DuPont takes exception to criticism of housewraps
In Joseph Lstiburek’s sidebar to the article “Rain-Screen Walls: a Better Way to Install Siding” (FHB #137, p. 89), Mr. Lstiburek claims that housewraps do little to reduce air infiltration. As an employee of DuPont, the manufacturer of DuPont Tyvek HomeWrap, Tyvek StuccoWrap and Tyvek CommercialWrap (all secondary weather barriers), I especially find this remark in error.
A study that was conducted in Charlottesville, Virginia, compared a vinyl-sided home wrapped with DuPont Tyvek HomeWrap to a vinyl-sided home that was not wrapped with any secondary weather barrier. The house wrapped with Tyvek showed a 28% reduction in air leakage.
A second study was conducted in Sacramento, California, comparing the effects of one layer of DuPont Tyvek StuccoWrap vs. two layers of grade-D 60-minute paper under EPS foamboard plus a one-coat stucco. The homes wrapped with Tyvek HomeWrap showed a 13% reduction in air leakage. In fact, this study was so persuasive that the California Energy Commission now provides builders energy credits when they wrap homes with Tyvek HomeWrap or Tyvek StuccoWrap.
Mr. Lstiburek then makes another incorrect assertion when he compares housewraps to carpets and claims that plain dirt can cause housewraps to lose their water repellency. The water holdout capability of DuPont Tyvek is a result of the unique fiber structure of the fabric, not the result of a chemical treatment of the surface. Scotchgard-coated fabrics are re-treated after cleaning to replace the coating that is lost during the cleaning process. DuPont Tyvek is inherently water resistant and does not rely on coatings for its water repellency.
—Todd A. Apple, North American business manager, DuPont Tyvek Weatherization Systems, Wilmington, DE
Joseph Lstiburek, author and building scientist, replies: With regards to the Charlottesville study and the effectiveness of Tyvek for reducing air infiltration, I’m afraid my experience just doesn’t support your conclusions. My company has been involved with the construction of more than 4,000 homes through the Building America program, and we’ve tested more than 1,000 of them for air leakage. These homes have experienced an average reduction in air leakage of 50% over conventional construction. Not a single one of these homes has housewrap. However, all the conventional homes we tested to establish the baseline for air leakage had housewrap. Many of them had Tyvek.
One of our builder partners is a production builder in Chicago who constructs approximately 1,000 houses per year. In 1998, we tested the air-leakage characteristics of this builder’s typical home. Approximately 25 of the builder’s houses were tested to establish a baseline for the study. The baselined houses all were constructed with oriented-strand-board (OSB) sheathing and Tyvek housewrap.
We then re-engineered the builder’s construction methods by instituting draft-stopping at fireplace enclosures, cantilevers, interior soffits and bathtub locations on exterior walls. We also relocated ductwork out of exterior walls and out of attic spaces into the conditioned area. And we draft-stopped the floor area of second-floor bedrooms over garages. We also replaced the OSB and Tyvek housewrap with foam sheathing taped at the seams.
A 200-house subdivision was constructed. Approximately 20 of these homes were tested for air leakage. These homes experienced an average reduction in air leakage of over 50% compared with the baseline homes.The builder subsequently adopted all of the draft-stopping measures and ductwork-relocation measures as a standard for all of the future construction. However, the builder dropped the foam sheathing and went back to the original practice of OSB and Tyvek. Approximately 20 of these homes were air-leakage tested. Guess what? These houses were just as tight (or just as leaky) as the homes with the foam sheathing that were 50% tighter than the builder’s original 1998 baselined product. The obvious conclusion was that it wasn’t the Tyvek or taped foam that was responsible for the huge improvement in air leakage but the draft-stopping and duct-relocation measures.
The Sacramento study cited by DuPont intrigues me. Now, I don’t want to state the obvious that air doesn’t leak through stucco, but I guess I have to. Air doesn’t leak through stucco. Whether or not a 60-minute paper or Tyvek is under stucco is irrelevant to the air leakage of the stucco. It’s the other components of the building that leak air. However, the fact that DuPont convinced the California Energy Commission that Tyvek was the causal factor is impressive.
