ATM Re: atm-digest V1 #346

[Bob Lombardi <blombard@iu.net>]

>
>From: Wolfgang Birkfellner <birkfell@merlin.pap.univie.ac.at>
>Date: Fri, 20 Sep 1996 16:25:27 +0200 (MET DST)
>Subject: shareware fem
>
>Dear ATMs!
>The idea of using FEM to design telescopes is of course great, and 
>altough I have some expirience with numerical simulations I have
>not yet seen (or sought) a shareware program for finite element 
>modelling. I spent so much time programming the weirdest things 
>that I am not too enthusiastic about adopting some routines from the 
>Numerical Recipes to suit the specific needs of telescope makers -
>it would be a long way from the algorithms to the solutions. Please 
>let me know about the ftp-server where CADRE is located. 

There are a few FEM analysis programs out there.  I've tried two called
PC-Stran and Grape 3D.  Of the two of them, I prefer Grape3D.  I don't know
if it's on the web; I bought mine a few years ago as a shareware disk from
Public Brand Software and registered it around then.  Last update I got was
version 2.0, and that was before Win95 hit.  Guess I need to go look them up
and see if there's anything newer.

Anyway, the nice thing about Grape is that it does lots of graphical things
that you really have to be a graphics programming wiz to do.  To program
something like this, you'd have to be an expert at graphics programming,
mechanical engineering, and numerical methods.  I modelled some truss
designs and you can both see the deflection of the structure under load and
animate the deflection.

>
>Another tool that would be helpful for us ATMs would be a program (or 
>a worksheet for an algebraic program like Maple or Mathematica) that 
>could numerically solve the PDE of thermal transport (parabolic initial 
>value problems) in three dimensions. I have been searching for 
>something like that some time ago when I had a discussion with 
>Bratislav on the method of insulating a piece of crown glass 
>during the process of polishing, but I did not find anything 
>for Maple (or Mathematica, which is not the hell of a program).
>I did, however, not search for a standalone application yet. 

I'm a Mathcad weenie, which includes a subset of Maple.  Mathcad is not good
at PDEs, at least in my experience.  But perhaps some simplifying
assumptions can be done to make it more solvable.

>
>If someone out there knows something about a shareware program that 
>solves this kind of problem (there are commercial solutions) it would
>be great to hear about a server for this program. It would be 
>interesting to quantify the effects of Your warm hands during 
>figuring or the temperature distribution caused by a fan in the 
>assembled telescope, wouldn't it?
>
>Regards
>Wolfgang
>
>------------------------------


Bob


Bob Lombardi            WB4EHS in Melbourne, FL         o         \---\---\
blombard@iu.net    or   blombard@freenet.fsu.edu       /\             |     
http://www.freenet.tlh.fl.us/~blombard               _ \ 7  Bicycling |
telescope making, optics, astronomy, piano, SW and  (*)/(*) ham radio |