×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Overstrength and Inelastic Response

Overstrength and Inelastic Response

Overstrength and Inelastic Response

(OP)
I'm a bit confused on the intent of overstrength provisions in the seismic AISC manual. Basically, I'm not sure if Seismic force resisting systems such as Moment Frames, Braced Frames...etc are used to take more inelastic response or to prevent it. In other words, are we including these systems to make the structure more inelastic or do we apply overstrength and demand that they carry more load so that the system can remain elastic throughout an event?

RE: Overstrength and Inelastic Response

You want them to be inelastic in an event. That absorbs energy in a controlled way. Imagine a chain being pulled. You don't know what link will break. Now imagine you put a plastic link in there. You pull on it and the plastic link will break, but now you have a lot more control and the break is gradual and absorbs energy. Same concept with high R values, you put in a controlled weak link, but certain parts have to be over designed such that they aren't the weak link.

For example, connections have to typically be way over designed so that they aren't the weak link because it would be sudden. Yielding of steel is what you want, not connection failure, etc.

RE: Overstrength and Inelastic Response

njlutzwe explained it very well. Another way of looking at it--we use various framing systems to resist seismic load, and take advantage of the fact that more ductile systems yield and relieve seismic forces that the structure would otherwise try to accept. This is the reason for the various R factors. But it is too risky to say that certain parts of the system, such as drag struts, can also be designed for the reduced load. So we design them for higher loads, to make sure they don't fail first.

DaveAtkins

RE: Overstrength and Inelastic Response

(OP)
By applying amplified loads to them and designing them to remain elastic for these loads, do we not ensure inelastic behavior in all the members BUT the overstrength members? Basically all the other members are designed to lower (non-overstrength) loads. Therefore they are less likely to remain elastic than the ones designed to higher (overstrength) loads. The way Im seeing it, because we are designing the Overstrength components to be so much stiffer, they will remain elastic during an event while all the members surrounding them become inelastic...

RE: Overstrength and Inelastic Response

I wouldn't say elastic. I would say members and connections designed with over strength factors experience less inelastic, possibly remaining elastic. Point is, the mechanisms you control are failing before the overstrength designed members do.

Imagine a braced frame where you design the brace to deform inelastically. Now imagine you didn't design the connection to transfer that higher load. At that point, who cares about the brace if the connection fails and the brace can't absorb the energy?

And remember, structures don't know what code they are designed to. R factors and over strength factors are approximate and educated guesses. Point of overstrength factors as I see them is to make sure the load gets dumped into the main lateral system, and that the lateral system is designed and detailed in such a manner to behave inelastically in a controlled manner, again the chain analogy. Connections, collectors, etc. that dump load into the system are critical, because if the load can't get into the lateral system it won't matter if you have a "weak link" there

RE: Overstrength and Inelastic Response

@awa: you've basically got the idea. I'd just refine it a bit to better reflect design intent.

1) We usually need inelastic response in order to dissipate seismic energy.

2) Inelastic response implies damage so we want inelastic response concentrated in certain places where that damage will not result in instability and hopefully be straight forward to repair. Typical choices are plastic flexural hinges in shear walls and moment frame beams and axial yielding in cross braces.

3) Seismic demand on our structures generally hits the lateral force resisting system first and then flows out to the collectors, diaphragms, gravity framing etc. By dealing with the bulk of the energy dissipation in the lateral system, close to the source (foundations), the structure further down the load path is shielded from having to sustain significant damage.

4) The primary way that we ensure inelastic damage is concentrated in the locations where we want it is by a)using R factors to ensure that these locations have nowhere near the strength required to resist the elastic level earthquake response and b) ensuring that these locations can sustain cyclic, inelastic damage without creating instability.

5) As designers, our goal generally is to keep our diaphragms and their connections to the lateral systems elastic. Two of the reasons for doing this are a) it's often hard to design collector/diaphragm path to be highly ductile and b) if our diaphragms yield then our assumptions about load distribution between lateral elements goes to hell in a hand basket and we may lose the ability to stabilize our gravity columns.

