2017-03-26 Project Update
After many long nights with the CNC, including a few marathon sessions that yielded just eight half-completed decks, I’ve decided that, reluctantly and with a lot of disappointment, I’m going to have to throw in the towel and issue refunds. Here’s why:A poor design
Knowing what I know now, I probably would not have attempted to commercialize this project. But if I did, I certainly would have made some big design changes. First and foremost, I would not have selected 30mm Baltic birch plywood. As I’ve noted in previous updates, this material is very tough to cut and requires 3-4 passes on the CNC machine, which really increases cycle time. Had I been smarter, my design would have called for very thin materials, like 6mm Baltic birch, to form the deck uppers and lowers, with dowels serving as the risers and to add rigidity to the assembly. Thinner materials would be cut faster, in a single pass, and wouldn’t have required specialized long-cutter-length bits to perform the finishing pass, which would have saved a lot on cutter costs as well. I should note that I have been experimenting with a thinner design, which does cut faster, but requires two decks be cut, so ultimately, there isn’t a tremendous amount of overall time savings even in a thinner design. Speaking of cycle times…
I was concerned during prototyping about the time it would take to cut the holes, and I thought that by selecting stepped drill bits and optimizations in CAM, I could reduce that time. And I could have, slightly. But with this project calling for producing about 160 decks, I should have taken a step back and really crunched the numbers. My goal was to get the machine time per deck down to 15 minutes, which still would have entailed 40 hours of CNC time. But I wasn’t able to get anywhere close to that. Realistically, to fulfill all of the orders here, I would be looking at 200-300 hours of CNC shop time. Which leads me to…Access and shop time…
I’ve been very fortunate to have access to a nearly full-size CNC machine with vacuum table and dust collection, and even though I don’t have to pay for its use, my colleague does. Whenever the CNC is running, the rotary phase converter is on to power the vacuum system and big dust collector. After the reality of this project started to sink in and I started to think about what my friend’s next power bill might be from Seattle City Light, I cringed. It’s not fair for me to use his shop and machines for the length of time this project would have required without significantly compensating him, which leads to my last epiphany…Costs
At the outset of this project, I didn’t have a full appreciation for the costs associated with producing this design en masse. My goal was to essentially break even here, but I didn’t do a good enough job of factoring in the cost of consumables. For example, router bits. After the Vortex 3184XP “extreme performance” compression spiral bit ($100 each) started fluttering on just the second panel, I knew this expense alone would sink the project. Based on some rough calculations and the router bit life I saw after experimenting with a number of other bits, I believe that each deck could have cost me $10-15 to produce. That’s after the $25 per deck you already paid.
I know this update won’t satisfy anyone, and I know many of you were very excited to receive your insert. I’m sorry that I wasn’t able to follow through with the production here, but I did learn some valuable lessons on this journey. If you placed a pre-order through Paypal, you should receive a refund notification today.
If you’re interested in cutting the inserts on your own machine—you’ve been forewarned—I’d be happy to share the DXF files with you. Just send me a PM.
And if you happen to have any expertise in injection molding, let’s talk. :-)
Obligatory photos, since, well, it’s all I have to show for this project:
Thanks again for the support, and sorry it has to end this way.