Subject: Re: FEA theory needed for FEA software?
From: smithj@I_should_put_my_domain_in_etc_NNTP_INEWS_DOMAIN (James Conrad Pope Smith)
Date: 11 Dec 1996 17:31:47 GMT
trimech@aol.com wrote:
: The FEA world is about to make a major turn-about. A new field is being
: broken off from FEA - the new field is DESIGN ANALYSIS. It's basically
: Finite Element for those without a PhD. I just received a Beta version of
: Cosmos/Works 2.0, the new program from Structural Research & Analysis Corp
: (makers of Cosmos/M) designed to run completely inside of SolidWorks 3D
: design.
Been there done that. Check SDRCs IDEAS. Does the same thing. Mechanica
integrated with Pro/E does it too. Nothing new.
: I was completely floored when I saw this, design analysis with
: Cosmos/Works is completely pushbutton. The program was written so that
: the engineer, at any point in the design, can run a quick analysis on
: parts or assemblies without exporting anything, without starting any other
: programs, just go down the menu or use their graphical interface to mesh,
: add boundaries/loads, and run.
This is not a good thing. For one, nothing is ever pushbutton. That's what
the CAD/CAE vendors would like for you to believe. Now there is much in the
analysis side that is pushbutton, but there is still quite a bit of hand
work that must be done. Try doing a model of an airplane with this pushbutton
approach and you'll fail miserably. For lugs and small piece parts, sure.
But real structures? No way.
: I don't want to ramble on and on about the implications of this, but now
: analysis is a part of the design process, totally seamless, with no time
: or $$ penalty!! If anyone wants any more details, you can send me e-mail.
I'd say it slows things down. We now analyze things to the nth degree, taking
more time. It may be cheaper for piece parts, but again, for intricate
structures where the pushbutton approach breaks down (and having been there
for over two years with Master Series I know it does), design changes kill my
productivity. The designer wants a new analysis of the new geometry.
James
--
----------------------------------------------------------------------
James P. Smith
NASA/JSC, Mail Code ES2
Houston, Texas 77058
smithj@es2linux.jsc.nasa.gov
