home brew rapid prototype machines

From: Brian Bartholomew <bb@...> wrote:
Subject: home brew rapid prototype machines

> The powdered machines (SLS) produce the "best" parts, but they are
> still not that strong.

[I'd heard that the "thermojet" (extruded like hot glue) parts had fairly
good physical properties as well- these machines (Sanders. etc) can use
engineering plastics as feedstock. The major problem seems to be
striations where the deposited threads join one another, although in the
newer machines these are less noticable.]

According to some recent vendor literature, the strength of parts is
improving. They claim to have a feedstock material now that's some
sort of metal, coated with plastic. After you stick the plastic
together with the laser, you fire it in an oven; the metal sinters.
This leaves a porous structure. You then put some copper stock on it
and put it back in the oven at a lower temperature, and the copper
melts and wicks into the pores. This claims to make injection molds
that last for 100K cycles. I don't remember if they said there was a
polishing step, but I would think there would have to be.

[Are you sure it's pure copper that's being used here, or is it some kind
of (much) lower-melting alloy? The oven would have to get awfully hot for
copper to work this way- and it would likely oxidize before it did, unless
it's in a vacuum oven.]

"bfp" <bfp@...> wrote:
Subject: Re: home brew rapid prototype machines

> To sum up, I think home shop rapid protoyping machines are several
> years away, and even if you could build one, what would you do with it?

I saw an article somewhere(design news?) about building up wax prototypes
then lost-wax casting a metal part from them
not exactly the machinist's way of doing things, not enough chips 8^)

[I've heard of machines (for producing jewelry models) that dealt with the
striation problem by combining the action of hot-wax deposition with that
of a milling machine, laying down a rough wax pattern and then refining the
surface with rotary cutters (thus producing chips). Z-corp, whose RP
process cements layers of starch powder with glue, claims that parts made
this way can also be burned out, but apparently the stuff actually leaves a
residue that can result in a high failure rate for castings.]

the machine they showed in the article didn't look like much more than a
flatbed
plotter with a modified glue dispenser in place of the pen and a table that
could
be lowered as the layers were built up
the hardware seems easy enough but the software would take some work
hmm,
solid model from cad system
routine to slice the model into layers
plot layer,lower table,repeat as needed

Brian

[It sounds like one could make a head for a small milling machine that
would substitute the modified glue dispenser for the spindle, and just
raise up as the part grew. Deriving the contour lines is fairly simple (in
Rhino, anyway), and converting them to g-code isn't that tough, but some
sort of batch program would be needed to feed these successively to the
machine with the correct vertical interval interposed. Then swap back to
the spindle head and machine the wax to specs.]

Andrew Werby
http://www.computersculpture.com



Andrew Werby - United Artworks
Sculpture, Jewelry, and Other Art Stuff
http://unitedartworks.com