Rail-Structure Interaction for Horizontally Curved Decks

[iabhiuce]

Dear All,

I am involved in a nonlinear rail-structure interaction (CWR) analysis for a horizontally curved transit bridge. The analysis that I am doing is as per UIC-774-3. I cannot find any clause on how to deal with interaction for acutely curved bridge decks. The curvature in my bridge is about 100m (almost L shaped). Can anybody please suggest any method for analysing this type of bridge for rail-structure interaction? Is there any standard that specifies something for track-structure interaction for curvatures?

Thanks,
Abhishek

 

[Dougt115]

Unique issue that I have not had to deal with. Sounds fun.

Have you thought of breaking it in half?

One is a just a straight transit bridge and the other is a curved rail. The trick will be combining the results into a defendable answer.

The curved rail analysis would give you the loading for the transit bridge in three axis.

 

[VoyageofDiscovery]

Do you have ballast?

 

[iabhiuce]

@VoyageofDiscovery: The rails rest on concrete bed. There is no ballast. I have read in AASHTO code mentioning something about reduction in allowable stresses in the rail based on the radius of curvature. But it does not say anything about the concrete bed properties. Will there be any transverse concrete bed properties that need to be defined apart from the longitudinal nonlinear ones?

 

[VoyageofDiscovery]

Of course transverse "bed" properties are required to properly distribute the forces to the bridge girders. It would be best to model this entire bridge with FEA including the longitudinal rails on adjacent spans/or approaches. Bearings orientation will require special attention as well.

 

[rickfischer51]

How are the rails fastened to the concrete bed? At the very least there will be an elastomeric tie pad. More likely, there will be a fastener body that bolts down to the concrete. It will be a pair of plates with rubber bonded between them (eg, Hixson), or perhaps a pair of castings joined by cast rubber (Cologne Egg). These are the fasteners in use a while back, not sure what's available today. I worked on a fastener design for the Vancouver light rail project back in the '80's where a maximum rail slip force was specified so as to prevent excessive rail forces from damaging the elevated structure. You will have to consider thermal loads and dynamic loads. I assume speed is very slow with a 100 meter curve radius.

Rick Fischer
Principal Engineer
Argonne National Laboratory