civilstudent

Hei!, is finite-element work fun and exciting? Is it intelectually challeinging, or does the computer just do all the work for you?

mcim

From what I understand, you set the parameters, equations, limits etc and the computer does the rest. I've had a workshop or two in it before but never did anything with it. Its a pretty good tool though.

gthopeful

There're commercial FEA software for most applications. Actually coding one is pretty time consuming and not the best use of your time.

Blah2009

Computer does the computation work but you can understand exactly whats going on by taking finite element methods courses which describe the differential equations used to model deformation. When I use it in my structural engineering research, I define the geometry, boundary conditions, contact deformation effects, loadings and nonlinearities all through the creation of a customized input file that is then run by computer FEA software (ABAQUS in my case). You'll soon find that setting up the geometry and nonlinearities through a mesh or equivalent loading approach is the hardest part. Finite element as a research tool itself isn't groundbreaking or exciting but the problems you use it to solve might be.

Mr Payne

My friend had a job using Ansys. He hated it and quit. You have to want hardcore analysis to do that job well - if you don't like it, you'll quit.

BoelterHall

Is FEA a graduate level course, or undergraduate? It's not part or my Mechanical Engineering curriculum for my B.S, so should I be worried? It also seems like a lot of internship positions would like you to have some knowledge about FEA programs too.

civilstudent

"My friend had a job using Ansys. He hated it and quit. You have to want hardcore analysis to do that job well - if you don't like it, you'll quit."

Was it too hard for him or just boring? What kind of analysis did he do?

Blah2009

Finite element courses are graduate level. You need element theory at the graduate level and some other graduate prereqs ideally before taking finite element courses.

aibarr

Coding one is how you learn what the commercial FEA apps are doing.

FEA software is very prone to garbage-in, garbage-out. FEA experience is precisely the reason why I've gotten offers in aerospace and sub turbine design, and why when I was casually talking to a NASA-Ames engineer (at a wedding shower, no less) a month or so ago, I was offered a job on the spot. If you don't know exactly what you're doing with boundary conditions, you're going to screw it up and to someone who doesn't know material behaviors extremely well, it'll look right. That's why FE analysts are in such high demand.

FEA is definitely graduate work. Don't worry, nobody's going to ask you to do finite element theory without some pretty heavy theoretical mechanics courses first, which you'll start in grad school.

I think it's fun and exciting. It's very challenging. It requires a lot of really hard-core analysis, and if you get a job doing that kind of stuff, your brain's gonna feel gelatinous at the end of the day... That's the exhausting part, and it's why a lot of people give it up.

Finite element is extremely applicable, and Payne's buddy could've done anything from designing missile nosecones to designing engine blades... My non-academic experience was in blast-resistant enclosures (basically explosion-proof guardhouses) for a government contractor, using ABAQUS and ADINA.

FEA is just breaking things into itty-bitty, highly analyzable and very regular pieces and making the computer do all the heavy lifting for you. The theory behind it is very complex, though, and like pretty much all of engineering computing, you can do things very badly if you don't actually know your jazz. It just *looks* deceptively easy to use...

Mr Payne

He was doing turbine design. He just found it boring (very bright guy). If you've used ANSYS before, you'd know exactly why it's boring.

keefer

At Michigan, Finite Element methods is a popular technical elective among juniors and seniors. (ME305) It gives you some perspective on how engineering analysis is done in the real world. I also took a graduate level Fluid Dynamics course, which is basically fluids finite element modeling, they are time consuming classes, and whether your model is right or wrong is often a fine line, this is especially true for fluids. You have to know the theory behind it to use the tool properly, ie. knowing the right parameters to set, the right modes, and different settings for each individual problem.

I know a few grad students who ended up doing analysis for big defense contractors, doing flow analysis for commercial and military planes, and they seem to like it. I personally thought it was way too one dimensional of a career.