Let the botfarming commence!

Let the botfarming commence!

In this article, Snowshoe, located in Madworks Coworking at University Research Park, discusses its innovative approach to 3D Printing.

Let the botfarming commence!


3D Printing is HOT

Everyone is talking about 3D printing — from food, to fashion, to organs, and, perhaps unfortunately, even weapons. You name a physical product, and we’ll find someone who has tried to 3D print it.

It’s a technology that’s been around for decades, primarily being used for rapid prototyping. It wasn’t until a series of patents expired, the open source community got rolling, and Makerbot released it’s iconic home and hobbyist printer kits, that a new group of users — artists, designers, and engineers — began creating innovative indy products using additive manufacturing.

Few companies, however, utilize 3D printing as a manufacturing process. For most mass-produced products, the goal is to manufacture thousands of the exact same object. When that is the case, 3D Printing is a pretty terrible manufacturing technology.

Injection molding offers massive economies of scale (the more objects you make, the cheaper each additional object gets). This is because each item takes just seconds to produce. While the upfront tooling costs can be tens of thousands of dollars (or more), the variable costs of injection molding are so low that, by the time you have produced just a few thousand objects, your total cost per item is likely lower than if you were printing each item and had no up-front costs whatsoever.

In contrast, because each product is printed one-at-a-time, and each print takes minutes (or even hours) to finish, there are few (if any) economies of scale in 3D printing. Your first object costs about the same as your 10th, which costs about the same as your 100th, 1000th, and 100,000th.

While the up-front costs can be very low — zero if you already own the printer(s) — most items being manufactured at-scale will quickly reach an inflection point where it is more cost effective to invest in tooling for injection molding than to continue 3D printing parts.

Making something different

Here at SnowShoe, we make magical pieces of plastic that hold secret digital identities. Each plastic stamp can be uniquely identified by a touchscreen device, yet it has no batteries, no power, no antenna, no moving parts. The stamps are 100% ABS, a thermoformable plastic best known as the stuff used to make these rather awesome construction toys. Our stamps can be used in everything from coffee shop loyalty apps to toys that interact with mobile video games.

The secret sauce in our technology is pretty simple: when you touch a stamp to the screen of a smartphone, the phone’s touch sensor thinks it is detecting 5 human fingertips. It isn’t. It is actually detecting a pattern of conductive ABS plastic that is embedded within the stamp.

By slightly modifying this pattern, we can slightly alter the configuration of the five touchpoints detected by the phone screen. Each stamp has a slightly different pattern of conductive ABS embedded within its footprint, and, thus, each stamp can be uniquely identified by our software when it is touched to the screen of a touch-sensitive device.

Did you catch that?

Each stamp has to be unique. As a result, traditional injection molding isn’t a viable manufacturing option. We have to 3D-print each and every stamp we send out, because each one has to be slightly different from the one that came before it, and there is no economical way to slightly alter an injection mold between each injection.

Economies of scale be damned!

With 3D printers, we can make each stamp’s digital fingerprint unique, and that is worth $.10-.15 additional cost per stamp at our current production volumes.

Scaling a 3D printed manufacturing line

When we launched the public beta of our developer-facing platform in early February, we had two Makerbot Replicator 2X 3D printers. These had mostly been used for prototyping to that point.

Thanks to a successful (and serendipitous) post on producthunt.co, demand for our free developer kits quickly outpaced our two printers’ manufacturing bandwidth.

We’re now up to a total of five 3D printers, and we’ve expanded staff to the point where they are all running at least 18 hours per day. We can now produce at least 500 stamps per week.

As demand for our developer kits continues to rise, we will be expanding our bot farm to make sure we can fulfil all dev kit orders within 24 hours. We’ve learned a lot about how to optimize our 3D printing process, and, for the time-being, we are confident that we can make all our stamps (and make money) with our little farm of robots.