seismic for multi-span bridges

[mshani]

I am designing a multi-span steel composite beam with concrete deck. I have few questions on seismic design of piers/abutments in multi-span bridges.

Below is the bridge restraint information

Single-span
*End abutment 1 Free in longitudinal direction
Restrained in transverse direction

*End abutment 2 Restrained in longitudinal direction
Restrained in transverse direction

Per AASHTO 4.7.4.3 seismic analysis is not required for the single-span bridges as long as the CONNECTIONS are designed for the seismic forces. I could not find a mention of providing battered piles to resist this seismic load.

Multi-span bridges (2-span/3-span)
*End abutment 1 Free in longitudinal direction
Restrained in transverse direction

*Interior Pier/Piers Free in longitudinal direction
Restrained in transverse direction

*End abutment 2 Restrained in longitudinal direction
Restrained in transverse direction

For multispan bridges, using above restraint conditions, for seismic loads in longitudinal direction, we are detailing the connections such that bridge "rides" over the interior piers in longitudinal direction and transfers all the load to the End Abutment 2.

For the multi-span bridge since the condition in the longitudinal direction is similar to a single-span bridge (free in longitudinal direction) do I need to just make sure that connection at the End Abutment 2 is designed for seismic force and not worry about providing battered piles at the fixed abutment to resist it?

I am providing battered piles for bracking force only. I am also designing the piers and its foundation for lateral load in transverse direction.

 

[bridgebuster]

Generally, battered piles are discouraged for seismic resistance. Vertical piles are preferred as they are more flexible and less susceptible to damage. Battered panels can be damaged in a seismic event and inflict damage to the footings.

 

[IDS]

What level of seismic loads are you talking about?

If the seismic loads would control the design if you designed for them, then you should design for them.

This includes the piles, and as bridgebuster says, you will get a more robust structure if you use appropriately detailed vertical piles to resist the lateral loads, rather than raking piles.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[mshani]

SEISMIC ZONE 3.

Are you saying that we should design the abutment and piles for the lateral load even for single span bridge? If yes, how do I find the lateral load on the abutment when it clearly says no seismic analysis required? Please bare in mind that i am not asking this question with respect to transverse direction. My question is along longitudinal direction.

 

[IDS]

It's a long time since I have worked with the AASHTO code provisions for seismic design, but I presume that single span bridges are estimated to have enough reserve capacity in passive resistance at the abutments that a separate seismic analysis is not necessary.

A multi-span bridge, fixed longitudinally at one abutment only, is a different situation, and clearly the provisons for single span bridges do not apply.

A seismic analysis of the structure will give you loads in both the longitudinal and lateral directions, and if these loads are resisted at one abutment, then that abutment needs to be designed for them.

I should probably leave any more specific advice to those familiar with current US codes.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[bridgebuster]

mshani,

On a multi-span bridge you have to analyze for the transverse & longitudinal forces.

I don't have an AASHTO handy but for Zone 3, aren't you supposed to do a multi-mode analysis?

 

[mshani]

Yes we do have to do a multi-modal analysis. However for preliminary analysis we are using uniform load method. Can anybody explain me the reason why we are not required to do any seismic analysis for single span bridge? AASHTO only requires to provide connection of superstructure to the substructure for the seismic load. Why are we not designing the foundation for that load?

 

[BlastResistant]

I am sure that you have, but have you verified the requirements for seismic analysis with your state DOT? I am sure you have but I am just kindly reminding you just in case you haven't, because in my state that part of the AASHTO has been superceded and an appropriate seismic analysis is required for all bridges regardless if single span or multiple span. On a personal note, I would do a quick seismic analysis for the single span bridge anyway since it would be very simple for a regular non-skewed single span bridge. Also, you are designing the fixed end of the bridge connection for a certain minimal load to ensure that the superstructure does not slide off the abutment seats as this connection design goes hand-in-hand with a minimum seat depth calculation. Also, the seismic connection load you are referring to is not a true seismic load. I am at home right now and don't have the AASHTO in front of me, but the "seismic" load you are referring to is just the product of the site coefficient times the weight of the tributary area of the superstructure times some other variable that I don't recall off the top of my head. It is not the "seismic" load that you keep referring to it as.

Good luck...

 

[crossframe]

"AASHTO only requires to provide connection of superstructure to the substructure for the seismic load. Why are we not designing the foundation for that load?"

Where do you expect that force to go once it passes through the connection?

Simple span bridges do not have a "structure response" because there is nothing else to give the bridge a different period (like a column). The abutments move with the soil, and the superstructure moves with the abutments. I suspect this, as well as observed performance of single span bridges, is why the analysis requirements have been relaxed.

You will still have dynamic (e.g. Mononobe-Okabe) and inertial forces on the abutments. This is where I include that connection force, because it is still an "Extreme Event."

 

[mshani]

crossframe...

thank you for the explaination.

"Simple span bridges do not have a "structure response" because there is nothing else to give the bridge a different period (like a column). "

Keeping this in mind, if in multispan bridges we releasae the movement in longitudinal direction for the internal piers and one abutment end, is it not similar to a single span bridge? Will the period in the longitudinal direction be similar to a simple span bridge as it will ride over the interior column?

 

[crossframe]

Theoretically, for solely the longitudinal direction, you may be correct as long as you can fully isolate the longitudinal direction, guarantee the seismic forces will have no transverse component (which will be awfully tough to reconcile with the specifications for several reasons), and guarantee 100% frictionless bearings that will transmit no horizontal forces to the piers/columns. With a 75-year design life, frictionless components get real expensive, real fast.

There is nothing wrong with having your Abutment 2 as the "fixed" support of a multi-span - you just need to design/analyze the support conditions and keep in mind what will happen as the horizontal forces become larger and larger.

You do not have a single span bridge and it is not located in Seismic Zone 1. Follow the code and move on. Seismic is just load case - no more, no less.