Unfortunate Shear Wall Location 10

Can you assume a flexible diaphragm that spans the full length of the building and ignore any resistance offered by the short segment? Of course it would have to be detailed accordingly.

Edit....missed your last sentence, guess you already considered that. Surprised item 10 did not have much effect. If its stiffness is (presumably) less than the longer walls, I would not expect it to attract that much load.

Are your joists running parallel with these "short shear walls" or perpendicular to them?
If parallel, are there girders framing into the "middle shear wall", tagged in your sketch? I assume there would be girders framing into the "inside shear wall", but how many?

I`m very interested in the question and think that there will be a different answer depending on the configuration of the gravity system.

Typically my buildings use non load bearing precast. When you use load bearing, how do you deal with chord forces? Is there a panel-panel connection at EVERY panel joint to transfer this diaphragm T/C?
Would it be possible to detail a diaphragm chord running EW at the N end of "outside shear wall" and artificially limit your diapgrahm depth to the length of that wall? This will increase your diaphragm forces, but if your loads are small that may not be a big problem.

The things that come to my mind:

1. Do the "slip connection" (at the header/top plate). You might could justify the release based on a max gravity load attributable to that location (using a ultra-conservative coefficient of friction and being sure there is no physical mechanism that could make it happen anyway).

2. Add another "chord" to the diaphragm (or beef up the ones you've got) to stiffen it up.

3. Perhaps add another shear wall on the interior. As far as the users are concerned, it would be a partition.....but you would know better.

What about the foundation level for that short segment? If that is detailed to allow rotation of the wall then it shouldn't attract much shear.

Does that wall have to be concrete or could you frame it from something else more flexible?

Could you make it a spandrel panel where it doesn't extend to the foundation and is just along for the ride? Nice spot for some windows at grade?

Since you mentioned wind loads, I assume you don't need to accommodate the large deflections associated with seismic design.

If this is true, I would treat the diaphragm like a beam, uncoped on the left end, and coped on the right end. The reaction at the right end will generate a moment on the diaphragm which must be "developed" into the diaphragm, using roof framing members. The short wall segment can be ignored, if you can justify the midspan diaphragm deflection at the short wall segment.

DaveAtkins

We have run into similar cases before and attempted to isolate the short segment with horizontal slots and detailed the adjacent walls to isolate the problem segment as much as possible. I think we also checked how much deflection there would be if the short segment had zero stiffness and judged how much damage there might be to it and the attaching deck if the isolation didn't work as intended.

Pretend that the short bits are non-existent for your analysis and come what may. Seriously, that's my personal recommendation. We make messy buildings, not meticulously crafted pianos. There's a limit to our "knowing". Chords & drag struts... you bet.

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.

I like KootK's practical answer. it works until something goes wrong and you get visited by the four horsemen and their lawyers.

another day in paradise, or is paradise one day closer ?

rb1957, I'm not sure what your intent is, are you being sarcastic saying you like KootK's answer? As an aerospace engineer, you likely design things that are closer to "meticulously crafted pianos" compared to buildings.

Since the end walls are significantly stiffer, I am surprised that any load is going to the short intermediate wall. Have you tried modeling a horizontal truss in the roof plane to force the load to the exterior wall?

If you do not provide a drag link in the roof diaphragm, the load will not be collected to the wall, regardless of any wall connection.

Put a window in the short wall segment to make it weaker...



Mike McCann, PE, SE (WA)

Thanks for all the replies.

Some other facts:
a. The roof joists are clear span across the entire width.
b. We have modeled the diaphragm as a series of "horizontal beams" along the edges, so to speak, with a stiffness both in flexure and shear that attempts to mimic the deflection behavior of a metal deck diaphragm under the proposed loads.
c. We have monkeyed around a bit with the wall stiffness, proportional to the individual wall lengths and also set to assume a range of shear wall deflections normally expected for rigid concrete shear walls on shallow spread footings....so a parametric analysis solution where we vary the diaphragm support stiffnesses (the shear walls) to again mimic expected lateral deflections in those systems across a range of values.
d. The idea of separating the small shear wall from the diaphragm occurred to us but seems impractical as the wall is comprised of two panels and we must connect back to the diaphragm perpendicular to the building at least.
e. For some other comments/questions:

There is small seismic.

