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
Musings of an Energy Nerd

All About Larsen Trusses

A detailed history of John Larsen’s system for building thick superinsulated walls

Larsen trusses are non-structural. These lightweight trusses are tacked onto the sheathing after a house is framed and sheathed. In most cases, conventional 2x4 studs hold up the roof load, and the 2x4s are sheathed with plywood or OSB. The Larsen trusses get installed late in the construction schedule, after the roof is on.
Image Credit: Bruce Coldham
View Gallery 19 images
Larsen trusses are non-structural. These lightweight trusses are tacked onto the sheathing after a house is framed and sheathed. In most cases, conventional 2x4 studs hold up the roof load, and the 2x4s are sheathed with plywood or OSB. The Larsen trusses get installed late in the construction schedule, after the roof is on.
Image Credit: Bruce Coldham
This detail drawing of a Larsen truss appeared in John Larsen's 1982 pamphlet, “The Larsen Truss System.” The lengths are shown in millimeters; 210 mm equals about 8¼ inches, while 600 mm equals about 23¾ inches.
Image Credit: John Larsen
This illustration of Larsen trusses appeared in John Hughes' article, “Retrofit Superinsulation,” published in the April/May 1984 issue of Fine Homebuilding.
Image Credit: Fine Homebuilding
This photo appeared in John Hughes' 1984 article. Polyethylene was stapled on the exterior side of the clapboard siding before the Larsen trusses were attached.
Image Credit: Fine Homebuilding
In 1985, an article in Harrowsmith magazine featured this superinsulated house built by Alan Day and Sandra Day in Alberta. Equipped with R-40 Larsen truss walls, the Day house achieved an air leakage rate of only 0.89 ach50.
Image Credit: Harrowsmith
This illustration of Larsen trusses appeared in a 1985 book by Ned Nisson and Gautam Dutt, "The Superinsulated Home Book."
Image Credit: The Superinsulated Home Book
This photo of a house with Larsen trusses was used to illustrate Jim Young's “High Efficiency at Low Cost,” an article published in Fine Homebuilding #87 (Spring 1994).
Image Credit: Fine Homebuilding
This illustration of Larsen trusses appeared in Bruce Coldham's article, “Six Proven Ways to Build Energy Smart Walls,” published in the December 2009/January 2010 issue of Fine Homebuilding.
Image Credit: Fine Homebuilding
This illustration of Larsen trusses appears on a Web page maintained by the Building Science Corporation. Unlike many other energy experts, engineers at the Building Science Corporation are leery of Larsen trusses because they fear that the outermost layer of sheathing may be cold enough in winter to allow moisture accumulation and rot.
Image Credit: Building Science Corporation
John Larsen used plywood gussets installed in dadoes cut into the vertical 2x2s; the gussets were secured with air-driven nails and glue. Other builders have chosen simpler details; these Larsen trusses on Topher Belknap's house have OSB gussets secured with drywall screws; there are no dadoes.
Image Credit: Topher Belknap
These air-sealing details for a home with Larsen trusses were prepared by Coldham and Hartman Architects of Amherst, Massachusetts.
Image Credit: Coldham and Hartman Architects
If you are using blown-in cellulose or fiberglass to insulate a Larsen truss walls, the insulation tends to migrate sideways, making dense-packing a challenge. One solution is to install air-permeable netting on each truss to separate the stud bays.
Image Credit: Coldham and Hartman Architects
This detail drawing of a house with Larsen trusses was prepared by David Delaney of Ottawa, Ontario.
Image Credit: David Delaney
Inspired by the Larsen truss system, some builders use 2x4s and stand-offs to create a thick wall for insulation when retrofitting an existing home. That approach is shown in this photo of Gordon Scale's house in Adolphustown, Ontario. While Scale did not use Larsen trusses, the idea is similar.
Image Credit: Gordon Scale
Aidan and Johannah Van Dyk built a house near Ottawa that includes 9¾-inch-deep Larsen trusses.
Image Credit: Aidan and Johannah Van Dyk
Architect Rob Harrison took this photo of wall framing at a Passivhaus project (the North residence in Olympia, Washington). Although Harrison refers to this type of framing as Larsen truss, the load-bearing framing is actually a wall truss, not a Larsen truss.
Image Credit: Rob Harrison
Robert LaPorte of EcoNest offers workshops on building walls using a mixture of clay and fiber. The walls include framing inspired by Larsen trusses.
Image Credit: Eco Nest - Robert Laporte
A German manufacturer, LignoTrend, manufactures wall trusses that include air-permeable netting to simplify the installation of blown-in insulation.
Image Credit: LignoTrend
Although LignoTrench wall trusses are designed to be load-bearing, they can also be attached onto existing wall sheathing in a retrofit job, just like Larsen trusses.
Image Credit: LignoTrend

A Larsen truss is a type of wall truss used to build a thick wall — thick enough to provide room for above-average amounts of insulation. It was developed in 1981 by John Larsen, a builder in Edmonton, Alberta.

In honor of the 30th anniversary of the Larsen truss, the time has come for a definitive article on the invention. This report includes an interview with the inventor of the Larsen truss, a history of its use, and a discussion of its advantages and disadvantages.

Defining a Larsen truss

A Larsen truss is usually site-built. Because the truss is not required to bear any roof load, its components are light. The original Larsen truss consisted of two parallel 2x2s connected by small rectangular gussets of 3/8-inch-thick plywood. The gussets measured 6½” x 8¼” each and were spaced 24 inches apart. A completed Larsen truss looked like a ladder with rectangular plywood rungs.

Although early Larsen trusses were 8¼ inches deep, they can be built to a variety of depths. Many builders have made 12-inch-deep Larsen trusses.

Larsen trusses are designed to be attached to the exterior surface of the wall sheathing of a new home. In most cases, these homes were framed with conventional 2×4 or 2×6 studs. Larsen trusses can also be used in retrofit work, in which case they are installed on top of the existing siding.

Many builders confuse Larsen trusses with wall trusses. If a truss is designed to bear the roof load, it is not a Larsen truss; it’s a wall truss. For example, some builders create double-stud walls with the inner studs bearing the roof load. They may connect the two rows of studs with gussets in order to allow the outer studs to cantilever off the foundation. Such trusses are properly called wall trusses, not Larsen trusses.

