Insulate attic rafters without spray foam
TNelson
| Posted in General Questions on
I have a poorly insulated existing attic that I would like to improve. The attic is ventilated with soffit and ridge vents. The floor of the attic is insulated with cellulose. There is a furnace and ductwork in the uninsulated space. There is a poorly insulated and sealed pull-down attic stair. The attic is also used for storage.
We would like to avoid closed cell spray foam. I have read a pile of information on this site and related ones, many thanks for all the great info! I believe the following design gets us where we need and want to be. Goals are code compliance, moisture control, reducing cost where feasible, maintain headroom, and ease of sourcing and install.
The house is in Louisville, CO, climate zone 5B. We have had issues in the past with excess moisture in the attic in the winter. This has been partially addressed by adding more ventilation, but now the furnace freezes in very cold weather, which is obviously not optimal!
Proposed Assembly – Conditioned attic with vent channels
* Asphalt shingles (existing)
* OSB roof deck (existing)
* 2×12 rafters, 24″ oc (existing)
* AccuVent baffles, 1.5″ depth, soffit vent to ridge vent
* Dense pack cellulose 9.75″, ~R-36
* 1.5″ foil-faced polyiso, R-9.6 (continuous)
* Polyiso fully seam taped with 3M 8067
* 0.5″ gypsum wall board, tape and mud, no paint
* Assembly U-factor 0.023, from Ekotrope calculator
* Should meet 2021 IECC code for <= 0.024
Many thanks in advance for any responses. Are there are any particular flaws in the plan, or special trouble spots to look out for?
Thanks, -Todd N
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
Interesting approach. My questions would be:
1. What is the intended sequence?
2. Typically, with dense fill cellulose on vertical walls they want no more than 16" oc spacing because the gyp bows out too much if the spacing is wider. My own experience is that it bows out even on 16" oc walls. I would be concerned as to whether 24" oc, partially horizontal gyp on the underside of dense fill, thus taking not just the density pressure but also the weight, would just bow out the gyp too much, and perhaps eventually pop it off.
a. I wonder if you could solve some or all of this issue with damp applied cellulose. No idea if you could use damp applied instead of dense fill - I'm not sure if damp applied can be applied to the underside of a partially horizontal surface.
b. Alternatively, perhaps you could add horizontal strapping to the underside of the rafters at 12" oc and attach the gyp (through the polyiso) to that . The side benefit is it gives you a bit more thickness of insulation.
3. Can you tell us the plan for the detail at the transition from the soffit/eave to the underside of the rafter slope?
4. Can you tell us the plan for the detail at the ridge? ie - how will you seal the end of the AccuVent to the ridge beam to prevent the dense fill from clogging the vent?
Todd,
It sounds like it a good plan.
Another alternative would be to fur down the rafters to get a bit more depth and just use cellulose. You could then choose whether to drywall or just cover the interior with a variable perm membrane as an air-barrier / vapour-retarder.
Edit: Akos makes a good point about the baffles.
Mostly good except for the vent baffles. These are meant as edge baffles in attics with loose fill fluffy. Since they are not permeable, there is no way for moisture to move into the vent space so they don't work for cathedral ceilings.
You need something permeable for the baffle. This can be netting, house wrap or fiberboard stapled to 2x2 strapping in the corners.
Dense pack cellulose is great but sometimes for cathedral ceilings batts are much simpler. Even if you do go with batts, you still need permeable baffles. My preference is to skip baffles and use high density batts which don't suffer from wind washing.
Thanks all for the replies, very educational. Since I don't have any experience with dense cellulose, it never occurred to me that 22.5 inches across might be too bowed out for drywall, but it makes sense when you think about it; on the underside of a sloped roof there's gravity in addition to the dense pack pressure, and the netting is only going to do so much.
I would think damp applied cellulose is more of a specialty product, and I'm trying to keep things fairly vanilla so I have better luck finding insulation contractors to bid on my project.
For the transitions, I was a little fuzzy on how the soffit and ridge design was going to work. The title of my question said "without spray foam". But I'm not completely opposed to the stuff. Certainly the top plates probably need foam sealing, and I wouldn't be opposed to spraying the narrow bits down by the eaves with closed cell to both air seal and insulate.
I was aware that the AccuVent is non-permeable, but I was hoping a good enough job on the air control layer (polyiso and tape) would mean there wasn't much moisture coming up from the house. But I'm aware that the general idea is to not rely on a perfect install, and provide some drying if the insulation does get wet. I was trying to avoid site-built baffles, based on a guess that they will cost substantially more in labor than any commercially available baffle. I'm just not aware of any permeable baffle that you can readily purchase.
From the responses here, maybe a Rockwool Comfortbatt has some advantages. It costs more than cellulose, but perhaps pays some of that back in savings. The savings come because you skip baffles entirely; so no cost to buy, fabricate or install the baffle material. There's a Comfortbatt 9.5 inches thick for 24" oc at R-38. If that was installed flush with the bottom face of the rafters it would leave a 1.75" air gap under the roof sheathing. Akos, is this what you mean when you say skip the baffles? Looks like I could still meet code with 1.5 inches polyiso and GWB.
Alternately, what about Comfortbatt between the rafters, a smart membrane, and then a non-flammable continuous insulation like Rockwool Comfortboard, or something else I'm not yet aware of? I wonder what the building inspector would think about that. The space will only have just over 6 feet of headroom at the peak, and I know I'm going to clonk my head on whatever is up there, so the drywall has some appeal in that regard.
