Flash-and-Batt Insulation

musingsheader image

Flash-and-Batt Insulation

Spray foam between the studs can reduce air leaks, but a continuous layer of exterior rigid foam makes more sense than flash-and-batt

Posted on Sep 11 2015 by Martin Holladay

Closed-cell spray polyurethane foam insulation has several desirable characteristics. It’s an excellent air barrierBuilding assembly components that work as a system to restrict air flow through the building envelope. Air barriers may or may not act as a vapor barrier. The air barrier can be on the exterior, the interior of the assembly, or both., an excellent vapor retarder, and it has a high insulating value per inch (about R-6). Unfortunately, it’s also expensive.

Insulation contractors who want to obtain at least some of the desirable performance characteristics of closed-cell spray foam at a lower cost than a full spray-foam job sometimes promote the “flash-and-batt” approach. This involves spraying one or more inches of closed-cell foam against the interior side of the sheathingMaterial, usually plywood or oriented strand board (OSB), but sometimes wooden boards, installed on the exterior of wall studs, rafters, or roof trusses; siding or roofing installed on the sheathing—sometimes over strapping to create a rainscreen. and filling the rest of the framing bay with a fiberglass batt.

A variation of this method is called “flash-and-fill” or “flash-and-blow”; these terms refer to an insulation job that combines closed-cell spray foam with a blown-in insulation material like cellulose or blown-in fiberglass.

The flash-and-batt approach can be used for walls, floors, or ceilings.

Some insulation contractors wonder whether they should compress fiberglass batts that are installed in flash-and-batt cavities. The answer is simple: Compressing fiberglass batts is a good thing to do, since (a) compressed batts have a higher R-valueMeasure of resistance to heat flow; the higher the R-value, the lower the heat loss. The inverse of U-factor. per inch than uncompressed batts, and (b) compressed batts are more likely to be in direct contact with the cured spray foam — and that type of direct contact is a code requirement for flash-and-batt roof assemblies.

A history of the flash-and-batt method

My first article on the flash-and-batt method was published in the November 2007 issue of Energy Design Update. At that time, the flash-and-batt method was still a little-used approach. I contacted a range of insulation manufacturers — both spray foam manufacturers and fiberglass insulation manufactures — and none of these manufacturers were willing to endorse the flash-and-batt approach.

Even though flash-and-batt wasn’t widely known in 2007, a few spray foam contractors claimed to have been using the method for a long time. Matt Momper, president of Momper Insulation in Fort Wayne, Indiana, told me, “It’s a foolproof system, and we’ve done it for over 35 years.”

Most spray foam contractors use rules of thumb when implementing the flash-and-batt method. Back in 2007, Momper told me confidently, “A half inch [of spray foam] is all you really need.” These days, however, most experts recommend that a successful flash-and-batt job requires a spray foam layer that is at least 1 inch thick. (The reason: a layer of spray foam that is less than an inch thick doesn’t provide a dependable air barrier.)

The first North American insulation manufacturer to publish a document endorsing the flash-and-batt approach was Johns Manville. In 2008, the company recommended the use of “flash-and-Spider” between rafters to create conditioned attics. (The recommendation was made in Johns Manville Technical Bulletin T07-016.)

Code requirements for flash-and-batt walls

Back in the rule-of-thumb days, before any building codes acknowledged the flash-and-batt approach, some, but not all, flash-and-batt installers realized that the spray foam layer needs to be thicker in a cold climate than it does in a warm climate. (If the spray foam layer is too thin, this type of wall can accumulate moisture over the winter. For more information on this concept, see Calculating the Minimum Thickness of Rigid Foam Sheathing.)

Now that building codes specify requirements for flash-and-batt jobs, contractors can throw away their rules of thumb.

The 2012 International Residential Code (IRCInternational Residential Code. The one- and two-family dwelling model building code copyrighted by the International Code Council. The IRC is meant to be a stand-alone code compatible with the three national building codes—the Building Officials and Code Administrators (BOCA) National code, the Southern Building Code Congress International (SBCCI) code and the International Conference of Building Officials (ICBO) code.) requirements for flash-and-batt insulationInsulation, usually of fiberglass or mineral wool and often faced with paper, typically installed between studs in walls and between joists in ceiling cavities. Correct installation is crucial to performance. in walls can be found in Footnote “a” to Table R702.7.1. The footnote reads, “Spray foam with a minimum density of 2 lb/ft3 applied to the interior cavity side of wood structural panels, fiberboard, insulating sheathing or gypsum is deemed to meet the insulating sheathing requirement where the spray foam R-value meets or exceeds the specified insulating sheathing R-value.”