“Larsen trusses are installed on the outside of the wall sheathing,” explains Bruce Coldham of Coldham and Hartman Architects. “In most cases, after the trusses are installed, you then install another layer of sheathing on the outside of the trusses. A double-stud wall only has one layer of sheathing.”

Robert Riversong is a Vermont builder who has developed a type of wall truss that he refers to as a “modified Larsen truss.” However, because the inner chord of Riversong’s truss is load-bearing, and because his system does not include two layers of wall sheathing, it differs significantly from the Larsen truss system and is more accurately described as a wall truss.

If you aren’t sure whether you are looking at a wall truss or a Larsen truss, here are some distinguishing signs:

  • In most cases, wall trusses bear the roof load and have just one layer of sheathing.
  • In most cases, Larsen trusses don’t bear the roof load, and homes with Larsen trusses have two layers of sheathing.

An interview with John Larsen

John Larsen still lives in Edmonton, Alberta. After working for decades as a renovation contractor, he is now retired. When I recently called him up for an interview, I felt like a medical reporter offered the opportunity to interview Dr. Heimlich about the origins of his maneuvre.

MH: How did you get the idea for your trusses?

John Larsen: “Back in 1979 or thereabouts, there was a fellow in Jasper who wanted to build a superinsulated house. The site was up in the mountains, and we were working in the winter. The sun didn’t reach the bottom of the valley until 10:00 a.m. and it set at 3:00 p.m., so we had a short window to do work. We were building a house with double wall framing, and I was trying to think of a way to get the framers to frame a house quickly, so we could do all the heavy work quickly, without having to build two walls. I got the idea of a truss that you could just tack on the outside of a newly framed house, and away you go.

“After I had this idea, I think it was in 1981, I made a set of trusses and put them on a house and the system worked. The truss wasn’t structural, so it didn’t have to be engineered. It’s a stand-off truss, attached to the structural components of the building. We never did have it engineered. We would install them on the outside of houses, and they were incredibly strong — they could be used for ladders.

“I had a farm outside the city, and I set up a little manufacturing line to start making up the trusses. I had a big saw to rip 2x4s in half to make long 2x2s, and then I dadoed the 2x2s out and put the gussets in — 3/8-inch plywood gussets. We attached the gussets with brad pins driven by an air nailer and glue.

“The standard depth of the trusses was 8¼ inches. All the trusses were the same depth; I would stick to my standard size. They were sized for 8½-inch fiberglass batts.”

MH: Why did you bother to dado the 2x2s to receive the plywood gussets?

John Larsen: “The dado made it easier to install the insulation. The insulation comes a full 24 inches wide. Having the plywood in a dado made the trusses real easy to insulate.

“I probably did about 50 houses myself using the trusses. Other people, the framers, would frame the house, and then I would come in to install the trusses as a sub-trade. It reduced the work of building a thick wall to a sub-trade.

“I sold information packages on the trusses — a brochure and a sheet of blueprints that you could insert into your plans for getting a permit and showing the builders. I also I sold a bunch of trusses to other contractors. I was making the trusses until the late ’80s. After that there wasn’t a big demand anymore.”

MH: What makes your system easier than building a house with double stud walls?

John Larsen: “With a double-wall house, you still have a problem with sealing the vapor barrier. What we liked about my system was, you could frame the house normally. Let’s say it’s a two-story house with an outside wall that’s 19 feet high. You could wrap the whole house with one piece of 20-foot wide polyethylene, and then put the trusses on top of that.

“Regular framers didn’t get the concept of getting a house airtight. It was hard to re-train them sometimes. But with my system, you could get regular framers to frame the house, and we just came in later and put on the vapor barrier and the trusses.”

MH: Did you use your trusses on retrofit jobs?

John Larsen: “I did 2 or 4 retrofits. When we looked at the truss system we realized it would work either way, for new construction or retrofits. For a retrofit job, you can put them right on top of the siding, as long as there’s enough structural support to nail them on properly.”

MH: Do you have any other stories to share about the early days of the superinsulation movement?

John Larsen: “Not really. My story is a simple one: going to Jasper and trying to do a double stud wall and realizing it was a lot of work, and trying to come up with a better way.

“Nowadays people are talking about net-zero houses. I’ve visited Peter Amerongen’s Riverdale net-zero house. My feeling is there’s a lot of overkill in a net-zero house. My version of energy efficiency is to conserve my own energy. It seems to me they are designing for the three coldest days of the year. Designing a house so it will keep you warm without any heat at minus 40 might be overkill. I say, design for 345 days of the year, and the other 20 days just burn some junk mail in the woodstove. I still have my farm out west, and these days I try to build with recycled materials. That’s being energy efficient in a different way — not using as many new materials.”

Magazine articles from the 1980s

Several magazines reported on John Larsen’s invention. For example, an article in the November 1983 issue of Popular Science, “Superinsulated House Trusses,” included an accurate description of a Larsen truss (see the sidebar, “An article from November 1983”).


Excerpts from “Superinsulated House Trusses,” an article published in the November 1983 issue of Popular Science:“A superinsulated house needs at least a 12-inch space in the walls for fiberglass insulation. That gives an R-value of 35. The normal way to do it is to build double stud walls of 2x4s. But John Larsen, a contractor in Edmonton, Alberta, has come up with another way — and his requires much less site work. “His secret is a special wall truss he calls the Larsen truss. He makes the trusses in the shop by ripping 2x4s in half. He then cuts a series of dadoes in one side of each resulting 2×2. Then he joins two 2x2s with gussets of 3/8-inch plywood, each 6½ by 8 inches and spaced 24 inches on center. The gussets slip into the dadoes and are glued and nailed. At the building site, these trusses are nailed to standard 2×4-stud walls. (A polyethyelene vapor barrier and sheathing have been installed over the studs) … The trusses can also be installed over the siding of an existing house. “The Larsen truss sells for 55 cents (Canadian) a foot, but delivery beyond a few hundred miles of Edmonton is uneconomic. However, it is not patented, and it’s easy to build.”

The following year, John Hughes, a designer of passive solar homes and an associate of John Larsen’s, published an article, “Retrofit Superinsulation,” describing the use of Larsen trusses in the April/May 1984 issue of Fine Homebuilding. Hughes explained how builders could use 12-inch-deep Larsen trusses to add R-42 insulation to the exterior of an existing house.

