GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Audio Play Icon Headphones Icon Plus Icon Minus Icon Check Icon Print Icon Picture icon Single Arrow Icon Double Arrow Icon Hamburger Icon TV Icon Close Icon Sorted Hamburger/Search Icon
Guest Blogs

Building an Airtight Envelope

In Connecticut, Michael Trolle continues construction of his Passivhaus home with a roof, insulation, and high-performance windows from Europe

Image 1 of 2
The attic of the author's home got as much as 24 inches of cellulose insulation, with a corresponding R-value of about 86. The roof is built with trusses to minimize thermal bridging through framing materials.
The attic of the author's home got as much as 24 inches of cellulose insulation, with a corresponding R-value of about 86. The roof is built with trusses to minimize thermal bridging through framing materials. A section view on the left shows the multi-chambered frames of the author's windows. On the right, an isothermal image shows the frame interior stays warm when exposed to 14 degree cold outside.

Editor’s note: This is the third installment in a series of blogs by Michael Trolle about the construction of his Passivhaus home in Danbury, Connecticut. The first part was published as “Building My Own Passive House.”

I chose to frame the roof with engineered wood trusses, which form both an attic and vaulted ceiling. I chose a truss, rather than traditional 2x roof rafters, to make it easier to achieve my efficiency goals. Over the vaulted ceiling area, I wanted the truss to accommodate 18 inches of cellulose insulation — about R-65. Because the truss has separate elements — a 2×6 top chord and 2×4 bottom chord and supporting elements — insulation can fill the spaces in-between, which prevents thermal bridging, or “heat bleed” through the wood, which is a poor insulator.

The attic area is too small for storage space so we simply blew in cellulose to a depth of 24 inches — about R-86. On one side of the attic, the cellulose is continuous with the cathedral ceiling. On the other side, it terminates against the front wall where I have 12 inches between the attic floor and the roof sheathing. Twelve inches of cellulose works out to R-43, sufficient to keep the roof cold, thereby avoiding ice dams in winter.

To satisfy the airtightness requirement, we attached 1/2-inch plywood to the interior underside of the roof trusses and then taped all of the joints with a European flashing tape which will adhere tenaciously to the plywood for as long as the home stands. Where this interior plywood meets the exterior plywood sheathing on the walls, we used a wider flashing tape to seal this critical joint.

We used flashing tape at all other intersections of floor, wall, and roof planes, and to install the windows and doors. The end result is a very, very airtight house, as demonstrated by a third-party blower door test result of 0.46 air changes per hour at a pressure difference of 50 pascals (ach50), which easily meets the Passivhaus certification standard of 0.6 ach50. The current code requirement in Connecticut is a maximum air leakage rate of 7.0 ach50, nearly 12 times the Passivhaus standard. This is yet another instance where the code standard for this critical efficiency element is well behind the times.

A continuous air barrier is essential

Floor, wall, and roof areas that adjoin unheated spaces make up the thermal envelope of a house. Building the house with an airtight thermal envelope is one of the most difficult Passivhaus requirements to achieve. Maintaining air barrier continuity at transition points in the thermal envelope is critical.

I began by wrapping the concrete slab foundation with a reinforced plastic sheet, which was installed with great care to eliminate possible holes. The plastic was cut to lap the plywood wall sheathing and plywood subfloor (by the old foundation) by 3 to 4 inches. These joints were taped with a high-quality construction tape to provide a durable and airtight connection.

The same tape was used to seal all joints and intersections of the plywood subfloor, exterior wall sheathing, and underside of the roof trusses (where plywood was installed to help guarantee the continuity of the airtight thermal envelope).

The intersection of the subfloor and wall sheathing could not be taped, so this intersection was sealed with EPDM gaskets. Finally, the windows and doors were installed, using the same tape to seal them to the framing.

A blower door test delivers the verdict

With the airtight shell complete, a preliminary blower door test was run to determine airtightness and to locate air leakage points that could be sealed. A blower door test involves a fan and special instruments to find out how much air the fan has to suck out of the house to create an air pressure difference of 50 pascals between the indoor and outdoor air. Our first test proved that the house was pretty tight, although not yet tight enough to meet the Passivhaus standard.

