Issue link: https://iconnect007.uberflip.com/i/1315894
10 DESIGN007 MAGAZINE I DECEMBER 2020 This is hardly a new topic. We've been talk- ing about eliminating PCB design respins for decades. When I first started covering this industry over 20 years ago, DFM was all the rage. Cut- ting respins was just one of the many benefits that were slated to accrue when PCB designers adopted solid DFM practices. Designers were no longer going to throw designs "over the wall." Designers would talk to CAM departments throughout each design, design data would arrive at CAM optimized for that fabricator's processes, and songbirds would sing in perfect harmony. Lee Ritchey and John Zazio even penned a textbook, Right the First Time: A Practi- cal Handbook on High-Speed PCB and System Design, full of tips and techniques for eliminat- ing respins. We could almost see the finish line. It sounded too good to be true…and it was. After 20 years of constantly talking about DFM, the disconnect between designer and fab- ricator is wider than ever. Some OEMs routinely go through two, three, or more PCB designs. Just try to find a CAM engineer who doesn't have a dozen recent horror stories regarding bad designs and design data. Lee and John could probably edit them into a much larger book called Wrong the First, Second, and Third Times. A September 2018 article in Lifecycle Insights put the average number of PCB design respins at 2.9. If this figure is accurate, I imagine that a few companies have eliminated respins com- pletely, but some OEMs are experiencing five or six spins per design. But we really can't blame designers for a situation that they didn't create. Many times, the designer's company has budgeted for two or three respins. The design gets out the door faster, and the designer can go to work on the next project. So, why focus on reducing respins by one? It all started during a meeting to plan an issue on new DFM techniques. "What if every OEM in the world reduced their respins by just one? What would that save, not just in wasted reve- nue, but in other resources like hours or time to market?" After a long pause, we did some back-of- the-napkin calculations and realized that we'd need a bigger napkin. It's impossible to gauge the savings from cutting one respin worldwide, but it would have to be a huge The Shaughnessy Report by Andy Shaughnessy, I-CONNECT007 What If We Reduced Respins by One Globally?