6) We do our best to ensure elasticity in the structure beyond the designated inelastic zones by designing the important parts of those systems to a load level intended to represent that corresponding to complete, whole system mechanism formation (the point where the structure would stop absorbing additional seismic load).

7) Recognize that elastic earthquake loads reduced by R represent the load at which the first designated inelastic zone in the structure yields. The structure as a whole does not yield until all of the designated inelastic zones yield which will occur at a load significantly higher than (Elastic Level EQ / R). How much higher? ((Overstrength - 1) x 100) percent higher.

In summary, we use overstrength factors to increase the seismic load to maximum level we expect to see in the system and thereby ensure that the critical parts of the structure structure beyond the designated inelastic zones remain elastic. It is, of course, a rough science.

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.

RE: Overstrength and Inelastic Response

It's no small feat but, if you can stare at this diagram long enough to full understand it's nuances, you'll have a pretty good handle on overstrength (and several other things).

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.

RE: Overstrength and Inelastic Response

One of my favorite diagrams. It should be pinned to the cork board in my office.

"It is imperative Cunth doesn't get his hands on those codes."

RE: Overstrength and Inelastic Response

I prefer Figure C12.1-1 in the Expanded Seismic Commentary to ASCE 7-10.

I am wondering about the source for KootK's figure. Somethings are unclear. For instance the "x Ω0" dimension. Shouldn't the dimension be from 0 to Vy , rather than Vs to Vy? As it stands, the figure makes it look like Vy = Vs + Ω0Vs instead of Vy = Ω0Vs. I have similar problems with the other dimension shown. Perhaps the commentary that goes with the figure would clear up my misunderstanding.

RE: Overstrength and Inelastic Response

Agreed with wannabeSE that diagram from the commentary is a clearer version in my opinion (actually the whole ASCE 7-10 commentary on seismic is pretty good).

njlutzwe I starred you. For payment I'm stealing your analogy of a chain with a plastic link. That's such a fantastic explanation of seismic R factors.

Professional and Structural Engineer (ME, NH, MA)
American Concrete Industries
www.americanconcrete.com

RE: Overstrength and Inelastic Response

That one does seem to have some logical flaws if you are trying to follow it very strictly. The linked graphic is better. Good comment.

"It is imperative Cunth doesn't get his hands on those codes."

RE: Overstrength and Inelastic Response

Quote (wannabeSE)

I am wondering about the source for KootK's figure.

I think the source of KootK graphics is the Structure Magazine article entitled "A Brief Guide to Seismic Design Factors", authored by the SEAOC Seismology Committee back in 2008. Link

Quote (TehMightyEngineer)

...I'm stealing your analogy of a chain...

Tom Paulay used the chain analogy way back when...but no 'plastic' links in his!

RE: Overstrength and Inelastic Response

I definitely stole the chain concept from someone. Can't remember where, but someone taught me that and it made perfect sense so I haven't forgotten it!

RE: Overstrength and Inelastic Response

From Paulay and Priestley's book "Seismic Design of Reinforced Concrete and Masonry Buildings":



RE: Overstrength and Inelastic Response

That graphic seems similar to springs in series, where the least stiff element controls the majority of the deformation of the system.

"It is imperative Cunth doesn't get his hands on those codes."

RE: Overstrength and Inelastic Response

Awesome, thanks for the source; I remember that book being on my "to buy" list so I think I'll finally get around to picking it up.

Professional and Structural Engineer (ME, NH, MA)
American Concrete Industries
www.americanconcrete.com

RE: Overstrength and Inelastic Response

In one version of Paulay's chain analogy, I'm pretty sure that it was presented with cast iron links and a single mild steel ring which I like even better.

