RE: [rp-ml] SLS info and experiences

Edit: Vanguard is a 2500+ repainted. The upgrades cost extra either way.
So either a 2500+ or Vanguard would be a good option for a
SinterStation.

Gregory M. Paulsen

Prototype Productions, Inc.

Office 703.858.0011 x314

Cell 540.974.1348

gpaulsen@protoprod.com <mailto:gpaulsen@protoprod.com>

www.protoprod.com <http://www.protoprod.com/>

From: owner-rp-ml@rapid.lpt.fi [mailto:owner-rp-ml@rapid.lpt.fi] On
Behalf Of Greg Paulsen
Sent: Tuesday, December 20, 2011 3:05 PM
To: Biermann, Paul J.; rp-ml@rapid.lpt.fi
Subject: RE: [rp-ml] SLS info and experiences

Hi Paul,

New contributor here. I have 4 years of experience running a DTM
SinterStation 2500+ (now 3D systems). I also have used FDM (Dimension)
and 3DP (Z-Corp 450) for about a year.. and toyed a bit with a RepRap.

I can tell you that with SLS there is only one way to make a part
correctly and a whole bunch of ways to mess things up. Process control
and routine cleaning is the key to repeatable build successes. The 2500+
we have at PPI is a workhorse; it's been running for around a decade
with continuous operation (3-5 builds/week) with what I would consider
"normal" maintenance and part replacement. It does not have high-speed,
high-q, or multi-zone heater upgrades.

What I like:

1. Parts can be used in end-use mechanical applications.

a. A lot of our work is DoD related, meaning that the parts see
harsh environments and shock/vibe.

b. Tolerance is tight (as long as you scale frequently) so you can
make first-article parts and electronic housings before (or instead of)
getting a mold

c. Generally the Nylon 11 & 12 (and various filled powders) can
fit the particular need for the job.

2. What I don't like.

a. Constant cleaning of machine, work area, and self. This is
dusty and you want some sort of venting or vacuum for powder mixing,
sifting, etc.

b. Breaking out part requires multiple workstations (sifting, bead
blasting, air blasting).

c. Requires its own room/lab - our is about 20x40' (this is also
where we store powder)

d. Build setup, mixing, etc. can be time consuming.

e. A lot of opportunities for human error.

3. What is most cost effective.

a. I'm assuming you're looking into buying on a research grant. Go
for a used Vanguard because it's essentially a repainted 2500+ with the
high speed and temperature control upgrades.

b. PROCESS CONTROL - again, your material use and quality control
has everything to do with how consistently you run the process. You
should be treating every job as if it is for a end-use customer. I have
managed to get a great refresh rate for my material by double sifting
cake powder, keeping overflows and cake separate, and running scales
after making a new mix. It is okay to get rid of cake when it shows
signs of significant degradation.

                                                               i.
Also, as a side note - put everything on wheels... it allows you to
clean around things and reduces strain of moving materials from place to
place.

                                                             ii. A
general process rule for after a build has cooled is to get the chamber
ready to run another build BEFORE you go to break out the parts just
made. I usually remove the cake and set it on the sifting table, then I
go to work on the machine in the morning. The afternoon I program/start
build then move to the parts to break out (unless they are next-day
priorities).

c. Maximize parts per layer Z. You'll get some intuition about
your machine that will tell you when a Z layer becomes too much or too
complex, but if you have high-speed and temperature upgrades you
shouldn't have trouble doing this.

4. Dust.

a. Unfortunately I don't have experience with newer systems from
3D or EOS. Dust can be controlled using vacuums/vents and a steady hand.
Personally, I drag my shop-vac to every work center in the SLS lab.

5. Powder Recycling.

a. Go back to Process Control. I've created a system where I have
a mix of %Virgin/%Overflow/%Cake that works and simplifies things into
55 gallon drums. I'd like to automate this over the next few years. A
melt-flow indicator is a handy tool to determine/quantify mix quality.

6. Changing materials.

a. Cleaning the chamber isn't too bad of an issue. I have more
beef cleaning out the sifter when changing mixes... especially because
I'm using a black powder and sometimes have to go back to a white
powder. Overall, just add an extra hour or two to your routine and wear
a respirator =]

One more thing - there are Cons with SLS, but the parts are much more
useful for my company's application (applied electromechanical
engineering) than SLA, 3DP, or FDM. Everything else would just plain
fail over time by peeling, snapping, or shattering.

Regards,

-Greg Paulsen

Gregory M. Paulsen

Prototype Productions, Inc.

Office 703.858.0011 x314

Cell 540.974.1348

gpaulsen@protoprod.com

www.protoprod.com <http://www.protoprod.com/>

From: owner-rp-ml@rapid.lpt.fi [mailto:owner-rp-ml@rapid.lpt.fi] On
Behalf Of Biermann, Paul J.
Sent: Tuesday, December 20, 2011 10:50 AM
To: rp-ml@rapid.lpt.fi
Subject: [rp-ml] SLS info and experiences

Could anyone on the list who has used SLS, either 3D Systems, EOS or any
other system share their thoughts? What do you like, what don't you
like? Which is the most cost effective? I know older systems let a
lot of dust into the environment around the machines, are the newer ones
any better? How well can you recycle the powders? How difficult is it
to change materials?

I really appreciate any feedback you can share.

Thanks,

Paul J. Biermann
Principal Professional Staff
Composites / M&P Engineer
Research and Engineering Development Dept.
15-W132

Johns Hopkins University / Applied Physics Laboratory
(240) 228-6911 (w)
(310) 266-0098 (c)

paul.biermann@jhuapl.edu