Collector requirements in rigid wood diaphragms

[idecharlotte]

Despite the popular consensus to design wood diagrams as flexible, in most cases with new construction they are almost always rigid. With irregular shaped buildings the most common method of handling diaphragm horizontal offsets is by using collectors with drag strut connectors to the shear walls. While this is appropriate with flexible diaphragms, with rigid diaphragms, collectors can be overly conservative. Stresses can be high at reentrant corners, but I am wondering with a rigid wood diaphragm how to analyze the stress distribution through the plywood, glue, and floor screws when the collector is absent. The SPDWS is a joke when it comes to "real world" diaphragms and the only literature I can find on more complicated irregular shape buildings always assumes a flexible diaphragm analysis, which in my opinion is not accurate, and usually over conservative. Thoughts?

http://www.idecharlotte.com

 

[XR250]

I think I would rather be overly conservative than try to figure this out. I doubt we know the real stiffness of anything with enough accuracy to make a difference.

 

[idecharlotte]

^In my experience the difference can be thousands of dollars on an average large and complicated custom home, and when you are competing with engineers that ignore LFRS almost completely because of lack of understanding, then it makes it tough to compete in an already very cost sensitive industry.

http://www.idecharlotte.com

 

[XR250]

My experience has been that the customer would rather spend $2000 more in materials than $500 more in engineering. I don't get it but it keeps me from sharpening my pencil too pointy.

 

[NorthCivil]

Disclaimer - I'm not sure if I'm interpreting the question properly.

But how I approach these things.

Unless the aspect ratio is large, I design for a rigid diaghram too. (I've heard arguments that the gluing of the subfloor might give more stiffness than the nails, and thus diaghrams are more rigid than we design for)

The CWC gives tables for shear resistance, in lbs/ linear foot for different diaghram types, and shear wall types. I ensure that I run a drag strut long enough above every shear wall, that the diaghram strength in lbs/ft at that drag strut has 20% more capacity than the shear wall.

As long as there are no enormous openings near that drag strut, the diaghram load should effectively transfer into the drag strut.

As for evaluating things more meaningfully than that - I couldn't apply first principles to figure out a shear strength of plywood nailed to studs/joists. I assumed the values given in the tables in CWCs wood design manual came from testing...

Hope this helps

 

[KootK]

I don't see that the use of rigid diaphragm assumptions changes anything much about the general demand for collectors, drag struts, and chords. It's still the same shear panel, equilibrium satisfying design methodology. It's just applied to a different set of support forces. If anything, the need for chords and collectors in a rigid-ish system is probably greater on account of the role that those elements play in keeping the diaphragm contiguous and justifying the rigid diaphragm assumption.

Even with glue and screws, resisting tensile stresses in a diaphragm without boundary elements usually means pulling the tops of the supporting framing members apart in tension perpendicular to grain. And that's a no-no in wood design.

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.

 

[MotorCity]

idecharlotte - can you explain your rationale for treating wood diaphragms as rigid? I have not heard of this practice and am interested how you justify it. Of course, any diaphragm is somewhere between flexible, semi-flexible, and rigid. Admittedly, I have never attempted to put numbers to it, I've just always been on the flexible diaphragm bandwagon for wood diaphragms. I guess it seems intuitive that wood diaphragms would be "more flexible" than your typical rigid (i.e. concrete) diaphragm

 

[DETstru]

I don't see how your collectors are going to change significantly in flexible vs. rigid.

Most jurisdictions I work with require we envelope the design for both flexible and rigid diaphragms. This is conservative but solves the issue of how will the diaphragm really behave.
You can't just assume flexible when you have cantilevered diaphragms all over the place (as architectural layouts seem to be nowadays).

 

[idecharlotte]

MotorCity, Flexible diaphragms probably make more sense for diagonal board sheathing but not for modern OSB such as Advantek, which can be twice as rigid as conventional plywood. But even with plywood diaphragms, the math usually indicates that it is rigid or semi-rigid and hardly ever flexible. In fact, even though the code limits wood floor diaphragms to a 3:1 aspect ratio, you can even get a pretty rigid floor at those limits as well.

NorthCivil,
I'm not sure what you mean by "I ensure that I run a drag strut long enough above every shear wall". My understanding that a drag strut (or collector) basically accumulates the diaphragm shear forces to the shear wall from any open boundary edges. So on reentrant corners, basically 'L' shaped diaphragms, by flexible analysis you would have two collectors (beams) in the X and Y direction running the full length of the diaphragm at the bend in the L to accumulate the shear force and then a drag strut connector would be required to connect the collector to the shear wall in some way by either running it along top of the shear wall and attaching it, or with a connection of some kind to the corner of the shear wall. In a rigid analysis, in theory, so long as the diaphragm forces unit shear is not exceeded at the shear wall connection and the stresses at the reentrant corners don't exceed the capacity of the diaphragm, then no collectors are needed.

