More insulation on top of 3″ or 4″ foam?
![](https://secure.gravatar.com/avatar/9327e36c68084020be89758960e76c53?s=52&d=https%3A%2F%2Fimages.greenbuildingadvisor.com%2Fapp%2Favatars%2F200x200%2Fe.png&r=g)
If I spray 3 or 4″ closed cell foam insulation in the attic (2nd floor), should I also include bat insulation to add to the R value? For example adding R15 Comfort Batt (Roxul) on top of the R21 or R28 from the Foam that I would get?
Do people do this? is this wise?
I’m in zone 5 and will live in this space. Would like to achieve air tightness with foam + the high R values that are recommended.
Foam installers are trying to convince me that 3″ or 4″ closed cell is more than enough and nothing else is needed.
Would love some advice.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb1.png)
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb2.png)
Replies
I think this is the Flash and Batt method (https://www.greenbuildingadvisor.com/blogs/dept/guest-blogs/flash-and-batt-method).
EldarChik,
We need more information.
1. Is the existing spray foam insulation on the attic floor, or on the underside of the roof sheathing?
2. When you tell us that you will "live in this space," are you saying (a) that you will live in the attic, or (b) that you will live in the space under the attic?
At 3-4" of closed cell foam is at most R28, and more likely to be under R25, might be under R20.
Code minimum in zone 5 is R49. It takes ~7-8" of closed cell foam to hit R49, not 3-4".
But any closed cell foam thermally bridged by rafters pr joist is having it's higher performance cut off at the knees by the thermal bridging, and is NOT equivalent to R49 of a lower R/inch product between joists or rafters (or not thermally bridged at all), due to the comparativey higher heat transfer through the shorter bridging path.
Given that it's substantially below the code required minimums, how are they able to support the claim that "that 3" or 4" closed cell is more than enough and nothing else is needed."???
In short, they're not giving you anything like a straight answer. Reading between the lines, "...that is all you need..." from foam installers usually means is:
"that is all that has an easy financial rationale when using this extremely expensive product"
...and even then it's a fairly short term economic analysis.
There are reasons to avoid using closed cell foam other than cost (mostly environmental reasons) but depending on the particulars there can still be a rationale for using SOME closed cell foam. In new construction it's easy to design it out, retrofits, maybe not.
EldarChik,
For more information on the issues raised by Dana Dorsett, see this article:
“It’s OK to Skimp On Insulation, Icynene Says”
Martin,
The foam will be on the underside of the roof sheathing. Rafters are 6". We will live in the space under the attic, but the whole 2nd floor (just one big room) + attic is exposed to the rafters at the moment. This is not new construction.
I definitely agree that the only reason they are saying is because it makes financial sense for them to put less foam.
I had a 3rd foam installer just say only 1.5" is enough (quoted close to 3k).
Would you advise to go with less close cell ? 3" and then add insulation on top of that?
Eldar,
As Dana explained, you are aiming for R-49. If you want to follow the flash-and-batt method, building codes require that the spray foam layer have a minimum R-value of R-20 in your climate zone (Zone 5). That means that you will need at least 3 inches of closed-cell spray foam.
To bring the spray foam insulation up to minimum code requirements, you would need to add at least R-29 of fluffy insulation (for example, fiberglass batts) under the cured spray foam (and in direct contact with the cured spray foam). You'll need about 8 inches of fiberglass to achieve that goal.
For a full explanation of the flash and batt method, see Flash-and-Batt Insulation.
Are the rafters milled 2x6, which are 5.5" deep, or are they full-dimension 2x6, fully 6" deep?
Is it possible to add depth and still have enough head-room, so that you can end up at R49 (or higher)?
Are you planning to re-roof some time in the next decade?
