Issue link: https://iconnect007.uberflip.com/i/1108006
26 FLEX007 MAGAZINE I APRIL 2019 Matties: One thing that we commonly hear is that there's not enough collaboration from design to fabrication to assem- bly. This is a case where col- laboration certainly paid off. Flannery: Yes. Moody: To jump in for just a second, Rich asked me a cou- ple of minutes ago to look at the DFM—the report that our CAM group does once the design is submitted to us in a final state for us to review and prepare for manufactur - ing. At that point, we had done so much work together back and forth that there was only one item on that DFM that we submitted to GraftWorx, and it was a sol - der mask opening for some components. That was primarily driven by our assembly group after they reviewed it. We wanted to review it from an assembly standpoint because the next aspect after you get it fabricated involved somebody putting components on it. We asked that group to take a look at it, and the only thing that came back after all of the col - laboration was a recommendation to open up the solder mask around certain components to facilitate the assembly; these things don't often turn out this clean. Matties: And that's the point of collaboration because otherwise, you may have gone through a few different spins to get to the product qual- ity that you achieved in your first pass. Flannery: Definitely. My background is in MEMS design where design and process development must go hand in hand. Contrast this with more conventional semiconductor ASIC designers who can design in a very refined, static CMOS process. Collaboration is an absolutely essen- tial component for this kind of innovation. If you try to do something new or you're trying to push the envelope, and you're not getting your designers talking to your manufacturing and process development engineers, it's a rec- ipe for disaster. One of the things that we're really thank- ful for is that Lenthor was willing to schedule meetings; we would sit down and talk about issues. I write about some of them in the accompanying article; it's not a complete his- tory of everything we've done, but there were some specific elements, such as handling sen- sors. With every specific detail that we worked on, they were willing to sit down and walk us through that. What we ended up with was this gorgeous, high-yielding board that looks great. We could not have done that by just throw- ing it over the wall like a standard process; it wouldn't have happened. For example, one of the challenges with MEMS sensors or other kinds of sensors that you want to put into a wearable—and the wear- able space is chock full of sensors—is they're sensitive and exposed to the world. They're not encapsulated in a nice epoxy shell like a standard integrated circuit. Your assembly techniques have to be different. We sat down with Lenthor's assembly line staff and worked through all of the details on our microphone, accelerometer, optical sensors, etc. Matties: In your process, is there anything that you learned that you would do differently next time? Figure 4: Patch applicator.