“Flash and batt” is an insulation technique that combines the air-sealing superiority of spray foam insulation with the cost benefits of fiberglass batts. An inch or two of polyurethane foam seals the cavity and the batt insulation adds R-value without costing an arm and a leg.
That’s roughly the plan Dave Frank is considering for the roof of a house — presumably his own house — in Climate Zone 5. But his plan contains a twist: He wants to spray the underside of the roof deck with foam and install the batts between the joists at ceiling level.
“My reasoning for this is to (a) get HVAC in the attic within the envelope where rafters can’t be properly vented, and (b) still have insulation on top of the flat ceiling drywall because there is some ceiling radiant heat,” he writes in a Q&A post at Green Building Advisor. “And there’s the added benefit of getting some of the R-value from a cheaper product.”
Frank wonders whether the code requirement for a total of R-38 in the roof will allow the insulation to be separated in this way, as long as the R-value totals for the different layers all add up to the right number.
Or is this a case where 2 plus 2 equals 3?
That’s the topic for this Q&A Spotlight.
The code is clear: Your idea won’t work
There’s not much wiggle room here, says GBA senior editor Martin Holladay.
He cites a provision in the International Residential Code (IRC), which says the air-permeable insulation (the fiberglass batts) must be installed “directly under” the air-impermeable layer (the spray foam).
“The air impermeable insulation shall be applied in direct contact with the underside of the structural roof sheathing as specified in Table R806.5 for condensation control,” the code reads. “The air permeable layer shall be installed directly under the air-impermeable insulation.”
And so, Holladay says, if you want to meet the IRC requirement for R-38 insulation, the foam and the fiberglass batts must be in contact with each other. Of course, that shouldn’t stop Frank from adding more fiberglass in the ceiling once he’s installed batts correctly between the rafters.
Is this passage open to interpretation?
AJ Builder isn’t so sure that Holladay has it right. He points out that the code provision uses the words “direct contact” in explaining where the impermeable layer of foam has to go. But the batts can then be placed “directly under” that.
“If you stand directly under me and I drop my hammer, it will hit you,” AJ Builder writes. “So the language is open to interpretation [in my opinion]. Not saying whether the idea is good… Not against someone seeing if the idea works or not.”
Come on, replies Dana Dorsett, “even a first-year weasel lawyer would have a tough time concocting an interpretation that the air-permeable insulation didn’t need to be in VERY close proximity, even though code dos not demand a compression-fit contact with the air-impermeable foam.”
But in a broader sense, AJ Builder argues that widely held beliefs about certain building practices aren’t always accurate.
“Codes aren’t perfect,” he says, “as some above have stated. In all actuality there are many assemblies that work that one person will say not to do and another one will do and works just dandy. And then there are assemblies that are pushed here, such as putting rigid foam on OSB properly, yet I have taken apart some rotted messes built as such.”
He specifically points to the work of Bruce Brownell, a builder in New York’s Adirondack region who uses layers of foam to insulate houses at R-values that shouldn’t work in that climate zone but somehow do.
“There are so many times here that R-value of a wall is mentioned, where the R-value is actually way lower due to the fact that it is discontinuous and has a multitude of other problems,” AJ Builder says. “So R-value, folks, is very much a slippery, slippery number to [the] point of losing real value if not fully explained every time it is used…”
What’s the science behind the code requirement?
Whatever the code might read, the building science in this case is key, Dorsett explains.
“Whether it creates a problem depends a lot on how the intervening space is used (if at all), and the ratio of foam-R/fiber-R at the attic floor,” he writes. “Airtightness at the ceiling plane below the floor becomes hyper-critical if the R-ratio isn’t super robust. You can’t go with the IRC minimums for the air-impermeable layer and get away with it.
“Foam installers are fond of removing all floor insulation when foaming the roof deck (presumably to meet code), but that isn’t truly necessary in many (or even most) cases in climate zones 5 or lower. (Good luck arguing that with the code inspectors though.)”
Atticus LeBlanc would like to bore further into the science. It wasn’t that long ago, he points out, that code required vented attics. “I can’t figure out why the distance between R-12 closed-cell foam under the roof deck/rafters and sidewalls and fluffy R-38 on the ceiling would have an impact on condensation, if any,” LeBlanc writes, “and the insulation to the heated cavity would still be more than required… at least for me in Zone 3.”
Holladay responds, “If the attic were perfectly sealed — few attics are — it’s probable that the proposed (illegal) assembly would work well. But if the ‘sealed’ attic had air leaks — infiltration and exfiltration — you wouldn’t get the R-value you expect, and you would have opportunities for condensation or moisture issues.”
Our expert’s opinion
GBA Technical Director Peter Yost had this to say:
First, on the code: while the wording of the code may not be crystal clear, the intent is. The insulation layers need to be contiguous (in direct contact). Now, it is true that you can find cases where an air space to the interior of combined insulations makes sense (for example, see Joe Lstiburek’s piece on contiguous insulation schemes, energy efficiency and thermal comfort). But in this case, leaving an air space between them simply introduces a huge convective loop between the ceiling batt insulation and the rafter spray foam insulation. Indeed, 2 + 2 = 3 (or in this case, R18 + R20 â‰ R38).
And that is if the ceiling is an airtight plane, only. If it is not, and chances are it is not even close, then the big convective loop in the attic will connect with air leakage from below, and now you have a moisture problem to add to the eroded thermal performance of the attic insulation system.
This scenario also assumes that you are willing to pay the cost of spray foaming not only the rafter bays of the roof, but both gable end walls of the attic as well. Depending on the roof pitch, this added gable wall insulation could be just a bit, or quite a bit.
This noncontiguous insulation configuration does not meet the code, and for good reason; it won’t work, or at least won’t work as intended or with sufficient energy and economic efficiency.