Seismic vs Wind: Collector Forces

[LearningAlways]

Hello.

I've worked the majority of my career in Florida, which means I've never encountered seismic design. We priced a few jobs out-of-state, so I know the general framework, but have never gotten into the details on a live project which would be constructed. I work in the precast field.

That said, I'm reviewing another engineers calcs. I now work for a different company who has offices across the country. This engineer included collector forces for the wind case, this building is in FL so again no seismic. I've reviewed and performed dozens and dozens of wind analysis calcs in Florida but have never seen wind collector forces calculated.

My understanding is that collector forces act like drag struts and transfer load from elsewhere in the building into the shear walls. There's logic as to why he would include collector forces in the wind diaphragm design but it goes against what I've seen, and ASCE doesn't mention collector forces in any wind chapters only in seismic.

What is the difference in the way forces are transferred in a seismic vs. wind event that one requires collector forces and the other does not? Its not easily obvious to the non-seismic eye.

Please and thank you.

 

[phamENG]

I've reviewed and performed dozens and dozens of wind analysis calcs in Florida but have never seen wind collector forces calculated.

Yikes. You just admitted to leaving a HUGE gap in the load path of every structure you've designed. Granted, in some cases it's maybe not that critical to consider them explicitly. There are some cases where typical construction practices just "work." Going out on a limb here, though - Florida and precast concrete structures are not areas where I consider that to be the case.

Codes are not cookbooks to be followed step by step - in most cases. ASCE's silence on the matter of collectors for wind isn't tacit approval to not have them. I'm pretty sure it doesn't mention shear walls or braced frames, either, but we all know we need them to resist wind. ASCE 7 is all about loads. They have special loads for collectors in seismic design because they are critical elements of the load path that need special attention and additional capacity over and above what typical analysis would show you to ensure a building designed using the ELF procedure behaves as it should. Not because it's the only reason to have them there.

Besides, just because the wind load is higher doesn't mean you can ignore seismic, even in Florida. It may not govern the overall loading, but the appropriate detailing for seismic must still be considered. Ever been to the St. Augustine lighthouse? If I remember, correctly, you can still see the big crack from the Charleston earthquake. And don't forget about the episode in downtown Miami a few years ago with everyone flooding the streets as the buildings all swayed back and forth to the ground motions of a Caribbean earthquake.

 

[skeletron]

"What is the difference in the way forces are transferred in a seismic vs. wind event that one requires collector forces and the other does not? Its not easily obvious to the non-seismic eye."

Seismic = cyclic motion. Back and forth. Side to side. Repeat.
Wind = single direction motion that can happen in all directions. Blows into the building. Releases. There is no reverse load cycle

You want ductility and redundancy in your load path, regardless. But it's especially important in seismic because of the load cycles.

 

[KootK]

1) While the character of the loads are different, both wind and seismic loadings tax diaphragms and create ostensible demands for boundary elements such as chords and collectors.

2) Historically, most buildings in most places have performed well without explicit design attention having been given to the design and detailing of diaphragms. What "historically" means in this context is dependent upon the the type of building under consideration and where the building is located. In some regions, the diaphragms of commercial buildings were receiving explicit design attention 30+ years ago. In other regions, like mine, they're still not getting explicit design attention on a consistent basis.

3) There is one situation in which "undesigned" diaphragms have proven themselves to be frequently inadequate in the absence of explicit design and detailing attention: significant seismic events. The cyclic nature of seismic events and the need for most buildings to respond to such events inelastically tends to bring out the worst in diaphragms, particularly those constructed of prefabricated elements lacking inherent continuity between elements (precast). So the need for the deliberate design and detailing of diaphragms in high seismic regions is quite real.

