Reponse Spectrum Method Applicability 2

[gotlboys]

Hi everyone,

I am tackling response spectrum method for seismic analysis. I found that it is a linear elastic dynamic method which has limited use in actual design.
How applicable it is if our building plan is symmetric in plan and height but is in the range of 5 - 10 storey?

I wonder if anyone has used or seen Engineers use it to design design structures of those storey or height.

Thank you
Gotlboy

 

[KootK]

You're certainly allowed to use RSA for structures of that sort. That said, RSA's primary benefit is in capturing complex behavior that would not manifest itself in an Equivalent Lateral Force code analysis. For a structure without significant irregularities, that benefit may be limited.

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.

 

[gotlboys]

Hi Kootk,

Great to hear from you about RSA.
Some references emphasize that nearly all structures behave inelastically in reality when subjected to moderate and high ground motion or strong wind. Since I am here in South East Asia where moderate and/or strong earthquake do happen, so I wonder if RSA is still applicable.
Last week, my prof said when they design buildings of that sort [here in Australia for moderate earthquake] it is desirable to keep the building in its elastic behavior. This implies that RSA could be used.

Gotlboy

 

[KootK]

In general terms, I believe that both ELF and RSA are elastic analysis methods applied to inelastic problems. As such, your degree of intended elasticity should not, alone, govern your choice of method.



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.

 

[RobertHale]

KootK is correct. We modify the elastic analysis through the use of the Response Modification Factor, R and the Deflection amplification factor, Cd in order to capture the nonlinear behavior of the structural system. We are designing for forces from an event much less than MCEr. I would go to the professor and ask to clarify what was previously said. It is very difficult to design the building for the MCE and keep the system elastic (think no cracking in concrete and no yield in steel). The only systems that I understand are designed this way are nuclear power facilities. Everything else is designed to sustain damage in specific parts of the structure because that is the main way we can dissipate energy. Modal analysis is applicable to the structure you are talking about, but as KootK said you probably won't see much benefit beyond ELFA. The ELFA is basically a boiled down version of MRSA assuming that the first mode of vibration dominates the behavior of the system. In the American codes, the irregularity checks are designed to find structures where the higher response modes will contribute and force you to really look at a more accurate response.

Robert Hale, PE

 

[gotlboys]

Hi RobertHale,
I did confirm and my prof said the same way you mentioned that we only design for fully elastic system in power plant (highly exceptional category), not other structures.

You're certainly allowed to use RSA for structures of that sort
Modal analysis is applicable to the structure you are talking about, but as KootK said you probably won't see much benefit beyond ELFA. The ELFA is basically a boiled down version of MRSA assuming that the first mode of vibration dominates the behavior of the system

1. Philippine code called NSCP is drafted mainly based on ACI, ASCE, UBC/IBC and AISC. I noticed that NSCP allows ELFA to be used for building height up to 75m, provided no vert and hor irregularities and the building is in low seismic zone. Also, during my internship, I noticed that a sr. Designer made use of a couple of shearwalls in different locations to draw the CR closer to CM which, according to him, validates the use of EFLA for a seven-storey building irregular in plan. What is your comment on this?

2. Say EFLA is not applicable in that seven-storey bldg, will RSA be applicable with that irregularity? I myself haven't used RSA in actual design nor witnessed anyone use it, so it would be great to hear from you all. If possible, direct me to a good source to learn more.

 

[KootK]

Also, during my internship, I noticed that a sr. Designer made use of a couple of shearwalls in different locations to draw the CR closer to CM which, according to him, validates the use of EFLA for a seven-storey building irregular in plan. What is your comment on this?

I think it's appropriate assuming that the modifications successfully eliminated the irregularities. Moreover, the strategy may well have produced a better performing and more predictably behaving building, which is great. The primary intent of the code irregularity provisions is to incentivize better building layouts. This sounds like that intent coming to pass.

Say EFLA is not applicable in that seven-storey bldg, will RSA be applicable with that irregularity?

From a code perspective, I believe that RSA is applicable to all structures.

If possible, direct me to a good source to learn more.

I've struggled to find a good primer sort of document myself. I have a copy of the book shown below which is a bit old and a bit theoretical. Still a solid reference though. I also have some extensive notes from a seminar on RSA that you can find here.

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.

 

[dik]

When subject to high wind, most structures remain elastic and do not enter the plastic region. Earthquakes have the rapid transients that can cause plastification.

Dik

 

[gotlboys]

Thank you Dik and Kootk.