The phrase “insulating sheathing requirement” is a reference to requirements for minimum R-values for rigid foam insulation installed on the exterior side of wall sheathing. This footnote informs builders who use the flash-and-batt approach that spray foam “is deemed to meet” the requirements established in Table R702.7.1 for exterior rigid foam on walls.

In other words, this footnote states that flash-and-batt jobs (a) must use closed-cell spray foam, not open-cell spray foam, and (b) need a spray foam layer that meets the minimum R-values for insulating sheathing shown in Table R702.7.1.

Assuming that closed-cell spray foam has an R-value of R-6 per inch, and assuming that you don’t want to install less than 1 inch of spray foam for a flash-and-batt job, here are the minimum thickness requirements for the spray foam layer when following the flash-and-batt approach, according to the IRC:

  • Marine zone 4: 1 inch for 2x4 or 2x6 walls;
  • Zone 5: 1 inch for 2x4 walls, 1 1/2 inch for 2x6 walls;
  • Zone 6: 1 1/2 inch for 2x4 walls; 2 inches for 2x6 walls;
  • Zones 7 and 8: 2 inches for 2x4 walls; 2 1/2 inches for 2x6 walls.
    • Although the fiberglass batts in a flash-and-batt stud bay will be thinner than the fiberglass batts in a wall with exterior foam sheathing, thinner batts move the wall in the direction of more safety rather than more risk, since thinner fiberglass keeps the interior surface of the cured foam warmer (and therefore less likely to collect condensation).

      Because the layer of fluffy insulation in a flash-and-batt job is thinner than the layer of fluffy insulation in a wall with exterior rigid foam, a contractor might choose to contest the requirements of Footnote “a,” arguing instead for a thinner spray foam layer. (Such an argument would alter the required R-values listed above, using the percentages shown in the table reproduced as Image #2, below.) The argument has merit, but any deviation from Footnote “a” requirements are best made with the approval of your local code enforcement official.

      Code requirements for flash-and-batt roofs

      Code requirements for flash-and-batt roofs can be found in section R806.5 of the 2012 IRC. That section declares that “Unvented attic assemblies (spaces between the top-story ceiling joists and the roof rafters) and unvented enclosed rafter assemblies (spaces between ceilings that are applied directly to the underside of roof framing members/rafters and the structural roof sheathing at the top of the roof framing members/rafters) shall be permitted” as long as a number of conditions are met.

      The code requires the spray foam layer in flash-and-batt roof assemblies to be “applied in direct contact with the underside of the structural roof sheathing.” Moreover, the code states that “The air-permeable insulation [for example, fiberglass batts or cellulose insulationThermal insulation made from recycled newspaper or other wastepaper; often treated with borates for fire and insect protection.] shall be installed directly under the air-impermeable insulation [that is, the spray foam].”

      Finally, the code requires that the spray foam layer meet the requirements “specified in Table R806.4 for condensation control.” The code-mandated minimum R-value requirements for the spray foam layer are reproduced below.

      • R-5 foam for Climate Zones 1-3,
      • R-10 for Climate Zone 4C,
      • R-15 for Climate Zones 4A and 4B,
      • R-20 for Climate Zone 5,
      • R-25 for Climate Zone 6,
      • R-30 for Climate Zone 7, and
      • R-35 for Climate Zone 8.

      Use closed-cell spray foam, not open-cell spray foam

      Most experts, including experts at the Building Science Corporation, recommend that contractors use closed-cell spray foam (foam with a density of 2 pounds per cubic foot), not open-cell spray foam, for a flash-and-batt job — especially for walls. There are several reasons for this recommendation:

      • When installed as a thin layer, closed-cell spray foam is a more dependable air barrier than open-cell foam;
      • The fact that closed-cell spray foam is less vapor-permeable than open-cell spray foam tends to make assemblies with closed-cell foam more robust and less prone to moisture accumulation than assemblies with open-cell foam; and
      • As noted above, the IRC requires that flash-and-batt insulation installed between wall studs use spray foam insulation with “a minimum density of 2 lb/ft3.” That requirement excludes the use of open-cell spray foam for flash-and-batt walls.