With respect to the other matter that is raised in the letter, I would like to remind everyone that dirt is a contaminant. Contaminants on surfaces affect surface tension. This statement is true regardless of whether the surface is treated with a hydrophobic coating such as Scotchgard or if it is “inherently water resistant” such as Tyvek. Dirty Scotchgard and dirty Tyvek lose water repellency.
Preprimed siding is a must
Your February/ March 2001 issue has an article about rain-screen walls and moisture-related paint problems (FHB #137, pp. 86-91). I experienced a similar problem with peeling paint on a house we built and traced it to wind-driven rain and cedar siding that had weathered. We had applied unprimed cedar lap siding in the summer and painted it one month later. It peeled like mad after the first few rainstorms. To solve the problem, I contacted the U. S. Forest Service’s Forest Products Laboratory (608-231-1361) and found that they publish a book titled Finishes For Exterior Wood. Their research indicates that wood left untreated will quickly lose its ability to hold paint if exposed to sunlight due to ultraviolet radiation. After re-siding the entire house (gulp!), we got it right. Now we use only preprimed materials and keep a spray can of primer available for the cut ends. The rain-screen system is a good method for windy areas, but every house needs preprimed siding (all sides) for long-term durability of the paint finish.
—Thomas R. Payne, via e-mail
The great collar-tie debate
The only thing worse than Mike Guertin’s answer to the collar-tie question (FHB #137, “Q&A,” pp. 24, 26) is the arrogance with which he presented it. I hope he is prepared to pay for all the roofs that collapse because of his answer. Whether mentioned in CABO or UBC or not (there are many things not mentioned in the codes), collar ties can serve two purposes.
Yes, they can help to prevent outward thrust. Although that help might be minimal, it could make the difference in a situation where the ceiling joists are not adequately fastened to the rafters. Assuming the collar ties and rafters are of sufficient size, are sufficiently nailed and are placed low enough, they could actually replace the ceiling joists, rendering a higher ceiling without higher walls.
More important, collar ties also can add strength to the rafters, allowing for a smaller rafter. Assuming the collar tie is placed in the middle of the rafter, the effective span is reduced by one-half, kind of like a simple roof truss. (That’s why they can make trusses out of 2x4s.) Up here where the snow can get quite heavy, a collar tie, even in a situation where the ceiling joists are adequate for outward thrust, can give a pair of rafters the added compression strength to make it through a heavy-rain-soaked snowy winter.
To offer a blanket answer to the usefulness of collar ties, especially given the lack of information about all reader situations, is to give a foolish answer. And any structural engineer worth the title would tell Mike that.
—Peter K. Martel, Bennington, NH
Thank you and Mike Guertin for setting the record straight on collar ties. I am a structural engineer, and I do a lot of residential work. I have spent the past 20 years telling every architect and framing carpenter I meet that traditionally installed collar ties are at best head-knockers and at worst a waste of good lumber.
Traditional collar ties are placed well above the plate line—usually about two-thirds to three-quarters of the distance between the ceiling and the ridge—and are too high to do any good. In fact, collar ties can actually overstress rafters if the collar ties are the only system preventing lateral translation (spreading) of the rafter ends.
To demonstrate this, I performed some analyses of a typical roof system. I assumed a house 20 ft. wide, with a 5-in-12 pitch gable roof, with 2×6 rafters spaced 2 ft. o. c. If there are ceiling joists parallel to the rafters that provide adequate ties, the rafters are stressed to two-thirds of their code-allowed capacity under normal live and dead loads. If the ties are raised to one-third of the way toward the ridge (about 1 ft. 4 in. above the plate) and it’s assumed the ties occur at every rafter, the rafters are overstressed by 47%. If the ties are placed two-thirds of the way up (a location that is more typically found), the rafters are 235% overstressed.
Obviously, collar ties do not work. The key to the CABO allowance of “rafter ties” is the statement that they must be “located as near the plate as possible.” In the above example, they would have to be less than 1 ft. from the plate line to meet code.
