Are structures to remain elastic under wind loading? 16

[amengr]

Just a quick question as I am new to wind engineering. For the MWFRS are the wind loads we calculate per ASCE7 such that we are designing the structure to remain elastic or is it similar to seismic design where we allow structures to develop plastic hinges/deformation?

 

[CELinOttawa]

As far as I understand, plastic design was originally conceived for regular structural loads (ie: gravity). Ultimate Limit States are just that, the once-if-ever peak loading that you cannot allow collapse.

Fundamentally, whenever you use a plastic section modulus to size a beam, you're planning for plastic.

I do not know your code. I don't practice in your area. I will be very interested to see other people's answers, but I cannot imagine you would have to stay elastic in your ultimate case, ie: 1 in 50 year or 1 in 150 year wind events (depending on your code and the return period, etc., etc.).

 

[ChorasDen]

I would need to glance through ASCE 7 again, as it's been a while since I've looked at the wind provisions, but my gut is telling me that yes, MWFRS is intended to maintain elastic design. Why do I say this? Because there are deflection limits for wind design in ASCE that differ from the required story drift limits for seismic design. These deflection limits are intended to protect the finish and waterproofing, and thus, in my mind, are intended for elastic response. Just my 2 cents, I've been wrong before.

Seismic design can also be analyzed elastically, but it's rare for your average building owner to be willing to spring for the increase in construction cost.

 

[KootK]

In general, and for your global lateral system, that system should absolutely be kept elastic. Seismic systems can only go plastic because of the cyclical and inertial nature of the load. We generally envision wind to be a quasi static load. Consequently, if your global lateral system goes plastic, the building simply flops over.

 

[CELinOttawa]

Going plastic does not mean, nor denote, collapse.

A simple beam that develops a plastic hinge does not continue to deflect and then collapse, even under the loads at which the hinge was developed. This is a very common misconception, and is wholly erroneous.

The best book explaining this, in my opinion, is one of the oldest: AISC's "Plastic Design in Steel", 1959.

A beam that develops a plastic hinge automatically hardens. Yes, if you had sustained (and increasing!) winds, you could get a "flops over", but that isn't possible if the peak wind load is the point at which the plastic hinge forms. The hinge will form, the steel will work harden, and the structure will remain safe.

Ultimate limits are just that: Ultimate.

NOW: I am also a huge fan of Capacity Design, wherein we design a system to never allow collapse no matter the load. So, I am not saying that this wouldn't require attention to detail(s), but I see no reason to code-blanket-proscribe plastic in a wind frame.

Anyone know this code well enough to tell us?

 

[JoshPlumSE]

I'll say it a little differently. You've got a few different considerations here:
a) Do the "ultimate strength" limit states rely on the members or the system undergoing some inelasticity? The answer here is YES.

However, most of the time the loads will be at "service" level loads and will NOT dip into their inelastic strength levels. So, at service level loads, the structure is assumed to be elastic.

At their FACTORED level ultimate loads they will experience load levels that will push them towards ultimate strength levels of member inelasticity.

b) Is the structure expected to have significant load re-distribution or ductility like we would expect in plastic design or seismic? The answer here is a definitive NO. So, you could say that the structure is essentially elastic.

c) Many structures (masonry, concrete) have lots of inelasticity that we ignore when we do our analysis. So, we pretend like these structures are elastic and the code allows us to do that. However, the reality of the structure behavior may be a little different.

 

[CELinOttawa]

Nicely said @JoshPlumSE.

Also remember that as you approach plastic, your probability of moment redistribution goes up like mad.

The only one who knows the "true" state of stress is the structure itself. There is plenty of reserve strength in most systems, so long as they are detailed with robust load paths (and ideally redundancy).

 

[KootK]

OP's new to wind engineering so I feel that it's important for us to provide her with some clear, definitive guidance on what is a very simple question. Yeah, I get that all kinds of things can and do go plastic without collapse. I don't, for a second, think that is what OP is asking about however.

I believe that what OP is asking is this: can/should a vertical, wind force resisting lateral system form a plastic frame mechanism as many seismic systems are designed to. And the answer to that is a pretty resounding "no", at least for anything remotely resembling routine design.

For the love all that is holy, surely someone can back me up this? This really is not a different perspectives / 10,000 nuances thing unless I've grossly misjudged OP's question.

 

[CELinOttawa]

Again, I would ask: Why not?

You cannot have a hinge in your columns, but a hinge in your beam elements is simply going to give you access to the additional 12% capacity increase typical of Z/S for many sections.

It isn't going to keep going. There is no more logic to denying access to plastic capacity in a wind frame than in a gravity beam... And we allow it in the gravity beam all the time.

 

[WARose]

I believe that what OP is asking is this: can/should a vertical, wind force resisting lateral system form a plastic frame mechanism as many seismic systems are designed to. And the answer to that is a pretty resounding "no", at least for anything remotely resembling routine design.

For the love all that is holy, surely someone can back me up this?

I got your back Kootk (and I agree).

 

[CELinOttawa]

Again, I would ask: Why not use plastic in wind design?

I am honestly wanting a clear, technical, reason why we would not allow plastic hinges in a wind frame.

I used to think this was the case... I have changed my mind after a number of years given more reading about both wind loading and plastic analysis.

I don't see a reason to expect the building to be lacking in stability or safety given plastic design of a wind resisting frame. Plastic does not equate to added movement/deflection/collapse mechanism without loads that continue to increase. By definition, that means the winds exceeded the ULS. If they exceed the ULS, they could simply exceed your elastic design and cause the same plastic issue(s).

There is no reason to fear the plastic design of an MWFRS unless the code has been tuned to expect elastic (and thus the statistical tuning is conceptually based on elastic for wind design).

