FEA package for machine design?

[MechatroPro]

i have worked a bit with ProE/Mechanica and found it a good package for machine design work. it works reasonably good in extracting mid-surfaces and creating shells from assembly components.automatically joins them and creates bonded interfaces between components that are touching. however i'm not familiar with capabilities of ansys or abaqus in this area.do they offer the same level of flexibility and automation? i'd like a comparison of these packages for machine design or structural analysis work. ofcourse ansys or abaqus offer alot more capabilities for researchers or academic people, but what about engineers working in the industry?
thanks in advance.

 

[ESPcomposites]

This may somewhat of a dated view, but ProE/Mechanica work very nicely together as you have found.

You are also right that ANSYS, ABAQUS, and NASTRAN are more often used by researchers, but are also more often used in industry as well (at least in the major aerospace companies). I do not think Mechanica is as common since it is not as good of a general purpose tool. It does have some nice advantages with the integration to Pro/E as well as the use of p-elements though. It will be hard to find a better package that links the CAD and the analysis, though I do think ABAQUS is making strides in that area.



Brian

http://www.espcomposites.com

 

[kan123]

As long as you are designing for operation in the linear regime, Mechanica can be a very good tool as you have found. But if you ever have to consider nonlinear response (material or geometric), that's when you'll have run into a brick wall. ANSYS is for users who may have to consider a wide variety of loading conditions and coupled physics effects. I haven't used ABAQUS, but I've heard that it is especially good at nonlinear analysis. I stopped using NASTRAN several years ago because ANSYS was more user friendly, but as the first to the dance it is still the code of choice for structural analysis at many large companies.

 

[EngAddict]

You can't really go past solidworks simulation for this type of work. It works directly with the CAD model making design and optimisation very easy. It is quicker that most packages to set-up and run a simulation, so for machine design it is ideal. Doesn't have some of the solvers for research type work but for general FEA the results are quite good.

 

[Twoballcane]

We use Pro M for the first few design iterations just because it easier to transfer the model from Pro E. However, once we get close to cutting metal, we use ANASYS. But, I have found both software packages do come close with results. Pro M can do non-linear models

Tobalcane
"If you avoid failure, you also avoid success."

 

[kan123]

The last comment was interesting. So I checked the link below to see that Creo from PTC does have plasticity capability. I don't see the typical kinematic or isotropic hardening options, but at least they have made an important advancement.

 

[Twoballcane]

When you define your study in Pro M, there is a check box for non-linear study. It is not as indepth as ANSYS. But, if I know I will have some plastic bending along the way to ultimate, I'll go to ANSYS.

Tobalcane
"If you avoid failure, you also avoid success."

 

[FeX32]

I suggest Comsol. It is by far the best FEM software I used. Very powerful nonlinear analysis in all domains of physics.


Fe

 

[FeX32]

Not to mention the learning curve is far less then most. They really spent a lot of time making it user-friendly. Something companies like ANSYS is terrible for.


Fe