earthquake design 2

[oneintheeye]

Hi uk based engineer here. Eorocodes are coming our way and seem to be including seismic actions. Can any of you US based engineers give me a quick overview of how you design for earthquakes?

 

[JAE]

Here's the simpleton overview and concept of the static method used in the US:

1. Seismic events put horizontal and vertical accellerations and velocities on structures.

2. To mimic the effects on these movements, the code uses the characteristic accellerations and velocities of typical seismic events superimposed over the natural period of the structure to estimate how the structure would respond to those movements in terms of distortion.

3. Once the distortions are understood, this equates to an equivalent static load, called a seismic demand, which is applied to the structure based upon story height and mass location. These are static loads that are based upon the full, actual, seismic event.

4. Depending upon the selected lateral force resisting system, the structure has different capacities to absorb the energy. Some systems are stiff (x-braces) like oak trees that would fracture easily....some are flexible (moment frames) like palm trees, that can absorb large deflections without fracturing (more ductile behavior).

5. To reflect the different bracing systems, the code allows the actual seismic demand to be reduced more for the palm trees than for the oak trees. Thus, we actually design for a reduced static force. We know the actual seismic event will produce a force F, but realize that we can't design to that full level and remain economical. So we design for a force F/R where R is our reduction depending on the type of bracing system we have.

6. So with the reduced F/R force we know that the true force is F so we are allowing the system to be permanently deformed and go into the plastic range but ensure that the structure doesn't fracture or fall down. We are thus designing for life safety and not to the level that would keep the structure in the elastic range. The building will be a piece of junk after the event, but it will be a stable, standing piece of junk with no lives lost.

7. To ensure that the system doesn't fracture, we identify parts of the structural system that might be brittle, such as rigid connections. We use overstrength factors on the force F, to increase the capacity of these connections so that we ensure that other areas would go plastic before the connection fractures. So in an X-brace, we want the diagonal brace to go into the plastic zone in tension before the gusset plate end connection fractures.

8. If the building has irregularities such as soft stories, irregular plans, etc., sometimes the static load method doesn't apply and the designer must use more involved methods such as a dynamic analysis of a 3D model.

I'm sure there are other points to be made, but this is the raw, basic concept.

 

[cvg]

don't forget evaluation of liquifaction potential on the foundation and the differential settlement caused by it and also remember to tie the structure down to the foundation so it doesn't "hop" off.

 

[oneintheeye]

so is this static load equivilent applied in combination with the wind lateral forces?

 

[asixth]

The code will take into account there is a very minor probability that the design wind event (500-2000yr) and earthquake occur simultaneously and not to combine them.

The horizontal force will be a function of the seismic mass, if the load isn't tied to the structure it is not assumed to contribute to the equivalent lateral force.

How seismically active is the UK, intra/interplate?

 

[Ussuri]

Not very, traditionally earthquakes are not usually considered as a design condition. There are exceptions to this such as nuclear power stations and dams.

 

[JAE]

herewegothen,

No, the wind and seismic load cases are never combined in one load combination.

 

[oneintheeye]

how about the geotechnical aspects? If the superstructure stands how do you evaluate the loading on ground and/or piles?

 

[kslee1000]

cvg has answered your last question in his earlier reply.
From structural point of view - simply apply the resulting seismic forces/reactions on top of the foundations, then perform structural analysis as usual (for laterally loaded substructures). Geotechnical engineer needs to be involved heavily to interpretate soil profiles/ground water conditions and provide proper soil parameters.

 

[oneintheeye]

thats my query how is seismic events handled by the sub structure. Surely an earthquake will alter the nature of the ground particarly in susceptable soils.

 

[cvg]

An earthquake may induce settlement, sometimes caused by liquifaction or collapse of the foundation soils. this should be analyzed by a geotechnical engineer experienced in this. since the structure may settle unevenly, a structural analysis may be necessary to determine if the structure will tip and/or remain standing.

 

[msquared48]

Really isn't as complicted as it initially seems to be from reading the code.

Mike McCann
MMC Engineering

 

[kslee1000]

Again, for structural design on substructures subjected to seismic force, please see the post "seismic load on basement wall", posted by stevegregory on 12/2/08. It contains an excellent reference for this matter.