Markus,
You are right in assuming that curl can be minimized, but the machine
controls on a SLA 250 won't allow you to accomplish it across the entire vat
or over multiple files with varing cross sections. In other words, the
adjustments you make to eliminate curl will fix one part and distort the
next.
The laser beam width, hatch spacing, penetration depth, and even the angle
of the laser from true vertical are factors that influence curl. To improve
on the current process you would need to re-adjust the hatch spacing as well
as the cure depth based on the laser profiler results for each layer, and
that would require a "slice on-the-fly" setup and build program.
I adjust my hatch parameters on a regular basis to adapt to changes in laser
profile and resin age. To help me do that, I created a set of diagnostic
parts that take a few hours to run and can be cleaned and measured quickly.
I wish you luck, as I've been on this road for quite some time...
I have found that you can normalize the stresses to minimize the curl that
takes place over time, by heating the parts to their glass transition
temperature and holding them there for 20 minutes.
To do this, I hit on a trick for fixturing SLA parts before thermally post
curing them. This was developed by an old sand casting pattern maker...
1.)Fill a metal box with fine sand and carefully embed your parts in the
sand.
2.)Completely fill the box with more sand and tamp the sand until it is
packed solid.
3.)Cover box and place the entire box into an oven and bring it to the
proper* temperature.
(*consult resin info)
4.)Hold it there for 20 minutes.
5.)Remove box from oven and allow to cool to room temperature.
The stresses are greatly minimized without the parts warping because the
sand acts as the fixture and holds the parts from moving.
Larry Blasch
Design Engineer
SLA Administrator since 1990
OPW Fueling Components Voice: (513) 870-3356
P.O. Box 405003 Fax: (513) 870-3338
Cincinnati, OH 45240-5003 USA
*********************************************************
"Always remember you're unique, just like everyone else."
*********************************************************
-----Original Message-----
From: Markus Hitter [mailto:hitterm@fh-trier.de]
Sent: Monday, January 10, 2000 3:05 PM
To: rp-ml@bart.lpt.fi
Subject: Re: Beefy SLA part warpage question.
At 8:20 Uhr -0800 10.01.2000, Joe Allison wrote:
>This is definitely sounds like curl.
>
>The name curl is used because it best describes the shape of the
distortions
>where the extremities curl upwards.
[...]
It seems that the curl normally goes upwards. In our SLA250 with
SL5170 the curl definitively goes downwards. Depending on the hatch
parameters it curls more or less, but always downwards.
So the engineerings brain combines that if it is possible to curl
upwards and to curl downwards there must be circumstances where the
curl is completely compensated.
Now I'd like to do some research and ask for somebody to send me a
part complete with all build parameters or better the complete build
files (I need the .r, .l, .prm and .v file fitting for the
traditional buildstation software). I whould build that part on our
machine and compare it with one built with my own parameters.
Perhaps there are other parameters (Temperature, humidity, resin age
etc.) which have an influence in curling.
So, any commitments are welcome and I'll post the results.
Best,
Markus Hitter
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Dipl. Ing. Markus Hitter http://www.fh-trier.de/~hitterm
RET - Markus Hitter Technologie der Produktentwicklung
Fortschritt in der Entwicklung Konstruktion
Stereolithografie
fon: +49 - (0)651 8103-530
fax: +49 - (0)651 8103-258
For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/
For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/
This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 23:02:35 EEST