The non-permeable baffles don't work, no way to make it into a safe assembly:
https://www.greenbuildingadvisor.com/question/moisture-between-insulation-and-baffle
For the rest either option would work. Semi rigid MW for a ceiling would a be a tough install though because it is still squishy, I would stick to regular foil faced rigid. If you are tight on space, you can't beat the R/inch of polyiso.
You can also look at the high density fiberglass bats for cathedral ceiling, they also work for a bit less cost.
You would still want a section of vent baffle by the soffit area to channel airflow into the vent gap. Since it is a small section, this can be any type of baffle readily available.
Akos, thanks for all your thoughtful replies, you have very patiently and thoroughly answered my questions and provided a real help.
To be clear on what's been suggested here in post #3: in a vented cathedral ceiling, omitting an airtight baffle between vent channel and fluffy insulation? Even if it's a high-density batt, that's not airtight. I get it has to be permeable, but my understanding was that airtightness was critical at this location, in addition to the airtight drywall at the bottom of the assembly.
AndyCD,
The standard detail here is a 3" ventilation gap and no baffles except at the exterior walls. The secondary air-barrier (sealed baffles) above the insulation is there to reduce wind-washing. Without them you take a small hit on the R-value due to the wind-washing, but can mitigate that by using either cellulose or high density batts, and increasing the depth to compensate - in much the same way you do with insulation on the floor of an attic.
Malcolm, I'm confused. Isn't the OP proposing moving the thermal layer from the attic floor to the roof rafters, creating a de facto cathedral ceiling? Where is this 3" air gap?
AndyCD,
The gap is the code mandated 63mm deep vent channel from the eaves to peak between the insulation and the sheathing above, on a cathedral roof with permeable insulation. Usually the details here on GBA show the vent channel separated from the insulation by baffles. Most vented roofs here omit those sealed baffles.
Ceilings need only one air tight layer. For vented assemblies, the simplest is at the ceiling plane. The rest is just belt+suspenders.
As for wind washing issue look at fig 10 here:
https://www.greenbuildingadvisor.com/app/uploads/sites/default/files/Wind%20Washing%20Effects%20on%20Mineral%20Wool%20Insulated%20Sheathings.pdf
It is a very small issue even with regular FG batts at the air velocity you'll likely see in a ceiling.
Akos, thanks for the link to the wind washing article, that's very helpful. My takeaway is that at 9.5 inches thick, wind washing is a negligible factor for mineral wool insulation.
Would it be possible to fix the stupid and get the HVAC out of the attic? Seems likely to cost about the same and should work better.
If you insulate at the roof line you have increased the area that needs to be insulated 30-100%!
The new larger area will always loose that percentage more energy in perpetuity and likely more as it gets very expensive to install the same R value at the roof line and generally less gets installed.
You did not say it but my guess is your plan is to leave the existing insulation on the attic floor and not install registers to fully condition the attic. If so, that would be very risky game because the dew point in the semi conditioned attic will likely be very high and often higher than the surfaces in the attic. When that happens, things get wet. The risk is that things will get and stay wet long enough for mold to grow and rot to happen.
If you can’t bring yourself to move the HVAC consider air sealing the attic floor and ductwork then burring the ductwork.
https://www.energyvanguard.com/blog/buried-ducts-allowed-2018-building-code/
Walta
Walta, I considered removing the HVAC from the attic, but it runs into territory that looks like the perfect as the enemy of the good. I currently have a dino-gas powered furnace, but the insulation is step one of a plan to convert to heat pumps. Yes, I plan to remove the attic floor insulation and condition the conditioned attic to avoid condensation issues. I have tried, as best as possible, to do my homework before posting. While I am not a builder by trade, I have learned a ton of useful information on GBA, including the part about a supply register in a conditioned attic.
I have 9 existing registers to supply, 3 bedrooms, 3 bathrooms, 1 hallway, 1 crawl space, and 1 for the attic. I have a small house and low ceilings at 7'9". If I were an HVAC technician, I would much prefer to put some slim ducts near the existing registers from the attic side. The alternative is a lot of work in the stairwell and hallway ceiling with soffits and dropped ceiling. And it will feel pretty cramped in places with a low ceiling. Some of the obvious duct runs would go through a door header, and that's just not going to work. So there's more wall demo and duct bends to make it work. There's a drain in the attic and not in the stairwell, and the attic location is gravity drain vs ejector pump in the drop ceiling.
I understand that attics with storage and HVAC are a bad idea to begin with. But that doesn't change the fact that I've got one, and I need to figure out how to make it work as best as possible. Back to the small house thing, we do utilize the storage space now, and will continue to do so in the future. So that has tangible value to us, even at the cost of increased heating and cooling. This article puts the operational cost increase at 10% over what you suggest, with the HVAC in the envelope and below the ceiling: https://buildingscience.com/documents/published-articles/pa-are-you-doing-somethig-stupid/view
My HVAC bill now is $100 to $150 per month. It will be less with better insulation, air sealing, heat pumps, etc. So the increased operational cost of the conditioned attic is less than 10 dollars a month. I'm willing to bet the payback on moving the HVAC below the ceiling is approaching never, and that's before ascribing any storage value.