Time and Resolution

I was discussing ways of making better prints with a guy in my office. What follows is the result of that discussion.

A big problem for 3D printing is that you have to trade resolution and time. If you want a cleaner and more accurate part you have to wait longer for it. This is because the printers can’t cover a large area at once. Whatever method is used for the smallest areas has to be used for the largest ones. Maybe you can change tools, but that’s got issues.

Wouldn’t it be nice to be able to form an entire layer at once? Yes. Yes it would. Here is a theoretical method for doing that.

Start with a build surface. Fill that build surface with a grid of tiny holes and in between those holes put tiny electromagnets. Next, flip that surface upside down so that the electromagnets are on the bottom. Now when you inject ferrofluid through the holes you can move it around on the underside of the build surface with the electromagnets. By turning the electromagnets on in a controlled order you can arrange the ferrofluid so that it outlines the exact shape of your first layer.

Now put that into a tank so that the bottom of the ferrofluid, which is hanging off of the build surface, is just touching the bottom of the tank.

Inject the printing substance through the build surface into the cavities formed by the ferrofluid. Even if the substance is liquid, it will be constrained by the floor of the tank and the ferrofluid.

Harden the printing substance in some way. Heat, UV light, catalyst…kind words. Whatever works. Then raise the build surface by one layer height. Fill the tank with enough of a support substance to just reach the depth of one layer height. This support substance needs to be denser than the printing substance.

Inject more printing substance through the build surface surface to fill the cavity formed by the ferrofluid and the previously hardened printing substance. Any overhangs will rest on the support substance.

Repeat this procedure for each layer, rearranging the ferrofluid when necessary.

This process, or something similar, could open up a paradigm in which you don’t have to trade time and resolution. Each area of detail can be resolved at the same time by just controlling all the relevant electromagnets, then the open space can simply be filled with whatever it is you’re using to print. It doesn’t seem like the control electronics would be all that complicated, either. Basically you’re just drawing on an LCD readout. The complicated part of this idea is the various substances. It’s more of a chemistry problem than an electrical problem.

Lower Entry Barriers For 3D Printing

The basic technology required to make a 3D printer work isn’t particularly groundbreaking, so it’s nice to see a brand new design as opposed to yet another copy. Origo is a project (company?) started by Artur Tchoukanov and Joris Peels with the goal of producing an $800 mass-produced 3D printer specifically for kids. It looks like they’re planning on using a double swing arm for the X Y motion, which greatly simplifies the physical construction, lowering the cost. They are also integrating the software so that kids can design things in 3DTin and then have their creations automatically printed. Also, a recycler, but I doubt that idea will work.

More 3D printers, particularly cheaper ones, is great. However, I would like to see the cost and complexity drop even farther. Here are some ideas for how that might happen:

• The printer itself shouldn’t require powerful or precise tools.

The Thing-O-Matic achieves precision with laser cutters, the Mendel achieves precision with another 3D printer, and they both use special steel rods. The designs depend on expensive and difficult-to-maintain manufacturing tools.

I think it’s possible to avoid the use of things like laser cutters, precision ground steel rod, and pre-existing 3D printers. Anything that’s going to be expected to recreate precise movements is going to need some precision parts and assembly, but that’s almost entirely about the layout. For example, instead of laser cut parts one could print a template on a desktop printer, attach it to the wood, and cut/drill by following the guide. The typical solution to linear motion is some kind of bearing riding on precision rod, but some things like aluminum angle and drawer sliders are nearly as precise while being far cheaper.

The real barrier to entry, however, are the precision manufacturing tools. You CAN download the blueprints for a Thing-O-Matic, but they are specifically designed to be produced on a laser cutter. For example, the T-slots aren’t something you can accurately reproduce on your own. Likewise, while you CAN download the parts files for the Mendel, good luck carving them. “Many of the mendel parts are quite difficult to make from wood, and could do with a re-design. They are all created with 3d printing in mind, so there is no consideration for access to internal spaces, or grain or any of the other things to keep in mind when working in wood.” Designs with these requirements create a speedbump which sucks down money, time or luck. Instead, the goal could be for the printer design to require nothing more than a hand saw (cuts) and an electric drill (holes).