Joists are spanning the short direction (up and down on my sketch), no girders.

Joists do not align, currently, with the short walls so right now no particular collector.

We might add one but that would not solve the problem of high shear attracted to the short wall.

Slip connection: diaphragm to wall is tough to detail right and questionable in true behavior I think.

Walls are all interconnected so if the overall diaphragm deflects, the small shear wall will naturally want to deflect with it.

Kootk's suggestion along with haynewp's thoughts might be the only true answer...i.e. detailing around presumed pseudo deformation/damage near the short wall but ensuring adequate overall safety.

Just thought I'd poll others here to see if we are not seeing a forest for the trees or something.





Check out Eng-Tips Forum's Policies here:
faq731-376

For what's its worth, we use KootK's approach with our precast walls where necessary. All diaphragm connections to said walls are slip and our diaphragm is assumed semi-rigid.

I wanted to get a quick answer off to JAE while I was busy but, now that I've got a little more time...

- I think that this space serves us best when we tell the down and dirty truth about what we actually do rather that pitching high complexity solutions that some other, fictional, more competent, more ethical engineer might prosecute. I like to delve into the theoretical of "what might be done" as much as anybody but I try hard to keep it real too. The lawyers can take a long walk, I ignore those little walls.

- As engineers, we're trained to think in a somewhat binary fashion. Pass/fail. As this example illustrates, some problems do not have wholly satisfying answers. I think that the key in these situations is to go back to the fundamentals and take an honest, holistic look at what the real risks and real alternatives are.

- One option available to JAE is to tell his archi-pals that, from now on, they can only have perfectly rectangular buildings. I wholly support this as I could really use the reduced competition.

- In terms of public safety, I think that the biggest risk here is associated with erection safety. I spend time in this space now so I know something of which I speak. Come up with some goofball detail where things are loosely tied together out of "cleverness" and some poor bastard is gonna wind up getting his foot crushed or worse. I think that the complicated / questionable efficacy veto is spot on.

- I think that the second biggest risk here is that cleverly isolating the offending walls from their neighbors might lead to walls that move differentially and create building envelope problems. Then some precast typical, non-redundant connection rusts out and a poor gal in shipping and receiving winds up with a 16,000 lb wall panel on her desk.

- I put seismic damage at a distant third in terms of risk. Perhaps, sometime in the next 2500 years, there'll be an MCE earthquake and then the roof deck and/or mini panel will get torn up a bit. Boo hoo. With things nicely tied together, it's unlikely that the wall panel will come loose and hurt somebody. The panel gets fixed and a future generation marvels at what a sage professional that JAE fellow was back in the silicon age.

- I put wind damage in last place. I've done enough wall panel work to know that this building form, and ones with similar problems, are very common. If service level winds were damaging buildings in a meaningful way, I think it safe to assume that we'd have heard about it by now.

- This is not a public/property risk but, rather, a risk to the caliber of our profession. If JAE details some goofy connection, that'll be expensive. Taken over the arc of a career, such decisions will drive work away from JAE and towards some yahoo down the road. That would make me sad as I'd take JAE's thoughtfully considered half assed engineering over some other boob's full assed engineering all day, every day.





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.

KootK said:

some other boob's full assed engineering

Click to expand...

Thanks for the laugh. I'm going to use that in the office in the future.

Doing some emergency peer review work today on something that really does stand a high probably of crushing some folks during construction. The truth is out there... and so are the boobs unfortunately.

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.

KootK said:

Perhaps, sometime in the next 2500 years, there'll be an MCE earthquake and then the roof deck and/or mini panel will get torn up a bit. Boo hoo.

Click to expand...

I agree 100 percent. Probably not a good idea to put that in writing, however

DaveAtkins

DaveAtkins said:

I agree 100 percent. Probably not a good idea to put that in writing, however

Click to expand...

A foolish and perverse part of me looks forward to my day in court. As long as I'm allowed to say my piece, I feel that I could explain this successfully. I want my Jack Nicholson moment. Lilly livered Joe Publics out there don't know how soiled our hands and souls get trying to keep them safely and economically sheltered. We are the men in the night on the wall. And the gals too of course,

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.