 

[MrHershey]

Should note that it's not just irregularities that would trigger MRSA being more beneficial or useful. MRSA tends to yield lower stresses and drifts for buildings that have their behavior significantly influenced by higher modes. That includes irregular buildings as noted, but it also includes long period buildings (high rise and/or moment frames) as longer period building behavior tends to be influenced more by higher modes in comparison to shorter period buildings. MRSA captures these higher modes, while ELF does not and assumes the building is responding essentially in the first mode only. This is conservative in most cases but when influence of higher modes gets large enough it can be overly conservative, especially for drifts.

Typically for shear wall buildings the break point in the buildings I've designed has seemed to be roughly a height of 100 ft/30 m where running MRSA starts to be beneficial. For moment frame buildings we had a 2-3 story building recently where it actually made sense to do MRSA. The benefit wasn't large, but we did get a little bit of benefit looking at MRSA instead of ELF.

 

[jsharkbait]

Hi guys,

We talk about MRSA being beneficial when there are irregularities or the first mode isn't 'dominant', but is there a rough guide for that figure (dominant/not dominant)? I've recently been modelling an 18 storey shear wall building. It's only SDC B and is very regular, but etabs was showing me the first mode of participation for the x and y directions was around 60%. The modal analysis didn't greatly reduce the forces for me but it did get the overall building deflections down to a point I was much happier with.

Also, I've been trying to do a bunch of research as to how the RSA analysis is used for design and in load combinations but haven't found much too definitive. Given the forces that come out of it are always positive, for a basic shear wall would it just be a matter of resolving the moment into a push-pull and then it's your basic 1.2D+L+EQ or 0.9D-EQ for any tension?

Thanks!

 

[MrHershey]

As I noted, very rough break point based on my experience is ~100 ft for shear wall buildings. Benefits mainly seen in drifts rather than forces as you noted because your base shear has to be scaled up to 85% ELF (or 100% ELF in ASCE 7-16) while drifts typically do not need to be scaled.

Generally for design forces you can resolve to push-pull like you've noted. Some documentation for how ETABS handles this can be found here: Link. They basically do as you do.

 

[JasonMcKee71]

Response spectrum method is very flexible as it enables the use of linear superposition, ie load combinations etc, due to its linearity. Alternatively, non linear seismic methods can be used (non-linear direct integration), but is such cases the evaluation of the results is not an easy task and requires experienced professional structural engineering skills.

Jason McKee
proud R&D Manager of
Cross Section Analysis & Design
Software for the structural design of cross sections
Moment Curvature Analysis
Interaction Diagrams
Reinforcement Design etc.

 

[Rahul Leslie]

Hi KootK,

Your book suggestion is excellent -- it does go into the fundamentals of RSA.

Hi everyone,

I happened to read a few of the earlier thread (I have joined the forum only recently), and happen to come across the thread on doing RSA with P-Delta effects included.

http://www.eng-tips.com/viewthread.cfm?qid=408235

I have often thought about it and concluded it to be impossible (due to the SRSS or CQC kind of combining requirement getting in the way. It is a bitter pill to swallow -- or rather like a coconut to a dog: desirable, but unbreakable.

Anyway, I learned a couple of things there.. primarily that P-Delta can mean either

(1) the P-Delta as I knew it, ie., additional moments developing in the structure by the gravity loads, due to deformation by the lateral loads (as done in STAAD.Pro, in which case the gravity as well as lateral loads are applied together and P-Delta analysis done in a few iterations); or

(2) the reduction of the structure stiffness due to gravity loads (the stress-stiffening and stress-softening effect), and taking mode shapes of the stiffness-reduced structure for RSA (I think ETABS does this).

The 2nd approach above is new to me (and still doubt whether that is what the term 'P-Delta' is meant by) -- only after reading the thread it struck me that this must be what ETABS has the facility for.

Also I'm also not sure which one of the above is intended by the DAM (to me it seems it's the 1st one). There was a suggestion in that thread to analyze for each mode individually and accounting for P-delta effects, and then to combine the modal results as per standard RSA procedures -- though not clear as to which method above it was meant for.

Concerning the 1st one, what I wish to clarify goes like this: if G represents gravity loads and M1, M2 and M3 represents the modal seismic forces, then the design load combination, can be represented (using SRSS instead of CQC for simplicity) as...

Design forces = G + SRSS(M1, M2, M3)
= G + (M1^2 + M2^2 + M3^2)^(1/2)

... (without denoting the factor of Safety). But when each modal forces are solved for, separately of course, if one does that with P-Delta included (the 1st approach above) , it means each of the analysis will have to be done with the gravity loadings on, thus resulting in the above load combination being...

Design forces = SRSS({M1+G}, {M1+G}, {M1+G})
= ({M1+G}^2 + {M1+G}^2 + {M1+G}^2)^(1/2)

... (omitting the G outside the SRSS since it's already included in each modal case) which when expanded will be considerably different from what we started with.

So the question still remains concerning the 1st method: Is there some other approach to include P-Delta?

Rahul Leslie