—Ashton B. Avegno Jr., P.E., via e-mail
Thanks for encouraging women in the trades
I am writing to compliment your magazine in general and in particular to respond to Tom O’Brien’s article in “Cross Section” where he both acknowledges and encourages women in the trades (FHB #138, pp. 54, 56). For nearly seven years, I have been working for a local contractor. On a daily basis, I tackle tasks such as electrical installations, framing, flooring, drywall, painting and overall troubleshooting. I also happen to be a 135-lb., 35-year-old woman. I absolutely love my job and am proud to be a part of this profession
I am an avid subscriber to your magazine for several reasons. My work is so varied that I find FHB to be an invaluable reference source. Also, in my experience your tool reviews are right on the mark. I have frequently based my purchases on your recommendations. Last, and most important, the fact that you do include women in your publication encourages me to believe women are beginning to be accepted as a legitimate part of the building trades.
—Kerry J. Kurtz, Flagstaff, AZ
Painter’s plea: Please don’t grout my baseboards
I would like to point out a pet peeve in the article “Grouting Tile” (FHB #138, pp. 68-73). The photographs show grout slopped all over the finish-painted surfaces. As a painter and perfectionist, this sight just makes my skin crawl. Many a time I’ve been told that “it just washes off,” but what you are actually doing is wet-sanding those beautifully painted baseboards, scratching them and dulling the gloss, or leaving nice sponge-wide stripes around the shower enclosure. I’ve lost count of the miles of trim and acres of walls I’ve repainted, mostly for free, always after I’m “done” with a job because of this technique.
I don’t get paint on your tile; please don’t get grout (or mud, or boot marks, or ink) on my paint. A tape gun costs about $25 and is well worth the investment. And if you’re on one of my jobs, I’ll gladly either give you one or mask off your work for you.
—Jon Tobey, Monroe, WA
Putting new shingles over old shortens the life of the shingles
Despite my reservations for roof-over jobs (new shingles over old), Stephen Hazlett’s “Reroofing With Asphalt Shingles” is a good article (FHB #138, pp. 84-89). I have shingled homes for 27 years and seen many roof-over jobs. Most have failed within the first five to eight years. The only reason to do a roof-over is cost, and the customer will usually go for the cheap price eight out of ten times. Your readers should know that many shingles are not recommended for roof-over jobs, and some manufacturers will void their warranty if shingles are used this way. Roof-overs almost always shorten the life of new shingles. Here’s an example.
A number of years ago after a severe hailstorm, I reshingled a 10-year-old home with new 25-year three-tab shingles. Three years later for a new owner, I added an addition to the back of the house and reshingled the whole house with a different color of 25-year, three-tab shingles. Because the roof was only three years old, I shingled over the existing shingles because the new owner did not want to pay to remove them. Eight years later, I was called back after a severe hailstorm to reshingle the house again. The insurance company wanted me to shingle over the existing roof. When I refused to put a third layer of shingles on this house, they finally agreed to stripping this roof. What I found is typical for my area, but the out-of-state adjuster could not believe his eyes. The new addition had minor damage. If it were not for valleys, I could have spent $45 for repairs (what the adjuster originally allowed). The older part of the roof was worn out (after only eight years). The top shingles had lost much of the granule covering. They were curling from heat buildup between the layers. The cupped shingles were destroyed by the hail.
Also, it was evident from the tearoff that parts of the roof had been leaking between layers for some time. There was little left of the original shingles (now 11 years old). They were hard and broke easily to the touch of a hand. In some places, there were even piles of shingle pieces under the top layer of shingles.
The moral of the story? If you want those asphalt shingles to last the 20 to 45 years my clients expect they will, you need to do the job right from the start—no roof-overs. On some brands of shingles, I can offer extended factory warranties. No shingle manufacturer offers extended warranties for shingle-over applications. This should tell you something.
—Jim Glover, via e-mail
Fine Homebuilding Recommended Products
Fine Homebuilding receives a commission for items purchased through links on this site, including Amazon Associates and other affiliate advertising programs.
Affordable IR Camera
Reliable Crimp Connectors
Handy Heat Gun
