I'm working on analysis of a bridge wall-pier for dead and live loads. When modeling the pier in finite element software (LARSA), I'm running up against several modeling questions:

  1. Is it valid to model the wall as a single column element? (Going to a plate mesh seems like overkill for a run-of-the-mill highway bridge.)
  2. How should the connectivity be modeled between the girders and the wall pier? Are rigid links appropriate? Joint constraints? Beam elements with really high moments of inertia?
  3. If I wanted to run modal analysis, would this change the modeling considerations?

Wall Pier


1 Answer 1


I'll go through your questions one by one:

  1. A simple model like you have shown is fine to get model loads. Once you get those loads, you can then use a more specific technique to design the pier. (strut and tie).

  2. Rigid links are ok. Just make sure that you are capturing the various induced moments. I usually have rigid links from the bearing locations to the pier nodes and rigid links from the center of girders to the bearings. That way you can see individual bearing axial loads and shears (and set moment/rotation releases properly).

  3. All of the above should work for modal analysis. In modal analysis you will have to be extra careful that members are drawn along their centerlines so that you can apply mass accurately.

  • $\begingroup$ Just adding a detail, the validity of rigid links depends on the real connectivity of the beams and column. If the beams rest on bearing pads (neoprene, for example), then total rigid links would transmit the beams' rotation to the column, which would be incorrect. Partial rigid links (linking only the appropriate degrees of freedom) should work. $\endgroup$
    – Wasabi
    Sep 21, 2015 at 1:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.