Even back in 1984, energy-efficient builders understood whole-building concepts. Hughes wrote, “The new vapor barrier that is installed around the house will decrease the building’s air-change rate substantially. You might need to contact a home-energy consultant who can measure how tight your retrofit is with special pressurizing equipment [a blower door]. If your air-change rate is 0.4 per hour or less, you should install an air-to-air heat exchanger [an HRV] to exhaust stale indoor air and bring in fresh air from outside.”

Useful hint: 2x2s can warp

The following year, Harrowsmith magazine published an in-depth article on Larsen trusses, “Suitable for Framing: Superinsulation Made Simple,” in its April/May 1985 issue. The author of the article, Len Milne, interviewed John Larsen, who provided tips to prevent warping. “You shouldn’t cut more 2x2s than you can assemble right away, and if you are not going to use the trusses immediately, they should be strapped or wire-tied into tight bundles,” Larsen explained. “Keep them dry and out of direct sunlight. Badly warped trusses are not much fun to work with when you are doing the install.”

One of the Larsen-truss homes described in the Harrowsmith article, a 4,500-square-foot home built by Alberta school teachers Alan Dan and Sandra Day, achieved a high degree of air tightness. A blower-door test revealed that its air leakage rate was only 0.89 ach50.

Construction details

Site-built Larsen trusses usually vary from the classic details developed by John Larsen. Most builders don’t bother dadoing the chords to receive the plywood gussets, for example.

John Hughes preferred to follow Larsen’s recommendations. “The trusses we use have 2×2 chords and intermittent webs of 3/8-in. plywood,” Hughes wrote. “The plywood webs are dadoed into the chords. Most builders in our area use an 8¼-in. deep truss. Adding these to a sheathed 2×4 stud wall yields a frame that’s close to 12 in. thick, not including the new siding. This is a convenient thickness because the plywood used to line the rough openings for windows and doors can be ripped from 4×8 panels with little or no waste.”

Ten years later, in another Fine Homebuilding article (“High Efficiency at Low Cost,” FHB #87, Spring 1994), Jim Young wrote, “As Larsen originally built them, the trusses were made from 2×2 vertical chords that had been dadoed to accept 1-ft. squares of 3/8-in. plywood spaced every 2 ft. or 3 ft.

“I made the trusses a little differently: first, by using 2x3s straight from the lumberyard instead of ripping 2x2s from larger stock, and second by skipping the dadoes in the truss chords. The carpenters … attached 1-ft. squares of ½-in. plywood to the sides of the 2x3s with waterproof yellow glue and ring-shank nails.”

Young’s article includes many helpful details. “Trusses were nailed on the walls through the 2×3 chords into headers and rim joists [and were spaced] 2 ft. o.c.,” Young wrote. “Along top and bottom outside edges, the trusses were joined to each other with short lengths of 2x3s toenailed at each end into the 2×3 chords. Outside corners were a little tricky. An L-shaped assembly made from 2x6s was attached to trusses on both sides with 2x3s to make a solid outside corner. To anchor each truss at the top of the wall, the uppermost plywood gusset was extended 1½ in. and nailed to the adjacent roof truss. At the bottom of each truss wall, sections of plywood were nailed in from the bottom to complete the box that would hold the blown-in fiberglass insulation.”

Aidan Van Dyk, a Canadian software engineer, built a Larsen truss house near Ottawa in 2006. Because he wanted to space the fasteners used to attach each Larsen truss 4 feet apart (vertically) rather than 8 feet apart, Van Dyk planned ahead: he included horizontal blocking at the midpoint of his 2×4 wall. Van Dyk’s 9¾-inch-deep trusses hold R-31 fiberglass batts.

Many builders who build their own Larsen trusses end up scratching their heads when it comes to outside corners. In The Superinsulated Home Book, published in 1985, authors Ned Nisson and Gautam Dutt provide a caption to an illustration showing an outside corner on a house with Larsen trusses: “Floating external corner: Floating 2×4 — attach [the 2×4] first to [the projecting] sheathing, then apply and nail second sheathing [from the intersecting wall] to it.”

Most experts recommend the installation of a plywood “bottom plate” connecting the bottoms of the projecting Larsen trusses. In “Six Proven Ways to Build Energy Smart Walls” — still another Fine Homebuilding article discussing Larsen trusses, this one in the December 2009/January 2010 issue — author Bruce Coldham wrote, “The truss exterior is not supported by the foundation and is sealed with a plywood bottom plate. A plastic air barrier and a vapor-retarding membrane are wrapped around the entire outside of the sheathed 2×4 frame and are sealed to the sill plate or directly to the concrete foundation wall.”

Almost everyone who writes about Larsen trusses is tempted to comment on their surprising ruggedness. “The Larsen trusses are very strong,” wrote Jim Young in 1994. “When nailed to the house, they became a sort of ladder or scaffolding that wouldn’t budge, even when we jumped up and down on them. Although the original purpose of the truss system simply was to hold insulation, it became apparent that it adds a lot of rigidity to the building envelope, too.”

Whole-wall R-value is 38.5

In 1997, engineers at Oak Ridge National Laboratory measured the whole-wall R-value of a classic 8-inch-thick Larsen truss wall insulated with 8″ fiberglass batts attached to a sheathed 2×4 wall insulated with R-11 fiberglass batts.

As reported in “Clear-Wall and Whole-Wall R-Values for Tested Wall Systems,” an article published in the March/April 1997 issue of Home Energy magazine, the wall had a very respectable whole-wall R-value of 38.5 — better than might be expected by those who dismiss the insulating ability of fiberglass batts.

Poly in the middle of the wall? What’s up with that?

Twenty-first century builders may be put off by John Larsen’s descriptions of wrapping the outside wall sheathing with a very wide roll of polyethylene. After all, in the 30 years since Larsen developed his trusses, the reputation of poly has slipped a few notches.

It’s worth remembering, however, that the pioneering developers of superinsulation techniques in Alberta and Saskatchewan were all using polyethylene as an air barrier in the early 1980s. In a cold, dry prairie climate, interior poly works well. Other air-barrier methods — specifically, the Airtight Drywall Approach and the current darling of energy-efficient builders, taped exterior sheathing — came later.