@wannabeSE: agreed on all counts regarding the superiority of the ASCE version and the flaws in the SEOC/structuremag doc. Nothing in the text of that reference improves matters. Many of the dimensions are sloppily presented by omitting the primary multipliers. Even at that, the definitions of omega and Cd are incorrect, not just sloppy.

Thank you for pointing out the errors and supplying the better version. The inclusion of CuTa adds value to the ASCE version as well. I wonder if there was an editorial issue at Structuremag. It's hard to imagine SEOC mucking seismic up that way.

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.

RE: Overstrength and Inelastic Response

Quote (KootK)

...cast iron links and a single mild steel ring...

Cast-iron = brittle, mild-steel = ductile.

RE: Overstrength and Inelastic Response

Were you really under the impression that I was confused about that? You spent some time in Canada. You know our book learnin' ain't that bad.

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.

RE: Overstrength and Inelastic Response

Quote (KootK)

Were you really under the impression that I was confused about that?

Hehe. No, but just in case it was 'lost in translation' from the Kiwi authors. Eh!

RE: Overstrength and Inelastic Response

Quote:

I prefer Figure C12.1-1 in the Expanded Seismic Commentary to ASCE 7-10.

WannabeSE,

If you found that chart easier to understand than the one Kootk posted then it's because you actually understood the principle in the first place. I did not, and possibly never would have, but for Kootk posting his chart a while back on a different thread. If you're wondering what the distinction is just look at the figure you referenced.

In my copy it reads as follows Vs x (omega-1), Vy x (Rd-1) and Vs x (R-1).

What they actually meant, of course, was Vs x (omega-1), Vy x (Rd-1) and Vs x (R-1).

Or, equivalently and far more simply, Vs/omega, Vy/Rd and Vs/R.

I find it exceedingly frustrating. If I prepared construction documents with such seemingly little care for or understanding of what they communicated you would be reading of me in the the inevitable litigation that would follow. I wistfully pine for the days of old when drafting and presentation were treated more seriously and I dream of future days when code-writers have to practice the same standard of care as do the practioners charged with implementing their decrees.

I know they [the code-writers] know what they're doing. Contrary to how it may sound I am in awe of them and their research and I am very grateful for it. I am standing on the shoulders of giants without whose work I would not be able to design so much as a swing set. I know they are brilliant; I just wish they would make it easier for little people such as myself to follow along. Or "climb aboard", if I'm to stick with the "shoulders of giants" metaphor.

RE: Overstrength and Inelastic Response

Quote (Archie)

What they actually meant, of course, was ...

Actually no, it's correct as shown below and as you wrote in this statement:

Quote (Archie)

In my copy it reads as follows Vs x (omega-1), Vy x (Rd-1) and Vs x (R-1)

Give it another stare and, if it doesn't come together for you, report back and we'll get it sorted.

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.

RE: Overstrength and Inelastic Response

Quote (wannabeSE)

For instance the "x Ω0" dimension. Shouldn't the dimension be from 0 to Vy , rather than Vs to Vy? As it stands, the figure makes it look like Vy = Vs + Ω0Vs instead of Vy = Ω0Vs. I have similar problems with the other dimension shown.

So it turns out that the graph below is quite correct, if somewhat oddly presented. The trick with the dimensions is to realize that there are no dimensions! Very Matrix-esque, I know. If you look closely at the "dimensions", you'll see that they only have arrows at one end. What the dimensions are, in fact, are the ratios required to get you from the starting values to the ending values. For example:

Vs = Ve x 1/R
Vy = Ve x 1/Rd
Vy = Vs x Omega

There, all is right with the world.

Quote (KootK)

I wonder if there was an editorial issue at Structuremag. It's hard to imagine SEOC mucking seismic up that way.

Naturally, I rescind these libelous statements.

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.

RE: Overstrength and Inelastic Response

Kootk,

Thanks, I see it now. Apologies to AISC/SEI for not looking at it carefully enough and for my little tirade.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources