. Search the site for more.

    And as to foam, some foam fits better than others. Foam needs to be fit as perfectly as possible at seams. Thermax I think is one foam that is used. Have to check.

    And I do believe the R rating is Bruce's way of letting the public know that his 4 inches continuous is much more R than if in a standard stud wall.

    The NYS study has great ideas on how to build an even better Bruce home.

    Bruce still designs the homes. Other builders build them. I know three.

  10. user-1115477 | | #10

    Bruce's R36 Walls
    I looked at one of Bruce's houses online a while back. Not sure I remember, but isn't the R36 a nominal value that also inclues some insulation in the wall cavities, in addition to the layers of external rigid foam? For example, I consider my house walls to be nominal R41, somewhat conservatively, as follows:
    ---2 layers of 2 inch polyiso: R24
    ---1 layer of EPS: R4 (came with the original house before retrofit and left there)
    ---original fiberglass in wall cavities: R13

    Ah, sorry, I think Dustin already basically said that. Also, as curt said, infiltration is huge for heating/cooling energy savings.

  11. Expert Member
    Dana Dorsett | | #11

    Answer for Sonny
    The center-cavity R of the framed portion is offset by the fraction of the area that is framing. In "typical" 16" on center 2x4 construction it's a whopping 25%.

    With fire-stop blocking and seismic zone structural elements in Dustin's CA it's often in the 30% range.

    Assuming some siding, half inch plywood sheathing plus half inch gypsum interior adds up to about R1, a 2x4 wall with a 25% framing fraction and R13 cavity fill averages about R9. At 30% framing fraction it's about R8. Sub-20% framing fraction balloon framed houses without fireblocking usually come in at about R10 with R13 cavity fill, as can 24" o.c. houses using advanced framing methods to minimize the thermal bridging.

    In Bruce Brownell's Adirodack region it's best to derate the polyiso to about R5.7/inch in that stackup, assuming an R13 cavity fill on a 2x4 wall. The ~R6/inch nominal for iso is valid when the center of the foam layer is at 75F. January mean temperatures in say, Plattsburgh NY is about 20F and in higher elevation locations it's lower still. With R9 average on the interior and about 3/4 of the R on the exterior, the temp at the mid-point of the foam is going to be about 45F, a point where the performance of polyiso is measurably lower. (At a 25F mid-foam temp it runs about R5.5/inch.). So, at a derated R5.7/inch that 4" of iso is about R23, add R9 for the studwall/sheathing/gypsum at you're looking at a nominal mid-winter whole-wall performance of about R32 for Bruce. If it's 2x6 framing he's probably hitting his nominal R36 performance as a seasonal average, if not his January average, due to the higher performance of the foam at somewhat warmer temps.

    In your case you left the EPS on the warm-in-winter side of the polyiso, so it's really still about R4.2, maybe R4.3. If your mid winter temps are similar to Bruce's your nominal whole-wall performance is about R36. If you're in a much warmer/milder climate it could be north of R37, but not R41 (but that's only about a 10% difference in expected performance.)

  12. user-1115477 | | #12

    Hee, hee, "R" should be "r" because it's a little ridiculous

    In your last paragraph, I think you misuse the word "nominal." My nominal wall value is just what I said, R41, because, historically and conversationally, people just add up the nominal R values of the layers and that's the nominal value. Not meant to be scientific.

    You were trying to pinpoint the actual value, not the nominal value. Such intense R value scrutiny is actually pretty funny. Our ancient ancestors from the 1990's would laugh outright at such detailed analysis.

    Anyway, I'm wondering if you would calculate my actual wall R value for me, taking into account the approximately 2,000 7.5-inch-long Headlock screws used to attach my layers of polyiso to studs? I started to do it, but I was unable to determine how many actually broke out the sides of studs slightly.

  13. itserich | | #13

    interior insulation
    This description of the building process does not mention any interior insulation.

    As a novice I may be wrong, but the amount of time and money to insulate the interior seems much greater and less efficient than adding more exterior insulation.

  14. Alex Wilson | | #14

    AAE's wall system
    I believe that Bruce Brownell relies only on the exterior foam insulation with no cavity-fill insulation to compliment it. If that's the case (and I'll try to confirm it) the R-36 figure is obviously significantly exaggerated.

  15. wjrobinson | | #15

    Alex, the R value is what it
    Alex, the R value is what it is because R value yapped about and on plans and codes for Bruce's life were stated for discontinuous very leaky horrible builds. I do not like your tone Alex. Bruce is quite the guy. I can see why you argued with him.

    And no, his homes do not have insulation in the walls. No. If you would go to a building site of his you would know this instead of postulating here which is mighty poor journilism is it not?

    Not far from Vermont you could come over and walk through a Bruce home. I might be able to arrange such Alex.

  16. Alex Wilson | | #16

    Come on, AJ
    I have great respect for Bruce, which is why I wrote the article about him. I admire what he does, and I always have. I respectfully disagree on a few issues, and I believe that his claimed R-values are exaggerated. But at the same time I believe that the R-values most designers and builders quote are even more exaggerated.

    The solution to inaccurate R-value claims, in my opinion, is not to inflate those R-values that aren't affected by thermal bridging due to framing, but rather to discount those R-values that are affected by that thermal bridging. Having long read your perspectives here, I'm surprised to hear you argue otherwise.

    We should not be listing the R-value of a 2x6 wall insulated with fiberglass as R-19; from data I have, we should report that as more like R-15 (assuming 16" on-center framing). Bruce's walls have a true 4" of polyiso insulation, which is great. But that doesn't make it R-36. It might perform better than a cavity-insulated frame wall with nominal R-36 (before accounting for thermal bridging), but that doesn't mean we should call it an R-36 wall.