In his 1984 Fine Homebuilding article, Hughes explained why it was perfectly acceptable to wrap poly around a house with Larsen trusses: “Research in Canada has shown that as long as there’s at least twice as much insulation (in R-value) on the cold side of the vapor barrier as on the warm side, no significant condensation will occur in the wall under normal circumstances.”

Advantages of the Larsen truss system

For builders of new homes — at least those who followed the traditional Canadian recommendation to use interior polyethylene as an air barrier — one of the most obvious advantages of Larsen trusses is that they greatly simplify the installation of the poly.

“When we were using double walls, we used to run the polyethyelene air barrier on the inner plane of the wall,” said architect Bruce Coldham. “More recently, we began putting the air barrier on the outside, by taping the sheathing, and we’ve gained confidence with that approach. However, we used to have an interior air barrier. With a double wall, that required a tortuous folding the polyethylene around the second floor rim joist. But putting up the poly is simple and clean with the Larsen truss system. It has a nice look about it. It’s aesthetically pleasing.”

Putting up with the learning curve

Larsen trusses have several disadvantages. One that is frequently mentioned is the steep learning curve. The problem was mentioned back in 1985 by Nisson and Dutt in The Superinsulated Home Book: “The main disadvantage of the Larsen Truss and similar systems is that they are new and very unconventional. They are commercially available only in western Canada and will usually have to be site-built. The first installation for any builder will undoubtedly be slow and probably quite expensive.”

More than two decades later, the problem persists. “Neither my general contractor, nor my builder had ever heard of a Larsen truss,” said Topher Belknap, a green consultant who built a Larsen truss house in Edgecomb, Maine. “I had to educate them, and we had many discussions about various details. This required not only knowledge on my part, but also a firm belief that this is what I wanted. It also required paying for someone else’s learning (both in terms of uncertainty and lost time).”

Another possible disadvantage: if you decide to fill your Larsen trusses with blown-in insulation, it may be hard to find an installer with enough experience to do the job well. “In 2008, I was able to examine the entire house with a borrowed infrared camera,” Belknap wrote. “Doing so pointed up many places where the insulation had settled (or never got) and there were now gaps in the insulation coverage. This required reapplying insulation (blown in from the inside).” When I talked to Belknap about the issue, he said, “The problems I had with the insulation were attributable to the fact that the cellulose would drift from one bay to the next, and some of the bays would never get filled.”

Experienced installers of dense-packed insulation have developed techniques to fill deep walls, including Larsen-truss walls and double-stud walls. However, depending on the skills of the insulation installer may be risky; some architects and builders prefer to install air-permeable netting on each Larsen truss to separate the stud bays (see photo below), a technique that facilitates insulation installation.

Another issue is the controversy about many types of thick walls, sometimes called the “cold OSB” problem. A document on the Building Science Corporation Web site states, “The truss wall system can achieve a very high whole wall R-value … and would be perform well in extreme climates provided the air barrier was detailed perfectly minimizing the high risk of air leakage condensation durability issues. The Larsen truss is time consuming to construct and susceptible to premature enclosure failures resulting from poor construction and detailing. … This wall system has greater risk for severe air leakage condensation since the sheathing is considerably colder than standard construction.”

However, there is a contradiction in this document; while the text refers to problems with Larsen trusses, the illustration depicts a wall truss, not a Larsen truss. It’s hard to tell whether these BSC criticisms actually apply to Larsen trusses, or whether they only apply to wall trusses.

One of the principals at the Building Science Corporation, Dr. John Straube, explained: “With a Larsen truss, cellulose, and sheathing on the outside, the sheathing is actually colder than ambient air because of radiation transfer. … Therefore there is a condensation risk for the exterior sheathing in this situation.”

Not all experts are as worried about the cold OSB problem as John Straube, however. To learn more about the issue, see How Risky Is Cold OSB Wall Sheathing?

Would you do it again?

Aidan Van Dyk is unequivocal. “I was quite pleased with the Larsen truss system,” he told me. “I’d do it again.”

However, some builders who have used Larsen trusses have moved away from the system. According to architect Bruce Coldham, “At the Moomaw house [in Williamstown, Mass.], using Larsen trusses turned out to be more expensive than my contractor imagined. I think the lesson we learned is that Larsen trusses tend to be more expensive than a double wall.”

Topher Belknap also learned a few lessons. “If I had to do it again, instead of using site-built trusses, I would just buy I-joists,” Belknap told me. On his Web site, Belknap wrote, “The trusses did represent a substantial chunk of labor (on site) and may have caused some problems with the insulation. I-beams are now commonplace.”

Wall trusses in Europe

At least one German company, LignoTrend, is selling wall trusses that can be used for load-bearing exterior walls in new construction or for retrofit jobs (attached to the exterior of existing wall sheathing like Larsen trusses).

The trusses are sold with air-permeable webbing already installed to facilitate the installation of blown-in insulation (see photos below).

Foam-free superinsulation retrofits

There is one type of project where Larsen trusses still make a lot of sense: a superinsulation retrofit job for a client who wants to avoid the use of foam insulation. If you have crossed spray foam and rigid foam insulation off your list of acceptable green products — if you’re trying to stick to insulation products like cellulose, blown-in fiberglass, or mineral wool — then Larsen trusses will work for you.

“Larsen trusses are a fine way of retrofitting,” says Bruce Coldham. “If you aren’t using foam — if you are using an insulation that is only R-3.5 or R-4 per inch — then you have to construct a containment cavity to get the R-value you need.”

Such retrofit jobs are expensive, of course. In many climates, predicted energy savings won’t be high enough to justify an R-40 retrofit job. John Larsen addressed the payback issue back in 1982, in a pamphlet called “The Larsen Truss System,” and his analysis is still relevant. “The question most often asked by people considering a superinsulated house is, ‘Is all this extra insulation and tight vapor barrier worth the extra cost entailed?’ In a location with high energy costs and a large winter heating load, the answer in an unqualified yes. In areas with low-cost fuel still available, or in areas with a smaller winter heating load, the answer appears less certain.”

Last week’s blog: “Job Sites in Maine, Part Three.”


  1. Bob Manninen | | #1

    use of sheathing over insulation
    Very nice article. I have a question as to the common practice of putting sheathing over the trusses. Are there are other common practices other than using sheathing? Could something like Rockwool be used for insulation (it's a little more moisture resistant than fiberglass and/or cellulose) and something like tar paper or a permeable building wrap be used instead? Or, is the sheathing used to prevent racking of the trusses.

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

    Response to Robert
    Some builders have installed housewrap and siding directly over the insulation, without sheathing the Larsen trusses. I don't recommend this practice, however, unless you are using a panel-type siding like T-111.

    Sheathing helps prevent the Larsen trusses from wiggling back and forth and improves the airtightness of the assembly.

  3. Doug McEvers | | #3

    Larsen Truss for energy retrofit
    It seems the Larsen Truss is the ideal candidate for superinsulating existing housing, more information on this system being used on retrofits would be welcome.

  4. User avater GBA Editor
    Martin Holladay | | #4

    Response to Doug McEvers
    If you want more information on retrofitting houses with Larsen trusses, see if you can get a hold of a copy of the the April/May 1984 issue of Fine Homebuilding. (You might be able to buy it used online.)

    You should also check out the first link in the "More Information" box. It's a link to a paper by Neal Carter with many details on retrofits.

  5. David Argilla | | #5

    Neal Carter Paper missing figures
    Great Article! I was wondering how to do the corners, most figures of the larsen truss leave them out. The linked paper by Neal carter is missing figure 5 and figure 7. Think you can ask if they are available? Pictures/figures are very useful.

  6. J Chesnut | | #6

    great article
    I was first introduced to larsen trusses by Tim Eian just a couple of years ago when he brought me into his office to help him with his workload. Tim detailed them for a retrofit design and my first impressions were great idea but (1) is an engineer going to have to sign off on their shear strength for holding up cladding and (2) contractors are going to price this through the roof. According to this article field experience seems to show no need for the engineering stamp but there still is the problem getting competitive bids.

    We have moved towards the wall assembly with the I-joists acting as the larsen trusses and the first sheathing layer functioning as the air barrier and vapor retarder (with a redundant vapor barrier paint on the inside for good measure.) We then require vapor open fiberboard to be the sheathing layer outside the I-joists.

    I don't know if it is anymore cost effective but I'm interested in the idea of using Homasote as the outside most sheathing layer. It is a vapor open material made of recycled paper and is readily available in our market. The only drawback I can imagine when using this product is whether there is a problem if it gets wet during the construction process. I imagine it only being used with a siding material over a vent space and covered with a WRB. Any thoughts?

  7. User avater GBA Editor
    Martin Holladay | | #7

    Response to David Argilla
    Concerning the missing figures from the Neal Carter paper: the link I provided brings up a scanned copy of the original Neal Carter paper from the 1980s. This historic (and rare) document was scanned by Marc Rosenbaum, and I don't know of another copy. If any GBA readers have an intact copy of the document that includes the missing figures, please let me know.

    Concerning outside corners: here's what John Hughes wrote in his 1984 FHB article: "To cope with outside corners, we've found it easiest to prefabricate a corner truss from two straight trusses. One side of this right-angle truss is then nailed directly along one edge of the corner of the house. This leaves a gap on the other side of the corner that we bridge with short truss sections nailed horizontally to the wall every 2 ft."

  8. User avater GBA Editor
    Martin Holladay | | #8

    Response to J Chesnut
    Experienced builders have a low opinion of Homosote sheating because it is so flimsy and susceptible to water damage. Of course, Passivhaus builders who follow European methods are always in search of vapor-permeable sheathing options -- these builders are likely to be attracted to some of the cheesier American products like Homosote.

    Personally, I'd rather use diagonal board sheathing than Homosote.

  9. Robert Riversong | | #9

    Larsen or Wall Truss

    Robert Riversong is a Vermont builder who has developed a type of wall truss that he refers to as a “modified Larsen truss.” However, because the inner chord of Riversong’s truss is load-bearing, and because his system does not include two layers of wall sheathing, it differs significantly from the Larsen truss system and is more accurately described as a wall truss.

    Most people are now referring to my modification as the "Riversong Truss", but the distinction is not as clear cut as Martin suggests. A structural truss is engineered to transfer gravity and wind loads through compression struts and tension ties to inner and outer chords. An insulation truss, whether Larsen or Riversong or any of the other varieties, is a non-load-distributing truss that is designed to carry only cladding and window loads on the exterior (and transfer wind loads to the structural frame).

    The primary modification of my system is to use the interior, load-bearing stud frame as the inner chord of the insulation truss by attaching an outer chord with gussets after the roof is in place. So the primary stud wall frame is the load-bearing assembly, just as on a conventional house, and the insulation cavity is extended outward. I always use a wide enough foundation to carry both inner and outer frame members so the tricky and often ugly cantilever is avoided.

    This hybrid system eliminates the extra layer of sheathing, uses half as much lumber for the deep cavity, creates a single insulation cavity to dense-pack (with cellulose), and puts the air barrier where I believe it belongs in a cold climate – on the inside, to stop the warm, humid air at its source – and makes it contiguous with the vapor retarder (VR primer). This does require the Air-Tight Drywall approach, but makes the entire construction process both quicker and far more resource-efficient (and hence less costly).

    The exterior can either be sheathed with diagonal boards or CDX (or fiberboard but never OSB), or simply enclosed with taped housewrap and clad with ¾" shiplap boards (pre-finished on all sides). This creates an extremely vapor-permeable assembly, with a high moisture storage and redistribution capacity, and virtually eliminates any moisture accumulation problems (assuming exterior detailing is well done). No rainscreen is necessary for this wall assembly as there are no moisture vulnerable sheathings and the entire assembly dries easily.

    My Riversong truss system can be seen here:

  10. Robert Riversong | | #10

    Sheathing or Not

    Sheathing helps prevent the Larsen trusses from wiggling back and forth and improves the airtightness of the assembly.

    If the trusses are tied together at the bottom with a plywood plate and tied to the roof framing at the top, as well as connected laterally around doors and windows, there should be no "wiggling" and no need for sheathing for rack bracing.

    The primary issue that determines the need for sheathing (other than code officials) is the type of siding. I would not install clapboard (especially fiber cement) without a solid backing, but 3/4" thick wood siding can easily span 24" oc framing without distortion - even with cellulose dense-packed behind it.

    The fewer exterior layers, the more vapor-permeable the assembly. The only other reason to consider sheathing is to make future repairs or renovations easier.

  11. Thorsten Chlupp | | #11

    Thanks Martin
    Thanks for honoring John Larsen's work Martin!

    He surely is an icon of superinsulation and I got a great chuckle out of his advice to burn junk mail in the wood stove instead of designing to the coldest day of the year. I will remember that this winter when it hits 40 below and I don't have to fire up my masonry heater yet - what was I thinking? Overkill - but I am still enjoying the hell out of it.

    Exterior sheeting - you know the problems related to high R-Value walls and low perm on your exterior sheeting. I have seen plenty issues in very cold climates at least. Also let's not forget that ideally the inside Sd-value must be at least ten times higher than the outside Sd-value for optimal drying potential. And - there is certainly no wiggling going on without exterior sheeting. What to use? I concur with Homesate not being a good choice. High perm European fiberboard is still unobtainable but would be ideal. Tyvek works but is not ideal either for obvious reasons. We are testing out SIGA Majcoat (34 PERM) right now, which might be the right solution. And even in an assembly like this I am not giving up my air gap/rain screen between the membrane and the siding as it makes things work just so much better (Overkill again?). And I tried it both ways ...:-)

  12. Dan Kolbert | | #12

    Ship lap? Sheathing?
    Robert - have you found the ship lap holds up well? Are you buying stock product of making your own? I've seen enough ship lap curling at the laps that I get nervous, but maybe that's a function of poor prep.

    Thorsten - tell us more about the "plenty issues" you've seen.

  13. User avater
    Albert Rooks | | #13

    Good stuff!
    Great article Martin.

    Thanks for both the interview and depth. There is a lot to learn from the development history and It's really sad that the momentum stopped in the 80's. If it had kept up we would probably have companies like Ligno Trend in the US and be able to pull good load bearing, or non bearing Larsen Trusses "off the shelf".

    Having watched two site made Wall Truss (bearing) jobs it seems that site built trusses are not efficient enough to last as a practice. It comes down to the fact that a site built truss is usually handed. The gussets being glued and nailed on the outside rather than center dadoed makes another sequencing step in keeping track of how the gussets fit window and door openings. It's cumbersome to handle the bottom plates when they cantilever over exterior foam.

    It seems clear to me that the issues would be solved by a good off the shelf product like the original John Larsen model: Non handed with centered gussets and the addition of "new models" of load bearing wall trusses with standard framing members inside in a few wall depths.

    It's interesting to note that Ligno Trend does offer both non, and load bearing trusses. They don't use a ply or OSB gusset, but instead use dimensional solid stock. Since there are no metal fasteners, you can pull a length and can cut it straight, or at an angle, at will. it's a great design and with more clients & projects demanding super-insulated envelops, we'll have some "hardy soul" venture into the business of offering them someday soon ( I hope).

    Attached are more pictures that I shot of the Ligno Trend trusses. (I've not cropped them. sorry for the poor quality). You can see the joinery and build-up method. they are really good looking components. I wonder how many builders would really be interested in the level of quality of Ligno Trend or the original John Larsen center dadoed truss?

  14. Thorsten Chlupp | | #14

    "Plenty Issues!?"
    Issues? Martin has blogged about this plenty. John Straub cautions about risky “cold sheeting” often. Brilliant Bill Rose educates us with his great work about sorption and basic laws of materials and moisture – “Cold stuff is wetter than hot stuff”. My ever curious mind has led me to take the time to physically inspect wall assemblies of old and new buildings for many years. And once too often I have seen wet plates and mold growth – and not just on old homes but also barley two year old 5-Star+ build homes. Bad and ugly stuff, conveniently hidden away behind a pretty wall. I am of course fortunate and live and build in a very extreme and unforgiving climate which create the worst conditions for building assemblies. Issues I can see in two years might take ten years somewhere else (or maybe they never develop? I am doubtful about that…).
    Plywood or OSB are a class 2 vapor barrier with a PERM under 1 in a dry cup state. And it is mainly made out of wood and will react to environmental conditions of its closest surrounding and try to find an equilibrium. Great features on the warm side – inside – not so good on the cold exterior.

    Take a basic analogy: Go for an easy run on a cold winter day. What will you wear? First we try a thick layer of fleece. Off we go, we are warm, we sweet and next thing we know there is a layer of frost on the outside of the fleece – but who cares, we are comfy. But bummer, it gets windy and now we are shivering and freezing cold. We need to stop that wind quickly to be able to warm up again – so let’s layer a windproof rain jacket! Great, now we are warm again and can run on. A mile later however we don’t feel so comfy anymore and will notice that a layer of ice is forming between our fleece and the inner jacket layer. Yikes! Okay, we are smarter than that. We get rid of the frost layer, take the fleece of and place the jacket under the fleece (VB!)– no moisture can get out, nothing can freeze and all will be good. Brilliant. Down the trail we go, warm and frost free – but ugs…it becomes uncomfortable pretty quick. Ok, time to quit, obviously running in the cold sucks. But wait, maybe there is a better solution after all…mmh? Screw the rain jacket, let’s use a permeable windbreaker on the outside. Now, that wasn’t that hard was it?

    Wet sheeting and walls are bad. Mold is bad. Dry rot is bad. We can be ignorant - but stupid? We know the answers for the longest time but choose nevertheless to build flawed systems which have to fail as they defy simple laws of nature. Our health is priceless … that is about the least we should consider when we design and build buildings in which we will spend a lot of time in and that should last a min of 80 years.

    Anyways, time to go for a run ….

  15. Robert Riversong | | #15

    Ship Lap

    I've used rough air-dried hemlock that I had milled 3-sides into 13/16" thick shiplap, and I've used standard pattern #105 spruce drop siding - with no problems. I use solid-color latex stain and pre-stain all six sides before installation with a second face coat immediately after installing each "lift".

  16. David Argilla | | #16

    Using 1x4's
    Any thoughts about using 1x4's sandwiching the gusset for outer leg of Larsen truss? I have a bunch of 1x4's stored away.

  17. User avater GBA Editor
    Martin Holladay | | #17

    Response to David Argilla
    If I understand correctly, you propose using two 1x4s instead of one 2x2 as the outer vertical member of a Larsen truss. These 1x4s will be separated by about 1/2 inch -- the thickness of the plywood gusset.

    I wouldn't recommend it, if only because it's hard to nail sheathing or siding to the 3/4-inch-thick edge of a 1x4.