• It doesn’t actually need a .5mm nozzle.

Creating sub-millimeter holes is kind of a problem. The only really reliable way to do it is to put the nozzle blank and drill bit into a lathe. If you allow the nozzle diameter to rise you start to run into pre-manufactured components like needles. Or, at a minimum, things like 1/32th (.8mm) drill bits that are cheaper and easier to obtain than .5mm bits (at least in the States). Even 1/16th (1.6mm) bits would be small enough to do something useful (probably equate to a 2.7mm wide track) and are pretty much free.

More importantly, moving up to something like a 1/16th bit wouldn’t require any special tools to use. Even holding a sub-millimeter bit requires a special tool. A 1/16th will fit into a standard drill chuck. Sure, you CAN get sub-millimeter bits that have expanded shafts, but that’s starting to raise the barriers again as people need to special order them and they can’t be easily replaced.

• Make each part do more than one job.

The Mendel (and derivatives) is a good example of not doing this. As great a design as it is, there are metal rods being used for the structure and then different metal rods being used for the linear motion. Using the frame material for linear motion would create more synergy (do you have your innovation Bingo card?).

• Simplify the positioning system.

The reason 3D printers tend to use linear motion is that it’s really simple to program for. 3D files record things in XYZ Cartesian coordinates, so it’s just a matter of calculating the steps. That’s great for the programmers, but not so great for the mechanical engineers. They have to figure out how to create a 3-axis linear motion system on the cheap. Switching to something more like the Origo (a double swing arm) would make the bot a lot easier to build. All you’d have to do is slap a couple arms onto the shafts of a couple motors. The programming would be more complex, and maybe the motors would be more expensive, but the physical construction would be much simpler.

• Reduce the number of electronic components.

Electronics are expensive, even when you build them yourself. The electronics package is around 1/4 of the cost of a Thing-O-Matic, for example. It might be possible to create a “board on a chip” design inside a Field Programmable Gate Array (FPGA) that would literally do ALL the calculations. It could even replace the stepper boards. FPGA’s, instead of running software, are reconfigurable hardware. You program their logic gates and then they just do what they do. They have hundreds of I/O ports, allowing all the windings of the stepper motors and all of the sensors and all of the heaters to be controlled directly by the FPGA (via some form of amplification). Also, and this is important, they are truly parallel. If something needs to happen on one I/O pin it doesn’t have to wait for the software to get to that part, it just goes in and right back out.

FPGAs aren’t exactly main stream, and there aren’t any open-source solutions yet, but you can get the software you need to program them for free (just sign up for a license). I’m not sure how many gates you would need to replace an Arduino and four Pololus, but seeing as how you could do it with a single chip it’s worth looking in to.

• Why not design things without a computer?

Origo is on the right track in terms of making it easier for people (kids) to design the 3D models that 3D printers construct. Why not make it even easier? OpenSCAD, a program that has already proven itself in the open-source 3D printing world, already has support for generating 3D models directly from 2D pictures.

With just colored lines and some labels OpenSCAD can generate a water-tight 3D model. Kids (anyone) could use crayons or colored pencils to draw a blueprint of their design, scan it, and have it start printing automatically.

I’m Conflicted About Buildatron

I have a few keywords like “reprap” and “makerbot” on my google alerts list, so I heard about this new company Buildatron. The link in my email took me to a press release that is responsible for the inner turmoil.

See…I want 3D printing to move into the mainstream. There’s only 4 years left until my inter-family prediction of “this stuff is totally going to be everywhere” becomes premature. I want to go to Wal-Mart and pick up a $100 3D printer that makes cell phones, and I want it now! On the other hand, I am convinced that the old people who own all the stuff and write all the laws are going to fill up their Depends when they realize the same people who pirate music and movies are now free to pirate physical products.

