What is the future of FEA softwares? 4

[Mech5656]

Hello All,
I am thinking to invest to train myself in FEA (mechanical, thermal, drop test etc). So, before doing that I have two questions.
1) Which FEA software is mostly used in industry? In other words, which one is more in demand?
2) Where do you see FEA in the next 15 to 20 years? Do you think it will actually become a separate field of engineering? Or it wont be used that much? Is the demand for it increasing?

I hope I hear from you all.
Thanks

 

[GregLocock]

1) which industry? A big company in just one industry might use 3 different pre processors and 6 different solvers.
2) To be honest I'm waiting for the shoe to drop on this one. I think the integration of FEA into CAD is a really bad idea, and at some point somebody is going to get killed because a design was signed off on the basis of an FEA done by a draughtsman who missed something in his analysis. That's not to say that experienced FEAers don't make mistakes, or kill people (I suspect), just that if they are any good they know that the they need huge factors of safety, or a lot of correlation data. I'm also guessing there will be a lot more co-simulation, and I hope there will be better optimisation tools, or at least more widely used ones. Looking at the questions that come up, improved modelling of composites would be nice.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[IceBreakerSours]

While I second everything Greg has said, let me give you a different take: Do you want to be a designer, analyst, or a developer? Keep in mind that this naming convention used in one industry might have a completely different meaning in another but I hope the idea gets across.

For instance, a developer might be interested in coding advanced elements, materials, implicit-explicit co-simulation, homogenization based multiscale modeling, contact (elements?), isogeometric analysis, meshless methods, advanced damage modeling, .. .. .. So, a developer is the one who makes the tool. They are the ones you bug when you discover a bug

An analyst is one who likes to use the tool to get a job of their interest done. While they must understand how the tool was put together (at a higher level, at the very least), they are more interested in the application of the tool.

A designer is, perhaps, the furthest from the 'source' i.e., the developer. They might care more about the various design requirements of the device they are working on. Whether it will actually fulfill those requirements is for the FE analyst to test and find out.

Once you know where you fit in this spectrum and which industry you are most interested in, the answers to your questions might be more meaningful to you.

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[Mech5656]

Greg and IceBreaker:

I am interested in FEA for pressure vessels, pipes, storage tanks and other related mechanical parts.
I want to work as an analyst. Please suggest me a the right FEA software. I want learn it and apply it. Thanks

 

[ShellsPlatesMeshes]

masihtukhi2012: I would suggest that you have a look at the following software; it is a good point to start from:

http://members.ozemail.com.au/~comecau/quad_shell.htm

The codes and models are developed to cover your specific areas of interest; there are several FEA models with numerical results and a discussion of the latter.

 

[Mech5656]

Hello all,
I am very new to simulation and FEA, could you please recommend some basic books that I can read to understand and learn FEA simulation. I think Cosmos is what I want to learn.

 

[stressebookllc]

Cosmos is decent, there a lot of choices.

You can check this if you are interested:
[URL unfurl="true"]http://www.stressebook.com/finite-element-analysis/[/url]

http://www.stressebook.com

Stressing Stresslessly!

 

[TGS4]

If you are, add you say, interested in FEA for pressure equipment, I would prefer you to NOT focus on any specific piece of software, and rather focus on learning and understanding the appropriate Codes and Standards.

With zero experience, you will want to be hired by a company so that you can be supervised and receive mentorship. They will want you to use their software, and marrying yourself to one specific piece of software may end up making yourself less marketable.

 

[stressebookllc]

Fact is that for a given solution type, most reputable codes are doing essentially the same things other than some special features developed by their respective developers based on customer feedback.

ANSYS, FEMAP, ABAQUS come to mind and they all are mature products.

http://www.stressebook.com

Stressing Stresslessly!

 

[gfbotha]

Hallo Greg

I fully understand and agree with you about the concern that a draughtsman, or designer (not experienced with structural analysis as such), might cause expensive errors to happen. And, that CAD integration increased the chances of this to happen (purely because of the FEA being available). But, I strongly disagree with you generically stating “the integration of FEA into CAD is a really bad idea”!

CAD integrated FEA in the hands of a sufficiently trained/experienced design engineer really made life much easier w.r.t. pre-processing and evaluating changing/evolving design geometry. It also promotes early-on analysis which I think most of us would agree helps to prevent more expensive changes later on during the product life-cycle. I believe one should not shoot down the easier-to-use software, but companies/engineers/managers should rather be aware of the danger and lay down internal rules and procedures to avoid such misuse. There are well-educated/experienced engineers performing both the design and the required analyses (think small to medium sized businesses). That said, more complex (say, non-linear) analyses might typically still call for more specialist FEA software perhaps without this tight CAD integration we are talking about.

Regards
Gert

 

[IceBreakerSours]

Couldn't agree more with what has been said above. You must learn the FEM, not one or the other code. Even if you do start with a given code, it will take a *LONG* time before you can even think of yourself at an intermediate level.

The best thing to do is to ensure your mechanics (in your area of interest, specifically) is solid and then take a course or two on the FEM. If you can not take courses, then try NAFEMS workshops/workbooks.. If that does not seem to work out either, then there are a LOT of freely available online resources including open source finite element codes.

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083