The Switch Doc rebuilds his cat toy launcher with 3D printing

Prototyping hardware has always been a difficult business. Hand-wiring boards, breadboarding, and cutting foam and wood to mock up cases was a time-consuming process. In the past couple of years, everything has changed. Printed circuit boards (PCBs) are easy and inexpensive to fabricate. Either make them at home or through one of dozens of fast and accurate prototype board manufacturers.

It no longer makes much sense to hand-wire boards, although during circuit development, it still makes sense to use a breadboard. I expect that will change in the near future. In one of my other lives, I wrote software that would simulate large digital and analog VLSI (very large scale integrated) circuits, which would often allow the design engineer to get it right the first time. The US$  200K price tag for a set of VLSI masks is a huge incentive to get it right the first time. In my IC design career, I burned huge amounts of computer time, simulating these circuits to work out the bugs before building them. I look forward to being able to do that with my own board designs in the future without purchasing a very expensive CAD system.

With the advent of relatively inexpensive 3D printers, makers can now apply a whole new toolkit to building one-off designs and prototypes.

What is 3D Printing?

3D Printing usually refers to a method for building a three-dimensional object from a software model. The consumer products generally use PLA or ABS plastic and build the model one layer at a time – usually 1mm or less per layer – from the bottom up. Dozens of 3D printers are available today and many are priced under US$  3,000.

I borrowed a MakerBot Replicator 2 to start building prototypes in a more modern way. Compare the foam and wood build prototype of MouseAir V1 of my previous article [1] to the 3D-printed MouseAir V2 in Figure 1. I will never return to foam and wood to build projects again. By the time Tom Daugherty  [2] (a great source of what is going on in 3D Printing) needed his 3D printer back four months later, I was hooked. I had to have one for myself. I also got to print with his prototype potato-based filament, which was a very interesting experience. Especially being from Idaho.

Figure 1: MakerBot replicator, 5th generation.

Selecting a 3D Printer

I spent a lot of time looking at the market as I selected the 3D Printer. I looked at printers from US$  1,000 to $4,000. I looked at inexpensive kits and fully complete and tested printers. I decided not to buy a kit because I wanted to use the 3D printer as a tool and not as the project itself. I also know from long experience that, when you want something very accurate and reliable, it is hard to build it from a kit and get all the tolerances right. Although that would make a good SwitchDoc Labs article, the point was to get a tool that would work out of the box and get me building quickly.

The software that comes with the printer (I call this the "tool chain" – the data path from the 3D design software to the 3D printer itself) – needs to work well and fit my criteria to build prototypes with engineering accuracy. The tool chain is critical to making all this work. After looking at a number of different printers and, most importantly, their reviews on the Internet, I came to the conclusion that I would buy the new MakerBot Replicator 5th Generation [3]. My reasons for choosing this printer were:

• Manufacturer is proven in the market (and by me) – I liked the MakerBot Replicator 2; it was pretty reliable. I did have problems with the extruder and had to replace it, which is not uncommon in all 3D printers. Extruder problems abound. MakerBot customer service made it a very good experience: They were very helpful.
• Reliable Tool Chain – I found I could use a number of 3D CAD software packages with the MakerBot. Every package I tried had a path given to the MakerBot that didn't take me down the black hole of Mac and Linux software compatibility.
• The features of the new MakerBot Replicator 5th Generation appealed to me. The Webcam seemed cool; they had a new smart extruder design, and the self-leveling Build Plate looked good.

I bought one for US$  2,800 (Figure 1) after trying to get a discount from the marketing department. They were not very forthcoming with any kind of a deal beyond a $50 educational discount (I am an adjunct professor at a local university), which was disappointing.

Overall, after over 70 hours of printing, I am pleased with my purchase. In addition to providing reliable printing, a good tool chain, and solid technical support, MakerBot offers many help and design sources on the web and great ideas for projects through the Thingiverse site [4].

Of course, there were also some downsides. Although the technical support is good, you do have to pay for it. I paid US$  350 for a year of MakerBot Care, which was worth it for me, but it might not be for you. Also, the software for support of the Webcam is next to worthless. The iPhone app can only be used on a local network (they promise support for the Internet at some point). I was expecting at least a time-lapse program, as well as remote viewing and control, but you can't do any of that. You always have to go to the printer to hit the Go button. Hopefully they will fix these issues in the future.

Lastly, using any filament other than MakerBot filament voids the warranty. Hmm … sounds like razor and razor blades to me, except they are charging US$  2800 for the razor.