As is pretty obvious to anyone who bothers to look ahead: 3D printers are going to create some legal issues. With the Supreme Court ruling that corporations are people, and Congress worried that existing businesses might not pay for their reelection, it’s not outside the realm of possibility that 3D printing would end up strictly regulated. The exciting thing about desktop additive manufacturing is how much more efficient it makes people. That’s great for individuals, but not so great for existing corporations. Combine the threat of copyright infringement with the decreased revenue from simple (and not so simple) little gizmos and you’ve got a recipe for a new version of the RIAA.

It seems to me the best way to make that possibility unlikely is to push 3D printing into schools where it can give a whole new generation a reason to learn STEM. About the time industry lobbyists are marching around demanding regulation I want school kids marching around showing off how they invented a new pacemaker or something. That will even get the attention of industry (they need all the STEM geeks they can hire). Thus, creating a strong argument for allowing 3D printing to flourish with modest, safety-based regulation.

That’s my point of view, and I’m working on it (slowly; I have a day job). But then I see stuff like the Buildatron 1. The company’s press release stops just short of going “YOU get a car and YOU get a car and YOU get a car”…but then you click on the link.

They put a box around a reprap. Then they put their name on the box. Then they promised a revolution.

The best thing is that there are already companies selling reprap parts and kits assembled or disassembled. There’s makergear and reprapcentral and reprapstores and reprapkit and XYZprinters and botmill and techzone and I’m sure I missed some. Last but certainly not least there is Dr. Bowyer himself. And you know what? None of them bothered to brand an open-source project. Why does the Mendel need a box around it? I assume it was the only way they could create enough space to FIT a brand name in there.

I love open-source for exactly that reason. Since no one had ever thought of putting a big title on the machine there isn’t even any room to add one if you want to. It’s a marketing department’s nightmare.

See, Buildatron instantly invites comparison to Makerbot. Makerbot is from New York, so is Buildatron. Makerbot had 3 techie founders, so does Buildatron. Makerbot was based on the reprap project and so is Buildatron. Makerbot looks like a box…and so does Buildatron. Even the name looks copycat (build-a-tron, mak-er-bot). However, all that stuff is just surface. What seems significant to me is that Makerbot looks like a box because those guys actually designed a new printer. It makes sense for them to put their name on a product that was merely inspired by the reprap project. The Buildatron 1, however, IS the reprap project. With a box around it.

They didn’t even have the decency to make the box printable.

A  few of my favorite quotes from the press release:

Today Buildatron Systems announced the international launch of the Buildatron 1 3D printer…

Their offering represents a new paradigm in 3D personal manufacturing technology…

…social network driven DIY (Do It Yourself) 3D printer kit.

We worked … industry leading customer support via Buildatron’s social network gateway…

Buildatron’s opening of the 3D printer market to millions at industry shattering prices makes no bones about the impact they will have…

Buildatron is working hard to put you in the drivers seat by developing a new generation of tools unparalleled in history.

It’s just sooo cookie-cutter and over-the-top. They’re international because they have a website. They’re a new paradigm because they have a box. They’re social networking because they have a facebook page. They’re industry leading because there’s no industry to lead. They’re opening the market by sitting next to Makerbot at Makerfair. They’re unparalleled because they’re taking what everyone else is working on in a weird direction. The best part is that they could actually lower the price of their kit if they didn’t bother to put it in a box.

All of this leaves me conflicted. I want people to get involved in 3D printing who can hype it. Bre Pettis is awesome because he can get in front of a camera and really evangelize 3D printing. I don’t think an external control panel is all that important or useful, but I’ll let Makerbot do anything they want without complaint as long as they’re putting so much effort into building public awareness. However, the approach Buildatron is taking, right from the get-go, makes me want to tell them to stop.

You know who else saw what Makerbot was doing and thought it was a good idea? Well, Rick Pollack of Makergear and Erik de Bruijn of Ultimaker come to mind. They are both selling their own (branded) boxy 3D printer. But THEY DESIGNED NEW PRINTERS! The Mosaic and the Ultimaker are both brand new and innovative designs that take INSPIRATION from the open-source projects that birthed them. It makes perfect sense for them to give it a new name because it’s new. Buildatron just put a Mendel in a box.

Is Makerbot Going To Make It

CES has come and gone, and as usual a lot of people have done “journalism” on topics they don’t understand. For example:

Glen Stantos of Geeky Gadgets seems to be unclear on the concept of a “work in progress,” describing the Thing-o-matic as “working exactly like your trusty printer.”

David John Walker of Seer Press News took Makerbot a little too literally at their word in describing just how “maintenance free” their extrusion system is.

Gizmag managed to get their description right by simply parroting Makerbot’s description.

At least the Tested guys , who have been running Makerbot products for a while now, were coherent.



Here David M. Ewalt of Forbes thinks that 3mm ABS filament only costs $5 per pound, when the Makerbot store he linked to actually sells it for $10-45 per pound. Makergear sells it for more.

Anywho, that stuff’s not the point. The point is that Makerbot is starting to approach that time in the life-cycle of successful startups when they go from small to large. Or, at least when they have the chance to go from small to large. All this press, and it is well deserved press even when it is inaccurate (reporters overstating something is par for the course), is a good indication that Makerbot has “made it” and needs to seriously consider their business model.

The way I see it, Makerbot can remain true to the open-source “maker” idea they started with and keep putting out a product that appeals to DIY enthusiasts. A big part of that model is continuing to let their customers do a lot of their testing for them. This ongoing extruder problem is a perfect example. Since they can get a dozen bots to print for days without any problems, and there aren’t hundreds of people crowding the forums asking about their dead extruders, this failure mode is reasonably rare … for a DIY project. But (and I don’t have any numbers) it seems like this failure mode is too frequent for a larger enterprise. If Makerbot starts selling an order of magnitude more bots they are going to run out of early adopters. While they might be able to get the general public to assemble a Thing-o-matic (at least a full day of work) they are going to have a really hard time getting them to be patient when it doesn’t work or gradually stops working over a couple of weeks. But even ignoring the assembly time (maybe they sell them pre-assembled), you can’t sell something for $1500 that only works for a few weeks. Not to the general public at least.

Which brings me to the alternative business model. Makerbot could go for the expansion that interest in their product supports, but they’d have to dramatically change the way they operate. Selling thousands of bots per quarter will require an obvious expansion of physical and personnel resources, which (since they’re in America, more specifically New York) will start to cost an awful lot. There’s a reason so many things are manufactured in China. The DIY, early adopter, solder fume addict community will pay a premium for something they can hack … the general public won’t. Also, Makerbot’s current clients will “pay a premium” in the extra time and effort required to make the product work properly … the general public won’t. In fact, selling something designed to be hackable to the general public pretty much guarantees that the first people to buy it will be the members of the general public who THINK they can hack, but actually can’t. These people will probably just poke it until it just breaks, and then demand a replacement. Additionally, since Makerbot’s business is based on open-source designs, if they start to create a new market they will find themselves with competition. Even more fun, they’ll find themselves with their first lawsuit as soon as someone leaves the bot printing overnight and their kid reaches in to grab the extruder nozzle, or someone opens their bot up and leaves it lying on the carpet with the z-motor powered up and starts a fire, or someone sticks their eye under the extruder to see if it’s clogged.

I don’t think it’s practical for Makerbot to actually pursue that second option. Not only do none of the founders seem interested in the corporate mindset, there are already tons of pre-existing corporations that can fill up the consumer marketplace with plug-n-play 3d printers. The fundamental technology really isn’t particularly complex or expensive, so as soon as a demand exists it will be supplied. But that does put Makerbot in an interesting position. They can run full speed to become a large corporation and fill the consumer market’s demand for 3d printers, or they can stay where they are with the open-source early adopter thing. They already pretty much guaranteed that they can’t get their company acquired (bought out) since they release all their IP to the public for free. If they wanted to do that they’d pretty much have to rely on the value in their brand.