KootK,
I appreciate your answer and I generally agree but I guess my point is the sheathing, glue and members have to have some capacity, and probably quite a bit, otherwise houses would fall down in a stiff wind all around us, and yet they don't.

Thanks for everyone's input by the way. I really do appreciate your thoughts.

http://www.idecharlotte.com

 

[rn14]

If you don't want to run full length collectors you need to detail out the transfer diaphragm. Whether it is a flexible, rigid, or semi-rigid makes no difference. My concern is more how you are determining the diaphragm is rigid. I agree that in reality screwed and glued diaphragms with heavy sheathing are in the middle somewhere but to say they act rigid is quite a leap. Just as idealizing wood diaphragms as flexible is a simplification of the problem so is idealizing concrete on metal deck as rigid. They are all in the middle somewhere but unless you want to model everything you have to pick one or the other. If anything you would need to do both and use the most conservative answer in my opinion.

 

[sandman21]

Not exactly sure what you want thoughts on. Of course SPDWS does not have design examples of "real world" diaphragms, the code lays out the restriction for diaphragms. If you want examples you will need to get another reference, I would suggest "The analysis of irregular shaped structures". Your thoughts on collectors are off, it does not matter that the diaphragm is rigid or flexible. If you provide sufficient capacity and equilibrium at the wall neither would require a collector. Since wood structures have historically preformed well in earthquakes, we have decided to allow some of the past assumptions, flexible diaphragms. However, if you compare stiffness between the wall and diaphragms, you will find that most diaphragms are rigid.

 

[KootK]

I appreciate your answer and I generally agree but I guess my point is the sheathing, glue and members have to have some capacity, and probably quite a bit, otherwise houses would fall down in a stiff wind all around us, and yet they don't.

And I generally agree with that. I guess my point is that there's a substantial chasm between what we both strongly suspect to be true and what we're able to prove based on the formal guidance currently available. I don't actually think that small wood residences should even be designed laterally. The BS factor is just too high to justify the effort. I'd much rather just have something prescriptive (min length & spacing of shear walls). That way at least we all be playing by the same rules.

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'd much rather just have something prescriptive (min length & spacing of shear walls). That way at least we all be playing by the same rules.

Don't we have that already in the IRC? Does not apply to about 50% of the house am at tasked at designing as they seem to be all glass.
Unless I am cantilevering a floor diaphragm, I simply base my shearwall designs on tributary area. I could care how flexible or rigid the diaphragm is. Is it correct all the time - probably not. Is it ever going to be a problem - probably not.

 

[idecharlotte]

Don't we have that already in the IRC? Does not apply to about 50% of the house am at tasked at designing as they seem to be all glass.
Unless I am cantilevering a floor diaphragm, I simply base my shearwall designs on tributary area. I could care how flexible or rigid the diaphragm is. Is it correct all the time - probably not. Is it ever going to be a problem - probably not.

So out of curiosity let's say you have a 24' deep by 33' wide garage with only three 9' openings in the 33' side (18" lugs.) By the tributary method you would need steel moment frames around the garage openings. By a rigid analysis you would likely need nothing. I don't think detailing out a moment frame will be a big time saver because I decided to treat everything flexible, not to mention a pissed off builder because now he has to order steel.
The IRC and IBC have prescriptive methods but like you only about 25% of houses we design meet that. In the IRC method, very little about collectors mentioned, especially when horizontal offsets are less than 8 feet, so even they acknowledge some account for the strength of the diaphragm without collectors.

http://www.idecharlotte.com

 

[XR250]

So out of curiosity let's say you have a 24' deep by 33' wide garage with only three 9' openings in the 33' side (18" lugs.) By the tributary method you would need steel moment frames around the garage openings. By a rigid analysis you would likely need nothing.

I do those prescriptively using wood portal frames (if possible) or detail a three side building design or I call the arch. and have them change the dimensions to make it work without going to extreme measures.

 

[rn14]

I do those prescriptively using wood portal frames (if possible) or detail a three side building design or I call the arch. and have them change the dimensions to make it work without going to extreme measures.

Exactly. Even as a rigid diaphragm you would have to resolve the torsion created by the offset of center of mass and center of stiffness which is what you are doing with a three sided building with a flexible diaphragm.