FEA software for plastics part 2

[Berserk]

We are looking at purchasing an FEA software
that would analyze molded plastics parts.
I do not have any experience on FEA.
I have been reading the thread about "FEA software ratings/comparison" and I read about the different kind of analysis that you can do on a part.
What kind of analysis is done on a automotive plastics parts? I am thinking that it would mostly be about strength and loads (dynamic and static).
Any suggestion what kind of software to use?
It has to be easy to learn and run on windows XP.
(and cheap ;-))

We are using Unigraphics v18 for CAD.
Any opinions about UG/strenght and UG/scenario?
We also do moldflow on the part using MPA6.0

 

[brad]

In general, most FEA codes are well-suited for plastics analysis--most of the mechanics do not radically change between metallics and engineering polymers (I distinguish these from elastomers).

Engineering polymers tend to more often operate with noticeable nonlinear geometric effects, and generally show significant anisotropy. For these reasons, you may want to consider codes which can address both of these (and many codes do).

As you mentioned Moldflow, I do want to note a few things--I believe that Moldflow sells its own linear FEA solver for shrinkage analysis and some rudimentary FEA analyses. I have not used it myself, but have heard from others who have (with varying opinions as to its effectiveness). Additionally, ABAQUS has an interface to Moldflow which can be used for shrinkage analysis and/or to capture the anisotropic effects of the short fibers in a fiber-filled structure. This is used by several plastics automotive companies and consultants.

Answering your UG question--UG/Scenario a few years ago was very clunky and not a good FEA code, even for simple analyses. I have heard that it has improved a lot lately, but I have not myself used it in recent years.

Brad

 

[ramshree]

When you mention that you dont have any experience with FEA, I assume that you have never used any FEA softwares before as well. Considering you are a beginner, I think you will most benifit from using a software that is simple, easy to use and is more likely meant for "design engineers" rather than analysts.

Yes, ANSYS, ABAQUS and other top of the line can probably be the best softwares, but they are not meant for FEA beginners and also that they are very steeply priced. Their learning curve is not that fast and I think will need a background in FEA. I would rather suggest you to have a look at products like ANSYS Design Space, Cosmos Design Star, etc.
I know Design Space works very well with UG and I think so does Design Star. You might want to check these out for demos.

 

[Berserk]

ramshree, you hit the nail right on the head.

What we are thinking of is doing some preliminary FEA on our parts and redesigning the part if needed, before sending it out for validation.

Parts that we are going to do FEA are handle assemblies,
where we need to validate strenght of handle when a certain load is applied (like opening the door), and make
sure that the spring would not let the handle activate by accident when encountering 30g of force (like in a car accident).

How much does design space cost?
annual maintenance?
Need to figure out ROI.

 

[brad]

The opening of the car handle due to a 30g force is not the type of analysis which one would use a linear code (such as DesignSpace) to do. This is a nontrivial analysis.

Actually, I've done this very analysis in the past, and to do it properly one must account for significant rotation of the part.

A nonlinear FEA code is necessary for FEA, but I've also seen this done using mechanisms codes.
Linear FEA will not fulfill the requirements. You'd be better off farming the work out, rather than throwing away money for a cheap but ineffective solution.

Brad

 

[ramshree]

I agree with bradh. Its better to outsource on occasions where you think it might need an expert opinion. By the way.. it has been a while. Did you actually buy Design Space or did you out source? Just curious.

Rob.