Provisions of Chapter 21 in Linear Elastic Range of Response 2

[cliziagatto]

Hello all!

I'm designing a structure in seismic zone disregarding the provisions of 21 Chapter of ACI 318-08 because I'm using the Elastic Response-Spectrum to evaluate the seismic forces. So I don't need to assign a seismic design category to the structure (SDC). My structure won't go in inelastic range of response. Am I right?

thank you in advance

 

[JAE]

I'm not sure where it is permitted to do that. Can you point to a code provision that says you can design for the full seismic event and stay elastic?
That would generally be one very strong (and costly) structure.

 

[Nonlinearer]

Inelastic response is necessary. The approach of all seismic codes including ACI/ASCE requires to design for inelastic behavior. If we compare moderate wind and moderate EQ forces for a 5 story framed building, ELASTIC EQ forces will be approximately 8 to 10 times higher than wind. Thus making virtually impossible to obtain economical design.

IR

 

[cliziagatto]

Thank you Gentlmen!

you are right in buildings... but what do you think in case of underground structures like metro-station or hydro powerhouse? I'm designining a shaft-powerhouse and is virtually impossible to obtain the right ductility from the beams and slabs and theretoo the structural global right ductility. But many times I've found in the technical specs: "the concrete structure shall be designed according to ACI 318-08....."

 

[cliziagatto]

for JAE,

in Eurocode8 1998-1:2003 "Design of structures for earthquake resistance -
Part 1: General rules, seismic actions and rules for buildings"

2.2 Compliance Criteria
2.2.2 Ultimate Limite State
......
(2) As a limiting case, for the design of
structures classified as non-dissipative, no account is taken of any hysteretic energy
dissipation and the behaviour factor may not be taken, in general, as being greater than
the value of 1,5 considered to account for overstrengths.
......

 

[TXStructural]

There are structures being designed to stay elastic through the MCE. However, the ones with which I am familiar are designed to minimum code requirements, then additional measures are taken to avoid reaching plastic response. Remember that the building codes are MINIMUM code, and will not result in an optimum design in may/most cases. There is nothing in the code which says your structure must fail when subjected to forces greater than those specified in the code. What is required is that you design the way the failure will happen once it does occur.

 

[cliziagatto]

Thank you all for interesting responses.
Yes I know the codes prescribe minimum requirement but never part of ACI 318 specify this. As usual on site, the Engineer wants to see justifications to "shortcut" the ACI. I'd like to find a comment in the ACI as well as ones in eurocode previously posted.
For belbos the project is central America.

 

[shin25]

In order to maintain the structure within elastic range, the response modification factor, R will need to be 1. If this is your design approach, then special ductility detailing per ACI Ch 21 will not be necessary.

 

[cliziagatto]

Tks shin25.

I'm agree with you. In Europe I usually adopt this approach. Could you suggest a commentary in ACI 318 where is written this?

 

[shin25]

I don't have the code infront of me at this moment. However, I know that ACI special detailing requirements kicks in when R is greater than or equal to 3. You should as a matter of fact look into ASCE 7-05 chap 12 where building system requirements are contained.

 

[frv]

I don't have the IBC in front of me, but in the US, you are simply not allowed to design a structure "elastically" (and I'll define that simply as no special detailing requirements) in anything higher than SDC C (I'm sure of that in steel, not so sure in concrete). Anything above that requires special detailing to provide ductility.

It is important remember that that the code approach, generally speaking, is to design for large earthquakes in the same way that modern cars are designed for head-on collisions. It's not meant to be pretty, just meant to allow the occupants to remain alive.

 

[TXStructural]

First, please take the time to read ACI 318 Chapter 21, all of it, including commentary. If you are designing to comply with it, you need to really be familiar with it. Whatever someone tells you here needs to agree with or inform your own understanding of the standard. You are the one responsible for (1) a safe design, (2) compliance with applicable codes and standards, and (3) using engineering judgement in turning codes and standards into a design meeting the location and client requirements.

ACI 318-08 (and I am certain -10 is the same) R21.1.1 says, in part, "...The integrity of the structure in the inelastic range of response should be maintained because the design earthquake forces defined in documents such as ASCE/SEI 7, the IBC, the UBC, and the NEHRP provisions are considerably less than those corresponding to linear response at the anticipated earthquake intensity."

This tells us that even when we think a design is robust enough to stay elastic during greater than code-level events, we still have to design so that the failure of components and members is known and planned to avoid collapse and other similar events. For instance, you would still want to design such that a concrete flexural member would initially fail by (ductile) yielding of steel reinforcement rather than (brittle) crushing of concrete. You would also need to be sure that the distribution of stronger and weaker components was examined to assure that the entire structure remains ductile.

