3D printing s the current darling of the engineering world, and with good reason. Now everyone can have a machine shop in their lab or garage, allowing you to turn out prototypes or very limited production runs in a wide variety of plastics. [UPDATE: Staples, the office supply chain, has announced it will offer 3D printing as a service. Much like you can order a short-run printing of your report or pamphlet on on of its full-color, high-speed paper printers, you will be able to send you CAD files to a 3D printer. Staples is using Mcor Technologies printers, which use layers of paper, a cutting blade, and a lamination process to create a wood-like cellulose-based object.] (Some printers can also create sintered-metal objects.)
Optics for LED lights are a possible applications for 3D printing. (Here’s a recent article that covered the possibility of embedding light pipes into objects.) True, you can go to a optics design company like LEDil or Fraen which have a catalog of readily available off-the shelf lenses. However, if you need a custom lens for a unique LED package or lensing capability, you might have to go with a custom lens requiring a custom mold.
A big advantage of 3D printers is that they can create an object directly from the CAD design files, circumventing the need for any molds or tooling. – there’s no need to go through the investment of designing molds. Molds, in addition to their initial capital investment, entail risk – what if there’s a problem with the mold, or, what if the the lens doesn’t work? You’re out the tooling cost and have to start again. 3D printers eliminate tooling costs.
But there are several drawback to 3D printing: For one, it doesn’t scale in either costs or speed. It takes as much money and time per object to print one as to print a million. With mold-based manufacturing, you amortize the mold costs over the life of the product, and molds allows mass production of shapes: None of this one-at-a-time stuff.
Another problem: In general, the tighter the tolerances, both dimensionally, and in surface smoothness, the more expensive the printer. While hobbyist printers can be had for several hundred dollars, professional-level printers start at $10k and go straight up with correspondingly expensive materials. And optics require high-precision lenses with tight tolerance: Light is not a forgiving medium – think how even a slight ripple is noticeable in a glass. Yet, because 3D printers lay down the material a bead at a time, surface distortion is almost a given.
And another concern: Making a product that has an overhang is a challenge with a 3D printer because it relies on always having a layer to build on. What some of the more expensive printers do is to put down layers of a substance that acts as a support structure and can later be dissolved by a solvent.
But in spite of these drawbacks, 3D printing seems likely to find a home in lighting design and development. Perhaps the most intriguing use of 3D printing for LEDs is for embedding LED chips — the bare emitter with no primary lens — and encapsulating it with the printed plastic into whatever lens shape you need. That’s in effect what the Christmas-star-encapsulated RGB LEDs and control logic covered in yesterday’s post. With 3D printing you could make the decorative shape and the encapsulated electronics all parts of one package that was both decorative and lensing. Would it save money in volume manufacturing? Probably not. But it would enable creative iterating and rapid prototyping.
(MAKE magazine just published a $9.99 Ultimate Guide to 3D Printing. It’s worth buying for the manufacturer’s table alone.)
Here’s a video from LUXexCeL Group of what can be done with a top-of-the-line printer, in this case from Roland, and its Printoptical Technology. (Compare the Roland to the printer in the photo at the top of this page; There’s about an order-of-magnitude difference in cost.) These optics are beautiful and show the direction that 3D printing can take with optics. [Via @SuleymanTurgut1 - thanks!]