Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations SDETERS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Washington State---Broken Roof Trusses

Apr 1, 2025
1
Looking at a job for a homeowner in Washington State. A tree limb fell on his roof a few nights ago. Pictured below are the damaged members. The first two pictured I'm fairly confident I can splint with 2x4 or gusset (let me know if you disagree). The last picture, the member on the right, is broke in two places, the obvious closest to the eve and the other under the connector plate. The one beside it is cut entirely in one place. Chords are about 14ft long.

Should I recommend full replacement of the trusses in the last picture? Can a 2x4 splint on either side of the chord running full length work? What do you think?
 

Attachments

  • 20250331_142134.jpg
    20250331_142134.jpg
    1.3 MB · Views: 35
  • 20250331_142102.jpg
    20250331_142102.jpg
    5.4 MB · Views: 33
  • 20250331_142059.jpg
    20250331_142059.jpg
    3.4 MB · Views: 36
Replies continue below

Recommended for you

Think you need to expose the entire truss without any of that insulation to check the rest of the connection and joints is still sound.

What is the plan for the roof? Remove the external tiles and what looks like plywood sheathing?

repair from inside looks tricky.
 
Just put it back the way it was.

If the top chord was #1 SYP (unlikely), then scab it with the same or better wood with sufficient engagement beyond the fracture plane. Use two members instead of one, to account for the workmanship of a typical handyman. Sawn lumber is cheap, and it’s usually not too tough to haul two lengths into the attic from below.

Mitek (and others) provide standard truss repair details. Look at those to get an idea. Don’t use them if their assumptions aren’t valid for your situation.

I’d probably be recommending replacement of that last truss. Yes, it’s invasive and expensive and done from above. Good luck analyzing that truss without proprietary truss-manufacturer software or your own TPI spreadsheet. Paragon is good but cost-gated for anything more complicated than D+L.

Splint is a medical term. Might be regional jargon in WA, but I’ve always used “scab.”
 
I feel like all can be fixed in place. Replacing a truss is easier said than done. Measurements have to be extremely accurate along with all the other logistical issues. There is usually an interior wall that can be made to work for support if needed. I have specified replacement a few times but that was when a bunch of trusses were severely damaged.
 
I'm also on team repair for all of the trusses. That said, it has to be done in a way that pays homage to the original design:

1) Moving the forces in an out of the joints between the webs and chords in a compatible way.

2) Acknowledging that the chord will have been designed with a flexural continuity that may be difficult to restore.

My first stop would be to lay eyes on the heel joints. Those are the hardest working joints on most trusses. If the heel joints are damaged, the effort of repair will increase markedly.

The plated joint that we see in the photos won't be working hard at all so that ought to be easy to deal with.
 
For what it's worth, I asked about the analysis because I'm very suspicious of this post. New member, classified as a student, talking about repairing a roof truss. Pretty sure this is a contractor looking for free engineering advice. I could be wrong and welcome proof to the contrary. But if the OP is a contractor, they should be hiring an engineer to design this repair or contacting a truss manufacturer that provides that service, not seeking free advice here.
 

Part and Inventory Search

Sponsor