I designed the two-piece enclosure to have mounting tabs on the base, so it can easily be attached to any surface with a couple of screws. The base supports the Arduino board and includes space so that you can solder wires on without wrecking the fit. The enclosure is open at the top, allowing you to mount connectors to the prototyping board. This part of the enclosure also provides a feature I wouldn't know how to fabricate any way other than with 3-D printing: an inner support for the cantilevered prototyping board.

Following the carpenter's motto—measure twice, cut once—I went over the design carefully to check the interior dimensions against the physical boards before using Alibre's software to generate STL (stereolithography) files, the standard format used to specify 3-D objects for printing. A little probing on the Web revealed the great variety of processes that can be used for this, including fused-deposition modeling, selective laser sintering, and stereolithography. More online investigation with some of the vendors also revealed the great variety in cost. The quotes I received for fabricating my little enclosure ranged from $30 to $150.

Figuring that my first attempt would surely have flaws, I simply chose the fabricator with the best pricing, a company called FastProtos.com, located in Edmond, Okla., which uses a process called PolyJet printing. If the end product was of poor quality, I reasoned, at least I could test it out a bit before my next iteration with some other vendor or process.

The online transaction with FastProtos went as smoothly as buying a book from Amazon, and the enclosure came about as quickly—one week. I was wowed. The curved surfaces showed no awkward faceting, the base and top fit together snugly, and both were very clean, which I understand can't be taken for granted in 3-D printing. Even more surprising was that I hadn't flubbed the design: The stacked printed-circuit boards slipped right in with no discernible slop. So I ended up with a very slick enclosure right off the bat.

Was it worth $30? Ironically, it's the professional result itself that gives me pause—I'm tempted to compare it with the zillions of plastic doodads I buy off the shelf, in which case $30 seems rather steep for so small an item. But were I to be shown something of the same size and shape machined out of a chunk of nylon or Delrin, I would immediately declare $30 to be a great bargain. In any event, there's no question that fabricating highly engineered widgets is nowadays easy and cheap enough to be within anyone's grasp.