4) To address the need for deliberately designed and detailed diaphragms in high seismic regions, standards have been gradually pivoting to explicitly require those things. This has created some confusion in aseismic regions where the demand for deliberately designed diaphragms (just equilibrium really) seems just as "real" as it does in seismic areas even though:

a) Standard practice in many regions has historically not included the deliberate design of diaphragms and;

b) In explicitly mandating deliberate diaphragm design for seismic loads it does create the unstated implication that such design may not be required where it has not been mandated (wind loads for example).

5) Diaphragm design can be especially awkward for precasters because it is very common for precast diaphragms to work like this:

a) As with wood sheathing and steel decking, it is commonly assumed that the precast elements of the diaphragm transmit only in plane shear and no in plane tension or compression.

b) The boundary elements of the diaphragms -- chords and collectors - are often elements of the base structure. Precast connection details are riddled with dowel-y looking connections into walls etc at 48" oc or whatever. In many of those cases, those connections are moving shear loads out of the precast decks and into chord / strut elements that are embedded within the base structure (rebar at the tops of walls etc). This is further complicated by the fact that many EOR's don't even realize that this is the case.

6) When I function as a precast engineer, one of my least favorite situations is when I have to deal with an EOR who is new to diaphragm design. Usually they've just watched an SK Ghosh webinar or something and decided to get serious about diaphragm design (with the best of intentions.) The drawings will say something like "Design for XXX PLF on the diaphragm. Precaster to design all diaphragms chords and collectors". This comes undone in the following ways:

a) You usually need a bunch of annoying load cases specified to delegate diaphragm design well.

b) EOR's rarely even tell me how much diaphragm shear is getting sucked out by the various VLFRS elements under the various load cases.

c) EOR's often don't realize that their own base structures need to be the boundary elements unless they want the precast to get crazy complex. As an example, it is a very difficult thing to build a tension chord into a hollow core deck perpendicular to the planks unless you have an uncommonly thick topping in which to embed rebar. You're not just going run a fireproofed HSS "chord" under the planks for a few hundred feet continuously without raising some eyebrows.

 

[driftLimiter]

Part of the reason these elements are addressed specifically for seismic is that the 'historical' way of doing things proved not enough.

Statics and equilibrium require that these forces be transferred into the shear walls in one way or another.
When you put a discrete element like a drag strut collector, then design its connections to the diaphragm and to the shearwall for the code required loading you have properly validated the strength of the load path.

If not, you rely on non-discrete elements and do not consider the detailing of the connections, I don't know how you can say that the load path is complete. Maybe it is working by some complex path but its hard to imagine someone managing to get the strength of that correct.

The logic of determining the collector forces for wind or seismic is to provide enough strength to resist those forces.

 

[JohnRwals]

Precast industry has been investing a lot of time and energy to develop seismic design including diaphragm analysis and design.
Therefore, some research papers led to revise ASCE and ACI.
I wonder these changes should happen ONLY in precast concrete design.
Can we accept CIP reinforced concrete or prestressed concrete design as done before for a long time
while precast design theories and codes change?

Is it true new findings are limited only to precast concrete design?

-JRW

 

[KootK]

Can we accept CIP reinforced concrete or prestressed concrete design as done before for a long time while precast design theories and codes change?

The monolithic character of CIP diaphragms certainly makes them much, much better at being competent diaphragms than precast decks in the absence of any specific design and detailing for diaphragm action. Personally, I'd go as far as to say that most CIP decks don't require much in the way of diaphragm design unless they are high aspect ratio, transfer level, or endowed with monstrous re-entrant corners, etc. Your typical high-rise condo slab usually does just fine with some drag bars at the shear walls.

In some regions, the situation is pushed even further and attempts are made to encourage precast detailing that emulates the redundancy and ductility of CIP systems.

Precast has many benefits but, as with most things, there's no free lunch. Precast is fast to erect because it's a kit of parts. And precast diaphragms tend to come apart when shook... because they are a kit of parts. A kit of parts that needs to be deliberately stitched back together by folks seeking to minimize the cost of that effort.

