Holger Wirtz,
Thank you VERY MUCH for the tip. This was a really good one. I read it and
said to myself "what an idiot I am, why didn't I think of that!". We need
much more of this type of thing on the news list to make us all better
builders. I'll tell Rick Davis about this so that the next time we have a
trapped volume, we won't have as much risk of failure.
There are other tricks that work pretty well with the SLA. The following is
one that I tried and you might be interested in exploring further. So, for
what it's worth...
PROBLEM: Eliminate de-wetting for 0.003" parts
OK, here it is. (BTW, I did this on an SLA-250, and am not sure how it
would work with a 500, etc...)
1) Support your part and orient on platform like normal.
2) Make a copy of the part with another name (leave in same CAD space)
3) Make 2 "dummy" parts that are simple cubes, 1 unit x 1unit x 1unit in
size. No supports needed here (don't worry).
4) Position the dummy cubes on two corners of the platform extents.
Scale the first one in the Z-direction, such that its min and max Z
correspond to your part min and max.
NOTE: This will cause you to be unable to make multiple copies of your
part using prepare in the SLA. You'll see the purpose of their size and
location in just a second...
5) Load the two copies of your part, the part support, and the two cubes
into your Partman spreadsheet.
6) Change the order of the parts in Partman, such that ONE of the part
files is FIRST, ONE of cubes is SECOND, and THEN your part support and part.
The reason the order is important, is that their order in Partman
determines the order in which they will be drawn in the vat. This will make
sense shortly.
7) Change the Z-wait to 1 second (I seem to recall this was the minimum
time I could enter, put 0 if it will let you). The point is to not have any
Z-wait, such that after recoating is complete, that laser will immediately
read its power and begin drawing the first copy of your part within a couple
seconds, thus curing material before it dewets. What about my Z-level?
Don't worry, we're not done yet, read on (if you care...)
8) Modify the draw parameters for the first part by editing the draw style.
The goal here is to draw a very thin, almost imperceptible "net" which
corresponds to your part on each layer. I did this by using a hatch spacing
of about 2mm, and the old triangular hatch style (I think Partman calls it
equilateral). Set it to NO FILL as well. The trick here is not only the
HATCH SPACING, but just as importantly, the CURE DEPTH. I don't remember,
but I think I used like 0.001" or so for cure depth, maybe less. The point
here is to change the properties of the resin by curing it just the tiniest
bit. This step is what creates the thin NET over your part. The net should
be free floating over the last layer, and only serves to stop the dewetting.
I tried to make it as thin as possible such that the resin could still
level over the last layer even after the net is drawn. But wait, I thought
we set the Z-wait to nothing?! That brings me to the next step...
9) Remember those cubes we created? Their purpose in life is to create a
Z-wait after some drawing has already taken place, but before it finishes,
and also to be located near the edge of the platform so they can be blocked.
The time it takes to draw the cube is your Z-wait. I calculated this based
on several factors. Cube size (X-Y), hatch spacing, laser power and beam
diameter, and cure depth are the variables which affect the draw time for
each layer of the cube, hence your simulated Z-wait. I took the Z-wait I
wanted as my solution, and varied the cure depth to obtain this, holding
hatch spacing etc. constant. You could easily use hatch spacing as your
variable as well. Inside the machine, I taped a bent pieces of sheet metal
(L shaped) to the wall such that they BLOCKED THE LASER from ever actually
curing the cubes (well towers after the scaling) in the vat.
Well, that's it. Quickly drawing a thin net over the last layer immediately
after recoating prevents de-wetting. The "dummy" parts serve as a "soft"
z-wait.
All I know, is I am grateful to 3D for designing a new recoater and a
machine mechanically capable of creating THIN layers. Maybe I'll get my
SLA-350 some day..................... :)
Any other SLA TRICKS out there?? Don't be a RAT, I expect an AVALANCHE of
posted material :) :( Go Florida Panthers!!!
E. Derek Smith
Engineering Prototype Center
Motorola Radio Products Group
Ft. Lauderdale, Florida 33322
(954) 723-4790
eds014@email.mot.com
________________________________________________________
To: rp-ml@bart.lpt.fi@INTERNET
From: wirtz@indy.ipt.rwth-aachen.de@INTERNET on Fri, Jun 7, 1996 2:07 PM
Subject: Successfully building parts w/ trapped volumes
X-Authentication-Warning: bart.lpt.fi: major set sender to owner-rp-ml using
-f
X-Mailer: Z-Mail (3.1.0 22feb94 MediaMail)
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0
Precedence: bulk
X-Mailing-List: rp-ml@ltk.hut.fi
Hello Prototypers,
I changed by subscription to this list a while ago to receive the digest,
but
never got even one ... this is why I am back to normal subscription. Maybe
the
list"master" could try and figure out what the problem is? Thanks.
About the subject: I'm not sure if someone has already come up with this
(since
I've been off the list for quite a while as mentioned above), but I'll post
it
anyways cause I think it has some significance. I've built parts with
trapped
volumes easily with the following trick:
I am creating a part that is somewhat similar to the trapped volume (i.e.
sthg.
like an offset of the walls of the trapped volume, e.g. with a support
software
like Magics by turning around the part, creating a solid support and
flipping
back again), positioning it inside the trapped volume and I am building this
part together with the original, except that I am building the dummy part
with
a hatch spacing of about 5 mm. This way, the trapped volume is divided into
a
lot of smaller trapped volumes, but wiping the material off of those smaller
volumes is not a problem at all. I have succesfully built lots of parts that
way; the only disadvantages are that cleaning up is somewhat more work, more
material is used and building the part takes a little bit longer than
usually.
I have not experimented with larger hatch spacings yet, but I'm sure larger
hatch spacings are tolerable, depending on the viscosity of the resin. I am
not
sure how SLAs handle hatch spacings of 5 mm, we are using a STEREOS 600 here
which allows separate values for hatch spacing, scan velocity and laser
power.
Comments are welcome. Happy building.
------------------------------------------------------------------------ "Opinions expressed are solely my own and thus do not represent or are in line with those of my employer ..." Dipl.-Ing. Holger Wirtz email: wirtz@indy.ipt.rwth-aachen.de Research engineer at the Fraunhofer Institute of Production Technology Steinbachstr. 17, 52074 Aachen, F.R.G. phone: +49 (0)241 8904-120
This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:37:23 EEST