A blower door measures how much air must be displaced to maintain a constant pressure difference between inside and outside. With careful air-sealing, the author exceeded the requirements of the Passivhaus standard.

At this point, we used the blower door to help locate air leaks in the thermal envelope. With the fan running, we could actually feel the air flowing where there were leaks. These were sealed with tape or a special air-sealing spray product (Knauf EcoSeal), which brought the air leakage down below the 0.6 ach50 level required by Passivhaus.

At completion, we achieved an even tighter result of 0.45 ach50. This means that with the house depressurized to 50 pascals, the air leakage in one hour is equal to 45% of the total air volume in the house. The energy code in Connecticut allows up to 7.0 ach50, meaning air leakage in one hour can equal 700% of the total air volume in the house, which would mean 15 times the air leakage in my house!

Adding windows and doors

The Passivhaus criterion for energy used to heat and cool the house is so stringent that it cannot be met by homes that use anything but the best performing windows and door. At this point in time, I am aware of only one small American company that makes windows and doors that meet the requirement for Climate Zone 5 (all of Connecticut), whereas there are numerous European companies that make such products.

I purchased windows and doors made by Klearwall in Ireland, where the company is known as Munster Joinery and dominates that market. My windows have a “tilt and turn,” two-way operation, which is typical of European windows. They tilt in from the top roughly 4 inches for ventilation, and they also open like an American casement window, only inwards instead of outwards.

The window-locking mechanism is inside the frame and locks all four sides of the operable window sash, unlike an American window which typically locks in one location.

Klearwall offers three types of windows. The most expensive are made from wood with an aluminum exterior skin; slightly lower in cost are windows that are made entirely from aluminum; and their least expensive windows are made from PVC, or polyvinyl chloride – in other words, vinyl windows. That’s what I purchased.

In the U.S., we think of vinyl windows as low quality. My PVC windows perform every bit as well as Klearwall’s more expensive products. The PVC itself is of premium quality, and the frame walls are much thicker than those of an American vinyl window. Moreover, the window frame construction is multi-chambered, insulated to prevent heat loss, and reinforced with metal (see the image below).

Two ways to open: European windows with a tilt-and-turn mechanism swing open from the top for ventilation (on the right) and also open like a casement (on the left).

The glass in my windows is triple glazed, which means that there are three separate panes of 1/8-inch glass, with two interior 3/4-inch cavities filled with argon for a total thickness of slightly more than 2 inches. The argon gas fill works much better than air to reduce heat flow. Two of the panes have an applied low-E coating, which reflects some of the sun’s heat to prevent overheating in the summer, as well as reflecting some of the indoor heat back into the room to reduce heat loss in the winter.

At this point in time, 99% of the American windows purchased for homes in the Northeast are double-glazed, which means that they have two separate panes of 1/8-inch glass, with one 1/2-inch cavity filled with argon, for a total thickness of 3/4 inch. One of the panes has an applied low-E coating. The few American triple-glazed windows on the market have a total glazing thickness of about 1 inch.

The performance difference between typical European and typical American glazing is enormous. The U-factor of my triple-glazing is 0.09, which converts to about R-11. The U-factor of American double-glazing is about 0.25, or R-4; American triple glazing is about R-6.

Why European windows are so much better

Why does the triple glazing in most European windows and doors have much better insulating values than the triple glazing in most American windows? After all, besides having the same three panes of glass, they both have low-e coatings, argon gas between the panes, and special spacers between the panes to stop heat flow.

The reason is that American triple glazing is about 1 inch thick and European glazing about 2 inches thick. With three panes of 1/8-inch glass, American triple-glazing has two cavities between the panes about 3/8-inch thick each. With the same three panes of 1/8-inch glass, European glazing has the same two cavities between the panes, but each one is more than 3/4 inch thick.

