Image Credit: Greg Labbé At the start of the job, the rundown 1940s home in Toronto had uninsulated block and brick masonry walls. The eave walls on the second floor were about 4 feet high, just as on a typical Cape Cod home. The interior of the 2-wythe walls are mostly made of concrete blocks, with occasional bricks. At the left, you can see the studs of the wood-framed kneewall on the second floor. The masonry exterior walls and the underside of the roof sheathing were insulated with 2-pound spray foam. Before the insulation crew arrived, the subfloor was cut back 3 inches at the perimeter of the building, so that the foam insulated wasn't interrupted by the subfloor. Blower-door testing revealed air leaks that were traced to blisters in the cured foam. The pockets were full of air, and it wasn't always obvious what caused the blister to form. In this location, a flash coat of foam adhered well the wall, but the second coat didn't stick to the flash coat and was coming off in large flakes. A smoke pencil revealed that the wood floor joists penetrating the new spray foam were still leaky at the newly furred out (non-structural) basement top plate. Builder Paul Clinkard checks out the foam job in the attic, at the junction between the gable end top plate and the masonry chimney. The gaps in the foam were touched up with more foam around the mothballed chimney. The attic floor was vacuumed clean and spray foamed. The blower-door test revealed the adverse effects of electrical wires and pipes on the airtightness of the attic floor. As a consequence, significantly more foam had to be used. The leakiest areas were at the perimeter of the attic near the eaves. This attic floor will eventually be filled with R-60 cellulose. This bar graph shows estimated annual heat loss in gigajoules.
If you’re retrofitting a vintage brick building without an air barrier, don’t count on the spray foam to create a perfect air seal. If you plan to use the spray foam as your air barrier, it’s important to test your work before you cover it with drywall so you can seal any air leaks.
I recently had the pleasure of working with a forward-thinking design/build firm in Toronto, Argyris & Clinkard Fine Homes. The company’s objective was to complete a deep-energy retrofit of a 50-year-old solid masonry home in Toronto. As with all high-performance homes, a fully ducted HRV was installed.
The goal was to make the house as airtight as possible. Medium-density (2 pounds per cubic foot) closed-cell spray foam was sprayed on the interior of of the concrete block walls. Studs were spaced off the wall to reduce thermal bridging and to ensure that there was a monolithic uninterrupted coat of foam against the concrete blocks.
Builders Paul Clinkard and Liam Argyris called us in to perform a blower-door test before the drywall went up. The results were an eye-opener, showing why we can’t rely on foam alone for a good air seal in solid masonry home retrofits, unless we include blower-door-directed air sealing after the foam is installed.
Using spray foam for a deep-energy retrofit
This story-and-a-half house had uninsulated solid masonry walls (3-inch concrete block on the inside and 3-inch decorative brick on the exterior). On the second floor, the masonry walls extended up about 4 feet from the subfloor, as in a typical Cape. The home had wood-framed kneewall partitions that created a narrow 2-foot-wide space between the kneewalls and the exterior masonry walls. The gable walls on the second floor had the same masonry construction as the first-floor walls.
This construction is typical for homes of its age in Toronto. The homeowner wanted to upgrade the house for a new century from the inside, so a computer energy simulation was performed to guide the decision-making process, based on the potential energy savings from a variety of possible upgrades.
Unfortunately, as in almost all renovations, the orientation of the home conspired against any improvements in solar heat gain. Obstacles to producing a continuous air barrier included structural penetrations through the planned spray foam.
The scope of work included the following measures:
- The attic floor was vacuumed clean and spray foam was applied to the back of the plasterboard ceiling.
- The sloped roof assemblies between the tops of the second-floor kneewalls and the attic above the second floor were unchanged, as they were already insulated with dense-packed with cellulose.
- The space behind the wood-framed second-floor kneewalls (formerly outside the conditioned envelope of the house) was brought within the conditioned envelope by spraying the underside of the roof sheathing and the interior of the exterior walls with spray foam.
- On both the main floor and the second floor, the ¾-inch diagonal tongue-and-groove pine subfloor (which originally extended to the exterior concrete block wall) was cut back 3 inches at the perimeter of the building to create a gap for the spray foam.
- New studs were installed on the main floor and in the basement; these were made of engineered lumber and were spaced about 2 inches off the concrete-block wall to provide room for a continuous monolithic layer of spray foam, reducing air leakage and thermal bridging at the studs. The engineered studs required less bracing to combat the tendency of expanding spray foam to cause studs to bow inward or to bow into themselves at inside corners.
- The gravel on the basement floor was covered with spray foam, and a new concrete slab was poured on top of the cured foam. The foam was brought up the wall to break the slab from the wall.
- The foundation walls were insulated on the exterior with 2 inches of XPS (R-10), from the footing to the mudsill. (Exterior foundation insulation was preferable to interior insulation because masonry partitions in the basement created hard-to-detail T junctions.)
