Sub-Diaphragms in Stick Framed Warehouse

[StructENG82]

We have some conflicting opinions in the office and I am hoping to see what others in the engineering community think. The situation is a 100' x 100' warehouse with stick framing, wood trusses and a plywood ceiling. The assumption that has been made is that for the ends of the building where the trusses run parallel to the wall a ledger board can be used used to transfer the load from the wall top plate to the plywood ceiling diaphragm. The diaphragm then transfers the load out to the corresponding shear walls. However, the difference in opinion is that 100' is to far to consider plywood ceiling to transfer out of plane forces and sub-diaphragms need to be created to transfer the load using blocking for the inward inward wind pressure and straps for the outward wind pressure. This wouldn't be needed in the opposite direction because the trusses are spanning perpendicular to the wall.

I know in seismic conditions for concrete and masonry walls with flexible diaphragms that sub-diaphragms are common (if not required) but I have not seen the need to create sub-diaphragms for stick framed buildings.

Do others commonly do this? Please let me know your thoughts.

 

[KootK]

Ughh. You may come to wish that you hadn't asked. Here's what I know/think:

1) By the book, wood diaphragms can only resist in plane shear. No in plane tension or compression allowed. That's one of the unspoken tenets of our commonly held diaphragm theory. It's also why the NDS insists that all diaphragms be surrounded by boundary elements on all sides. And it kind of makes sense. Your plywood sheets will be installed with small expansion gaps between them so the ability to transfer axial loads between adjacent sheets is dubious. See this thread for more than you ever wanted to know on this subject: Link. In particular, see the discussion around the sketch that I've reproduced below (my handiwork).

2) Because of #1, the "parallel to framing" edges of all diaphragms should theoretically have sub-diaphragm framing. But they don't. And that's a weird thing in my opinion. Sure, high seismic loads make having a definitive load path that much more important. But, then, by not asking for a sub diaphragm framing in conventional situations, are we not basically saying that we're okay with no demonstrable load path whatsoever at the "parallel to framing" condition?

3) It seems that we effectively base our decision regarding sub diaphragm framing on the magnitude of the load (plf) applied to the edge of the diaphragm. This, and not the magnitude of the shear in the diaphragm. As such, I feel that the size of your building has no bearing on whether or not you need sub-diaphragm framing in situations where it is not required by code. With regard to the need for diaphragm framing, I feel that these two situations are essentially equivalent:

a) A 20' x 20' diaphragm with framing at 2' o/c and 200 plf edge load.
b) A 1000' x 1000' diaphragm with framing at 2' o/c and 200 plf edge load.

As such, I'd be okay with omitting sub-diaphragm framing for your case. Personally, I'd be tempted to include it anyhow just based on the scale of the building. It just feels right. I do not, however, see a rational argument that would make such framing more necessary than it would be for a smaller structure.

4) "Ledger board" makes me nervous. Take care not to have any cross grain bending in your connection's load path.

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]

No in plane tension or compression allowed. That's one of the unspoken tenets of our commonly held diaphragm theory

Of course, this is ignored on a daily basis by most engineers (including myself). I am sure you have designed your fair share of structures using in-plane tension.
I guess I don't worry about compression as much as it is self limiting and, of course, the plywood clips will help :>

 

[StructENG82]

Thanks for the posts and the link to the other discussion!

 

[jayrod12]

FWIW, I block the last one or two joist spaces at exterior walls pretty well all the time. If I have tall walls it could be even more.

 

[Triangled]

I may use the jargon a little differently in my neck of the woods, but if I understand correctly, you're talking about wall anchorage to the roof diaphragm. I provide positive members at all walls, developing sub-diaphragms as necessary. I don't rely on cross-grain wood tension in any case.

 

[KootK]

I am sure you have designed your fair share of structures using in-plane tension

How dare you accuse me of such malfeasance! Tension strapped, full diaphragm depth blocking has been installed on each and every one of my wood projects, even the hip roofs and Dutch gables.

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.

 

[TehMightyEngineer]

1) I disagree with the statement that diaphragms can't take compression. My understanding (and I believe the Breyer textbook backs this up) is that the compression side of the diaphragm gets it's resistance from both the compression chord and a "butting up" of the panel edges under compression. Obviously the panels can't take much without trying to buckle but I believe a measurable and significant amount of capacity exists for diaphragms to resist in-plane compression loads.

Professional and Structural Engineer (ME, NH)
American Concrete Industries

http://www.americanconcrete.com

 

[KootK]

1) I disagree with the statement that diaphragms can't take compression.

I think that I need to clarify my stance a bit here. Obviously, diaphragms can take some in-plane compression and tension. Typical detailing requires it and I've taken advantage of the notion on nearly every light frame wood project that I've been involved with. My original point was simply that:

1) Our typical diaphragm theory (not whole building theory) assumes no in plane tension or compression and;
2) The load path is difficult to establish/quantify, particularly for tension.

I didn't mean to suggest that there is no in-plane T/C mechanism whatsoever in wood diaphragms. To the contrary, my original recommendation here was that the OP not install sub-diaphragms on his very large light-frame wood building.

Back when I was an eng-pup, I used to lose sleep over the fact that we design diaphragms for a uniform shear stress when any new grad knows web shear stress is parabolic in a thing with limited redistribution capacity. Then, someplace, I saw some testing results for wood diaphragms having boundary members and, low and behold, the monitored shear stress was uniform. Yay! Not so fast though. That uniformity of shear stress logically implies that wood diaphragms tend not to draw tension and compression. And that brings us back to this discussion.

Don't ask me for the testing documentation as I sadly failed to file it back in my pre-structural document curator days.

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.

 

[TehMightyEngineer]

Ah, understood. I missed that you were talking about diaphragm theory only.

Professional and Structural Engineer (ME, NH)
American Concrete Industries

http://www.americanconcrete.com

 

[KootK]

In retrospect, I did a lousy job of saying it. And, make no mistake, I still feel that where in plane diaphragm axial capacity is required, it's sketchy and generally under-tended to.

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.

 

[TehMightyEngineer]

I would agree on the capacity; however I believe it to be significant enough that some axial compression in-plane isn't unreasonable.

Professional and Structural Engineer (ME, NH)
American Concrete Industries

http://www.americanconcrete.com

 

[RFreund]

I generally take care to have a detailed diaphragm for every job (including precast planks, topped or untopped which I believe we be the next hot topic as it gets ignored worse that wood, sorry side track over), and I don't think the sub diaphragm is a major deal breaker for "non-seismic" criteria. There is no limit on spacing your of strapping/blocking perpendicular to your wall, you just need to make sure that you can span between it. You already have joists/trusses to create your sub-diaphragm chords so you just throw some straps across them at any interior bearing point and call out the required axial load.

Until the code gives some guidance on allowable diaphragm tension that can be taken by staggered wood panels...

EIT

http://www.HowToEngineer.com

 

[KootK]

You already have joists/trusses to create your sub-diaphragm chords so you just throw some straps across them at any interior bearing point and call out the required axial load.

Trouble is, by the book, you don't actually have a sub-diaphragm unless those sub diaphragms grab onto some drag members extending further into into the main diaphragm. All you've really got is a narrow diaphragm acting alone. Not many diaphragm to wall connections can get the job done over 4'.



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.