      Flash-and-batt walls are better than ordinary fiberglass-insulated walls

      Flash-and-batt fans list the following advantages to the flash-and-batt approach:

      • This insulation method is easy for contractors who are set in their ways, especially those who don’t want to learn how to install rigid foam insulation on the exterior side of sheathing.
      • It’s a good method for hot climates, since the spray foam layer interrupts inward solar vapor drive. (The same advantage can be gained by installing exterior rigid foam, of course.)
      • It’s a good method for creating an unvented conditioned attic. Most building codes require that spray polyurethane foam be covered with an approved ignition barrier, and many types of fibrous insulation, including blown-in fiberglass and mineral wool insulation, qualify as an ignition barrier.
      • Flash-and-batt walls usually have lower levels of infiltration and exfiltrationAirflow outward through a wall or building envelope; the opposite of infiltration. than walls that are insulated with fiberglass batts alone.
      • Substituting 2 inches of closed-cell spray foam for 2 inches of fiberglass batts adds about R-5 to a wall’s center-of-cavity R-value.

      Flash-and-batt approach doesn’t address thermal bridging

      Flash-and-batt skeptics list the following disadvantages to the flash-and-batt approach:

      • Closed-cell spray foam is relatively expensive.
      • Closed-cell spray foam is produced with an blowing agent that has a high global warming potential; that’s why most green builders try to avoid or minimize their use of closed-cell spray polyurethane foam.
      • Although a flash-and-batt job can address some air leaks, it can’t get them all. Many flash-and-batt walls still have significant leaks — for example, between the bottom plate and the subfloor, and between doubled studs and doubled top plates.
      • The flash-and-batt approach can be risky in very cold climates, unless the thickness of the foam layer is carefully monitored. If the foam layer is too thin, condensation can form on the interior surface of the cured foam, especially in late winter.
      • The flash-and-batt approach doesn’t address thermal bridging through studs or rafters. While the nominal R-value of a flash-and-batt 2x6 wall with 2 inches of spray foam is R-25, the actual whole-wall R-value (after taking thermal bridging into account) is only R-17. Because of the thermal bridging problem, many insulation experts say that the flash-and-batt approach is a waste of good foam. In most cases, builders will get a bigger bang for their buck by installing a continuous layer of exterior rigid foam than by installing a thin layer of spray foam between the studs.

      Pay attention to details

      There are several possible ways to mess up a flash-and batt job. Remember:

      • Don’t install interior polyethylene on a flash-and-batt job. Like wall and roof assemblies with exterior rigid foam, flash-and-batt assemblies are designed to dry to the interior.
      • Use caulk or gaskets to seal air leaks that aren’t addressed by the spray foam layer — for example, the gap between the bottom plate and the subfloor, the gaps between doubled framing members, and the gaps between window frames and window rough openings.
      • Once you install a layer of closed-cell spray foam on the interior of your wall sheathing, you shouldn’t install a layer of rigid foam on the exterior side of the sheathing (since doing so would trap the sheathing between two low-permeance layers). If you want a layer of foam, choose just one approach — either interior spray foam (flash-and-batt) or exterior rigid foam. Don’t try to mix the two approaches.
      • Pay close attention to minimum thickness requirements for the spray foam layer, especially in Climate Zones 6, 7, and 8. The colder the climate zone, the greater the risk arising from a spray foam layer that is too thin.

      Bill Abney, a spray foam contractor in Helena, Montana, knows from experience that thin foam is riskier than thick foam. In 2007, he told me, “Starting about 30 years ago, we were doing 1 inch of foam and then a 6-inch batt. Then we did a job for a homeowner, and a while later, in the spring, he called and showed us where there was about a shot-glass of water running out the bottom of the south wall and the bottom of the north wall. Once we saw that, we said we aren’t doing that anymore — we’re learning. We started doing 2 inches of foam with an R-11 batt over it. Then about eight years ago, at the home of one of my employees, we did the wall that way — 2 inches of foam and an R-11 batt. He was living in the house before the sheetrock was installed, in the winter. And there was a period of cold weather when it was getting down to 40 below, and he came home from work, and he pulled the fiberglass back to look at the wall, and he saw a layer of frost on the Corbond [spray foam]. And I know that it wasn’t a wet house — there wasn’t a lot of moisture on the windows. But the fiberglass was cooling the surface of the foam enough for frost to form. So now we just install three inches of foam and no fiberglass.”

      More minuses than pluses

      A Building Science Corporation article on the flash-and-batt approach sums up the pluses and minuses of flash-and-batt this way: “Using spray foam insulation can be costly, and while it reduces the risks of moisture related durability issues, the minimal increase in R-value due to the thermal bridging may not be worth the increased cost of the spray foam insulation.”