  18. Dan Kolbert | | #18

    Thanks, Robert. Having done several fairly high tech projects over the past several years we're trying to simplify our details (and houses). Nice to see you back.

    And I agree with Martin - doubled 1x4 sounds only marginally more trustworthy than single 1x4.

  19. David Argilla | | #19

    Ah yes, that's an important component. Didn't think it through, thanks.

  20. User avater
    Ted Clifton | | #20

    Why not just Insul-lam?
    The Larsen Truss looks like a lot of work for a retro-fit, with a lot of potential moisture issues. Why not just screw Insul-lam™ or Nail-Base™ (now sometimes called Retro-fit SIPS) panels to the existing wall? One operation, and you are insulated and ready for siding. The EPS insulation can be ordered up in any thickness you desire, we have used up to 9 1/4" of EPS, for R-40, outside of an R-15 HD f/g batt in a 2x4 wall. Costs a lot less, and does a better job, air-tight, also, no poly required. R-55 total wall construction.

  21. Thorsten Chlupp | | #21

    Insul-Lam? x-Factor
    Factor x64 in Embodied energy might be a motivator against the foam...
    Costs in a high R-value wall is another reason: foam is more expensive.
    If out-insulating with foam be sure to stay in your safe ratios. This is extremely important. No. 1 quick fix to boost wall R-value I see in new construction and retrofits is adding 2"of foam on the exterior. No moisture buffering and drying potential...guaranteed to fail in my climate.

    Picture from a two year old wall assembly in very cold climate; from inside - out:
    1/2"sheetrock, 6 mil VB, 2x6 Std, R-21 Fiberglass batt, 1/2"CDX, Tyvek WBR, 2"Foil faced EPS, Vinyl Siding. Mold, Bottom plate with very high moisture contend = unhappy wall.

  22. Skylar Swinford | | #22

    Larsen Truss "Chainsaw" Retrofit Animation

    You included one of Rob Harrison's photos in your article, but you left out his awesome animation of a Larsen Truss "Chainsaw" retrofit!

    For those that are interested check it out:

    Thorsten, I know you make the most of your short building season so thanks for taking the time to impart your wisdom.

    Riversong, good to have you around again!

  23. Skylar Swinford | | #23

    Lamdaplus Larsen and Passive House Conference Proceedings
    In addition to the Lignotrend truss system that Albert posted (I think that was my favorite booth at the conference), I saw a photo of the Lamdaplus System during a presentation:

    Also, if you missed the 2011 PHI Conference in Innsbruck, I recommend picking up a copy of the "Conference Proceedings 15th International Passive House Conference 2011" at the link below (Don't worry the book is in English). or search elsewhere with ISBN: 978-3-00-034396-4

    This 600+ page bible is a compilation of all the conference presentations. Books from past conferences are also available.

  24. User avater
    Albert Rooks | | #24

    A questionable admission
    To a building material nerd such as myself, there is nothing sexier than a high quality Larsen Truss. (Either wall or classical non bearing.)

    There. I said it.

  25. Robert Riversong | | #25

    Back at Ya
    Ted Clifton: "The Larsen Truss looks like a lot of work for a retro-fit, with a lot of potential moisture issues. Why not just screw Insul-lam™ or Nail-Base™"

    Reason: a lot of potential moisture issues.

    A cold-climate wall system should ideally:
    - be relatively air-tight (and easy to keep air-tight over time)
    - limit outward vapor diffusion (mostly controlled by mechanical ventilation)
    - buffer (safely store, redistribute and release) excess diurnal and seasonal moisture
    - breathe (be vapor permeable) to the outside with no mid-wall vapor stops
    - have as few discrete layers as possible (improves breatheability and moisture release)
    - contain no highly moisture-vulnerable materials (such as OSB)

    The focus on cold sheathing is a diversion (unless the cold sheathing is also highly moisture vulnerable, such as OSB), because cold and wet is not a problem - only warm and wet (and aerobic) can lead to mold and decay organism growth. If sheathing picks up some moisture during the winter but can dry out quickly in summer or from solar radiant drive, then it's not a problem. Cellulose so effectively stores and redistributes moisture that it protects sheathing and framing from reaching high (dangerous) moisture content.

    OSB, foam board and spray foams limit or prevent moisture diffusion - both liquid and vapor - while fiberglass insulation has no moisture buffering ability. These are not good materials to use in a highly insulated wall assembly. Even plywood limits vapor diffusion, but has the salutary ability to increase its (wet cup) vapor permeance as it becomes more saturated. OSB does not.

    And polymeric housewraps can become problematic since they present a condensation surface often adjacent to sheathing and will trap liquid water. 15# felt diffuses liquid water as well as water vapor, and it's shingled application style allows better drainage and better integration with door and window flashings.

  26. User avater GBA Editor
    Martin Holladay | | #26

    Do walls have to breathe?
    Most building scientists would dispute the statement that "A cold-climate wall system should ideally ... breathe (be vapor permeable) to the outside with no mid-wall vapor stops."

    A properly designed wall with an adequate thickness of exterior rigid foam will not have "a lot of potential moisture issues." On the contrary -- ongoing studies continue to show that such walls are among the driest and most problem-free cold-climate walls ever studied.

  27. Bill Dietze | | #27

    Building Science Corp. Comments

    In the text of the blog, you write
    " A document on the Building Science Corporation Web site states, “The [Larsen] truss wall system can achieve a very high whole wall R-value … and would be perform well in extreme climates provided the air barrier was detailed perfectly minimizing the high risk of air leakage condensation durability issues."
    But isn't the wall in the BSC article something other than a Larsen Truss? A Wall truss perhaps - there is no inner sheathing, so do the BSC comments apply? Seems not.

    Bill D.

  28. User avater GBA Editor
    Martin Holladay | | #28

    Response to Bill Dietze
    Good catch! I have edited my blog to reflect your observation. Thanks for your comment.