If the finished ceiling will be mounted directly to the rafter edges and you're NOT going to take it fully up to R49, from a moisture safety and performance point of view in climate zone 5 it's fine to go with just ~40% of the total R as closed cell foam, which would be about 2". With 2" of closed cell foam (R12-R14) in a 5.5" deep cavity it would leave 3.5" of space, which is ideal for installing R15 rock wool or fiberglass batts. The total R would only be R27-R29, but R12/R27 is fully 44% of the total. If it's full dimension 2x6 it would leave 4" of space. Compressing an R19-R20 fiberglass batt designed for 2x6 cavities down to 4" would perform at R15-R16 at that thickness, and the 2" foam would still be reliably more than 40% (with some margin) of the total. At that ratio the batts will not accumulate enough moisture over a winter to become a problem in your climate, and the vapor retardency of the foam (0.5-0.7 perms) will be sufficient to protect the roof deck, while still allowing some drying capacity toward the interior.
With R27-R30 in the cavities, when it comes time to re-roof adding just 3" of polyiso above the roof deck would be sufficient to meet code minimum thermal performance. While new stock is ~$60-65 for a 4x8 sheet, used 3" polyiso in reasonably good shape is commonly available from reclaimers at $15-25/sheet. With the foam held in place with an OSB or CDX nailer deck through-screwed to the structural roof deck it adds 3.5" to the roof thickness, which can still be made reasonably aesthetic with a 1x4 facia board on the ends. There will be some complextity and possibly new flashing needed around complex roof lines like dormers, etc, but unless it's hipped roof all cut up by dormers it's usually do-able, and "worth it" from an ice-dam mitigation point of view in climate zone 5A.
Take the time to read this document, but in particular take a look at the column in Table 3 labeled " 2" ccSPF + spray fiberglass":
https://buildingscience.com/sites/default/files/migrate/pdf/BA-1001_Moisture_Safe_Unvented_Roofs.pdf
Even at R38 the roof deck is moisture safe in zone 5A with 2" of closed cell, and you'd be at R30 or less, which is safer still. Even at R49 the roof deck, would be safe in Minneapolis (zone 6A) or International Falls (7A), as long as the roof isn't a solar-reflective light color. This is a safe stackup even without the exterior foam, but it'll be safer still if rigid foam gets added at the time you re-roof.
Martin,
Thank you for clarifying. And this method is the best method out there to achieve the air tightness and high R value? Or is there something else that you would recommend given the current situation?
You'll need the 2" closed cell for moisture control, and there's no way around that unless you can add depth to the rafters and add a vent channel between the roof deck & insulation.
A full 5.5" deep cavity fill of 0.7lb density open cell foam with "vapor barrier latex" primer on the ceiling gypsum would deliver comparable thermal performance, better air sealing than 2" of closed cell, but may still have moisture issues on north facing roof pitches even in zone 5A (the risk is pretty low though.) It's in the grey area of the code prescriptives for moisture control, and it's cheaper than 2" of closed cell foam + batts (roughly the same cost as just the 2" of closed cell foam.) To hit code-R with rigid foam above the roof deck later with open cell foam in the cavities would take 4" of exterior polyiso.
EldarChik,
Q. "And is this method is the best method out there to achieve the airtightness and high R-value? Or is there something else that you would recommend given the current situation?"
A. The best method, by far, is to install an adequately thick layer of rigid foam on the exterior side of the roof sheathing, followed by another layer of roof sheathing and new roofing. Of course, this approach is expensive, so you will probably have to use the second-best or third-best approach.
All of your options are explained in these two articles:
How to Build an Insulated Cathedral Ceiling
Creating a Conditioned Attic
Martin and Dana,
Putting insulation on the exterior of the roof is not an option at this time, have to work from the inside at this moment.
So, if I understood correctly the good option for me is to do 2" closed cell +3.5" of Roxul insulation.
What if I did 3" closed cell with 3.5" Roxul (I know I would have to build out an extra 1" on top of the current rafters).
Or stay with the 2" +3.5 and maybe add 1" Rigid Foam with reflective foil under the Roxul comfort bat? How would that stack up?
typo, NOT an option to insulate from exterior.
"What if I did 3" closed cell with 3.5" Roxul (I know I would have to build out an extra 1" on top of the current rafters)."
