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

Community and Q&A

Questions about the Larsen Truss for retrofit applications

GBA Editor | Posted in Green Building Techniques on

Using a Larsen truss system to retrofit an older home for increased wall insulation seems to have at least 2 primary benefits:
1. No foam insulation required, if you think that is an issue.
2. The ability to increase insulation without producing a vapor impermeable exterior wall surface. This could be a big benefit for a home that has an interior poly vapor barrier, and you don’t want to disturb the existing sheet rock, etc.
3. Reduced thermal bridging across the wall structure, again without a foam layer.

A few questions I have…

What other benefits have I left out?

Are there any reasons in particular that a truss system is used, besides to increase cavity thickness? For instance, consider a 2×4 walled home. To produce a wall that has a similar R value as adding 2 inches of foam sheathing to the exterior, would only need approximately 3 to 3.5 inches of densepack cellulose. Would there be any benefit to building trusses, and not using a solid strap, to achieve that size cavity? The truss would reduce thermal bridging over a solid strap, but the solid strap could be spaced to alternate with the existing wall studs.

Are the trusses attached to the existing sheathing, or only hung from the roof rafters, and stood in the bottom plate extension? If attached to the wall, are they aligned with, and attached to the studs, or just attached to the sheathing?

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. User avatar GBA Editor
    Martin Holladay | | #1

    I hope you track down copies of the articles and books I cited when you asked your last question. Used books are now readily available from dealers on the Web.

    Another advantage of Larsen trusses: They don't take up interior wall space.

    A disadvantage (in a retrofit situation): They don't work well on houses with stingy roof overhangs. Extending roof eaves and rakes can be a pain.

  2. Nathaniel | | #2

    Yup, already sent inter-library loan request for Superinsulated home book.

  3. Riversong | | #3


    Another advantage of exterior trusses over exterior foam, which you didn't mention, is that it provides solid nailing for horizontal siding, with or without additional sheathing.

    I don't know how you would build parallel-chord trusses that were only 3" deep, since you would need at least 2x2s (1½"x1½") lumber and that would leave no thermal break, in which case simply attaching 2x4's on edge would offer the same exterior insulation space and they can be staggered with the interior studs. Attaching to band joists and/or bottom and top wall plates would provide solid anchoring (might need long Timberlock screws).

    John Larsen imagined cutting off the existing roof overhangs and wrapping the entire building with trusses, including over the roof (creating new overhangs). This would provide a continuous air/vapor barrier (if installed exterior to the existing envelope) and a continuous thermal barrier surrounding the entire structure. Existing doors and windows could be left in place, if still serviceable, or replaced with new at the exterior wall plane (which would require only minimal interior finishing of the deep boxes).

    I can't imagine going to the trouble of building and attaching a trussed exoskeleton if it were not at least 6" deep (and preferably deeper). That may require extending roof eaves, but that could be done creatively, and it might offer attachment for rain gutters which are important for a durable structure.

  4. Riversong | | #4

    Larsen Truss References



    Superinsulation for the Masses, Mother Earth News, September/October 1986
    High Performance Wall Systems by Alex Wilson, JLC, April 1988
    Energy-Smart Walls by Bruce Coldham, FHB, Dec/Jan 2009/2010
    Comparison of Exterior Wall System Alternatives, Coldham & Hartman, NESEA 2007

    Superinsulated houses: A survey of principles and practice, William Shurcliff, 1981
    The Super-Insulated Retrofit Book, Brian Marshall & Robert Argue, 1981
    The Superinsulated House: A Working Guide, Ed McGrath, 1981
    Sun/Earth Buffering and Superinsulation, Don Booth, 1983
    The Superinsulated Home Book, Ned Nisson & Gautam Dutt, John Wiley & Sons, 1985
    Superinsulated Houses and Air-To-Air Heat Exchangers, William Shurcliff, 1988
    The Timber-Frame Home: Design, Construction, Finishing by Tedd Benson, 1997
    Energy-Efficient Building (Best of Fine Homebuilding), 1999

  5. Nathaniel | | #5

    You mentioned, " that it provides solid nailing for horizontal siding, with or without additional sheathing."
    What would you envision as a suitable barrier to blow cellulose behind in place of solid sheathing?

    Also, the link to the presentation is great, thanks.

  6. Riversong | | #6

    It was Bruce Coldham and I, working together on the design of a super-insulated affordable duplex community building project (Building Our Swords Into Plowshares) in 1993, who came up with the "modified" Larsen Truss idea. Bruce has gone back to the original Larsen Truss in his work, while I continued to work with the modified version.

    For retrofit, blowing cellulose from the outside is somewhat weather-dependent. You could theoretically blow through insulweb, but you'd have to get the housewrap up before the next rain, so using sheet-stock sheathing makes sense (it could be thin sheathing, however, if it's primary role is to contain the cellulose and the siding was nailed into the trusses).

    The alternative would be to use batts of Roxul, blue jean, or wool - but filling all the voids would be difficult unless the trusses were solid-web I-joists. Retrofits are always more problematic than new construction.