      In almost all cases, installing a continuous layer of rigid foam insulation on the exterior side of the sheathing makes more sense, and provides a better-performing assembly, than the flash-and-batt approach.

      Martin Holladay’s previous blog: “Cold-Weather Performance of Polyisocyanurate.”

      Click here to follow Martin Holladay on Twitter.

Tags: , , ,

Image Credits:

  1. Image #1: Pacific Northwest National Laboratory
  2. Image #2: Martin Holladay
  3. Image #3: Fine Homebuilding

Sep 12, 2015 1:45 AM ET

Mineral wool on the exterior?
by Christopher Welles

I'm hoping to do flash-and-batt with a couple of inches of ComfortBoard on the exterior. This is in Northern NJ, zone 5. My current ideal wall stack would be:

- Mineral wool or fiberglass batts.
- 2" of closed-cell spray foam.
- OSB/Plywood sheathing with WRB (Zip?) - the air seal.
- 2" exterior continuous mineral wool.

The idea would be to use the exterior of the sheathing as the air barrier. I'd like the closed-cell spray foam to provide a second line of defense there, but I'm unclear as to how you could meaningfully test that in a real-world build process. I would think you'd want to rush be seal up the outside first to keep up the moisture, and then you'd lose the ability to uncover any gaps in the CC air seal.

Keeping the exterior mineral wool down to 2" should simply a lot of the installation pain points involved in using it on the exterior, while still providing the key benefits. Also, the property taxes are kind of insane here, so wall thickness has an impact on long-term costs outside of what one might consider rational and sane.

I find it depressing to discover that no one is currently doing closed-cell spray foam with the more friendly HFO1234yf blowing-agent. (thanks for that update) I'd love to hear about it as soon as someone is.

I know the idea of a "foam sandwich" is extremely unpopular, and I'm not really such a fan of exterior foam myself, but I'm not really convinced it's so bad as long as you provide an appropriate gap. I know it's been heavily discussed, but I'll point out again from the BSC's Mind-the-Gap: "Unless you provide a small gap between the exterior face of the OSB and the back surface of the foam sheathing to provide for some hygric redistribution." Huberwood provides also provide a detail sheet utilizing this advice "Zip system wall detail w/ closed cell insulation & exterior foam". I'm not sure why it just gets boiled down here to "Don’t try to mix the two approaches".

Note: I've found the following article BSC report helpful in pointing out various quality-control risks regarding in relation flash-and-batt air sealing - "Hybrid Wall Construction and Quality Control Issues in Wyandotte" - http://www.nrel.gov/docs/fy14osti/58712.pdf

In case it's not completely obvious, I will point out that my experience in this area is extremely limited. I've managed a lot of large projects building software, but none building a building. I'm relying on family, friends, and sites like this for everything I know in this space. Thank you for all your insight.

Sep 12, 2015 5:09 AM ET

Response to Christopher Welles
by Martin Holladay

One thing I've learned over the years is that there are many, many ways to detail high-R walls, and every designer and builder has a different favorite approach. If you like the wall you have chosen, from a cost perspective, a buildability perspective, and a performance perspective, then you should go ahead and build it. It certainly works from a moisture-management perspective.

There may be some green builders reading your comment who will note that the environmental downsides to closed-cell spray foam (the blowing agents with a high global warming perspective) are serious enough to give your approach a thumbs down. After all, you say that you will be taping the seams of your wall sheathing to create an air barrier. So you might as well just fill your stud bays with cellulose -- you aren't getting much benefit from the spray foam, and you have a continuous layer of exterior insulation to address thermal bridging.

But, as I said earlier, everyone has their own opinion and their own favorite approach. If you like your system, use it.

Sep 12, 2015 11:38 AM ET

Edited Sep 12, 2015 12:15 PM ET.

Flat roof assembly
by Lowell Lo

Mr. Martin Holladay,

Thank you for all your green technical advice. I am an Architect in Toronto, Canada. I just submitted a design for a flat roof assembly for permit. I don't have that much experience in flat roofs but read your articles on the website.

Because the building is attached to other buildings on both sides so I think I need parapets that extend above the roof. So I thought venting it would be difficult. The size of the roof area is about 37' by 16'.