What is the ideal substrate?
Spray foam sticks really well to clean wood or masonry, as long as it’s dry and not frozen. It doesn’t stick well to dirt, oil, or water. The person spraying foam needs to understand these facts, and also needs to know where to apply the spray foam, and how the best location to apply the foam continuously changes as a spray-foam installer moves from the basement to the rim joist to the roof.
The ideal substrate for spray foam is scrupulously clean, openly accessible, and free of wires, cross bracing, 6-mil poly, plumbing, cables, or ducts — you get the picture. Ideally, there would be nothing in the stud cavities and the foam could be uniformly applied, producing a good air seal. Unfortunately, the reality is that stud cavities are busy places and spraying in them for a perfect air seal is really hard to do.
Spray foam is shot from a distance of 3 feet, and all the stuff in the stud cavity can create unfoamed “shadows” on the back substrate (just like the shadows created by a flashlight beam). Those shadows create voids or open blisters that can leak air. With each pass of foam, the substrate’s surface transfers its increasingly distorted shape to the next layer of foam.
The situation can be improved by training electricians to run wires on the back of the wall. Don’t move the plumbing pipes to the back of the wall, though, even if it would make things easier for the foam installers. Plumbing pipes need to be on the warm side of the insulation to keep them from freezing.
The trouble with spray foam
The trouble with spray foam is that it has to be sprayed onto a surface, and often we can’t choose the surface, as the situation forces our hand.
For the sake of discussion, let’s imagine that on this retrofit project, we wanted to use drywall as our air barrier, following the Airtight Drywall Approach. To apply spray foam to the air barrier, we’d have to spray the drywall from the back side — an approach that would be ideal but impossible. So we apply spray foam from inside the house, adhering the foam to the concrete block and hoping for an airtight result.
Avoiding voids in our building assemblies
The advantage of bonding foam to the air barrier is that it eliminates air movement between the two. If there is a gap between the back of the drywall and the interior face of the spray foam, the gap becomes a highway that connects all the small leaks from the foam’s imperfections and unfoamed penetrations. When a difference in pressure exists, voilÃ ! For leakage to happen, you need only a difference in pressure and a hole. We can’t stop physics from producing pressure differences, but we can seal holes.
To be effective, spray foam should be sprayed onto an exterior air barrier like wood sheathing. If you are renovating an old house, the usual method is to coat the inside surfaces of the exterior walls with spray foam insulation or a liquid-applied membrane like StoGuard Gold Coat. (The latter approach was described in a recent Fine Homebuilding article).
By the way, if a liquid-applied membrane is used as an air barrier, it still needs to be tested for airtightness before spray foam is applied. The purported advantages of liquid-applied membranes lie in their flexibility and strong adhesion which provide long-term performance.
For any deep-energy retrofit, the air barrier needs to be tested, whether the air barrier includes spray foam, polyethylene, a liquid-applied membrane, plywood, foam sheathing, Tyvek, Typar, or drywall.
It’s common on other job sites to see drywall hung as soon as the spray foam is installed. However, we suspect that for retrofits where foam is applied directly onto solid masonry, finding and sealing leaks with the help of a blower door will ensure significantly better performance.
On this project, the homeowner understood the value of performing a blower-door test to check the spray foam for leaks before the drywall went up. The spray foam was intended as an all-in-one solution that provided high R-value insulation and an air barrier. Had the drywall been installed over the air leaks, the gap between the cured skin and the back of the drywall would have short-circuited the insulation.
So what happened at the house?
After the spray foam was installed, we performed a pre-drywall blower-door test. The test showed an air leakage rate of 10.8 air changes per hour at 50 Pascals (10.8 ach50), with an equivalent leakage area (ELA) of 340 square inches.
This massive rate of air leakage would make the house uncomfortable and more expensive to condition; it could also lead to condensation problems. For purposes of reference, an Energy Star home should test at or below 2 ach50. At the Toronto house, the heat loss attributable to air leakage would have exceeded the conductive heat loss through the entire ceiling and above-grade wall area.
With the blower door running we were able to locate the air leaks in the foam, and the spray foam contractor came back to touch up the leaks. The smaller leak locations were caulked or touched up with canned foam.
After this work was performed, a second test showed that the air leakage rate had been cut in half to 5.4 ach50 with an ELA of 177 square inches (see Image #10, below). This was a significant drop in heat loss and condensation liabilities, and should result in improved occupant comfort.
As the project moves forward and the final two windows and doors are installed, we expect the air leakage rate to be halved again.
Greg Labbé is co-owner of BlueGreen Consulting Group, a high-performance home consulting firm that works with architects, builders, and homeowners to optimize the energy performance of new and existing homes through detailed energy modeling and site testing.
Get building science and energy efficiency advice, plus special offers, in your inbox.