 

[JohnRwals]

We know that seismic design is critical especially in earthquake active regions.
I think everybody agrees with that.
Problem is how we can implement (quantify) this PHILOSOPHY.
OP thinks it is excessive or strange design.
And, recent precast research papers say old diaphragm method does not provide enough diaphragm reinforcement.
Accordingly, construction codes change almost every year.
Isn't this environment related with 'quiet resignation' also?

-JRW

 

[KootK]

Problem is how we can implement (quantify) this PHILOSOPHY.

I'm always surprised when the codes single out seismic design for something that is just equilibrium when they could just as easily have said "everybody needs to design diaphragms with care regardless of the load source". The only explanations that I can think of for that are:

1) Incredibly poor coordination. Seismic researchers come up with the seismic requirements and just don't consider the implications for wind for a few code cycles or;

2) The code writers actually intend for "conventional practice" to be good enough for wind. If this is the case then, again, I wish they'd just say that outright.

I used to work for a firm that would put a big star beside stuff in their plans that was very important not to screw up. Then the trouble became the implication that anything without a star was not important enough to do right. This feels a bit like that to me.

 

[driftLimiter]

I don't know Koot. The code shouldn't need to state that sufficient strength in the load path to maintain static equilibrium is required. Its implied by our job role. We should be following the load path all the way through and providing strength as required.

I can't believe that the code writers intent was to allow for considerable gaps in the load path via your #2).

I hear you though that the over-emphasis on seismic in the code can be misleading. But that isn't an excuse to forgo the basics of statics and load path. My Breyer textbook (and others) from college has plenty of examples of collector design for wind and seismic. So I don't believe that this concept should be foreign to any practicing engineer.

IMHO there is no need to clarify what elements are needed for wind design, if the engineer is diligent and documents the load path, those elements will arise out of necessity, regardless of loading.

 

[KootK]

The code shouldn't need to state that sufficient strength in the load path to maintain static equilibrium is required.

Perhaps they shouldn't need to but, in my experience, they absolutely do. I've trained enough people now to know that what happens with most engineers is that, the minute they leave school, they rapidly forget most of what they learned in school and simply adopt the practices of the engineers that they work with. And that makes sense because we all need to find efficient ways to complete our tasks, get home to our families, and make a little profit for the man. I've seen many EIT's pretty much "forget" statics en route to figuring out how to get things done in the real world.

At the risk of giving offence, I feel quite confident that, if you submitted some of your drawings and calculations to me for review, I'd be able to identify a bunch of things that you're not currently checking simply because the other folks who handed their engineering dogma down to you haven't been checking them. The engineering dogma that we incorporate into our engineer selves becomes the water in which a fish swims: you don't even see it after a while. Yeah, there are some engineers out there questioning things but, in my experience, they are very much the exception rather than the rule. Just look at the vitriol I earned here for my efforts at questioning the standard dogma: Link.

So I don't believe that this concept should be foreign to any practicing engineer.

With respect, I suspect that you opinion is partly a function of your having worked extensively in a high seismic region where the attention typically paid to all things lateral is much greater. Again, it's the water in which you've been swimming.

My Breyer textbook (and others) from college has plenty of examples of collector design for wind and seismic.

The last version of Breyer that I digested still espoused inflection point bracing for LTB which, we all know, was proven to be nonsense some three decade ago. A little pee in Breyer's pool as it were.

 

[driftLimiter]

All fair points KootK (even about finding things I don't check ). I guess by discussing all this here more people will see it and perhaps think differently in the future. Perhaps as the OP is doing, people who don't do it very often might learn a lot by digging into seismic design, concepts that can apply to wind design as well.

Also to add to this, the value of professional development, webinars, and any opportunity to learn new dogmas from engineers outside your bubble are excellent opportunities to grow.

 

[KootK]

(even about finding things I don't check )

As would be the case if you parsed some of my work. If there's a fully legit chord system on any of my wood buildings -- ala Terry Malone -- you can be assured that was nothing more than an accident of geometry.