I have seen her work over the last couple of years, really slick stuff.
She uses Rhinoceros for a lot of her design work, which is where I
know her from on the Rhino newsgroup.

Tim Wescott wrote:
carl mciver wrote:

"Tim Williams" wrote in message
...
| "Richard J Kinch" wrote in message
| . ..
| http://www.bathsheba.com/sculpt2/
|
| Machine THIS
|
| I'm afraid that would be just a little difficult...
| Cool stuff though.
|
| Mmm, stereolithography...(alllllggggggrrhhhh, drool...)
|
| I wonder just exactly how hard it would be to make a machine that does
that.
| I mean, people make foam and wood milling machines, not to mention CNC
| conversions and so forth all the time. Use a knee for the Z axis
and add
a
| mechanism to spread the media evenly.
|
| Tim
|
| --
| "California is the breakfast state: fruits, nuts and flakes."
| Website: http://webpages.charter.net/dawill/tmoranwms

I just call California Granola Country!

Anyway, this isn't quite the stereolithography I'm familiar with.
Stereolithography takes a bath of epoxy or metal powder and holds a
support
grid at the top, just a fraction below the surface. A laser beam is
played
over the surface of the bath, hardening the media. As one layer is
completed, the grid lowers a bit and a new layer is added. I had a
handmade
prototype of a drill jig I used on the plane that had to fit exactly
into a
CATIA part. To machine it out using conventional methods would have cost
tens of thousands of dollars because the tolerances all around the
part made
it not possible to mill out in one step, so it would have to be set up
again, losing the tolerance that was critical. Cost about $250 and took
less than a week to deliver. Don't know if that was for the pair of
them,
but I was floored. They use it all the time for prototype parts that
allow
us to put them in hand and get a better idea of how they'll work out.

Apparently a similar process has been developed where they take worn
aerospace parts and put them in a bath of powder and build up the
material
right onto the part. I guess new parts are worked on that way also,
as you
can build up very dissimilar metals that way, allowing things like engine
compressor blades to be made to fit all the requirements of the demanding
application.

Seems to me that this type of machine isn't incredibly complicated in
the mechanical sense, but in software it would be a challenge. High
power
laser beam control has to be pretty precise, though.

I've seen it done with wax, too. I never found out if the wax pattern
could be sent straight to an investment casting house, but that was the
obvious question.