Image Credit: Harborside Design
Image Credit: Harborside Design Our client recently installed the bright pink insulation on the left. The dark areas on the pile to the right show why air-sealing is important, especially with a material sensitive to air movement, such as fiberglass. Trapped air is what insulates, and fiberglass batts do not trap air well on their own.
Image Credit: Harborside Design After the rest of the old blown-in cellulose is removed and the wiring is replaced in this house built in 1860, we will remove the roof framing, build a new floor system over this one, and frame a new roof. Canned foam will seal penetrations (with 26-gauge sheet metal, fire caulk, and rock wool around the chimneys), and loose-blown cellulose will insulate without blowing out the plaster ceiling below.
Image Credit: Harborside Design The rafters rest on a structural sub-fascia supported by beams 14 FEET on center, cantilevered over the top plate. In theory this may work, but a century and a half of leaky gutters and ice dams has left the decorative Italianate corbels supporting the roof structure. The expense of repairing this leaves little in the budget for energy upgrades.
Image Credit: Harborside Design
Two projects my company is currently working on involve a common problem: not enough insulation in the attic. Both homes are old; one dates from 1860, the other from 1705. In both cases we initially recommended insulating the rafter bays. In both cases, however, we were not able to get over homeowner biases against heating “storage spaces,” and instead opted for insulating the attic floor.
We had different energy auditors look at each house; interestingly, one always recommended unvented “hot roofs” and the other recommended strongly against them. We have had good luck with hot roofs, as long as the insulation is well done, but find that every remodeling project is composed of so many variables that there is never one right answer.
In the 1705 house, a two-story timber frame, not only did our clients reject the idea of heating the storage space (and could not accept our promise that it would cost very little extra to heat the space), but also it would have been a shame to destroy the beautiful hand-hewn timber frame with spray foam. The other option for insulating the rafters was dense-pack cellulose with foil-faced polyisocyanurate rigid foam facing the attic. That was probably the wisest approach, but posed other problems, including undersize rafters and a 1970s addition that would have to be cut off from the rest of the attic.
Our approach will be to remove the existing 6-in. fiberglass batts, which are no longer their original bright pink due to dirty air washing through them from the heated space, and to remove the poly-sheeting vapor barrier that was installed when the house was renovated in the 1970s. We plan to cover the joists with two overlapping layers of XPS foam and dense-pack the cavity with cellulose; then we will re-install the original board flooring. The challenge is going to be dealing with the stairs leading to the attic, which include a historic banister we cannot alter, and that makes installing a hatch nearly impossible. Our choices are to build a vestibule at the top of the stairs or modify the existing vestibule at the bottom of the stairs, either of which will be difficult to seal and insulate adequately—one of the primary reasons we would have preferred to insulate the rafters instead. We are leaning toward doing the best we can with the existing second-floor vestibule.
In the 1860 house, we are in the process of removing 6 in. of cellulose insulation to allow us to replace the mix of knob-and-tube and BX wiring and to seal any penetrations to the conditioned space below. Although we used a flash-and-batt hot roof on the addition we built onto this house last year, to save installation cost on the old part of the house our clients decided that insulating the attic floor would be adequate. In the colder months they live mostly in the new addition and will keep the thermostat turned down in the old part of the house. The existing roof system is framed in an unusual way and is dangerously inadequate; we will build a 2×10 floor system above the existing 2×6 floor joists, then we will use 2x10s to frame a new roof. Seventeen inches or so of loose-blown cellulose will provide more than R-49, and a well-insulated attic hatch similar to the one Mike Guertin built here will complete the thermal barrier.
A third project I am working on—designing a major overhaul of my grandmother’s former house, built in 1839—will involve spray foam. The new owners wanted to make use of the greatly underutilized second floor of the Cape Cod-style house, and they had already renovated one of the “ells” using open-cell foam. As on the 1705 house mentioned above, I tried to convince them to go with dense-pack cellulose in the rafter bays with a layer of rigid foam on the inside, which would preserve the beautiful hand-hewn timber frame for future generations. They would only consider sprayed foam, however; at least I think I have them talked into closed-cell foam instead of open-cell. It will ruin the aesthetic value of the frame but will block vapor. With a damp basement and inadequate wall insulation, open-cell foam would only provide an R-value in the mid-20s and almost certainly lead to serious decay due to condensation. Closed-cell foam will provide R-40 or so, and protect the structure while saving energy costs.
The common thread in all of these projects is an old house, a limited budget, and making the best of the situation. We consider many factors, including historic integrity of the architecture and consistent employment of our dozen or so exceptional carpenters, both of which often require compromises to accomplish. We employ independent energy experts when necessary to help make informed recommendations, and we do our best to meet our clients’ needs within their always limited budgets.
Weekly Newsletter
Get building science and energy efficiency advice, plus special offers, in your inbox.
13 Comments
roof
Mike - did you try talking them into rigid over the roof sheathing?
