4' overhang on shed roof with floor to ceiling windows. 1

[Rich Zimmerman]

I'm wondering if one of you could comment on my roof detail. I'm going for a shed-style roof with a 4' overhang that's really "thin" looking. So the main/interior roof is framed with 14" TJIs and then I've nailed a 2x6 on the side that cantilevers over to provide the overhang. Not really a huge deal except for two things 1. there's a section with a 7' wide floor to ceiling window and 2. I'd like to do rim board headers above other openings to maximize insulation in the wall. (those are really sort of the same issue).

So here's the details I've come up with:
Where there are no openings below, the TJI sits right on the top plate. There's a continuous 2x10 rim board with a 2x6 blocking above that. There's a 2x4 squash block that transfers the load from the cantilevered 2x6 to the wall. (I could have used the rim to support the 2x6 but I was worried that it would shrink differently than the TJI and cause problems)

When there's an opening below we add 1 or 2 2x10s and then a short joist hanger to support the TJI off the header. The cantilevered 2x6 rests right on the header and rim.

What do you guys think? Is there a better way to do this? (FYI slope of the roof is 1.25:12)

 

[JAE]

I would keep the 2x10's plumb and kerf their tops to the angle instead of installing them at the roof angle.

(minor pet peeve: blocking in wood framing is typically shown with one diagonal vs. the two "X" diagonals you show here. Continuous members, such as your rim board and the 2x10's are typically shown as you drew them with full"X" diagonals)

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[Rich Zimmerman]

I would keep the 2x10's plumb and kerf their tops to the angle instead of installing them at the roof angle.

Yeah, I had them like that but I found it hard to find a hanger that skewed down (or up on the opposite side), wide enough for the I-Joist, and was only 8-1/2" high. Any higher and you start interfering with the 2x6 on the side.

(minor pet peeve: blocking in wood framing is typically shown with one diagonal vs. the two "X" diagonals you show here. Continuous members, such as your rim board and the 2x10's are typically shown as you drew them with full"X" diagonals)

Ha. Yep. Was in a hurry to sketch something out...

 

[jdgengineer]

I generally, don't love TJIs for these situations and prefer a solid 2x or LSL/LVL as the main joist / rafter. All the labor with the web-stiffeners, etc and support with a sloped seat seem to me to be a pain. Also keep in mind you can't nail the rafter tail into the narrow face of the TJI flange so all load is being transferred through the web-filler.

Can you use LSL or LVL joists instead? The material is more expensive, but you don't need the web-stiffeners, etc so labor will be faster (I would think at least). I typically wouldn't put in the squash block either as that is a lot of labor. I would design the nails to the LSL or LVL to take the reaction and not worry about transferring it in bearing. I would probably use an engineered rim instead of 2x to avoid differential shrinkage concerns. I typically only use squash blocks below columns that are supporting axial load where the shrinkage could pose a problem. In other conditions I haven't been too concerned, but I don't typically mix and match sawn & engineered like you are doing.

I would also likely use a single full-height block and cut out a notch in it at the 2x6 tail rather than a partial height Rim with blocking above. Both could work, but you'll need to add LTP4 clips or some other mechanism to transfer the shear from the roof diaphragm.

 

[txeng91]

Agree with jdgengineer. Weyerhaeuser only recommends this detail with a 2’ max overhang.

https://www.weyerhaeuser.com/application/files/5715/3806/1534/TJ-9001.pdf

Going continuous to the edge of the overhang with a solid joist is the best option. I’ve done similar overhangs with up to 12” nominal depth and they look fine. But if the architect is insists on a 2x6, solid joist with the same scab extension your showing is the best option. Pay attention to the deflections at corners of overhangs, with only a 6” depth, it might be hard to frame those conditions without excessive sagging. I’ve seen it be an issue before with just dead load on a 4’ corner overhang.

 

[Rich Zimmerman]

Actually Weyerhaeuser has a detail on

https://www.weyerhaeuser.com/application/files/5715/3806/1534/TJ-9001.pdf

page 11 detail F1 where they go out 4' with basically my same detail.

I've modified my details based on your suggestions. I'm tenatively sticking with I-Joists, but I've got a quote request out to the local lumber supply to see the cost delta to go with an LSL or LVL, and I'll switch if it's not cost prohibitive.

Here's my framing. I originally boxed my corners but I'm worried about what txeng91 said about sag in the corners, so I doubled the 2x6 on a diagonal and minimized the amount of load the rim will see there. Hopefully it works out OK.

 

[JAE]

What exactly is the joist hanger doing for you?

I'd still have the blocking, etc. vertical and not sloped as you have it.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[Rich Zimmerman]

Thanks.

So the second detail is where we're generally going over typical top of wall situations (no joist hangers or headers there).

The first detail is for one of two situations.
A. where we are over a full opening (i.e. floor to ceiling windows), in which case there is no wall below.
B.Where we have a "rim board header" per R602.7.2 (i.e. no header directly over the openings (window/door) below, so we spread the roof load beyond the openings. I like rim board headers because you don't have that giant header that eats up insulation space above openings. However, you do have to add joist hangers in those sections because the wall below is no longer load bearing (or doesn't exist as in situation A)

I'll make them vertical if I can find a face mount sloping joist hanger that isn't a custom order for 6 degrees down/up for I-Joist widths that doesn't exceed 8.5" high (I don't think Simpson makes one like that, Maybe Mitek does). I agree that vertical is a better situation. That'll be my next task once I clear up some time later today...

 

[txeng91]

If those corner overhangs are 4 foot each way, I think you should really consider increasing the depth on the soffit. You should consider running a 3d model to check forces and deflections. The outriggers nearest to the corners don’t have adequate backspan to provide support so you need a rimboard that spans from the corner to the first outrigger with enough backspan, which may need to be double or tripled up. I alluded to it before I’ve seen a situation with the same overhang dimensions and 2x6 outriggers have significant sag. Let’s just say lawyers got involved and nobody wants that.

 

[AaronMcD]

Check wind uplift. It can easily be 100 psf or more. You may need clips for 600+ lbs depending on the span length of the main joists. It will also be higher in the corners where you have less backspan. Also, a quick check unless I'm mistaken, with that kind of wind uplift you may be getting bending stresses of ~1600 psi.

http://www.hellodwell.design

 

[Rich Zimmerman]

If those corner overhangs are 4 foot each way, I think you should really consider increasing the depth on the soffit. You should consider running a 3d model to check forces and deflections. The outriggers nearest to the corners don’t have adequate backspan to provide support so you need a rimboard that spans from the corner to the first outrigger with enough backspan, which may need to be double or tripled up

Yeah, that's probably true. I actually do have a callout on the main drawing stating that there can be no breaks for 12' beyond the corner. I'll re-run some of the calcs. None of them seemed out of line, but you get a bunch of deflection stack-up in these situations which might result in more deflection than we want. Looking at it right now I'm not sure how I got away with not doubling the rim at the corners.... I might try with boxing the corners rather than running the 2x diagonal to see if that makes things better or worse. My suspicion is that you're always better picking up the corner with a direct support rather than cantilevering the corner through the rims with a typical boxed corner, but I could be wrong. I also like the look of the diagonal if they decide to expose the underside, although right now it's called out to cover it.

Check wind uplift. It can easily be 100 psf or more. You may need clips for 600+ lbs depending on the span length of the main joists. It will also be higher in the corners where you have less backspan. Also, a quick check unless I'm mistaken, with that kind of wind uplift you may be getting bending stresses of ~1600 psi.

Wow, 100psf for uplift? I didn't get anywhere near that. I got 24psf for uplift on the roof in general (well, the outside edges), and 16 psf directly blowing upward, so 40psf total. That's Cat II, Exposure C, 115mph, 12' off the ground, some terrain having to do with being on a hill (Kzt=1.03). I think clips are a must though. Thanks....

 

[AaronMcD]

I'm getting 25 psf for velocity pressure:
V = 115
Kd = 0.85
Kzt = 1.03
Kh = 0.849
qh = 0.00256*0.85*1.03*0.849*115^2 = 25.2


For overhangs on low slope roof, GCp = -2.8 at corner and -1.7 at edge (10 sqft) (figure 30.4-2A)
The wall GCp = +1.0 (figure 30.4-1)
So for the overhang I get 67.9 psf at the edge and 95.6 psf at the corner.

Edit:
ASD - so I guess you get to reduce the wind by 40%
So yeah, 41 & 57 psf at edges and corners

http://www.hellodwell.design

 

[Rich Zimmerman]

So you did C&C, but (correct me if I'm wrong), for uplift on overhangs MWFRS is more complete(???) I got the same qh, but my Cp in the corner is -0.9 (27.4-1) and -0.8 for the directly applied (27.4.4). My G is 0.85 (26.9.4).

So we get Pext = 25.2*(-0.9 * .85) = 19.3 and Poh = 25.2 * (.8 * .85) = 17.1

The 24psf in my previous post included the Pint of -4.4, which shouldn't apply to an overhang. So the total is actually 36.4psf.

 

[Rich Zimmerman]

You know what, I think I'm just wrong here on MWFRS vs. C&C. I'm going back and redoing everything now. I now think 2x8s will be required at a minimum...

 

[AaronMcD]

I'm pretty sure C&C is for designing individual parts, like roof joists, studs, cladding. And MWFRS is for overall stability of the structure as a whole.

http://www.hellodwell.design

 

[Rich Zimmerman]

So you're correct.. Damn. I really thought I could make this work with a 5-1/2" deep overhang.

 

[HouseBoy]

If it's a stress problem, consider LSL outlookers (if thickness is really a big deal - like it is with many of my architect clients!).
I think you can be a bit "un-conservative" regarding deflection in this condition.