Here is the assembly from top down:

-Mechanically fastened mod-bit (torch down) roofing on 1/2" plywood on 5" XPS extruded poly insulation on #15 asphalt felt on 3/8" plywood on 2"x12" spruce roof joists 24" O.C. filled underside of plywood with 2" of closed cell with membrain smart vapour barrier on 1/2" plywood on 2"x3" chase space for recessed lighting on 1/2" drywall

My questions are:
1. Is mod-bit roof the right type of roofing to use, I have one interior drain in the middle of the roof area, can that work with an interior drain?
2. Because of the 2"x12" roof joists, I thought it would be difficult to fill the cavity the entire space with batt so I thought I would use 2" closed cell spray foam instead under the rigid insulation and have the void of the joists empty, is that not a good idea?
3. Do I need the 2"x3" chase space for lighting to avoid punching holes in the vapour barrier as it will be required in Toronto?

Thanks for your help again,


Sep 12, 2015 2:16 PM ET

Response to Lowell Lo
by Martin Holladay

Your roof assembly includes several vapor barriers, including the roofing membrane, the 5 inches of XPS, and the 2 inches of closed-cell spray foam. So don't waste your money on MemBrain. Your assembly is so vapor-tight that the idea of adding a "smart" vapor retarder that can open up and be vapor-permeable is a total waste of money.

I don't know much about modified bitumen roofing, but I can't think of any reason why you couldn't use it with a roof drain.

If you want to fill your 2x12s with fiberglass batts instead of installing 2 inches of closed-cell spray foam, you can. If you install 11.25 inches of fiberglass, the R-value of the fiberglass is about R-40. (Obviously, that's more than the R-12 you get from the spray foam.) You've got R-25 of rigid foam on top, so a total of R-65 -- making the rigid foam layer about 38% of the total R-value, which should work in Climate Zone 6.

I'm not sure I understand you last question. You don't need another vapor barrier, since your roof already has 3 vapor barriers. Whatever you do, don't put in any recessed lights that interfere with your insulation.

Sep 12, 2015 3:20 PM ET

Thanks for the reply.
by Lowell Lo

Thanks for the reply and advice again Martin.


Sep 13, 2015 10:11 AM ET

Compressed fiberglass batts?
by Lucyna de Barbaro

Hello, do I read correctly in this article that the compressed fiberglass batts have higher (?) R-value than uncompressed one? We've just finished correcting a bunch of fiberglass batts installations because we were told by Energy Star /Resnet raters that they were too big/compressed and would not perform correctly. Similar statements are made in the https://www.energystar.gov/ia/partners/bldrs_lenders_raters/downloads/En...

Sep 13, 2015 10:23 AM ET

Edited Sep 13, 2015 11:09 AM ET.

Response to Lucyna de Barbaro
by Martin Holladay

Compressing a fiberglass batt reduces the R-value of the batt but increases the R-value per inch of insulation.

If you are installing a 3.5-inch-thick batt in a 3.5-inch-deep cavity, you don't want to compress it. You want to fluff it up so that it achieves its full, labeled R-value.

However, if you decide instead to install a 6-inch-thick batt in a 3.5-inch-deep cavity, you will achieve a higher R-value per inch of insulation.


Compressing fiberglass.JPG

Sep 13, 2015 10:37 AM ET

Yes, that's right
by Dana Dorsett

As the density of the fiber insulation increases, so does it's R per unit of thickness (up to a point.) An R19 batt at ~6.25" thickness is R19, or ~R3 per inch. Compress it to 5.5" (in a 2x6 framing cavity) it's total performance drops to R18, but that is an INCREASE in R/inch to about R3.27/inch.

Compressing it even further to 3.5" (a 2x4 framing cavity) it drops to R13, but that's an even higher R3.7/inch. And...

Compressing it into 2.5" (a 2x3 framing cavity) it performs at R10, which is R4/inch. The density is more than twice that of it's manufactured & tested loft of 6", but it's R/inch has improved by 33%.

Cramming it into 1.5" (a 2x2 framing cavity) it's performance is about R6.6, or R4.4/inch, at about 4x the density of it's manufactured loft, but that's about the limit of what you can get out of fiberglass per inch.

Study the manufacturers' compression chart values carefully, and look up the weight per square foot of product and the thickness at which it was tested. You'll find that an R19 and an R13 weigh the same per square foot, but the lower loft of the R13 means it's higher density.


Whether the RESNET raters correctly assessed the situation or not is impossible to say without more detail, but if you were compressing an R19 into a 2x6 flash & batt and counting the R of the batt at it's full loft, that would have been an error, since it isn't R19 even when installed in a 2x6 bay without the flash-foam thickness. But whatever thickness it is, it's possible to interpolate the charts to estimate it's true performance.

