Pwdr 0.1 based SLS printer

Came across this forum whilst hunting for information. It's good to see there is a growing community out there! I thought I might share my own endeavours and hopefully get some feedback.

I've been playing around with the idea of building an SLS printer for a little while and have finally decided to commit to it. I've based my design on the Pwdr powder bed hosted on github. I've widened it slightly to accomodate a different roller mechanism and made some minor revisions so that generic parts can be used. Hopefully I should have this together within the next few weeks.

Meanwhile, I've been breadboarding my own controller and writing software to cater for the task of controlling the steppers / galvos. This isn't finished as such but is getting there. I really need an assembled rig before I can produce something of production quality. Currently the software will read a gcode file off a SD card and produce the corresponding signals for the galvo drivers. I'm just using cheap "laser projector" galvos as such but intend to either modify this unit with mirrors appropriate for the task or perhaps invest in a better quality system depending on how my current system responds to the additional weight.

I have a 40W CO2 laser on hand which I'll be using. I haven't determined what printing medium I'll be using yet; Nylon 12 seems difficult to procure at a reasonable cost / amount. I have a couple of questions regarding printing mediums; I see some people are using sand, is this the typical sand you would find at the beach? What considerations do I need to make when using this?

The more I read the more emphasis I see placed on how beneficial it is to use a heated build area. Unfortunately my design is a little too simplistic and won't be able to accomodate this. This doesn't concern me greatly as I'm more interested in developing the necessary electronics / software at the moment, DMLS is where my real interest lies. I thought perhaps as a workaround, rather than projecting the beam at a high intensity while executing each layer only once, I could take advantage of the galvos high scanning rate and project the entire layer as if it were an image whilst varying the temperature. I think this would give me control of the temperature gradient without compromising print speed too severly. Any thoughts on this?

Here's my modified version of the Pwdr rig:

First off, for SLS, you need to have your powder temperature in the bed at a few degrees below its fusion temperature, so that your laser power is used for fusing the powder, not for heating it up.

Otherwise, you will need to either vastly increase your laser powder, or dramatically slow down your scanning speed.

So a heated bed is pretty much required for SLS, unless you are trying to print powdered black wax (which has its own issues with curling).

The being said, you only need to heat the surface of the powder - see my BrundleFab design for a layer head with an integrated heater to give you some ideas. I was able to use this to keep my powder surface temperature at around 150C.

With a fully enclosed 'hood' above your printing surface (and yes, unless you are giving out laser goggles to everyone within a sight line of your printer, you definitely need an IR opaque hood!) you should be able to install a fixed heater element and a fan to keep the bed surface at your target temperature.

Also, consider mounting your motors under your pistons - that way powder won't get into the bearings.

Hello All,

Just posted a contact on powder threads but we can offer reasonably priced PMMA powders (Acrylic) 45, 85 and 140 micron or we can also offer polyamide (PA6) co polymer with PE in a sub 100 micron fraction.

If of interest get in touch info@aksechemicals.com

Thanks

Ezrec - Thanks for the comments. It's a little late for major design revisions now unfortunately. I'll persevere with what I have and see what I can achieve. Given that I've seen some people are sintering sand with the same laser I'm still hopeful. I doubt their build chamber is anywhere near melting point. We shall see any how.

aksechems - Can you give us a rough idea on price?