This mandates that you can use any level of design forces you like (in excess of the minimums) but you still must consider the failure modes, and effects of those failures. This requires that you assign a seismic design category, and assure that your design meets of exceeds the minimums for the category.

 

[cliziagatto]

Thank you for your interesting replies!

ACI 318-08 (and I am certain -10 is the same) R21.1.1 says, in part, "...The integrity of the structure in the inelastic range of response should be maintained because the design earthquake forces defined in documents such as ASCE/SEI 7, the IBC, the UBC, and the NEHRP provisions are considerably less than those corresponding to linear response at the anticipated earthquake intensity."

The integrity of structure is assured by the ductile yielding of steel and stronger/weaker compontents of course and is needed when you adopt reduced design earthquake forces due to inelastic response of your structure. Indeed less than forces due to linear response. My structure doesn't undergo into inelastic range. Stays elastic. For this reason I don't need to comply inelastic provision.

This tells us that even when we think a design is robust enough to stay elastic during greater than code-level events

Sorry TXstructural but again I'm not in agree with you. I'm not talking of greater than code-level event. I'm talking of the same code-level event: same earthquake intensity (same Magnitude, Distance, exceedance probability). I'm talking of different structural response to the same event. The earthquake event is clearly indipendent from structural response.

 

[frv]

No, no, no!

TXStructural has it right. This is a fundamental misunderstanding of the provisions and the intent of the code. To draw a parallel, you are arguing that you are going to design your car to survive a head-on collision in the elastic range. The code already anticipates that a structure will survive a minor event in the elastic range. But in a major event, you must allow for inelastic displacements. You are mixing up cause and effect. The displacements of the structure are what induce the seismic loads, not the other way around. Very simply put- if your structure is excessively rigid, it will fail.

And the code-level event (design) is NOT the MCE. The code accounts for inherent damping in the structure.

You cannot choose to design your structure "elastically". You MUST follow the detailing requirements. The purpose of these detailing requirements is not simply to "allow" you to design for lower seismic base shear; it is to ENSURE that the building will have the ductility to withstand the seismic displacements.

 

[shin25]

@frv,

How a portion of a structure, such as bridge foundation is designed to remain elastic during an earthquake? Due to difficulty of inspection and repair, AASHTO requires the critical bridge foundations to be designed to remain elastic. Elastic design is ensured by using R=1. AASHTO permits other methods, which probably is irrelevant for this discussion.

The response spectra that is developed using code provisions is an elastic spectra. Code requires extensive detailing of structure if it is designed for a high R value. Why do you think, this is the case? This is becuase, there is a trade off between high R factor and increase ductility. If higher R is used, the structure is pushed into much higher ductility requirement (due to use of verly little seismic force) requiring high level of detailing. On the other hand, if the structure is designed for low R, the ductility requirement is reduced due to the fact that the force which is used for design is closer to the elastic seismic force. That being said, if a structure is carefully designed can be designed to remain elastic. Although, this approach will be very costly and foolish for a high seismic design category. Elastic design may be attainable for a low to moderate seismic demand and for a smaller structure.

Damping of the structure does not depend only on post elastic deformation of the strucure. Soil, material, connection, etc all play important role in providing damping to the structure.

 

[frv]

shin25,

I will not address AASHTO, as I am only vaguely familiar with it and would at best be venturing an educated guess (and I have a pretty good one )as to why you are allowed to use R=1 in foundations (I'll take your word on that). As to the rest of your post, you are absolutely correct in the relationship between base shear, response modification criteria and ductility requirements. I wasn't arguing that at all. This is seismic design 101 and we completely agree that seismic design requires you design your structure for either the inertial forces (seismic base shear) OR the displacements (ductility), not both.

Please look at ASCE 7 Table 12.2-1. Ordinary Reinforced Concrete Moment Frames are not even allowed in seismic design category C, let alone anything higher. Why do YOU think that is? You are confusing what the code requires and what it allows. The code REQUIRES ductility above certain thresholds. The reason? At best, you are attempting an educated guess as to what the seismic accelerations will be. To sit here and argue that you KNOW what these accelerations will be is self delusion. Again, and this was thoroughly explained in a seminar I attended a few years back, the code intent is to allow your building to survive the seismic event; not to come out unscathed.

But I do get what you are trying to argue. The code, however, does not allow it (at least in the US).

 

[TXStructural]

To echo frv, the key is to design it as strong as you like, but also design it to fail in a planned, ductile way once it does reach failure. Do not design it with the expectation that it can never fail.