  7. Nathaniel | | #7

    Sorry, I had forgotten that you used Typar in the buildings described on the Build it Solar site. Clearly you felt that this would be a sufficient barrier to water infiltration into the cellulose, in combination with the horizontal siding, and well flashed openings, I'm sure.
    Have you had any opportunities to observe that that was true?
    Seems like that could be risky for a vertical siding, like Board and Batten. I suppose that leakage would eventually dry anyway, but repeated wetting of the cellulose can't be good.
    Have you used a typar type product for all the truss buildings you've done? Have any had a "hard" sheathing?

    I wonder if in a retro situation where super insulated was not the goal, merely well insulated (Like the 3 inch space I described above), if a layer of 1 in. EPS could be used as a thermal break, and the sheathing to blow cellulose behind it.
    It's fragile stuff though.

  8. Riversong | | #8


    I've used Typar on all my projects (brand loyalty, I suppose), but I've yet to sheath the exterior of any truss-wall houses, since I always spec 3/4" or thicker horizontal shiplap siding. I would never use board and batten siding on a house, as I don't consider it weathertight. I did use vertical shiplap (channel rustic) on my first Larsen Truss house (because of owner insistence), which required horizontal strapping and more lumber than necessary.

    You could use any number of sheathing options, however, including XPS (I wouldn't use EPS), fiberboard, or exterior gypsum.

    Cellulose insulation, just like solid wood, can tolerate regular wetting as long as its punctuated by periods of drying. Cellulose insulation can absorb and release up to 30% of its weight in water safely. Beyond that, it turns to paper maché.

  9. Riversong | | #9

    I should add that my favorite horizontal siding is 1x6 pattern 105 "novelty drop siding" (also called Dutch, German or Cove Lap siding) because of its excellent drainage pattern.

  10. Jon Senior | | #10

    I'm currently in the design phase of a timber frame (Post + Beam) house which will be "wrapped" with a Larsen Truss. I'm a little stuck with the roof though since the system needs to be structural in order to support the tiles and snow-loading. Trusses are the only option for the 260mm of insulation required, which leaves me with either I-Beams or some form of Larsen Truss. Trusses can be built on site and (since this is a self-build) will be significantly cheaper, but if I can't make them structural, then they are of no use. Any suggestions?

  11. User avatar GBA Editor
    Martin Holladay | | #11

    You have two choices:

    1. Buy roof trusses. Roof truss manufacturers have engineers on staff to meet any structural requirements you have.

    2. Build structural roof trusses on site. If you choose this option, you either need to understand the engineering parameters of truss construction or you need to have your truss design approved by an engineer.

    It's not clear that option 2 will save you any money over option 1.

  12. Jon Senior | | #12

    Thanks Martin. I should have explained in more detail. The roof structure is timber-framed, principal purlin, with a box on top containing insulation (essentially functioning like a SIP-based roof, but built up from parts on site). The structural requirements of the roof trusses are simply to transfer the load of the roof tiles (+ snow loads etc) to the purlins. The trusses thus need to be parallel beam. I've found wood I beams that will fulfil the role, but was hoping that I might be able to use a larsen truss in some structural way. As you rightly point out, this requires performing the structural calculations, and I'm not sure how to do that for a Larsen truss. I'll just chase prices for the I-beams I think. At least I can be confident that the roof will be suitably over-engineered that way.

  13. Riversong | | #13


    What kind of roof live and dead loads are you designing for? What is the o.c. spacing and size of the rafters and the purlins? And what is the required depth of the roof trusses (including ventilation cavity)?

    If you want to site-build your roof trusses, that should be easy. You can either use a Larsen Truss configuration with rabbeted-in plywood webs or a diagonal web configuration with plywood truss plates.

  14. Jon Senior | | #14


    European loading requirements (Eurocode 5). 0.9kN/m² dead load; 0.6kN/m² snow load, 0.6kN/m² wind load (down); 1.3kN/m² wind load(up). These are then combined with in various combinations with various multipliers and the proposed structure is tested against each. It's horrendously complex, but easily built into software. Finn Forest offer a free download of FinnWood which can be used to perform the calculations for their range of products.

    To answer those of your questions that I can (in a useful manner), the purlins are about 1.2m OC, but can be closer, the rafters are in the bents and are spaced unevenly at distances between 3.5m and 5m. A layer of wood (wainscotting) will form the internal surface of the roof structure. The trusses (sized to allow 26cm of fibreglass insulation + a 2cm air gap) go on top of this running up the pitch of the roof. Finally a water proof membrane is topped by the tile structure.

    These requirements can be met and exceeded by the FJI38-300 (38mmx300mm wood I-beam) at 90cm OC. I'm planning on 60cm OC to ease the installation of insulation, so we're well within tolerance.

    To be honest, I think I've talked myself into using the I-beams for the roof. If I ever get pushed for proof that the structure can support the necessary loads, I can point to the calculation sheets, which will be much harder with home-made trusses. I'm already planning on home-made trusses for the wall skin and that's a significantly larger proportion of the structure.

    Thanks for you help guys, and for the wealth of information that you've left lying around the net.A few years ago I would never have even considered this to be a feasible project.

  15. Riversong | | #15


    So you're on the other side of the pond. You're on your own, buddy. The only Newtons we deal with here are Fig Newtons (except when I wear my other hat as a rope rescue instructor).

  16. Anonymous | | #16

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |