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 much insulation is too much?

Nola_Sweats | Posted in Energy Efficiency and Durability on

I’m in Climate zone 2A, where the general code is R-38 for ceilings, but many state/local codes accept R-20 spray foam in a sealed attic as acceptable based on performance.  In my existing house (with ducts in the hot attic), I’m debating a sealed attic with R-20 spray-foam vs. R-27.  I’ve got a proposal for each, and the R-27 is about $500 more.  Both would encapsulate the rafters to help with thermal bridging, though at much less than R-20 thickness on the rafters.

I’m reading that the incremental increase from R-20 to R-27 only decreases actual heat transfer by about 2% compared to the currently uninsulated attic.  The simple payback period (avoiding complexities of time value of money) seems to be 35 years, if about half of my $120 average electric bill is heating or cooling.  I’d save $1.20 a month by reducing a/c and heat load by an extra 2%, $14 per year, $504 in 35 years.

I’m having trouble justifying the extra expense of R-27, which could go toward other improvements.  I recognize that 2″-thicker foam in the R-27 will mean the rafters will be better-insulated, resulting in less thermal bridging, but I don’t know how to quantify that.  Am I missing something?  

(And I’m aware of the potential moisture issues with open cell, but I’m going to have active conditioning of the space to keep temperature and humidity under control.)

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

    Try "none of the above". As popular as it is, half pound spray foam isn't a very good way to insulate a roof in zone 2A. Spray foam has very little thermal mass, resulting in a high thermal diffusivity. During the cooling season R20 cellulose will beat R27 open cell foam in reducing peak and average cooling loads. Compare the diffusivity of cellulose vs. any foam in Gutex's bit of marketing fluff:

    Even EPS or rigid polyurethane/polyiso runs 2x the diffusivity of cellulose, and half-pound polyurethane would be even higher due to the even lower thermal mass.

    It is reasonably safe to use R30 cellulose in an unvented assembly in that climate under a shingled roof. See the 2A-Houston-R30 row, Cellulose column of Table 3:

    If you have 2x10 rafters a full fill of cellulose brings it to R34-ish, in a fairly safe assembly. When re-roofing adding a layer of 3/4" polyiso or 2.5" of asphalted fiberboard above the structural roof deck would bring it within the IRC's prescriptives for both R-value and exterior-R for dew point control, giving it an even greater resilience edge. Alternatively, an inch of CLOSED CELL foam on the underside of the roof deck, would be enough to allow more than R30 fiber insulation below and still meet IRC prescriptives. Denim batts have roughly the same characteristics of 1.5lb density cellulose, and could be an easy enough DIY. Cellulose batts exist too. But damp sprayed cellulose is likely to come in cheaper than low diffusivity batt solutions.

    Stop thinking about it SOLELY in terms of simple payback on energy savings. Think about both resilience and comfort. A sealed conditioned attic is more resilient against hurricanes (it lowers the uplift pressures on the roof deck in high winds), and bringing most of the structural wood within the thermal and pressure boundary of the house lowers the average moisture content of the wood. Using a cellulose product that has only borate type fire retardents also reduces the wood boring ants/bees & termite risk substantially.

    So, how deep are your rafters?

    Are you up for doing any of it DIY?

  2. Nola_Sweats | | #2

    Thanks for the response. Answering, the rafters are 8" nominal, about 7" actual. I could do some DYI, but man, it is hot up there. I'd rather pay for someone else to do it. I do plan to add rigid foam outside when it's time to re-roof, but I think I'm about 10 years away from that. Yes, comfort and hurricane-resistance are big factors in the decision to seal the attic -- bigger than money.

    Air-sealing would be the issue with anything other than foam. There's no evidence of much effort to air-seal when the house was built -- I can see a sliver of light at the bottom of many rafter bays (no soffit vents, just random gaps between roof decking and fascia). The attic floor has at least 6 penetrations per bed room (LED can lights, ceiling fans, a/c registers) and the attic floor is decked over with plywood, meaning I can't caulk from the attic side. Probably can't reach those slivers of light at the fascia, either.

