Reinforcement detailing provision below the seismic base? 1

[hetgen]

Say I’m doing a multi-story concrete building in low seismic region (0.1g) with double basement floors imbedded in a stiff soil/rock capable to transmit lateral seismic load on to basement walls constructed monolithically with floor slabs. The seismic base is at ground floor.

With all the stiffness gained by the retaining wall and stiff soil /rock surrounding the lateral resisting system do I still need to consider seismic reinforcement detail at column beam joint below basement 1 to allow for ductility?

 

[ishvaaag]

Resorce to the code for answer. The seismic region is not too much exacting, but if anyway you are required for ductile design do so at least for the joints at ground level, since they are in the load path to bearing both lateral and vertical.

 

[hetgen]

Thanks ishvaaag,

Forgive my ignorance…. I practice in a region long been categorized as non-seismic, however the recent published code recommend 0.1g peak ground acceleration in some part of the region and I’m trying to refresh the little that I know about seismic actions.

Yes, I agree joints in ground floor and floors above require ductile detailing, but I can’t visualize any ductile deflection below ground due to seismic action. I refer to the code and it recommends ductile reinforcing for members designed to resist forces induced by earthquake forces and doesn’t state any different for structures below ground imbedded in stiff ground.

 

[ishvaaag]

You can refer to chapter 10 of the Earthquake Engineering Handbook to read "The results show a definite reduction in motion of embedded foundations compared to surface foundations or free-field surface values. Numerous comparisons are presented by Chang et al. [1985]"

In general and particularly for these moderate earthquakes such reduction of structural response supports some lesser ductility requirement for the underground levels. But this is a matter of engineering judgement, and the view of the pertaining code or those reviewing its application may differ, in which case the safe path is to provide the ductilty to the same degree established for the whole superstructure.

 

[SAIL3]

I am not so sure I would assume that the seismic base is at ground level.
Two overall questions come to mind.
1. At what level are the seismic loads imparted to the bldg
structure.
2. What is the response of the bldg structure to these loads.

Seismic base
a/ Base at bottom of second basement.
b/ Base somewhere between bottom of basement and grade. At best
this would be a WAG assuming sufficient contact and stiffeness
between soil and basement walls. Would also require design
of basement walls for this local seismic loads.

Bldg response
What really is driving all this uncertainty is that there is no marked difference in the bldg response structure from bottom of basement to top of structure.(from info as posted).
At a minimum, I would consider seismic reinforcing details all the way to bottom of 2nd basement. I would also attempt to bound the problem based on above assumptions and any others I may have missed.
Your saving grace is that this is a low seismic area.

 

[hetgen]

Ishvaaag,
thanks for your insight. ... I will look for the Hand book... Any other good books that you recommend for moderate earthquakes?

 

[ishvaaag]

Since soil structure interaction stays without complete code clarification (at least not in most codes) one can very well understand the good common sense and technical insight of others. One can see the movement of the above ground structure penetrating in the undergrond levels. This points to both points of some lower level where to apply the ground motion than the ground level and extend seismic details downwards in the basement. Yet Chapter 10 of the referred book above clearly recommends to use the ground motion -assuming the ordinary way of basement embedment- at ground level, if only because the data from which the ellaborated input for our seismic load proceeds has been mainly reduced to such level. So really in the end the answer to the question of the detailing remains as open as the governing code leaves it, and an accurate evaluation of the response means one or several state of the art soil structure interaction analyses, better done following expert advice or good guide, and this is even less forthcoming in places where the seismic risk is not bigger than 0.1g peak ground acceleration sites, or scarce seismic design was being done before.

I quote again from chapter 10 of the same book some references recommended for soil structure interaction analyses:

"Numerous guidelines exist defining the required steps to perform SSI analysis for design or evaluation
purposes. All methods of analysis are treated. Selected guideline documents are:
ASCE Standard, Seismic Analysis of Safety-Related Nuclear Structures and Commentary [1998]
EPRI Guidelines for Soil–Structure Interaction Analysis [Tseng and Hadjian, 1991]
Earthquakes and Associated Topics in Relation to Nuclear Power Plant Siting"

Quite surely there are good references out there and I may have some others, I just remembered chapter 10 because I read it some months ago. If I find something suitable to your question I will post. For example, it is quite likely that for your level of demand, you could design one reinforced concrete building in Spain for bigger forces from assumed lesser ductility and then soften much the seismic detailing of your concrete structure. If you contrarily wanted to use some available ductility, the structural response might turn somewhat less but the seismic details more cumbersome to prepare.

 

[ishvaaag]

In the intent of condone the ductile details, the last paragraph shows an alternative way permissible according to the NCSE-02 code approach, for the design the basements from some levels down, you would repeat the design for ductility factor 2 instead of 3 or 4 and use not the particulars of seismic detailing mandatory for ductility factor 3 or 4. Not precisely accurate nor wholly recommendable, this would still be neglecting entirely any soil structure interaction, but you at least would be seeing in the underground levels enhanced values of seismic demand corresponding to non ductile details.

So, the only proper way to proceed the best available soil structure interaction analysis.

By the way, one reference was lost above for SSI analyses:

"Seismic Design and Qualification for Nuclear Power Plants [IAEA, 1992]"