That works from an R-ratio/ moisture-control point of view too, but isn't necessarily the best bang/buck. HFC blown closed cell foam can only be installed in 2" lifts (for both safety and quality reasons) at a time with a cooling period before applying the second lift. So a 3" installation done right takes more time, and is more expensive. HFO blown foams can be blown in 3" + lifts (and it's a far greener blowing agent) but are also more expensive than HFC blown foams. Since the closed cell foam will be thermally bridged by the rafters, the additional performance improvement isn't going to be anywhere near what the additional R6-R7 at mid-cavity might imply.
Please verify- are they full dimension 2" thick x 6" deep rafters, or are they actually 1.5" x 5.5"?
Unless explicitly told otherwise I'm going to assume they are milled 2x6 rafters, with a nominal 5.5" cavity depth.
Do NOT install full sheets of rigid foil faced foam on the underside of the rafters, since that will effectively block ALL drying potential for the assembly. Adding an inch of foil faced polyiso foam at the RAFTER EDGES ONLY would work though, and double the performance of the framing fraction. If that takes the cavity depth from 5.5" to 6.5", a 2" shot of closed cell foam would then leave 4.5" of space for the fiber insulation. A 5.5" thick R23 rock wool batt compressed to 4.5" will perform at about R19. If the closed cell foam is rated R6.5/inch you'd have R13 out of a total R32, which is about 41%, which is still adequate for dew point control. If it's only R6/inch stuff (R12 @ 2") it'll be a bit marginal, but still good enough for most of zone 5, but compressing R20 fiberglass into 4.5" it'll be R17, again pushing the R-ratio into the 41% range.
The real performance improvement in that assembly would be the rafter edge strips which would cut the thermal bridging by half. Going for a marginally higher center-cavity R at the much higher cost of another inch of spray foam would be a waste.
Eldarchik,
Your final decision depends on your R-value goals. I've provided enough links to enough articles for you to be able to make your decision.
The basics of flash-and-batt are described here: Flash-and-Batt Insulation.
You can always deepen the rafter bays, either by attaching strips of rigid foam to the rafters, as Dana suggests, or by adding framing members to the underside of your rafters, at 90 degrees to the rafters, to create an insulation space as deep as you need.
That said, if your R-value goals are modest, you can just insulate the 2x6 rafter bays, even though you will end up with a roof assembly that doesn't meet minimum code requirements.
Dana,
After measuring, my rafters are 1.5x5.5
One of the foam installers said they cannot even quote me for 2" of foam because its not enough. The other did. How many inches is recommended for the wall foam? my wall cavity is 3.5".
Martin and Dana I will definitely go over the advice you are offering, just being compeltely new to this doesn't help with learning process and decision making, especially because winter is coming :)
I forgot to mention Dana, I'm being quoted for- 1. Icynene product 2. Covestro product. The difference in price is small - $140 (Covestro being a bit more)
1. Insulate roof to 3” (R21) and exterior walls to 2” (R14) with Icynene Pro Seal closed cell spray.
The other installer has this product:
2. Covestro - CLOSED CELL: Install 2 lb. closed cell Urethane spray foam 3" avg depth in roof rafters of attic, including smaller section off main attic (R21) and 2" avg depth in exterior / gable end walls of attic spaces (R14)
For the wall foam, go with half pound density open cell foam (~R13-R14). Even though it's below the R20 code minimum. At 3.25" it would hit R20 closed cell foam meeting code for R-value, but the shortness of the thermal bridge robs it of performance, yielding only another R1 of whole-wall performance.
Installing a full fill of open cell foam in the cavities and a continuous half-inch of rigid foam (any type) between the wall board and the stud/cavity foam would be higher performance (despite the lower center-cavity R) at lower cost, and would meet code minimum on a U-factor basis. A half inch of foil-faced polyiso would be highest performance, adding about R3 to the whole-wall R.
What type of siding do you have?
The installer that won't quote you 2" because "it's not enough" is just wrong., or needs to be asked "enough for what?". If you were going for the full R49 it wouldn't be enough for dew point control, and would not meet code prescriptives, but if you're adding no more than 4" of fiber insulation between the foam and the ceiling gypsum it's enough, as spelled out in previous posts.
