Structural Steels Modeling

[PLCKing]

I wasn't sure how to edit my post and move it to the right place, so I'm re-posting it in the right place instead. If an administrator could erase the original from the Structural->other thread. Thanks,

I'm using solidworks/cosmosworks to model a structural frame that's made up of structural beams. I'd like to run different scenarios using different structural members in the model, so for example the corner posts will be 6"X4"X3/16" HSS and the top beams will be 4"X4"X3/16" Angle. What and then compare if I change to a different size of HSS or change the angle to HSS, etc. When modeling in Cosmosworks and using the beam mesh then all the members get converted to beams, but I don't know how comosworks differentiates between and HSS structural member and a Angle or C structural member.

The other option is to use a solid mesh, but with the beams being 150" long, it's a bit impractical and the computing time is very long. Any suggestions?

PK

 

[GBor]

At the bottom-middle of your post, there is an "Inappropriate? If so, Red Flag it!". Click on that for the post that you want deleted and it will give you an opportunity to type a note that you have re-posted here. Make sure it is the one that you want to have deleted when you press the "inappropriate..." item.

Do NOT use solid mesh for this application. Not only is in impractical and with high computing time, but it is analytically incorrect...the assumptions driving the theories for solid elements are not appropriate for objects that do not fit within the parameters governed by the assumptions...

As for the practical application of CosmosWorks, you may want to post in the Cosmos Forum...but that would make this a triple post...not viewed highly by site management, but they forgive you if you keep "red-flagging" yourself

Garland E. Borowski, PE
Engineering Manager
Star Aviation

 

[PLCKing]

Thanks for the reply, I'll try to move it into the solidworks forum.

I agree that solid mesh is not the best, but then how do you go about modeling different types of strucutral members? Also, the model has solid blocks in it for the coner re-enforcements. Can you mix beam and solid meshes in cosmos works?

 

[GBor]

My familiarity with Cosmos is limited. Also note that there is a separate forum for Cosmos...don't lump it in with the SolidWorks forum.

I'm sure you can designate different groups of structural members in Cosmos. They should all look like line elements and the different groups will determine their physical properties. depending on your configuration, you may have to offset the beam elements in some manner to indicate how far off you are from the neutral axis.

I'm sure you can also mix beams and solids. The difficulty here is degrees of freedom. Typically, beams are 5 DOF and solids are 3 DOF. This means you have to make some modeling adjustments to handle the DOF differences. For beams and solids, this is often done by extending the end of the beam into the solid elements and possibly "tying" the through-beam in the solid to adjacent nodes. If you don't, the beam may "spin around" in the solid.

The easy way to check this is to run a modal analysis and see if you get a rigid body mode (infinite displacement). It may be the beam freely rotating about its neutral axis.

 

[PLCKing]

That's great information. I'll play around with it a bit more and see what I come up with.

For some reason in solidworks simulation 2009, if you want to do a non-linear analysis, then you can't change mesh types from solid mesh to beam(or I have't figured out yet how).

Thanks for the responses.

PK

 

[PLCKing]

The difference between a linear and non-linear analysis is the varience in the stiffness of the material as a load is applied. Since a HSS(square tube) beam has a different stiffness then a C channel, would this mean that you need to do a non-linear analysis to see the difference each structural member would make in a frame? And if a linear analysis is performed then the results would be similar in either case?