Sep 13, 2015 11:48 AM ET

by Malcolm Taylor

The first thing you need to do is read the two relevant sections of your building code. Requires all roofs to be vented "unless it can be shown to be unnecessary". The appendix cites "specialized roof assemblies... used on factory-built buildings". So you are going to need to develop a case for not venting if that is the route you want to go down.

9.26.11 Has all the requirements for flat built-p roofs. You need to at least incorporate some of them into your specifications.

More generally, my advice would be that if you are unfamiliar with flat roof construction you should employ a roofing consultant to develop the details and specifications, rather than relying on the advice we are giving you here.

Sep 13, 2015 3:11 PM ET

open cell?
by Charlie Sullivan

I was expecting Dana to comment on the choice of closed cell vs. open cell, and say that open cell is actually better at air sealing, and thus a good choice for this application.

On Martin's three reasons, the first is that closed-cell is better for air sealing in thin layers. Is that the key point--open cell does a better job of air sealing in thick layers, but closed cell does better in thin layers? Or is that still an open question?

As far as vapor permeability, I would think that open-cell's permeability would actually be helpful in allowing some drying to the exterior in the winter, keeping the cavity dew point a little lower than the inside air dew point.

But regardless of those points, if the local interpretation of the code is that open-cell can't be used in that application, it can't be used.

Sep 14, 2015 6:06 AM ET

Edited Sep 14, 2015 9:35 AM ET.

Response to Charlie Sullivan
by Martin Holladay

This topic (the rate of air leakage through closed-cell spray foam and open-cell spray foam) is the topic of an upcoming blog; stay tuned.

The short version of the story is that you need a minimum of one inch of closed-cell spray foam to limit air leakage enough to meet the definition of an air barrier material -- that is, leakage of no more than 0.02 l/sec-m² @75 Pa (0.004 cfm/sf @ 1.57 psf).

Most types of open-cell spray foam can't comply with that standard (and be considered an air barrier) unless they are installed at a thickness of 5.5 inches or more. A few manufacturers of open-cell spray foam claim that their products form an air barrier when installed at only 3.5 inches. Even still, these products aren't airtight enough to be considered for use on flash-and-batt jobs (where the spray foam is generally installed at a thickness ranging from 1 to 2 inches).

Sep 14, 2015 12:31 PM ET

Edited Sep 14, 2015 12:41 PM ET.

It's the vapor retardency, not the air sealing quality.
by Dana Dorsett

At only an inch or two open cell foam is more vapor-open than some housewraps, and even with a perfectly air-tight but ~5 perm painted interior the sheathing is going to gain quite a bit of moisture over a winter in a zone-4A/B & higher climate (zone 4C is more complicated.)

But with closed cell foam even 1" is closing-in on class-II vapor retardency at 1.5 perms max, dramatically reducing the moisture uptake, and at 2" it's well under the 1-perm margin.

At an equal R-value (rather than thickness) of R13, open cell foam appears to have a more consistent air tightness across outdoor temperature than closed cell foam, has a better overall air-tightness average according to the experimental data presented by the Building Science Corp, that Martin covered in this earlier blog:


Note how the infiltration & exfiltration numbers barely move between the 42C/108F data and the -18C/0F data for open cell, whereas with closed cell it nearly doubles:


But both are fairly good at air sealing, if not a perfect solution. The assemblies tested had no caulking under bottom plates, or between doubled-up top plates, places where the foam doesn't reach. But in those conditions (common in retrofits, and in a lot of new construction) here is a very real net performance difference. Looking at the average degradation of performance with imposed air flow, the R13 open cell is substantially less than with the R13 closed cell:


But R13 open cell foam is still on the order of 10 perms the very high end of class-III vapor retardency (about 2x as vapor open as interior latex paint), whereas R13 closed cell is on the order of 0.5-0.7 perms, a class-II vapor retarder. Ten perms is way too high to be of any use in limiting the moisture uptake into the sheathing. Even if the condensing surface (the foam/fiber interface) is enough for dew point control the high vapor permeance of open cell foam makes that fact pretty much an academic footnote. Depending on the climate zone and the construction details of the assembly it could still need an interior side vapor retarder tighter than 5 perms, even though no condensation occurs at the foam/fiber boundary.

Roof decks don't dry toward the exterior in most assemblies, and walls without rainscreens don't either. OSB or plywood are on the order of 1-perm when dry, and not great for flow-through vapor diffusion drying- they're the problem! A fiberboard sheathed assembly with an exterior vented gap might do just fine with an open-cell flash & batt even in a cold/very cold climate but there are a lot of particulars to consider before using open cell foam that way.