My head is spinning Dana. Is there an easy way to do this? lol.
I was trying to wrap my head around this first.
"Adding an inch of foil faced polyiso foam at the RAFTER EDGES ONLY would work though, and double the performance of the framing fraction. Can you show a picture example of this? I'm trying to understand what this means.
Cut strips of 1" polyiso 1.5" wide (the width of the rafters), and tack them to the rafter edges with cap nails. You'll be then looking up at empty cavity (or cavity with just 2" of closed cell foam on the roof deck), but with rigid foam on the edges of the rafters. Friction-fit the batts into the cavities so that the fluff is just proud of the foam edge strips so that it compresses a bit when the gypsum board is installed. With 1" foam you can long-screw or long-nail the gypsum board through the foam into the rafters.
If adding more than an inch to the rafter edges it's better to make edge strips with some 1x2 glued to the 1x, then trough-screw the 1x to the rafter edges, and mount the gypsum board to the furring.
It's pretty easy to cut 1" polyiso into 1.5" strips using a 4" steel taping knife that has been sharpened on the edge, as demonstrated in the video on this page:
http://www.finehomebuilding.com/2009/01/29/theres-a-better-way-cutting-rigid-insulation
Ok, so here is where I'm at right now. Please correct me if I'm wrong. My options are:
For the Roof Rafters (5.5" cavity)
1. 2" of Closed Foam + 3.5 inches of Roxul + sheetrock
or
2. 3" of Closed Foam + 3.5 inches of Roxul + 1" of Foil Faced Polyiso Foam at the Rafter Edges Only (I still need a picture example of what this means) + sheetrock
or
3. 2" of Closed Foam + 4.5 inches of Roxul + 1" of Foil Faced Polyiso Foam at the Rafter Edges Only (I still need a picture example of what this means) + sheetrock
The #3. option seems to be the best option for the money and R value? because it is the most cost effective, takes care of the moisture problem, and has the highest R value.
For the Walls (3.5" cavity)
I should fill the cavity completely with half pound Open Cell and then add half an inch foil faced rigid foam board
p.s. Roof was just redone, so no option of installing anything on the exterior :(.
Dana, siding is Metal.
Your #1 & #2 both work, but #3 is a bit marginal on dew point control if using Roxul IF the closed cell foam is only rated R6/inch.. It has to be R6.5/inch or higher,to use 4.5" of Roxul. If the closed cell foam is R6/inch it's better to use low or mid-density fiberglass (labeled R19 or R20)for a slightly lower R to the fiber layer. Almost all HFO blown closed cell foam products run ~ R7/inch and will be fine, but only a few HFC blown closed cell foam products hit R6.5, so you have to pay attention to which product is being installed. A couple examples of ~ R7/inch HFO blown foams are:
http://www.demilec.com/documents/Heatlok-HFO-High-Lift-Flyer.pdf
https://lapolla.com/spray-foam-insulation/foam-lok-2000-4g/
For pictures of edge strips with 1x furring attached (in a stud wall application) see:
http://www.finehomebuilding.com/membership/pdf/9750/021250059.pdf
For your application you can skip the 1x furring, cap-nail (or 1.5" roofing nail, if you're careful) the edge strips in a few places just to hold in place before adding the batts, then just use longer fasteners for the ceiling gypsum.
Roxul ComfortBatts in USA are sold in 3.5" (R-15), 5.5" (R-23) and 7.25" (R-30) thicknesses. There is no 4.5" thickness. You might extend the framing to accommodate the 5.5" thick batts. Or use fiberglass batts to compress more easily. Literature says ComfortBatts compress more easily at the edges for a good compression fit to studs, not as compressible throughout the batt apparently.
Rather than spray foam...
Old board sheathing could be too difficult to air seal. But if the roof sheathing is plywood, could you use canned spray foam to seal edges, then use batts only and a smart vapor barrier (cheaper DIY)?