Dan, two of the projects'
Dan, two of the projects' scopes did not include new roofing, and on the one that did involve new roofing we were trying to maintain the original rake trim, which meant that we couldn't build the roof up.
On one past project where we used rigid foam over the sheathing, we had to add another layer of sheathing as a nailer, and found the process slow and expensive. Nailbase insulation would be a good alternative to the taped-and-overlapped Thermax panels we used on that previous project. The 3-4 inches of foam we would need in this climate to prevent condensation on the inside of the foam would just not have worked with the historic trim details.
Also, in my opinion, a hybrid system, such as rigid foam or sprayed foam with a batt or blown insulation on the inside, should only be used with a vapor retarder on the inside, but not a vapor barrier such as poly sheeting. Latex paint on drywall is probably the best option, but we were not planning to sheetrock the underside of the rafters.
attic insulation upgrades
I am not familiar with dense packing cellulose covered by rigid XPS for inulating the floor of the ceiling. How does this compare with the standard loose blown cellulose?
Thanks.
It's an unusual system
Sue, there are a few reasons why we decided to use a hybrid system in this case. One advantage is that the foam provides more R-value per inch so it saves space. At approximately R-6 per inch, 4" of XPS will provide R-24, about the same as 7.5" of loose-blown cellulose.
Blowing cellulose would require building an entirely new floor system on top of the existing floor system. In the 1860 house that's exactly what we're doing; it makes more sense there because the existing attic floor system is inadequate anyway and the original plaster ceilings can not take the pressure of dense-packed cellulose.
On the 1705 house the ceilings have been redone with sheetrock and the floor joists are not overspanned, so we can use the existing structure.
R-value of XPS
Michael,
As far as I know, every manufacturer of XPS rates their insulation at R-5 per inch, so 4 inches of XPS provides R-20.
Right
Thanks Martin, 2lb spray foam is R-6 and Polyiso ages to R6.5, so I have 6 in my head for foam, but you're right about extruded polystyrene. With R20 foam over 5+ inches of dense-pack cellulose (totalling R19), we'll just squeak by with R39, but it will be a well air-sealed R39. Could be better but could be worse. Along with the insulating and air sealing we'll do in the basement, the homeowners should see a drastic change in their energy usage and in their comfort.
cellulose behind rigid
We've done densepack behind rigid (usually foil face polyiso) several times in the roof assembly several times on reno jobs - I think it's a nice system. We usually tack the rigid up, then cross strap for drywall and to hold the rigid in place, then the insulation subs come in and blow behind the rigid. Very airtight if you're careful.
question for Dan Kolbert
cellulose behind rigid
by Dan Kolbert
"We've done densepack behind rigid......" Roof unvented?
aj
answer
Yes
re: cellulose behind rigid system
Very interesting concept; we've been delaying this project for years because of conflicting building science issues...so, in summary, for a non-vented attic in an older home (like ours - 1932 Tudor), you are suggesting two layers of rigid foam secured to the underside of the roof rafters and then dense-packing cellulose within the cavity between the rafters, rigid foam, and underside of the roof deck, correct? and you have had no issues w/ moisture condensation, etc. Also, since we have a simple gable w/ dormers, would you extend this detail to the gable walls? We too have a pull-down stair situation and floor-boards i would prefer to not pull-up, besides only having 2x8 ceiling joists, so we would not have a lot of room for insulation, as well.
Thanks so much and i look forward to your response.
With caveats
Bob, it depends on many variables, such as your location, exterior details, interior conditions, etc. In general, yes, I think that the approach you mention can work well. It's important to use the right kind of foam with good installation details. I would wrap the dormer "cheek walls" too, treating them as part of the roof.
It's not the only answer, though. Closed cell spray foam or exterior foam are excellent alternatives. Vented roofs often perform better than unvented roofs, but again, it depends on your local conditions.
Roof insulation
Mike -
I'd like to ask your opinion on my roof - i have a 1890s frame house in new mexico with 2x4 rafters and 2x6 ceiling joists in desperate need of insulation. The insulation contractors who have looked at it recommend blown in fiberglass to R-28 at the ceiling after air-sealing the ceiling. the attic is not usable space, but blown in fiberglass would really limit access if necessary for whatever reason. House is approximately 1100 sf, in need of a new roof however i'd hope not to have to reframe the roof in the process. What approach might you take to this situation?
Thanks,
Response to Sara
Sara,
The most cost-effective way to insulate your ceiling is with blown-in insulation as recommended by your insulation contractor. Here in the Northeast, most builders prefer cellulose to loose-fill fiberglass, but either product can work if it is deep enough.
As usual, the mantra is: the deeper the better.
If you intend to use the attic for storage in the future, it's always possible to beef up your ceiling joists (to increase their depth) before the insulation job begins. After your ceiling is insulated, you could then install plywood or OSB over your beefed-up joists for storage.
Log in or create an account to post a comment.
Sign up Log in