    A flash-and-batt with a thin layer of closed-cell foam as a base might do the air-sealing, but would roughly double the cost, according to the installer. It would require a second visit with a different truck, and the logistics of getting the hose into my house are complex (no street access, no driveway, requires a few hundred feet of hose to get from nearest street to my house).

    The guy making the recommendation is also an energy auditor (he didn't audit my house) and his company does install spray cellulose, so I don't think he's recommending it just for the $. I wanted closed cell, and he recommended open cell for its better air sealing because of greater expansion.

    Oh, your first link doesn't seem to link to the correct article for thermal diffusion. I would like to read about the subject. Thanks again!!

  3. MattJF | | #3

    I don’t know a ton about warm climate roof design, but I have not heard anyone state open cell having better air sealing than closed cell. It actually has so much expansion that it sometimes bridges over features before they can spray into tight sections leaving voids. My installer actually prefers .75lb foam over .5lb as it is a little more predictable.

  4. Jon_R | | #4

    >How much insulation is too much?

    Too much to pay back while you live there? Too much to pay back during the life of the building? Too much to have a cost effective environmental impact? Too much to have a positive environmental impact? All of these will result in different answers.

    Do comply with International Residential Code (R806.5 - which doesn't allow only air permeable insulation unless you use a diffusion port).

  5. Expert Member
    Dana Dorsett | | #5

    >"Oh, your first link doesn't seem to link to the correct article for thermal diffusion. I would like to read about the subject."

    OOPS! :-) ( I need an editor, or at least another cuppa coffee... )

    Try this:

    >"...the rafters are 8" nominal, about 7" actual. "

    R30 rock wool is designed to fit into 2x8 rafters, and is comparable in most ways to the 1.8lbs fiberglass simulated in that BSC documented. Take look at the other columns in Table 3. With asphalt shingles it's not too risky, and that risk could be further mitigated with a broadsheet smart vapor retarder on the interior side. That would make it comparable to the "Kraft faced fiberglass batt" column, which appears to be very low risk in the simulation. But the diffusivity rock wool is still high, probably about the same as half-pound polyurethane foam, but (I haven't seen side by side comparisons of polyurethane foam at different densities.) The Gutex advert's diffusivity bar chart with rigid PU foam is probably at least 1lb density, but that is not stated explicitly.

    Air sealing the obvious leaks with polyurethane caulk (or can foam for bigger gaps), and air sealing + rock wool cheaper than going with a full fill of 0.75lb open cell foam (which also comes out to about R30 @ 7.25") or half pound foam ( R27-ish).

    The differences in air sealing quality between 3" or more of open cell foam and 2" or more of closed cell is pretty academic- they're both very good sealants.

    Hurricane resilience goes up when 2" or more of closed cell is used, since it essentially glues the rafters to the roof decking, and stiffens the roof deck, creating a wood + foam monocoque structure. That doesn't happen with open cell foam. Mind you 7.25" of open cell foam uses slightly less polymer than 2" of close cell- it's still not the absolute greenest solution from an insulation perspective, but it's arguable that the value-add of the structural aspect may be "worth it" in hurricane zones.

    A shot 2" of closed cell foam (R14, if HFO blown) leaves enough space in a 2x8 rafter bay for R19 denim batt or R20 celluose batt, (for the favorable thermal diffusivity) which is not quite R38, but north of R30.

    >"I could do some DYI, but man, it is hot up there. I'd rather pay for someone else to do it."

    I'm sure it's pretty hot up there right now, with just R1 of roof deck between the attic and hot shingles, but after it's air-sealed and insulated with R12+ on the roof deck the attic temps will moderate considerably, making a DIY on the batts less daunting. Start with the sun-baked south or west facing rafter bays first and it'll be even more comfortable for doing the rest.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |