Philosophical Hip Roof Question

[XR250]

I am engineering a 38'x50' house with a standard 7:12 hip roof.
If I can keep all of the rafters from spreading by using the attic floor diaphragm, do the hips need to be designed as beams and do they need vertical support at the peak?
The hips are approximately 27' long in plan and the roof will be sheathed in 7/16" OSB.
The IRC requires the hips to be designed as beams if the pitch is less than 3:12 but also requires them to be supported at the peak regardless of the pitch - which does not make much sense if they are not considering them beams.

 

[Jerehmy]

How do you build it and keep the jack and hip rafters in place before the sheathing is installed if it's not connected at the top? I also imagine in reality, the hip takes some shear at the ridge with the jacks not being installed perfectly and from some fastener slip. Plus, it adds some redundancy.

 

[KootK]

If I can keep all of the rafters from spreading by using the attic floor diaphragm, do the hips need to be designed as beams and do they need vertical support at the peak?

I vote no on both counts, philosophically. With a rectangular roof plan, however, you will be depending on the diaphragm for gravity load imbalance and gravity load stability in the longitudinal direction. That, in addition to your usual lateral loads.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[XR250]

@Jerehmy
They usually put in temp. supports. I would design the hips to not deflect too much under the dead load.

FWIW, in the thousands of house I have inspected, I only rarely see issues with hip roofs - usually older homes with low slopes and cut up with spliced hips and valleys.

 

[msquared48]

If the attic floor diaphragm intersects the bottom of the hip, you should be able to design and install a collar tie to support the upper end of the hip.

Mike McCann, PE, SE (WA)

 

[XR250]

@Mike;

I don't understand your response.

Thanks

 

[msquared48]

I was merely trying to describe a framing situation that I have used in the past to support a ridge beam and hip beam intersection using a collar tied truss arrangement at the intersection. If the geometry is there with the "attic floor diaphragm' as mentioned by the OP in the first post, this could work.

Mike McCann, PE, SE (WA)

 

[XR250]

ok, got it.
thanks

 

[XR250]

I vote no on both counts, philosophically. With a rectangular roof plan, however, you will be depending on the diaphragm for gravity load imbalance and gravity load stability in the longitudinal direction. That, in addition to your usual lateral loads.

Just to make sure I am clear - you agree that they do not need to be supported and they do not need to be designed as beams?

Thanks

 

[KootK]

Agreed on all counts. I did an interesting mental experiment where I imagined all of the hips ridge "beam" segments between rafters to be pinned at each rafter. Then I asked myself if the system would sill stable like that, with no flexural continuity in the hip ridge beam. It would in my opinion. And I interpret that to mean that the hip ridge beam does not require flexural capacity.

I believe that you'll develop a fair bit of axial load in your ridge beams. How you'll estimate that load, design the ridge beams for it, and deliver it to supporting members are all interesting questions.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[XR250]

I guess the reality is that there is a bunch of different mechanisms going on to keep this roof standing. Folded plate action, arching action of the jack rafters across the building, hips acting as beams supported by the ridge which is supported by the main rafter/ceiling joist triangle. Which likely explains the rare failures we see with these roofs -even with only a sheetrock ceiling diaphragm.

 

[KootK]

Back in 2006, I reviewed a 1950's attic framed this way. It was a perfect square where the ridge beams actually came together like a pinwheel. It showed no signs of distress whatsoever. Deflected shape was probably a bessel function of some sort. They wanted to install several new dormers but I vetoed it. I wasn't willing to upset the delicate balance of whatever the load path actually was..



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[RFreund]

I don't believe they would need to be designed as beams, but the load path gets a bit complicated, especially if you ignore folded plate action (and I'm not sure how to account for folded plate action). Here is one possibility:

Lets say your ceiling joists run in the transverse direction and are not turned at the ends of the diaphragm. The rafters framing parallel with the rafters are no problem as you have triangular trusses. For the rafters framing perpendicular to the CJ's you must develop the outward thrust into the attic diaphragm. This would require some sort of strapping from the rafter into the diaphragm. Now the question is how far into the diaphragm must you go? If it were a collector in your LFRS, it would need to extend between chords (which in this case would be your CJ's). This depth would then define your "transfer diaphragm" which would transfer this force to the shear walls on the exterior of the building.

Another likely load path working beyond the ridge (as you move down the hip) is the CJ acting as a collar tie (tension member), then the short piece of rafter being in compression then you have "sheathing strips" in compression along the hip side of the roof.

I have used the technique mentioned by Mike in the past but then the hip would still be designed as a beam as it is spanning from wall to truss.

Maybe there is a better way though?

EIT

http://www.HowToEngineer.com

 

[XR250]

@RFreund:

You bring up some good points. What I normally do is put a double ceiling joist 4' from the edge and then ladder frame from there. That way all rafters get a ceiling joist lap and all the ceiling joists are tied to the diaphragm. I ended up sizing the hips for the dead load and 10 psf live load and added a post at each peak as I had stacked load bearing walls below that area anyway in the lower stories. On a smaller roof, I probably would not have added the posts.

 

[RFreund]

I had meant to address turning the framing in my previous post.

I agree that the ladder framing is the way to go. However you still have some transverse rafters that would then need to be 'strapped and blocked' into the diaphragm.

I think in reality the diaphragm sheathing can take a fair bit of tension but this is not allowed by code (NDS) or atleast not for diaphragm design. Which makes me wonder - If you were to apply general analysis principles/standards to the prescriptive IRC designs, would you need to assume that the floor sheathing takes tension load? (I don't mean to side track here, I'd have to do some major research into this).

I always dislike the case where in reality it is probably OK, but proving this is just not worth the effort. I suppose this is one of these cases.

EIT

http://www.HowToEngineer.com

 

[XR250]

I always dislike the case where in reality it is probably OK, but proving this is just not worth the effort. I suppose this is one of these cases.

…or proving it is about impossible. Lots of these cases in wood construction.