Pinned at both ends vs fixed

[Antnyt23]

JoshPlum,

I was curious if you could take a look at this frame I was modeling. I wanted to confirm if my understanding of how to use pinned at both ends vs fixed was correct. It's a simple rooftop platform. All interior beam to beam connections are bolted clip angles or plates welded to one beam bolted to the other. Corner beams are plates on top with (like a gusset) bolting both beams together. Also was curious about my boundary conditions. I will add a picture of the site also. Just curious if you could give me any recommendations on how I can better accurately model something like this.

 

[Antnyt23]

Here is the picture I changed my loading due to the modifications that are going to be done.

 

[JoshPlumSE]

Antnyt23 -

The question of pinned vs fixed is usually one of engineering judgment. I don't generally like reviewing drawings or models and suggesting what type of connection to assume. Because, the reality is that all connections fall somewhere between fixed and pinned.

Looking at your model, those assumptions do not seem un-reasonable. The beam-girder connections where the beam lays on top of the girder might be closer to a moment connection based on what I see in that image file. But, it's not entirely clear as I cannnot clearly see the extent of the welding or bolting.

I've seen some retrofit projects where the engineer used a questional pinned vs. fixed connection assumption. If I'm not confortable with either assumption, then I might do one of the following:
1) Analyze the structure both ways. Then use the more conservative forces or design calcs.
2) Use a connection that is in-between. This is especially useful if you've got some Moment-Rotation curves from testing for this connection. In order to do this, you just set up a small "link" beam between the centerline of the column and the face of the beam. If you adjust the EI of that link beam then you can get behavior between a rigid/fixed connection and a flexible / pinned connection.

 

[Antnyt23]

Thanks I appreciate the info again.