  29. Steve Young | | #29

    Wood beams in Larsen structures
    I am starting to look into a Passive house and the Larsen Truss system sounds very interesting.
    I was thinking of a wall like this, from inside to outside.
    Drywall ADA - with vapour perm paints
    2x4 insulated "Chase " wall, with dense pack cellulose, sealed electrical boxes and, piping gaskets.
    OSB taped (and primed?) AIR BARRIER
    2x6 24" OS structural wall, filled with dense pack cellulose
    A permeable rigid sheathing of some sort. (Beaverboard-like substance?)
    2x14 I-beans filled with Cellulose, again)
    Another permeable rigid sheathing
    furring strips and siding

    Overall, does this sound OK? Any suggestions for easily obtainable, and inexpensive permeable sheathing? If I use wood I beams, do I have to worry about the OSB? Or should I assemble my own full height "C" beams (no dado) with plywood and a couple 2x's?

    PS I love this site. The articles, and subsequent discussions have been a real eye-opener as I wrestle with trying to come up with a super-insulated wall. Things have gotten a lot more complex since my days in Edmonton, Alberta in the late 80's when I last undertook a significant home insulation project.

  30. User avater GBA Editor
    Martin Holladay | | #30

    Response to Steve Young
    Q. "Any suggestions for easily obtainable, and inexpensive permeable sheathing?"

    A. There are several options, including diagonal boards (1x6s, for example) or structural fiberboard (for example, Buildrite fiberboard or G.P. structural fiberboard).

    Q. "If I use wood I beams, do I have to worry about the OSB?"

    A. Your question is unclear. There are several possible reasons to worry about OSB, but I'm not sure which worry you are referring to.

    Q. "Should I assemble my own full height "C" beams (no dado) with plywood and a couple 2x's?"

    A. I'm not sure what you mean by C-beams. If you want to build Larsen trusses, detailed information is provided in the article on this page.

  31. Steve Young | | #31

    Thank you for the links for the fiberboard.

    With regard to OSB, I am concerned that it will get very cold in the I beam of the larsen truss. I have read many times on this site and other's that OSB does not like to get cold because it absorbs water and expands/degrades - eventually failing. Are there other aspects of OSB that I should be worried about?

    "C-Beams". Larsen went through the trouble of rabitting the 2x2 to insert the plywood plates. Rather than going through that detail, I would simply attach the plywood to the sides of 2 2X2's with outdoor screws and construction adhesive), making a stretched "C" shape.
    I have concerns about dense packing these cavities, therefore, I might run the plywood the entire length of the truss - no open area between cavities. (although, I read about a mesh product in another blog at this site that is used to "harness" celulose - Insulweb, I believe).

  32. User avater GBA Editor
    Martin Holladay | | #32

    Response to Steve Young
    Many Passivhaus builders, including Katrin Klingenberg, have framed thick walls with I-joists. However, if you are worried about the OSB, why choose that route? The article on this page explains how to build Larsen trusses with plywood instead of OSB.

    It's possible to skip the step of dadoing the 2x2s if you want to simplify the construction of your Larsen trusses; the article and photos include examples of builders who skipped the dadoes.

  33. Steve Young | | #33

    Response to Martin H
    Thank you, again
    Yes, I have been in KK's house, along with the other PH's in Champaign-Urbana. I don't know whether she has had any problems with the I-beams' OSB over the years (I didn't know enough to ask at that time - it was all very new to me). I liked the wall used on the Stanton Farm, but without the expense that they incurred.
    I am trying to develop a wall that I can build on my own. I would contract the framing of a 2x6 walled house (using advanced framing, of course), then I could tackle the Larsen Truss system, additional sheathing, the inside OSB airbarrier and inside chase wall. At first, I considered the true Larsen Truss, but it seemed like a lot of extra work to make an I-beam that I could purchase. Then I remembered all the talk about how OSB has issues with cold weater, and figured that I would ask for other folk's expert opinion.
    These Blogs/Forums have been very educational. Every time I read something, it leads to another topic or detail. My tabs list keeps getting bigger and bigger.

  34. Mat Marcum | | #34

    Hello, All:

    I've got an 1887 home that is 2-course, structural brick that I'm completely renovating. The brick is in good condition with the usual issues that require pointing, but the bodies of the brick have no burst or crumbled so, it seems like a perfect candidate for Larsen trusses. I have a few questions, though;

    Off the top of my head, the most reliable air & vapor barrier for this circumstance is closed cell spray foam. Would it be better to use a 10mil poly draped and taped with a liquid applied window and door flashing as a tie-in? Then, just increase the depth of insulation to accommodate the missing 2" of spray foam?

    Is there a common alternative to the classic Larsen Truss that utilizes a different type of member on the wall-side of the truss? Something like a metal framing stud so that the surrounding area can be better insulated rather than having the entire contact area behind the wall-side stud being non-insulated? Perhaps a couple tiers of 2'x as stand-offs and then attach the LTs to that?

    I'm planning to use 4'x8' wood grain finish sheets by Hardie as the base of a board and batten system, and then attaching their 1'x for the verticals. That's a lot of weight. Should I create a small, perimeter footer and build a wall atop that? I'm reluctant to attach so much weight to just a two course brick wall of this age. Without the reliability of new brick and solid mortar, I don't want to risk the additional weight pulling my walls outward over time, or the bricks to which the Larsen's are attached being "Jenga'd" by the heavy-faced wall.

    Thanks for any/all advice/direction y'all are able to provide.

    1. User avater GBA Editor
      Martin Holladay | | #35

      Your suggested retrofit approach is unusual, and I don't recommend it. Every Larsen truss job I've ever heard of has been used on a wood-frame building, not on a building with structural brick walls. Your suggested approach raises structural issues, so if I can't dissuade you, your first visit should be to an engineer.

      The usual method for installing exterior insulation on an old brick building is to use continuous rigid foam, followed by synthetic stucco (a method called EIFS) or the siding of your choice. Here are links to two case studies of this type of retrofit:

      "Exterior Insulation for an Ugly Brick Building”

      “Deep-Dish Retrofits"

  35. Mat Marcum | | #36

    Hey, Martin:

    Thanks so much for your prompt reply! I have not seen any other bricks doing Larsen trusses, but I greatly dislike the appearance of stucco and my neighborhood association would never allow that to move forward as its not in the covenant to permit stucco. So, I'm stuck with brick, grey-tone stone or a fairly large selection of vertical siding. It's one of those "you can have any color you'd like as long as it's black" deals. Per your suggestion and an apparent lust for pain, I made some calls to structural engineers and I'm looking forward to meeting with one of them in the next few days. I'll try to share any interesting news on here in case anyone else would like to flog themselves. Thanks again!

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |