You know, with work like this already out there, and with the advent of a print-in-place gearbox, we’re really not all that far now from a clock you print, soak, and then operate. Prospects like that kinda blow my mind. Also, they make me want to see how small I can print one of these with a powder-based printer and still have it run…
I remember a comment a while back from someone who was pretty adamant that sharing everything would pretty much immediately ruin their business, and I have no reason to doubt that, in that case, this is a reasoned judgement. But that doesn’t mean I don’t believe in companies sharing everything, opening their doors, and in general collaborating with their customers, in the open.
But a lot of your customers aren’t used to this, and unless you’re directly marketing to consumers, nearly all of them will be locked in to the trade secrets, patents and trademarks model– and from experience, I can tell you that at least some of them will treat open source almost like hazardous waste.
Winning with Sharing in these cases is a heck of a lot more challenging than if you’re working in a field where what you sell goes directly to individuals, because where individuals are eager to form brand loyalty to the more open provider, institutions in this climate are broadly sharing-averse. But while a radical leap into deep sharing may very well be as dangerous as starting a new company altogether, there are things that aren’t, such as:
Of course none of this will fix the problem that these days bringing a product to market can require a war chest of patents, or that in a lot of sectors open source is a four-letter word, but the steps above are in the right direction, and to me at least they look like good business choices too.
Here’s a neat little Blender script for converting point cloud data (often available for landscapes or as the output of 3D scans) into a mesh in Blender 2.6. There are of course lots of ways to skin a point cloud, (not the least of which is meshlab) but I’m an old Blender nerd at heart…
Speaking of Blender, while most of the really exciting stuff happening in Blender lately has been to do with animation and video, there are a few new features that I’m looking at as potentially pretty helpful for modelers– BMesh is merging to the trunk soon, allowing faces with more than 4 edges, which combined with improvements to the boolean operators might really start to give Blender a real shot at becoming more useful for solids.
So a while back I talked about using OpenSCAD to tack on support struts to tough builds, giving them a better shot at printing properly. My printer’s still not fully calibrated yet, but as you can see, despite huge overhangs, this steggy printed out just fine with some support struts.
But the process of dragging the pegs around in OpenSCAD was a bit tedious, so I worked up a quicker way to place them: Blender! By creating a separate object with all the posts placed and aligned, OpenSCAD only needs to do a single union operation to merge the posts to the model, and an intersection to flatten the feet to the right height. Here’s an example of what I mean– I colored the posts blue so they’d be visible:
The model could probably use a fifth post in the center to hold everything up, but the important thing with posts seems to be that they are wide enough to print solidly, and close together enough that the plastic “bridges” from post to post easily. On my machine and in my climate this is a little under a centimeter– you might try building a bridging test object to see what works best for you!
I had some questions in comments on my peg placement, so here’s a snapshot of the model with the pegs in. When it printed, they snapped off easily and the head turned out pretty well considering how small I printed it!
I didn’t try supports for the belly and tail, and the underside of both came out a little scraggly as a result, but at this scale (less than 8cm long) those details ended up being less important than keeping the head from falling over during printing.
On more complex models I’ve found you can get away with surprisingly little support, but the cost of going too far with this can be anything from a ponytail that twists off before it reaches the connection point up top to total build failure. Like with all home 3D printing, you get a “sense” for what will work and what won’t over time…
Skeinforge has the ability to automatically create support meshing that comes away comparatively easily with a pen knife, but at some point it starts to feel less like printing and more like whittling, and while soluble support structures are less far off than they once were, if you’re like me you’ll still find yourself looking around for a more elegant solution.
Enter the support peg.
The idea is to basically extend a vertical support peg through your model, allowing the fairly common practice of bending or breaking the “45 degree rule” to stretch a little farther than it usually does. For my example I’m going to use this adorable but tough-to-print stegosaurus model:
I go on about how 3D printing and Thingiverse (and of course, the users who know their way around 3D tools) are well-aligned to provide things which aren’t just good enough to serve but which are precisely suited to their desired application, but this thing speaks for itself rather boldly.
It does what it’s for. More or less exactly.
Looking at the thumbnail for this, I literally thought, “oh man, that’s beautiful, no way will it print on a thermoplastic extrusion system though,” before clicking it to discover that that is exactly what has been done here.
What you’re looking at is actually the crystalline configuration of diamond, which is also the exact same configuration of the atoms in silicon semiconductors. In semiconductors, different atoms are pushed into the lattice, replacing silicon atoms, to alter the local average number of electrons, which in turn makes it possible to build diodes and transistors in high densities through a combination of technologies related to photography and, well, clay firing, which enables complex but inexpensive circuits like microcontrollers, which in turn enables low-cost 3D printers, which is where we get models like this one…
So it’s all connected really.
This is video of a 3D-printed, Arduino-controlled animatronic tail. It’s a great example of design following nature, with vertebrae and tendons being modeled by mechanical substitutes, and while you might be able to replicate this version of the design with more traditional prototyping methods, the 3D printing makes it both easier to duplicate and easier to reconfigure. And a few of the improvements I can imagine, such as making the vertebrae interlink like biological ones to be more firmly linked to one another would be really tricky with hand tools, but pretty easy on 3D printing…
So leaving out the fact that a nested Reuleaux Triangle that you can fit in your pocket is something that just oozes cool and, were you to whip it out at a party (at least the kind I go to) have everyone instantly fascinated even if it weren’t something you downloaded and printed, mechanisms of this type particularly interest me because they natively react well to extruded plastic 3D printing.
Captive parts aren’t always possible in a project, when it works, it’s magic.
Subscribe by RSS