Wall Panel Vertical Load Distribution

[JoshH726]

I'm trying to verify load distribution through a wall panel. Say I have a portal frame with distributed load. Moment makes sense. Now if I take the same load applied on top of a wall panel, the resultant moment to a beam supporting the wall panel is much less than anticipated.

Where is the difference coming from? Does the wall panel itself act like a deep beam and not trickle down the vertical load to supporting frame? Would it be better to attach the frame and wall using rigid elements at discrete points over the length of the beam?

The purpose of this thought experiment is to capture the removal of a first story load bearing wall and support it now with frame. A new frame would support a second floor load bearing wall. Handcalcs are fine, but I'm more curious than anything why the discrepancy here.

Thanks in advance

 

[JoshPlumSE]

All that's happening is that the wall is significantly stiffer than the beam that's supporting it.

Totally natural and understandable from a FEM standpoint. Now, as engineers we often want to design that beam as if it's fully supporting the wall, and the wall has little stiffness of its own. How to do this?

I can think of a few ways:
1) Don't model the wall, instead just model the loading that the wall imparts on the beam.
2) Model the wall but, soften it a good amount (by adjusting E and G) so that it is not so rigid. Or, maybe adjust some of the Icracked values for the wall to soften it another way.
3) Your rigid link method might work. Though the load will still probably "arch over" the middle of the beam. So, this would likely only be a partial solution.
4) Maybe do a hand calc for the beam design instead. This is an example where the "FEM reality" may be different from the "design reality" that you want based on engineering judgment. Sometimes there isn't a great way around that.

 

[JoshH726]

Thanks Josh. Make sense.