Wear and tear begin to show
Taller but more frail
I went to the 2015 Inside 3D-Printing Conference in NYC a week and a half ago and have been forgetting to recap what I saw there until now. This is an annual, travelling showcase of 3D-printing technology that rotates through various cities around the world. Meckler Media, the organizer, apparently runs a lot of different trade shows throughout the year. It seems like these things are mostly for either investors looking to learn more about the field or industry upstarts/leaders looking to showcase their newest developments. Luckily, they provide free tickets for access to the keynotes and the main trade show floor, so it's a nice opportunity for students to go check things out. A dump of the images I took can be found here.
It was really interesting to see how things have changed just in the past year. 3D Systems, which dominated the conference last year, was a much smaller presence this year, perhaps reflecting their shift away from the home consumer space towards more industrial applications. There were also a lot more non-machine-centric 3D-Printing companies popping up, like filament providers, printing service providers, and custom software add-ons. I thought Mattercontrol Touch was a nifty, bootstrapped solution that could be a really useful, swappable interface for low-cost machines, if only it would be possible to install their software onto mini-tablets we buy ourselves instead of through them. Overall, it seemed like there was a more concerted shift towards non-expert consumers instead of hackers, and that was reflected in the conference participants as well. There were definitely a lot of people milling around who were clearly very new to the field, based on the questions they were asking the vendors. On one hand, it's definitely nice (and important for future development/funding) to be entering the mainstream, but in my opinion, 3D-printing's not yet at the point where it's a turnkey solution that "just works," so I'm a bit worried about the backlash from inexperienced adopters who may be expecting a lot more than what's available.
Some of the other things that stood out:
Chinese Knock-offs and the Race to the Bottom
There were a lot more Chinese companies that came to show off their printers this year. Last year, it was just XYZ Printing and their Da Vinci printers, but there were several more this year, many of which showcased printers that were suspiciously similar to pre-existing, open-source designs with minimal modifications. So this is pure speculation on my part, but it seems like many of these companies are just content to take advantage of their access to low-cost components and undercut the price of printers released in the States. It makes a lot of business sense, especially if they're using designs that have been released open-source, but it's a tad boring, and I think it often results in printers that perform worse than the designs they're based upon.
The Chinese printer companies weren't the only ones aiming to build cheaper printers. I finally got to see the Micro printer from M3D up close. It had previously gotten a great deal of press for being remarkably, or suspiciously, cheap. More seasoned hackers were quick to criticize the price, arguing that it couldn't possibly deliver any sort of reliable performance at such low cost. Some accused the company of being a scam, pushing out vaporware to take advantage of Kickstarter funding. Similar complaints have been raised recently for the Tiko printer. So, it turns out the Micro is definitely real, but maybe not much more. The company representatives were either non-technical or under orders to be hush-hush, but they still wouldn't say what motors the Micro actually ended up using, and I haven't yet found anyone who's done a teardown of the machine yet. Odds are good that they're still using stepper motors but are using the 28BYJ-48 steppers in favor of the more common Nema models. Their linear rails were tiny and injection-molded plastic, so I really wonder how long the printer will last before those parts wear down. Rampant pessimism aside, it's totally possible that the printer would do a decent job if you greatly limited its print speed. Honestly, for a first-time user who doesn't know any better, this may not be that bad of a tradeoff. I think that it (along w/ the Tiko) should certainly be considered in a completely different class of printers, and the company should be proactive in lowering or at least readying consumer expectations. There are certainly more useless and more expensive paperweights than the Micro, and I would imagine there are certainly some, albeit limited, range of prints that the Micro is more than capable of producing (ie. no tiny features, components largely composed of simple geometries)
That said, there were also several Chinese companies that have been making strides towards more novel, or at least more robust, designs. In this conference, there was a Chinese company called Doogell, which I can't find documented anywhere on the interwebs, that showed off a really clean-looking printer with professionally-made, nicely-machined parts. I've also been tracking companies like Atom and Rapide Lite, who have essentially been following the Printrbot model of improving upon DIY printers with more properly machined parts, resulting in printers that could be considered intermediaries between DIY Repraps and more industrial printers from Stratasys. This is a subset of printer designs that I'm particularly interested in: printers that for the most part are as dependable as a Stratasys uPrint or Fortus, as long as the user can commit to somewhat regular but repeatable maintenance practice, such as re-tightening bolts, re-checking alignment, etc.
A nice addition to the conference this year was an additional space where artists, modellers, and other users of 3D-printing technology showed off the "products" they could make with 3D printers. It was still predominantly just 3D renderings printed via a high resolution method like SLS, which was disappointing, but companies like SOLS were also there to show off the benefits of customization. Functional and non-paperweight print applications are apparently still difficult to come by, at least outside of very specific use cases. There's still that big translation gap between the user's functional need and a finished, customized product. Many of the keynotes seemed to reiterate this same problem. Many groups seem to think that simplified design software may be the best approach, but I sometimes feel like this global, top-down approach, while arguably better for scalability and adaptation, will never be all that useful without a more thorough understanding of the very specific use cases, whether they're just simple replacement brackets or more complicated, articulated structures.
One guy that really stood out was the designer behind InsaniToy, which was developing a series of 3D-printed kits for articulated toys, similar to the old-school action figures from the 80s that had rubber-band joints. I'm certainly biased due to my research lab, but his use of rubber bands and tubes to make poseable and moveable joints seemed like a cool way of compensating for the performance limitations of printed joints at that scale. However, it seemed like he was more using 3D-printing as a means of doing low-volume production runs instead of taking advantage of anything customizable. His designs were certainly really novel and unlike anything I had seen prior, but the end product may as well have been just any other injection-molded toy.
I also got to finally see the e-NABLE hands up close. They're non-profit, but their product would definitely be described as low-cost, functional, and customized. Their prosthetic use case is in my opinion one of the best examples of maximizing the utility of 3D printers, and they're actively dealing with many of the mechanical challenges that so-called paperweight designers don't generally consider. The hands I saw also had a lot in common with the 3D-printed, prototype hands we have in lab in terms of trying to find common, off-the-shelf components to augment the overall system. Testing for robustness and long-term use seems to be an area that they (and we) haven't been able to focus on, and that's a shame. With 3D-printed objects, it's almost as if planned obsolescence has been replaced by expected obsolescence, and I think it's critical to figure out exactly what the useful time scales are for projects like theirs and ours.
Overall, I'm getting the feeling that the 3D-printing hype's finally beginning to plateau, and now we're finally dealing with the practical issues and potential, functional uses of 3D-printing. In particular, I think it's beginning to be more difficult to argue in favor of the utility of hobby FDM printers, even as they're increasing in both reliability and quality. Without a general use-case for non-technical users, 3D-printers remain primarily a prototyping tool, albeit an invaluable one.