I’m very excited to announce that we have put yet another designing mechanical parts tutorial live. The third tutorial follows the first one that deals with material properties and the second that shows you the design issues of an actual project. This third installment is extra special. This is some of the most in depth information ever published online with regards to Selective Laser Sintering(SLS). Furthermore it is not some marketing blurb that you can find all around. This is actual testing information that if you understand & use it will let you know more about 3D printing than most everyone.
SLS is the process behind our White, Strong & Flexible material. This tutorial is nothing less than initial design rules & in depth information on the detail resolution of the SLS process. This information is based on research conducted by Dominik Sippel for EOS GmbH. EOS one of the world’s largest rapid manufacturing machine manufacturers. They have decided to share the accuracy and technical information behind their technology, with you.
Detailed graphs and tables show you gap sizes, gap deviation, wall thickness, hole accuracy, etc. This is huge. An engineering driven company in the very competitive rapid manufacturing business wants to give you, the Shapeways community, accurate hereto internal data so you can use this information to design and make things.
With online and open source we have been coddled with documentation, API’s and outreach by the people who make a product to the people who use that product. We think nothing of downloading some source code here, messing about in someone else’s Bugzilla or reading release notes. Good luck though in trying to get detailed information on the limitations of your toaster or the schematics of your LCD TV.
In manufacturing it is virtually unheard of to let people in on your own research and to reach out to them as EOS is doing here. A lot of people are talking about opening up R&D, co-creation, reaching out to inventors and consumers alike. But, very few are actually doing it. So, once again, this is huge. It might seem a bit boring and a bit nerdy, but it is huge. On to the design rules tutorial.
Very useful stuff! I look forward to being able to apply this information. Thanks for getting us this data!
What I’m hoping for next, beside flying cars and housekeeping robots, is to see how WSF material behaves under dynamic cases: For example plastic and elastic deformation statistics for strips and rods of various thicknesses. Also I’d be curious in testing of the lifetime of the materials under stresses (repeatedly bend strips and rods of materials until their point of failure). Heck, maybe even such things like finding the coefficient of friction of WSF vs WSF as a function of contact area.
Thanks for the useful info, and keep up the good work. I wish I had these stats before I sent my models to order last week. 🙂
Psawhn,
We’re working on deformation statistics. We also have data on stresses and we will get that to you also. Friction is something we have not come up with a test for.
Joris
Sounds incredible! Looking forward to it!
I just noticed that pins were consistently thinner than designed – this is definitely important to keep in mind. I also find the negative correlation between the size of pin and magnitude of deviation interesting.
If I may make a suggestion to the graphs: I would find them easier to read if the independent axis was nominal wall thickness, not measurement number, so the graph would be Deviation vs. Nominal Thickness. This would flatten out the curves and make correlations and lines of best fit more intuitive to see. I could also be picky about choice of colour, but I won’t. 😉
Psawhn,
for this tutorial we do not have the raw data, we only have the output. For the other tutorials that will follow we do have the data so we could do things like that.
Joris
Great stuff guys!
One thing that’s not a 100% clear to me though: regarding the “Switch from edge to contour exposure”.
How I understand it is that for walls below 0.8mm the laser’s path goes back and forth but (because it doesn’t want to overlap itself?) the paths don’t fit and the actual wall thickness therefore is bigger than the nominal thickness.
Am I right? Maybe it could be explained in a different way. The image is quite confusing to be honest.