Bearing Stiffness affecting abutment displacement unexpectedly

[lprogers]

I am subjecting a single span precast prestress concrete I-beam bridge to longitudinal ground motions. when i specify a stiffness for the bearings in the transverse direction, the abutment longitudinal displacement becomes smaller than expected. when the transverse bearing stiffness is set to 'fixed' reasonable results are obtained for longitudinal abutment displacement. for expample; when transverse stiffness is fixed, abutment longitudinal displacement is 10-15mm, when I specify a transverse bearing stiffness the displacement is 1mm or less.

any idea as to why this is happening?

I am using multilinear plastic links to define the bearing, here is a screen shot with the transverse stiffness fixed.
Link

 

[stressed]

It might be useful to see a screenshot or two of your model to see how you modeled the bearing and abutment. Your multilinear link force-deformation input appears to have the same force with different displacements.

 

[lprogers]

Well i solved the problem of unexpected deformations, the foundation springs were incorrect. I had the vertical stiffness switched with the longitudinal stiffness.

My ultimate goal is to model the bearing as a elastomeric pad with 2 steel dowels. To do this I am referencing a thesis done by Bryant Nielson at Georgia Tech. I first tried to add in parallel the stiffness of the pad and dowels into one multilinear plastic link element. I am unsure if this will be accurate since when the bearing unloads it will follow the initial stiffness of the pad only, it will not consider the initial stiffness of the dowel at appropriate deformations. I did not get the expected results.

Here is a screen shot of Nielson's explanation of the bearing,Link.

I am thinking I will have to define the pad and the dowels separately in two different multilinear links. I had tried hand drawing in extra links at the bearing locations and defining them as dowels but i feel there is a better way.

Do you know if there is a way to add two separate links together for the bearings? I have 88 bridges to model for a graduate thesis paper so I am looking for the most efficient way to accomplish this.