It's going to be interesting trying to get these pieces and parts and connections to work via analytical methods. You could maybe get the rafters to work via the code if you add purlins and purlin braces, but the ceiling joist won't have the interior bearing wall you want it to sit on. So the purlin brace is engineered and that force will be large and difficult to transfer. You've also got the wind loads to contend with.

There are a lot of houses with that same type of construction. Rafters with a half-ass collar tie. Every time I see a sagging roof or leaning walls when I drive by, I immediately think that's the type of framing that caused it. I think your proposed solution is actually pretty good. You could also use a tension rod to "pull up" the roof if you've already got significant sag.


I might as well mention a similar situation I'm running into for my own house. The garage "truss" feels a bit under sized. 1970's construction, spindly rafters and bottom chord with a single vertical in the middle. Probably fine for just the relatively light roof loads. But, the desire is to "finish" the garage. Add a drywall ceiling, maybe some hanging storage from the trusses.... And, there's no way the bottom chord could handle the bending.

My plan is either:
a) Leave the existing rafters / trusses exactly where they are and don't touch them. But, add some new trusses in between that will carry any new load from the drywall and storage.

b) Double up the top chords and bottom chords with a gap in between where I'll put in new diagonals to strengthen the truss. This is easier if I end up doing the work myself.

One difference between mine and yours is that I've got good plywood on top of the rafters. You've got some good wood boards there, but I don't think it ties it together as well (at least for lateral force distribution).

Maybe I'm being thick here but I don't see any problem with the proposed solution, assuming everything calcs out
Seems like the dead load is pretty minimal so the deflection as-built should be acceptable
You're then adding in additional structure to ensure that the dead + snow case works
Putting tension ties across with hangers from the ridge seems reasonable and pretty easy to calculate out and this whole thing would be straightforward to model

Is there going to be a ceiling running through?
Is lateral load transfer a design requirement?

Had a similar situation about 15 years ago. A friend wanted to remove a load bearing wall running down the center of the house. For all rafters, tied the ceiling joists together and added a vertical at the center to create king post trusses. Still working as expected.

Adapting that to your situation would be pretty much what you described, just on a larger scale:



To upgrade load carrying capacity of the existing rafters, create queen post trusses:

Thank you all so much. Nope, no ceiling running through, for lateral, it's essentially heavy timber construction, client has 9x6 posts connected to 10x4 beams. We're adding 4x4 knee braces every 8-ish ft, thinking the 10x4 beam and bracing will be adequate for this single story (about 4 kip per wall or about 1kip per post/beam connection for seismic). I wasn't counting on the roof to resist any seismic/wind load. The walls are very sturdy.

Sort of the opposite of what you're asking for (going from rafters to trusses) but a way to upgrade the roof structure and open up the attic without removing the roof...

https://youtu.be/ZArMma4O1IQ?si=Y8qysjZ1tYyw294g&a...

Might be easiest to just add new trusses between the existing rafters as mentioned above.

Is the roof showing signs of deflection and distress?

Are the eaves beams absolutely massive to prevent roof spread?

Not saying you shouldn't bring it up to code but you would expect either distress or there to be a legitimate load path.