abutment modeling

[Muntasir]

Hi,

What is the best way of modeling the abutment of a bridge in seismostruct?

Thanks

[huffte]

Muntasir,

I would recommend a single link type element, linear in nature - i.e., the 'lin-symm' link type element. The vertical stiffness would typically be very, very large - large enough to be essentially fixed. The horizontal translational degrees of freedom should be based upon some rational estimation of stiffness provided by the piles and/or endwall, with or withour backfill stiffness, depending upon what you really wish to model. If you are doing dynamics, youmay wish to exclude the endwall and backfill stiffness as only one backfill can be active at any instant. Thsi also depends upon the presence - or absence - of expansion joints at the abutments.

Rotational degrees of freedom for the link could also be bsed on some rational analysis. You might could justify zero stiffness fro rotation about an axis perpendicular to the bridge centerline and essentially fixed for rotation about the other two axes.

You could also use a set of links at the abutment instead of a single link, but I'm not sure what that extra effort really acquires, again depending upon what exactly you wish to model.

Hopefully, this makes a bit of sense and you'll find the element type to model your situation perfectly Muntasir.

Regards,

huffte