Sep 14, 2015 12:41 PM ET

Edited Sep 14, 2015 12:44 PM ET.

Response to Dana Dorsett
by Martin Holladay

I'm writing an article on this topic, and I'd prefer to reserve the discussion until my article is published.

Briefly, BSC's Thermal Metric testing you refer to involved stud cavities filled with open-cell spray foam. That's 3.5 inches of open-cell spray foam.

Some, but not all, open-cell spray foams qualify as an air barrier material at 3.5 inches. The results of ASTM E283 and ASTM E2178 test results for any brand of spray polyurethane foam can be found by looking up the foam's Evaluation Service Report (ESR). The minimum thickness of open-cell spray foam needed to qualify as an air barrier material ranges from 3.5 inches to 5.5 inches. No open-cell spray foam qualifies as an air barrier at 1 inch or 2 inches.

For closed-cell spray foams, the minimum thickness is typically 1.4 or 1.5 inch.

Sep 14, 2015 12:50 PM ET

I'll be very curious to see that blog!
by Dana Dorsett

There are lots of moving parts to the story, and I'm happy to wait to read what you come up with!

A financial side note on the prior BSC testing: The installed price of 3.5" of ocSPF is typically 25-35% cheaper than 2" of ccSPF.

Sep 14, 2015 2:49 PM ET

by Charlie Sullivan

I thought that would an interesting discussion--it already has been and I look forward to reading more.

Sep 15, 2015 3:51 PM ET

Malcolm Taylor
by Lowell Lo

Thanks Malcolm for your advice.

Lowell Lo
Architect Toronto

Sep 16, 2015 11:57 AM ET

Flash and Batt in Regard to Vapor Barrier
by Bruce Rachel

Would you consider the flash and batt concept in Zone 1, 2 or 3? The issue being the vapor permeability of the closed cell insulation.

Sep 16, 2015 12:05 PM ET

Response to Bruce Rachel
by Martin Holladay

Yes, the flash and batt method can be used in hot climates. That's why I wrote, "It’s a good method for hot climates, since the spray foam layer interrupts inward solar vapor drive."

Sep 16, 2015 1:46 PM ET

More on vapor barriers
by Martin Holladay

You seem concerned that the closed-cell spray foam layer in a flash-and-batt wall creates a vapor barrier. While 1 to 2 inches of closed-cell spray foam has a vapor permeance of 0.8 to 2.0 perms -- meaning that it's either a vapor barrier or almost a vapor barrier -- it won't cause any problems in hot climates. Everything on the exterior side of the spray foam layer can dry to the exterior, and everything on the interior side of the spray foam layer can dry to the interior. The surfaces of the cured spray foam will never get cold enough in Climate Zones 1, 2, or 3 to become a condensing surface.

Sep 16, 2015 2:36 PM ET

The sheathing IS a vapor retarder!
by Dana Dorsett

A closed cell foam is between 0.8 and 1.5 perms @ 1",and 0.4-0.8 perms @ 2".

OSB or CDX is between 0.5 & 1 perm @ 1/2" (unless the average relative humidity of proximate air is high over a very sustained period of time weeks, raising the moisture content of the wood). The sheathing itself is as much of a vapor retarder as an inch or more of closed cell foam in a cooling dominated climate. A rainscreen gap or other back-vented siding (eg vinyl) keeps the sheathing from becoming vapor open, but a stucco or brick-clad assembly may still have fairly permeable sheathing if the cavity isn't properly vented, due to the high moisture retention of those types of siding, keeping the RH of the air in the siding gap high.

With properly vented siding the effect of the additional redardency of the foam may be measurable in a lab, but from a practical and latent cooling load point of view it doesn't matter NEARLY as much as the air-tightness of the assembly in a cooling dominated climate.

Paraphrasing a 1990s internal Clinton campaign slogan, "It's the air leakage, stupid!" It only takes but a miniscule amount of air leakage to move as much outdoor humidity into the assembly to equal the vapor diffusion through a full sheet of half-inch OSB. The cable guy's air leaks introduce a whole wall's vapor-diffusion worth in a typical installation.

Sep 16, 2015 2:53 PM ET

by Tim C

Martin, what's your take on flash and batt's "sibling", flash and loose fill, for unconditioned attic floors?

Sep 16, 2015 3:16 PM ET

Response to Tim C
by Martin Holladay

You are asking about the use of a thin layer of closed-cell spray foam against the ceiling drywall -- sprayed from the attic side -- before installing a thick layer of cellulose or loose-fill fiberglass. Right?

