Monthly Archives: January 2013

An Emerging Symbiosis

I’ve read plenty of articles that boil down to, “What is LEGO going to do in The Future, when anyone, ANYONE! can print their own LEGO kit in their basement?”

One can swap LEGO with whatever $manufacturer you choose: Tinkertoys, Lincoln Logs, K’Nex, Colt Firearms. (The 3D printing of firearms is a topic I’ll visit sometime soon, but for now let’s stick to things you can buy at Toys R’ Us.)

The implication of these articles is often that $manufacturer is ultimately screwed. Sooner or later, The Future will arrive and $manufacturer‘s revenue stream will dry up as we all print their products willy-nilly in our living rooms.

I see things differently.

This afternoon my son was playing with the new K’Nex set graciously provided by Grandpa and Grandma. He couldn’t find the one gray piece he needs to finish his roller coaster, so I volunteered to print him one.

Thingiverse doesn’t have the specific piece I need. Nor should they; K’Nex owns the rights to their designs and they deserve to profit from them. They’ve invested the capital in building their factories and hiring designers, and it’s their right to enjoy the fruits of their efforts. Thingiverse probably doesn’t want or need the legal headaches associated with hosting knockoff models of popular toys.

(I just realized I’m making arguments similar to those made by drug companies. Maybe there’s a path to The Future where designs go “generic” after a couple of years so that $manufacturer has an incentive to continue creating new ones? I digress.)

So Thingiverse doesn’t have the piece I need, but they do have pieces from the Universal Construction Kit, which lets one link toys that weren’t designed to be linked– Tinkertoys to Lincoln Logs, Duplo to K’Nex, and so on.

So I grabbed a K’nex -> Duplo connector, did a little vertex surgery, and now I have the piece I need as a 3D model.

knex

To print!

This is what I got, printed in red ABS. Note the difference between my bootleg copy and the original. Cue sad trombone.

knex_fail

You could argue that the MakerBot Replicator is *the* top-of-the-line home 3D printer. There are others in its price range that have more or less the same capability, and if you go an order of magnitude more expensive you can get yourself a small industrial quality prototyper with superior resolution.

Part of this failure is surely inability to configure my printer settings properly, but I think a larger part of it is that the Replicator just can’t print this precisely yet. K’Nex parts are cast in ABS, which effectively means they’ve got molecular-level resolution. No way I can get that on my desktop in 2013.

A further limitation of the additive printing process is those overhangs– even if I could get the resolution right, gravity would pull those long grooves down and make the part significantly less K’Nex-able.

You might get around this by printing the entire piece as a column, but my experience in printing toothpicks vertically is that they tend to fail.

So K’Nex gets a pass until the resolution of home 3D printers increases. Maybe the Form1 will be able to do it. We’ll have to wait until the Form1 is out in the wild to see. I’m hopeful, because right now a lot of my ideas are limited by resolution and gravity.

What about lower-resolution objects, like Tinkertoys? Tinkertoys are well-within the resolution range of the Replicator. I’ve already made a couple of Tinkertoy-compatible items.

apple

There’s nothing stopping me from printing a Tinkertoy clone right now except time, but the manufacturing process that makes Tinkertoys at gets them to the toy store (or, in my case, Amazon) is so much more efficient than a Replicator that it just doesn’t make sense, it can’t *ever* make financial sense to print these toys at home unless the price of plastic feedstock drops to near zero.

(Ultra-low-cost feedstock might happen with a descendant of the Filabot Reclaimer, but it’s not here now.)

So I don’t see home fabrication of toys taking off anytime soon, but here’s what is happening. People are using their 3D printers to extend the capabilities of their existing toys.

gears This is the emerging symbiosis: traditional manufacturing is now and likely will forever provide the base of the mass-market toy experience, cranking out LEGO bricks and Lincoln Logs by the millions. Home-based 3D printers will make small production runs for niche items– gears that extend your Tinkertoys, wacky Lincoln Logs accessories, and the like. Over time, $manufacturer will adapt to the proliferation of high-resolution home printers by offering certified premium models, guaranteed to be interoperable with their mass-manufactured cousins.

zheng3_penny_catapultIn short, I think 3D printers aren’t going to kill the toy industry, they’re going to make it much more creative. And haven’t even mentioned new toys like Dutchmogul’s Pocket Tactics, or my pet project, Seej.

Don’t let your wards grow up to be junkies

Modeling a 3D-printed boxing glove arrowhead in around 20 easy steps.

glove

6:29 AM Start with a cube. Riveting.

glove_001

Subdivide the cube and scale it in Y.

glove_002

Pull some points until the cube has the rough shape of a hand.

glove_003

Do a test smoothing to see if the basic shape is going to work out. Looks good.

glove_004

Go back to the low-resolution form and add a cube for a thumb.

glove_005

6:44 AM Scale the thumb and add a little detail at the tip and a knuckle joint.

glove_006

Boolean join the two forms. This is going to create some weird geometry at the thumb/hand junction. I wouldn’t do this if I was modeling for a video game, but it’s OK here since I’m not going to deform the model much. It bugs me on an aesthetic level, but I want to get this done fast and not worry too much about immaculate topology.

glove_007

Add a beveled cube for the wrist.

glove_008

Smooth the cube a bit and squish the insides in with a lattice deformer. I haven’t actually booleaned the wrist to the hand yet. Smooth the hand and use the Sculpt Geometry tool to further smooth out some sharp edges.

glove_011

Create a cylinder to punch out the arrow socket. Again, from a video game modeling perspective, this geometry is an absolute mess. It’ll print fine though.

glove_012

Test print! Looks pretty good, but it’s a little stiff. Back to Maya.

test_print

Deform the entire glove with a lattice to give it a little cartoony character.

glove_013

Uh oh. The arrow socket got deformed when I twisted the glove. Easy fix though. I wanted to make the socket just a wee bit larger anyway so the arrow was easier to remove.

glove_014

7:00: Breakfast break. Did some other things until 8:20 or so.

8:24 AM: A final once-over with the sculpting tools to introduce a little randomness in the geometry.

glove_015

Final print in red ABS, because (red funny > white funny.)

glove_with_arrows

Designed to fit these arrows, but will probably fit a pencil too.

If the title of this post seems nonsensical to you, check out Green Lantern/Green Arrow #85-86, wherein Speedy gets hooked on horse.

download