Adam Hutz

ramping without runway

It’s easy to get carried away with the iterative possibilities of additive manufacturing. An increasing arsenal of materials developed for FDM and SLA 3D printers has not only served as an end unto itself, but has made for new workflows in RTV casting. But there’s no escaping the intractable reality that, for mass production of plastic parts, injection molding is still king.

But tooling for injection molding, especially for products that incorporate multi-part assemblies, can run into the tens (or even hundreds!) of thousands of dollars.

It’s therefore very important for a small startup to de-risk the move to tooling as much as possible prior to taking the plunge. In order to make sure our final parts would come out as close as possible to our expectations, we endeavored to take injection molding into our own hands.

Meet the bantam tools desktop milling machine

We started by learning CAM, learning the Bantam Tools Desktop Milling Machine (DMM), and practicing on softer materials, like wood and HDPE, before making some molds in aluminum.

Early molds were made in the usual way: by subtracting material until the negative one wants is revealed.

After milling test molds in wood and HDPE, we began to experiment with 3D-printing various portions of the injection mold body on our Form 2 SLA printer. Milling aluminum molds assumes a commitment to one particular design, but we were still iterating, and didn’t want to generate a whole new mold for each future version of each assembly part. Around the same time, as if responding to this concern, Form Labs put out a white paper describing the possibility of injection molding using resin inserts. And so, once we felt confident with our abilities in CAM, and understood the limits of the Bantam Tools Desktop Milling Machine, we endeavored to make a variable-use injection mold case.

What we came up with was a carefully-designed aluminum body with a cavity that could fit injection molding inserts 3D-printed on our local Form 2. We printed the inserts in normal, clear resin, and tested them with RTV urethane prior to use.

Our first sprue, runner, and slug, were all too narrow. We returned to the milling machine to increase their diameters, and then fortified ourselves to injection mold.

BREAKING THE MOLD(s)

Throughout this process we had been on the lookout for an injection molding machine we could access and use. By the good graces of the machinists next door, we were introduced to just the device we needed: a Morgan Press, here depicted awaiting orders.

The Morgan Press is an extraordinary machine: it has a relatively low barrier to entry, as far as injection molding machines go, and can output 20 tons of clamping pressure and inject plastics at 12,000 PSI. After some practice, we were off to the races with polypropylene pellets.

As you can see, we learned a lot about both mold design and operations of the machine. For many types of plastic, including PC/ABS, moisture absorption can be a big problem, and so it’s important to desiccate pellets before use. The difficulty of timing this process, along with some less-than-effective part geometries, made certain pieces come out less than ideal. The translucent print above, for example, is a four-part mold that includes two inserts that handle part undercuts, and ended up deteriorating quickly given the high temperatures and pressures involved.

All of these learnings became changes in our final CAD models, which we eventually sent out to a tooling manufacturer for mass production. Because we were able to injection mold all of our parts ourselves first, and then make necessary changes, we were able to send six interfacing parts to manufacturing and receive them all back with zero necessary changes, saving money on tooling and compressing time to market.