  17. Expert Member
    Dana Dorsett | | #17

    Come ON Alex- let's get real!
    "We should not be listing the R-value of a 2x6 wall insulated with fiberglass as R-19; from data I have, we should report that as more like R-15 (assuming 16" on-center framing). "

    Who is exaggerating now!!? Show us your data!

    (No-really, this I'd like to see! I LOVE magic shows! :-) )

    It takes some careful design (and sometimes even code-violations on fire-blocking) to hit a 20% framing fraction at 16" o.c., and R19 batts in a 5.5" deep cavity only perform at R18 according to both Certainteed & Owens-Corning compression charts, eg:

    Assuming you can get it as low as 20% (which is usually a stretch), and allowing R1 or so for the sheathing + siding + gypsum, the best you'd hit is about R13.1, unless you're using some extra-special lower-U framing timbers.

    Using a more realistic 25% framing fraction that drops to R12.2

    MAYBE if you don't have any windows & doors, no fire blocking and singled top plates to the studwall you might approach R15, but it takes about a 12% framing fraction to get there (not too likely, given that just the studs adds up to over 9% framing fraction), but it reads like a ~23% overstatement of R-value to this casual observer. :-)

    OK calling iso that is only R24 @ 75F center temp R36 is a bigger exaggeration, but let's keep it real, eh?

    By my reckoning we should be callingcall 2x6 16" o.c. w/ bargain-basement-R19 batt walls R12, and non-OVE framing 2x6 24" o.c. walls R13.

    At 24" o.c. OVE/advanced framing, with a cavity fill that actually delivers R19 performance you might hit R15 (at least in theory) with a 14% framing fraction, which is still a pretty tough framing fraction to hit (and far from how most houses are actually built.)

    But if you have some data (rather than a napkin-math 2-D model using R1/inch for the framing) to show, let's have it, eh?! I'd put my napkin math model up against anybody's lipstick on mirror model, but actual data on real assemblies trumps all theory.

  18. Alex Wilson | | #18

    Whole-wall R-values
    I should have been more careful. I was looking for a quick number to use in making the point that the wood framing significantly degrades the R-value of insulation, and I went to an old article we had written on the topic--whose numbers I thought were reasonably accurate. In that July, 1994 article in Environmental Building News on thermal bridging problems with wood and steel framing, we used the parallel-path method to calculate the degraded R-value of a frame wall. According to that ancient article, with 2x6 wood framing 16" on-center the nominal R-19 FG batts drops to R-15.1, and with 2x6 framing 24" o.c. degrades to 16.1. (With steel framing those values drop R-7.1 and R-8.6, respectively, according to the same article.)

    When I went to that old article I thought that those numbers seemed high, and I should have taken the time to look for more recent data. I'm sure your numbers are more current.

    The point I was trying to make, though, is exactly the point you are making--that framing significantly degrades the performance of the insulation. So we're in agreement there.

  19. Expert Member
    Dana Dorsett | | #19

    I think we DO agree (on more than just this)...
    ...but it was fun twisting your tail on it, eh? ;-)

    With the kind of crazy bump-outs, dormers and extra corners that McMansion architects are in love with these days, it wouldn't surprise me if some code-min R19 average R10 for the entire wall surface, underperforming simple-line R13+ 5 walls by a substantial margin.

    The Building Science guys like to use R15 or R15.2 as the whole-wall estimate for 2x6 OVE framing with R19 cavity fill at a 16% framing fraction (see Assembly 1a in table 3: , and the isotherm diagrams in figure 30 ), based on 3-D modeling using THERM, but that seems optimistic to me too, if only a little bit.

    They're also calling 16" o.c. R19 batt walls with 25% framing fraction R13.7, not R12, which seems even more optimistic, but they may be using an actual R19 center cavity value (achievable with low density R22 batts) not product-labeled R19s that only measure that high when fluffed to ~6", not 5.5".

    I'd be surprised if a real assembly actually performed fully that well, but it might hit within measurement error bars on the 24" o.c. OVE version with R22 batts in the cavity.

    Any way you cut it, the thermal bridging takes a huge slice out of the cavity insulation performance, and the higher-R the insulation, the larger the nominal vs. whole-wall difference. Which is why I cringe when I've heard people brag about their R30 walls with 5" of closed cell foam in the cavities, and want to scream "Dude! That's only R15 after thermal bridging!" Saving the high-R foam budget for exterior sheathing the way Bruce Brownell does only makes sense. But cavity insulation still counts- do a good job, but only use the cheap stuff there.

  20. David_Gregory_CZ3_CA | | #20

    Mass and Moisture?
    Trying to learn more about how these houses work; any data on moisture?

    From the website ( looks like the radiator is just after the recirc fan, so perhaps that keeps the dewpoint below the surface temp of the slab-embedded ducts? Perhaps keeping temps relatively even, with an active fan, and having moderate surface area (and metal) for the heat exchange avoids condensation / mold problems (which I believe can be an issue with more 'convective' passive systems)?

    The images make me worry a little, showing a big open basement taking the warm air, though can't put too much weight on diagrams like that.

    I assume exhaust-only bathroom and kitchen venting, with other air changes coming from occupant entry/exit? So relatively low moisture loads (not sure how many occupants in that 4000sf home).

    Burning wood (as many of these houses likely do?) will help with moisture too, if not direct vented...though with all the issues that brings (including questions about the contribution of direct vs. imbedded solar energy).

  21. wjrobinson | | #21

    Actually visiting a Bruce
    Actually visiting a Bruce Brownell home is key to seeing if they put a smile on your face.

    The homes I toured were lived in for years and besides being quite comfortable with no stratified air temperatures they also were architecturally beautiful builds.

    David, Bruce's website is quite old as to some it and as to the slab system for sure.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |