Issue link: https://iconnect007.uberflip.com/i/1007258
JULY 2018 I FLEX007 MAGAZINE 13 The key point that I've learned is to have a really good relationship with a technical expert from that supplier who can basically be on call for us at that early stage. We've been lucky to find suppliers who had those resources for us, some sharp individuals. Also, I'm leveraging my engagement and involvement in IPC. There are lots of experts there and on the committees. It's really a good place to learn. It's reassuring to hear that other people are having similar challenges. Recently we've been attending flex workshops, profes- sional development courses, and I'm trying to get some of our designers now to start attend- ing these conferences and seminars, because there's a lot of good learning there. Once designers can start to think in a three-dimen- sional space and understand some of the limi- tations of these materials, and the capabilities, it opens a different world. Shaughnessy: What portion of your products have flex versus rigid? Do you have any idea what the breakdown would be? MacFadden: I find this to be a really unbeliev- able change in just the last four years. Four years ago, I would have told you that 90% of our overall spend was rigid, and 10% was flex. Now that has completely changed to the point where rigid-flex and flex each constitute roughly a third of our overall PCB design work and spend. That's happened all within the span of three or four years. We had to come up to speed with the design rules and then an entirely new supply base. It happened quickly. It's been inter - esting to be at this transition point in the company. Shaughnessy: Wow. Definitely a sea change. Now, tell me about the flex for wearables that you mentioned. What kind of chal- lenges do you see there? MacFadden: Our latest prod- uct goes around your neck. It's sort of a collar speaker that provides a nice sound curtain for the individual. All of these product concepts are based on trying to reduce mass so it's com- fortable enough for the wearer, and to put as much functionality as possible into the small- est package possible. The only way to accom- modate that has been with flex and rigid-flex to be able to minimize the number of connec- tor headers to be able to fold things up. It's also really coincided with the chip sets that are available off the shelf, and some of them are very small. They've been custom- ized for the cellphone market, where minia- turization is everything, and the rest of the consumer electronics market is both the ben- eficiary and also the victim of it. There can be a lot of functionality in a really tight BGA, but it yields a complicated board design. It means higher layer count, more HDI microvias to make these interconnections. At the same time that the form factor requires miniaturization, the functionality is requiring more complicated ICs and off-the-shelf-parts. In some ways, the off-the-shelf packages also have become more efficient, and we need fewer sub-components to accommodate these designs, so it's a trade-off. It's all about try- ing to make a wearable that's comfortable and lightweight, that is going to stand up to all of the abuses that you can imagine: drop tests, moisture exposure, leaving it on a dashboard in Phoenix, or leaving it in your car over- night in Minnesota. Sweat is a huge one, and it's obviously a very challenging environment to accommodate. Fortu- nately, flex materials are chemically robust and thermally robust. Shaughnessy: It's interesting that you mentioned that sometimes you find flex is cheaper than rigid boards and having to do a whole song and dance to make a rigid board fit in. MacFadden: Flex materials are vastly more expensive on a per- unit basis compared to rigid materials. But there are cer- Figure 1: The Bose ProFlight aviation headset features a dynamic slider flexible circuit inside the headband.