[Edited for clarity and spelling]

 

[WARose]

Again, I would ask: Why not?

Quite simply because we are talking (say in a hurricane zone for example) potentially way too many cycles of loading. The comparison to Dead & Live Loads ignores the fact that in every steel text I have that considers plastic design in that load case....it also cautions about the cyclic nature of live loads. In fact in Jack McCormac's 'Steel Design: ASD Method', 4th ed, he makes the point that the reason plastic design hadn't caught on in the profession is (in part) because of that fact. [EDIT: He lists several factors actually, but that was one.] Indeed, I haven't worked in a single office that uses plastic design for the D+L case.

 

[CELinOttawa]

An interesting point of view.

I've practiced in a number of countries, and I would say that common practice is to use plastic design in most situations.

Is this a continuation of the dominance of ASD (and tuned-to-ASD "LRFD") US practice?

Wouldn't explain why Kootk is against this, as he is another Canadian Engineer.

This may well be an "agree to disagree". I was really hoping for an actual reason, as I would prefer to have my mind changed than to wonder why we didn't all agree in the end.

 

[JoshPlumSE]

Again, I would ask: Why not use plastic in wind design?

I don't think there is anything that forbids us from doing plastic design for wind. But, this is a different type of design methodology than we typically use.

a) For Seismic Design, we have all kinds of detailing and ductility requirements that allow us to use a reduced force level than would otherwise be required. Because of this, we MUST design the connections to probable force / moment levels to ensure the type of ductility / energy absorption that was assumed when we reduced our force levels.

b) The design codes do NOT assume this with wind load design. Therefore, if you decide you want to account for moment re-distribution and such, then you're obligated to do a lot of things that aren't very well described in the code for wind. There's no reason why you can't do this. But, it's probably more work than you'd want to do for the vast majority of projects.

 

[GC_Hopi]

I would add from a practical standpoint that the owner of a structure does not want to have their structure undergoing plastic deformation everytime there is a stiff breeze. The general mindset is that structures should remain occupiable, operational, and have the building envelope integrity remain during a wind event. Obviously there are exceptions to this...tornados for example.

Adding to what Josh stated you could go to plastic design but you are taking on more design time, risk, etc.

 

[GC_Hopi]

Based on your structure there might be value in that design approach (seems like tall buildings are where this is valuable). Here is an article discussing performance based design for wind Link

 

[WARose]

Is this a continuation of the dominance of ASD (and tuned-to-ASD "LRFD") US practice?

I don't think so. Whether we are talking ASD/LRFD being their weapon of choice.....I don't know a single engineer that does this for D+L. Just seismic.

If you were to consider plastic for D+L, you would then have to investigate where the live load stress range puts you.....and that's getting things way too complicated for (what should be) a simple design.

It's also worth noting that I come from a industrial background.....and having that reserve capacity is something that is considered important for the main structural framing (and foundations) in that setting.

 

[JoshPlumSE]

I've practiced in a number of countries, and I would say that common practice is to use plastic design in most situations.

Perhaps there is a language barrier here. Or, a difference in terminology.

In the USA, using ultimate strength levels (like plastic moment strength for steel beams) is NOT considered plastic design. You still do an elastic analysis, it only affect the design strength calculated for individual members. This is usually called Strength design. Even today's ASD steel design is based on "allowable strength design" which is just a safety factor applied to the same ultimate strength capacities that you get with LRFD.

Plastic Design (in the US terminology) is generally related to a type of analysis where you use moment re-distribution to satisfy the total load requirement. So, if you've got a fixed-fixed beam (which would have an end moment of WL^2 / 12 and a mid-span moment of WL^/24 in an elastic analysis) you could instead design for a smaller fixed end moment and a larger mid-span moment. This type of design requires that you be willing accept a certain amount of plastic deflection and yielding at the fixed ends in order to re-distribute the rest of the load in a manner that's closer to a pinned end beam.

 

[CELinOttawa]

Hmmm... You may have the cause there Josh; You certainly could not, ever, have an uncontrolled/unpredicted redistribution in any frame.

In all the countries I have worked in, and in all of my work, ULS/SLS "Limit States Design" has been the norm.

Given that nothing can be affected in your SLS case, we don't care what happens in your ULS other than "does not fall down".

So: Redistribution, yes. Plastic, sure. BUT: Columns are sized to guarantee they never go plastic (overstrength factors are applied here, involving the ratio of Fu/Fy if you're talking about steel). Connections are likewise designed for increased loads. Your system, in short, is protected from the "bad" outcomes through diligent design.

 

[phamENG]

Where's dik? He preaches plastic analysis for just about everything. I'd be interested to hear his take.

I agree - you could design the frame for all the fancy hinges and connections for member capacities, but why would you? Ignoring PEMBs and gas station canopies, how many steel framed structures typically fail in a hurricane? Not many. They generally perform pretty well. The same cannot be said for steel frame buildings that have undergone significant seismic events without robust detailing to ensure proper strength and ductility. So we'd be looking at a drastically increased design cost for a shot at a slightly decreased construction cost (I realize a slight decrease in the latter can dwarf a quadrupling in the former) to solve a problem that is largely non-existent.

It's also important to consider the way wind and seismic events behave. We like to think of wind as a big hand pushing on the building, but it's really a series of random impacts all over the building surface of varied magnitude and duration. So while a seismic event gives us nice, pretty hysteresis loop with fairly consistent cycling, wind is more of a scatter plot.

So I say you should run an elastic analysis under wind loads and design the members in accordance with the relevant material code. Wood - better stay elastic. Steel - as long as you don't form any hinges you're fine, but AISC allows for partial plastification. Concrete - keep your steel elastic.