Performance based design of a bridge pier for a train collision event ?

[Structural_Thumb]

Hi all,

First time posting here.
Does anyone have experience with performance based design for collision loads on bridge piers ?
I'm part of a project where we're working on a viaduct design that has circular piers going quite close (~5m/16.5ft) to railway tracks. At such close distances, collision becomes a very real issue (at least in the client's mind). Furthermore, the lack of space also makes erecting suitable deflection walls difficult.

I feel that if - through performance based design - we're able to prove that the pier (and the foundation) can survive a collision event and remain standing, we can justify doing away with the deflection walls. I'm thinking of suggesting CFT piers since these are quite resistant to collision but I'm not sure how to go about this, if it even is a valid solution.

Thoughts ?

 

[pikku14]

Hey Structural Thumb, to put it quite bluntly, I think this is a really bad idea. Practically from a structural design perspective, the magnitude of load of a train collision gets very high, very quickly. The loads provided in the codes tends to be for a 'glancing impact' happening at an angle to a wall rather than a head on collision

Deflection walls are really about responsibility as an engineer to provide something that is as safe is reasonably practicable - which is enforceable by law in a lot of OHS acts (although of course it varies by country to country). Deflection walls are there to prevent head-on collisions which: prevents the structure from collapsing, but also from avoiding deaths and injury within the train due to the sudden deceleration of the train. Granted it's a low probability event, but the consequences are very high

 

[BridgeSmith]

Good luck with designing a pier to withstand the impact of a train collision. If the trains are moving more than about 15mph, the impact load is huge, in the range of several thousand kips (each car can weigh 400 kips). Even then, the railroad would likely want the wall to deflect a derailed train to minimize damage to it. The only "performance based" design that might have a chance of making it through the railroad's review process is demonstrating that the bridge does not collapse if the pier is gone. I seriously doubt that would be acceptable either; railroad are very picky, and they have the clout to insist on getting their way. Most likely, you're going to have to find a way to provide the crash wall. We have typically integrated the crash wall with the columns.

 

[SlideRuleEra]

Agree with pikku14 and HotRod10, include crash walls. As a former bridge contractor, we constructed several (see photo)... they are built like battleships. Also below, are typical general requirements for RR crash walls in the USA:

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[BridgeSmith]

You're going to need to get very familiar with the AREMA, and if it's a UP or BNSF railway, their Guidelines for Railroad Grade Separation Projects.

 

[damorim]

Agreed with the above, a crash-protection wall is definitely required. Reviewing and designing to AREMA is a must. Here are a couple of snippets from CNR (Canadian National Railway) standard details that provide some reference as to what we use up here in Canada. Piers fabricated from concrete filled steel pipes are very common but they must be designed with a large solid wall at their base meeting the crash loading and geometric requirements prescribed in AREMA.