I certainly agree that every ceiling needs to be air sealed before cellulose or fiberglass insulation is installed above the ceiling. I described the necessary steps in my article on the topic, Air Sealing an Attic.

However, it's almost always unnecessary (and unnecessarily expensive) to install a layer of closed-cell spray foam over the entire surface of the drywall ceiling in order to perform air sealing. It makes more sense to think about where the air leaks are -- for example, at the gaps between partition top plates and partition drywall; at plumbing vent penetrations; and at electrical penetrations -- and to address those air leaks with an appropriate sealant (two-component spray foam, one-component spray foam, tape, or caulk).

One other reason why you don't need to install a layer of closed-cell spray foam over the entire ceiling: closed-cell spray foam is manufactured with a blowing agent that has a high global warming potential.

Sep 16, 2015 3:41 PM ET

More on Vapor Barriers Response to Martin
by Bruce Rachel

Thank you, Martin. I have always been told than in zones 1-3 we want the vapor to go all the way thru the assembly. That is why I wanted to confirm.

What I really want to build in Zone 3 is this wall assembly:
The issue that has been confusing me is that all of the structural insulation sheathings I have found have a 0.X perm rating. So they are a vapor barrier.

Sep 16, 2015 3:42 PM ET

flash and batt
by Anders Lewendal

Martin: What about installing 1 1/2" of closed cell foam on the outside of the studs and spraying 2" of closed cell foam in the cavities against the exterior foam and then blow in 3 1/2" of cellulose to finish the insulation? Interior shear walls of either plywood or 5/8" drywall will substitute for not having exterior sheathing. Yes, an engineer is required. I have done it a few times with good success.

Sep 16, 2015 4:17 PM ET

Response to Bruce Rachel
by Martin Holladay

In Climate Zone 3, you can use any thickness of rigid foam you want on the exterior side of your studs. You can use 1/2-inch rigid foam or 3-inch rigid foam -- it all works. There is no condensation worry in Climate Zone 3, because your winters never get cold enough for you to worry about this issue.

If you end up with an exterior vapor barrier created by installing exterior rigid foam -- all the better. The rigid foam stops inward solar vapor drive, and that's a good thing.

For more information on these issues, see Calculating the Minimum Thickness of Rigid Foam Sheathing.

Sep 16, 2015 4:21 PM ET

Edited Sep 16, 2015 4:23 PM ET.

Response to Anders Lewendal
by Martin Holladay

Q. "What about installing 1 1/2 inch of closed cell foam [I think you mean rigid foam] on the outside of the studs and spraying 2 inches of closed-cell foam in the cavities against the exterior foam and then blow in 3 1/2 inches of cellulose to finish the insulation?"

A. From a building science perspective, that works just fine. From a financial perspective, it's a little expensive (because of the cost of the spray foam). From a green building perspective, it gets the usual thumbs down, because of the fact that the blowing agents used to make closed-cell spray foam have a high global warming potential.

I'd prefer to see you develop a way to use the exterior rigid foam as an air barrier -- for example, by taping the foam seams. That way you could skip the spray foam.

Feb 29, 2016 7:49 PM ET

Flash and Batt Air Barrier Requirements
by Robby Schwarz

Martin, We believe that fiberglass insulation needs to be installed full encapsulated with an air barrier on all six sides to meet manufacturer's installation requirements and to reduce convective loops within fiberglass assembly. I would like to get your opinion on if flash and batt meet the IECC air barrier requirement and installation requirements in attic knee wall applications or chase wall applications where drywall or other air barrier will not be installed on the fiberglass side of the assembly.

Feb 29, 2016 7:58 PM ET

Edited Feb 29, 2016 8:00 PM ET.

Response to Robby Schwarz
by Martin Holladay

You wrote, "I would like to get your opinion on if flash and batt meet the IECC air barrier requirement and installation requirements in attic knee wall applications or chase wall applications where drywall or other air barrier will not be installed on the fiberglass side of the assembly."

Usually, flash-and-batt is installed with the spray foam layer on the exterior side of the assembly. It sounds as if you are talking about backwards flash-and-batt -- with the spray foam installed on the interior side of the assembly (as it was in the photo used for this article).

I'm not familiar enough with IECC air barrier requirements to give you a code ruling -- and in any case, my opinion is irrelevant. (The only code interpretation that matters is the opinion of the local code official.) But I know the building science answer to your question: You always need an air barrier on the fiberglass side of the assembly.

Register for a free account and join the conversation

Get a free account and join the conversation!
Become a GBA PRO!