How to size this frame given the load distribution through the length?

[enghelp2806]

Hello,

I'm trying to size this frame, which is essentially a beam fixed/pinned on either end and loaded externally. The FEM is basically a global model (other components not shown) used to distribute loads and isn't used for stresses due to its coarseness. What I have below is the frame, plotted with the shell forces of each element along the length of the part.

What would I do with these shell forces to size the part? This is just a preliminary sizing, nothing detailed or final just yet, which is why I want to go this route.
1. I could take the max force, but would that be the max force of any one element, or the max of the summed forces of 3 elements along the z axis?
2. I could take the average force, but would that be the average force of all elements, or the max of the average force of 3 elements along the z axis?

Thanks

 

[WARose]

If this is "essentially a beam"....cannot a trial size be done with a hand solution?

What is causing this "external" force? Your force application looks like a moment that has been reduced to a couple.

What does this "frame" consist of? A plate? A beam? What are its dimensions?

 

[enghelp2806]

WARose,

I definitely could do a trial sizing by hand, but I'd like to understand how it could be done using the available FEM.

This is basically a keel beam in an aircraft, and the external loading is complicated, as it begins as pressures applied at the wings and distributed throughout the entire aircraft. The red lines on either side of this part are the areas that the load is introduced/dumped out.

The "frame" is really just an I-beam; you can see the web in the picture, but what isn't shown are the flanges, modeled as beam elements.

I created this post in the hopes that I could get some understanding in how to take the shell forces and create some "resultant" load to size with. More of a general question, really.

Thanks!

 

[rb1957]

Grid Point Force Balance would be one approach ... you have a bunch of point loads distributed on the section ... calculate the total axial load and the total moment on the section, then hand calc stresses. This is good for checking the FEM (and seeing that three elements is not a very good mesh for a bending distribution); compare your FEM internal loads (the total axial force and moment) with hand calc.

another day in paradise, or is paradise one day closer ?