What they could do, and what I think makes the most sense, is they could expand the “Makerbot umbrella” to include more products. Whenever they get their 3d printer’s mean time to failure low enough, they can start on something else. We can already see that they’re sort of pursuing this strategy since every now and then they mention that they also sell other things like a 3d scanner and an egg plotter. However, I’m unsure about how long that strategy could last. It seems to me that it would only work if they came up with a new flag-ship product that attracted just as much DIY community interest as the Cupcake/TOM did. See, once someone has a TOM, they don’t need a second one, and the market is just too small for them to continue to run a business off of consumables like plastic.

Basically, I think Makerbot needs to stay in the DIY market, but to do that they need to come up with a new flag-ship product. If they don’t, I fear their initial success will be short-lived rather than sustained. An initial expansion followed by a long, slow contraction. And my intuition tells me that the kind of guys who would start something like Makerbot are not going to stick around when their business becomes boring and gradually shrinks, but they can’t really sell it to someone else since all the information someone needs to copy their business is available for free. So, if they don’t come up with a new Cupcake-type-thing in a year or so they just might call it quits.

So, what could Makerbot work on next? I’d like to hear your suggestions, but in the meantime here are some of mine:

• I suppose the most obvious product is a plastic recycler. It’s a more popular topic in the RepRap forums, but pretty much everyone everywhere is already interested in the idea. Something that could take bulk plastic (even if it’s just failed ABS prints) and turn it back into filament for the printer, would be quite popular. Additionally, a lot of work on this topic has already been done in the open source community, so Makerbot wouldn’t have to start from scratch. That seems in keeping with what made the Cupcake/TOM successful (it was pretty much entirely based on the RepRap). Finally, they might even be able to run the recycler off of the same electronics kit already sitting in the bowels of the TOM. The Gen 4 electronics have extra actuator/sensor connections that aren’t being used. A firmware upgrade, a new version of ReplicatorG, a second power supply, a few wires and you’re up and running.
• They could introduce new models of 3d printers that don’t use plastruders. There are literally several dozen different methods of printing in 3d and not all of them require a million dollars. The most obvious method is ACTUAL 3d printing, where a powder medium is smoothed flat and an inkjet head prints a picture to bind it together.
• A complimentary subtractive manufacturing process might be good. If they could build a CNC machine that automatically took parts the TOM had finished, and cleaned them up, drilled holes, maybe even produced a smooth surface finish, it would probably be bought up by the same people who already have a 3d printer. So, not just Makerbot customers, but also RepRap owners and UP printer owners and whatnot.
• Perhaps they could expand into technologies that would compliment the desktop manufacturing paradigm. For example, I have a wet/dry vac that is totally plastic (printable) except for the motor, switch, and wiring harness. In fact, the limitation of “but what can a plastic part do” is the first thing people think of when I show them 3d printing. ABS is structural. To make it generally useful it needs to  hold electrical and mechanical systems in position. Since their market is (and should be) the DIY early adopter hacker type, they could create a sort of Legos system for DIY projects. For example a pick-n-place bot that could build circuit boards by putting components in position and soldering them, and then put them into a 3d printer to have the housing built around them, would be quite attractive to that community.
• One of the biggest limitations is that everything is in plastic. Metal is much more useful for many practical applications, but machining it is a pain. Building up the shape of the metal part you want in easy to handle plastic, and then casting it in metal, would make the machining of metal nearly unnecessary. Perhaps Makerbot could build a sandbox casting bot that takes the part the TOM spits out, dumps sand around it, melts some metal, and injects it into the mold … turning the plastic part into a metal part automatically.