Gap Element to model vertical bearing reaction on bridges during vertical ground motions

[lprogers]

I am applying vertical ground motions via non-linear time history analysis to a suite of single span bridges. to model the vertical properties of the bearing I am using a gap element to replicate the real world situation where there is virtually no resistance to uplift provided by the bearing.

Even though the gap link is set as nonlinear and the time history load case is set to nonlinear I am still getting incorrect results. the figure shows vertical displacement of the abutment compared to the superstructure. (mm on y-axis, seconds on x-axis) You can see how the superstructure just keeps going up! I experimented with different stiffnesses and also used a multi-linear link but still cant get expected results.

Any Ideas?

https://drive.google.com/file/d/0B8nwWtPD9JL6OEowcEtRMlhBWVk/view?usp=sharing