Where's the Diaphragm Chord?

[steellion]

I have a 160'x160', one-story warehouse bay with shear walls on the perimeter on 3 sides (the 4th side is a building expansion joint). The roof is an untopped steel deck. I have been able to check that the diaphragm meets acceptable deflection limits in the direction of one shear wall. My question is when the wind is blowing in the direction of the 2 walls, what is acting as the chord member of the diaphragm? And what force do I design the chord for? Typically, I would design the chord force for M/d, where d is the distance between extreme lateral elements, but there is only one lateral element in this case.

 

[JAE]

I would not try to design a 160 ft. square building using three shearwalls with a flexible diaphragm. I would always provide a lateral brace of some type along that open end.

Usually chord members for a diaprhagm are steel members running along the edge of the diaphragm, either edge angles or beams.

 

[a2mfk]

What are you shear walls?

As JAE alludes, it all depends on deflection compatibility of your diaphragm and the shear wall drift/defl ratios. There are code/spec limits on this relationship that determine whether you can assume a rigid or flexible diaphragm. If I guess from your description, you have CMU or tilt walls for shear walls, I doubt you will have such a compatibility. But I know engineers who neglect this and just model it as rigid.

 

[hokie66]

I wouldn't design a building 160'x160' which depends on "C" shaped lateral resisting elements. Brace the fourth side or provide a stiff rigid frame.

 

[BAretired]

I agree with JAE and hokie.

BA

 

[csd72]

Agree with the above.

 

[DaveAtkins]

I respectfully disagree with my colleagues--I am OK with your three sided diaphragm, if you can prove everything works.

The chord on the open end will be the line of joists or the line of beams along that edge of the roof. If they are joists--make sure you put on the drawings that the joist supplier must design the joists for the diaphragm chord axial load, and develop a splice detail (a plate works nicely) to splice joist to joist. If they are beams, check them for interaction of gravity load bending and diaphragm chord axial load, and design the end connections for the prying that will be induced by the axial load. I would not be surprised if you need 5/8" thick connection angles on these beams, due to prying.

DaveAtkins

 

[a2mfk]

I agree Dave, but that is a big IF, and he has to run the numbers. I think many engineers don't, they just say its rigid and go with it. Depending on how tall those shear walls are, and what they are made of, 160' is not going to deflect(story drift) much at all. So unless you can get your diaph deflection way down it will be tough.

If you do get it to work, like Dave said, pay attention to your chord design. I like to use WF there so I can do the design and detailing and not put that on the joist mfr (axial plus uplift). Also, depending on if you have a canopy or storefront, a WF may come in handy there as well.

Let us know how it calcs out.

 

[nuche1973]

Dave and a2mfk have it right. Sometimes we are not give the ideal structure to work with, then we have to get creative.

There are days when I wake up feeling like the dumbest man on the planet, then there are days when I confirm it.

 

[JAE]

Dave, I don't disagree that a three sided shearwall building is possible. But I simply stated that I wouldn't do it.

There are many times when a building "wants" to be three sided (retail store with solid back and side walls and glass front) but these are usually quite long where the two side wall shearwalls are short and the eccentricity is small.

But in steellion's situation there is a 160 foot SQUARE building.

Even when considering a design using the three sides only, the second order effects on the building columns might get large and many engineeers don't take the time to check this - or don't know how to. The building will twist under lateral loading and all those leaning columns will add to the lateral force applied.

It just isn't something I would do at 160 x 160.

 

[msquared48]

Ever ride a three-legged horse?

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[UnneutralAxis]

I don't have all the experience as some of these other folks, but I agree with those who say they wouldn't do a 3-sided shearwall blding with this footprint. As others have stated, the eccentricity is just too large. Brace this side.

 

[nuche1973]

I've seen a three legged mule: my grand-father was a farrier. It got around just fine.

There are days when I wake up feeling like the dumbest man on the planet, then there are days when I confirm it.

 

[SAIL3]

My understanding is that the load carrying capacity for steel decks as far as shear and beam action is concerned have been established by testing....I doubt if the deck in your application is capable of handling the kind of torsion you would have and have reasonable accompaning deflections....also, the torsional capacity may not have been established thru testing...could be wrong though.

 

[BAretired]

When wind blows parallel to the open end, one wall carries all of the shear. That would be 160*w where w is the load per foot on the diaphragm. Since its length is 160, the unit shear is w, the same as it would be for a 160' x 320' diaphragm with walls all around.

The torsion is 160*w*80. This is carried by the two walls perpendicular to the wind, so each carries 80*w (in opposite directions).

It should work, but I still wouldn't do it.

BA

 

[hokie66]

On an individual member basis, we wouldn't try to use a channel as a beam to resist torsion, so why use the same shape to resist a LOT of torsion?

 

[kieran1]

Unless I misunderstand the question, why not provide a moment frame on the open end?

Kieran

 

[msquared48]

Or an internal moment frame, still cantilevering the diaphragm to the front, but having a backspan as it were.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[SAIL3]

BA you are correct in the distribution of the shear loads, assuming the steel deck has the ridigity and capability to distribute the shear loads as indicated.

 

[DaveAtkins]

I like kieran1's idea--I was thinking a braced frame would be required, which is probably not desirable in a warehouse. But a moment frame would not cause any problems in a warehouse.

DaveAtkins