The R23s are 5.5", and are still easily compressible to 4.5", where it will perform at about R19. But R19 is already too risky from a dew point control point of view with 2" of R6/inch spray foam, still OK with R7/inch foam. If more than R19 of fiber is going to be installed it'll need another half-inch to inch of closed cell foam for dew point control. With less foam-R than 40% of the total there is risk is high wintertime moisture accumulation in the fiber insulation creating a mold hazard.
A vapor barrier on the interior and all-fiber in the cavity creates a moisture trap, almost guaranteed to eventually rot the roof deck in places, which is why code demands a vented assembly unless there is low permeance insulation directly in contact with the underside of the roof deck. The roof deck can be fine at a lower ratio than code prescribes, but it has to be less than 1 perm at the installed thickness. Installing a vapor barrier or smart vapor retarder on the interior side of an under-foamed assembly may still work, but it's straying into less conservative design territory. A WUFI simulation of the stackup would indicate the relative risk.
IIRC Owens Corning has been experimenting with fiberlglass insulation and smart vapor retarders in unvented roof assemblies, but if I recall correctly they were recommending for that use only for zones 1B-3B.
See: https://www.greenbuildingadvisor.com/articles/dept/musings/can-unvented-roof-assemblies-be-insulated-fiberglass
Certainteed's SmartBatt is in some ways a similar product, but is not being recommended for unvented roof assemblies, only walls:
https://www.certainteed.com/resources/30-45-064.pdf
I did same thing; 5 inch closed spray foam between rafters
and not planning to add anything else except cover it with
considering air tightnes you can't compare it with butt insulation
we had before R13 fiberglass and was livable
spray people say 3 inch is enough because with 3 inch or so you
acheving 96 % Of efficiency if you add another inch your insulation will be only 1% more effitient and another yet onnly half percent or so
Zbigniew,
I have heard that statement -- "3 inches [of spray foam] is enough because with 3 inches or so you are achieving 96% of efficiency; if you add another inch your insulation will be only 1% more efficient" -- before. The statement is made by spray foam contractors. It is meaningless.
"Efficiency" is a meaningless term when describing the thermal performance of insulation.
Here is a more accurate statement: Every time you double the thickness of insulation, you cut the rate of heat flow in half. The heat flow through 6 inches of spray foam will occur at only half the rate of the heat flow through 3 inches of spray foam.
Ok Dana,
So it seems I'm leaning towards the #2. 3" of Closed Foam + 3.5 inches of Roxul + 1" of Foil Faced Polyiso Foam at the Rafter Edges Only + sheetrock
I think I will do this without furring as you advised, since I can use longer screws to attach Sheetrock right on top of the 1" foam strips. This setup will be tightly packed since Roxul does only come in 3.5 inches so total 6.5 is what my total setup will include after installing the 1" of foam strips onto the rafter edges.
The difference between the 2" and 3" foam is roughly $1,000 in price. The question then, is it worth doing 3" of foam for extra $1k and having a tight cavity (+ additional R7) , vs doing only 2" + 3.5 of Roxul + 1" of strips and having the cavity only filled to 5.5", but saving $1,000). Thoughts?
p.s. Icynene product is rated R7.1 per inch for the Closed cell and R3.7 for Open cell based on spec sheet that the installer provided.
Robert,
Thank you for clarifying about the Roxul batts and their sizes. I don't want to extend the framing more than 1", we don't want to take away from the size of the room as its already small. So I think the setup of 3" foam + 3.5 Roxul comfortbatt will work out well, as Dana already confirmed in regards to Dew point control.
Zbigniew,
I found that people on this forum like Martin and Dana are much more knowledgeable than the installers who just trying to make a sale and not really educate the consumer. I had 4 different installers come in, and give me 4 different opinions. For a person such as myself (who is not as knowledgeable) this forum is a perfect place to learn and not take the installers "confident" words as the truth. At the end I need to educate myself here, make my decision and advise the installer to do what I need.
p.s. Not one installer mentioned Dew Point control or thermal bridging.