Flexible Steel Roof Deck Diaphragm Continuity of a Gable Roof

[jochav52802]

Good Day!

The attachment is a typical Vulcraft detail, however, the "RIDGE PLATE" is intended only to provide something for the rigid insulation and roof system to rest on/attach to, and not to provide diaphragm continuity. In my case, the ridge beam wide-flanged member is not a collector, so my intent is to just connect the (2) diaphragms together so that they'll act as one continuous diaphragm.

Your support with the following questions is greatly appreciated!

1) I'd appreciate any support in understanding what standard practice is for designing a flexible metal roof deck diaphragm to maintain continuity over a gable ridge. I would think the ridge plate could be designed to provide this continuity, but am not sure where to start on the design front; any tips are appreciated. It'd also seem best to place the ridge plate beneath the deck so that the diaphragm's shear plane doesn't shift up to the top of the deck.

2) Since the diaphragms are inclined, when a lateral load is applied perpendicular to the ridge beam, in addition to the lateral force traveling through the plane of the diaphragm, an additional upward and downward load component will be created on the nearside/far-side joist seats that will tend to twist the ridge beam. Is that usually torsion usually accounted for, and if so, how it is typically supported? I'm guessing it's probably small enough to ignore particularly with low-sloped gabled roofs.

 

[DaveAtkins]

I think the ridge plate will be 22 GA or 20 GA, so it should be adequate for transferring the diaphragm shear. If your ridge is at the center of the building and midway between shear walls, the diaphragm shear will be practically zero.

DaveAtkins

 

[jochav52802]

Thank you DaveAtkins,

That's my thought too, but I'm trying to calc it out to show this for sure, especially since my ridge is offset from the center of the building.

I'm thinking this is the approach to take, (see my attached updated sketch):

1) Provide a bent plate welded to HSS blocking at the side lap connections between the joist seats on the ridge beam to take the vertical force component of the inclined diaphragm force down to the ridge beam (otherwise that force would rely on the deck's out-of-plan weak axis for resistance), and connect the diaphragms to the bent plate so the lateral force can transfer directly between them.

2)Use the shear flow equation to determine the shear at the diaphragm-to-diaphragm connection and check that the side lap connection strength/spacing is sufficient, (I'm thinking it should work by inspection and be essentially the same result as if the roof was flat as it seems the only difference is the addition of the vertical force resultant created by the inclined roof, which I removed via step-1 above.)

 

[DaveAtkins]

I think the ridge plate on top of the deck is adequate. I don't see the need for the additional plate and tube. What vertical component are you talking about?

DaveAtkins

 

[jochav52802]

Hi DaveAtkins, please see my attached sketch. The vertical component is a result of the horizontal wall anchorage load passing through the inclined gable roof diaphragm.

Since my gable roof ridge is not centered on the building and since the windward and leeward wall anchorage forces are not the same, the vertical components from both sides of the ridge joists will likely not offset one another leaving some net vertical force left over for the ridge beam to support.

Additionally, since these vertical components will be applied in opposite vertical directions at the joist seats sitting on the wide-flanged ridge beam, this will produce a torsion on the beam, but will be taken out by the diaphragm, so the torsion shouldn't ever actually reach the ridge beam.

Many thanks for your time!

 

[DaveAtkins]

My response may not answer your question, or satisfy your curiosity, but here goes...when designing a sloped roof diaphragm, I design for the horizontal load imposed by the windward and leeward walls, and I ignore or neglect the vertical components which result (although I do admit a rigorous analysis would include them somehow).

Hopefully someone else will chime in!

DaveAtkins