Because the cavities of both American and European glazing are filled with argon or krypton gas, which greatly slows the flow of heat, the thicker cavities with more gas in the European windows and doors do a much better job of slowing heat loss. [Editor’s note: The performance differences between U.S. windows and European windows can be partly explained by differences in the testing protocols used by standards-setting organizations in Europe and North America. Since North American windows are tested at lower temperatures than European windows, North American window manufacturers size the gaps in the IGUs to optimize window performance for these test conditions.]

So why aren’t American window companies making units with thicker glazing? My guess is they aren’t convinced Americans are ready to pay the extra cost for them to re-tool their factories and then to make these vastly better products. Window sashes would have to be thicker as would the frames (which greatly improves the insulating value of the frames!). That said, for one of our current projects, we priced American triple-glazed windows against European and were shocked to find the European window quote was actually slightly lower than the American.

Remember how American car manufacturers were forced to improve the mileage of their cars when customer demand for energy-efficient cars exploded? I think the same thing will happen in the window industry.

Choosing windows based on what direction they face

For my Passivhaus, I chose between three different triple glazings with slightly different performance values. For my south-facing windows, I chose glazing with a relatively high 61% solar heat gain coefficient (SHGC), because I wanted the sun’s heat during the cold months — and there’s no summer overheating problem because my roof overhang was designed to fully shade the windows during the summer months.

For my east, west, and south-facing windows, I chose glazing with a lower 49% SHGC because I wanted the slightly better U-value (this is the inverse of R-value, so lower is better.)

The truly fascinating thing is that my windows, in total, supply much more heating energy than they lose on cloudy days and at night. There is at least some solar gain from windows facing in all four directions of the compass — even north — although there is a net gain only from the south. My south-facing windows help to heat a heavily insulated concrete slab under tile in my main living space. This prevents the house from overheating during the day and gently warms the house during the evening and night hours.

Here’s a link to Part 4 of this blog series: The Four Keys to a High-Performance Home.

Michael Trolle is a co-founder of BPC Green Builders, in Wilton, Connecticut. This post, and the ones to follow, were originally published in slightly different versions at The HomeMonthly.com but also are available at the BPC Green Builders website. On October 7, 2015, BPC Green Builders was named Grand Award Winner in the Custom Home category in the Department of Energy’s Housing Innovation Awards.

5 Comments

  1. User avater
    Dana Dorsett | | #1

    Not that it would be easy to measure...
    But the rolling dunes of cellulose in the attic undercuts the thermal performance of that layer.

    There appears to be level markings crayoned onto the king pins of the trusses, which eyeballing shows a depth variability of nearly a FOOT between the tops of the dunes and the bottoms of the dune pits. If it can be dragged level with a roof-rake or something it's worth it to gain that performance back.

  2. User avater GBA Editor
    Martin Holladay | | #2

    Response to Dana Dorsett
    Dana,
    Good point. For more on this issue, see Is There a Downside to Lumpy Attic Insulation?

    When I needed to smooth my attic insulation, I used an ordinary garden rake. (If you want, you can turn it tine-side-up so that you are raking with the smooth side of the rake.) I think a garden rake is easier to maneuver in an attic than a roof rake.

  3. User avater
    Dana Dorsett | | #3

    Agreed, garden rakes are easier but...
    A garden rake is easier to move around an attic, but it may not have the reach. If the trusses are 24" o.c. it looks like 16-18 feet to the gable from where the camera is located, which is why a roof rake came to mind.

  4. Trevor Fontaine | | #4

    Interior plywood meets the exterior plywood
    Can you explain in more detail (maybe with a diagram) how there is an airtight seal between the 1/2 interior plywood to the exterior plywood?

  5. User avater
    Michael Trolle | | #5

    Interior to exterior plywood
    The framer cut strips of plywood which he taped to the top of the plywood wall sheathing at the perimeter walls, leaving the strips unfastened. (In my "Tearing down to start again" blog entry, there's a picture of the side of the house where you can see the tape line I just described. It's 18" down from the top of the roof.) Then he installed the roof trusses above the strips and nailed the strips to the bottom flanges of the trusses. Later on, after installing the balance of the interior plywood, it was a simple matter to tape the plywood strips at the edges to the interior plywood.

Log in or create an account to post a comment.

Related

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |