Issue link: https://iconnect007.uberflip.com/i/1007258
28 FLEX007 MAGAZINE I JULY 2018 these boards and the reliability of our products has gone up as well. Shaughnessy: What do you think are the biggest challenges for anyone designing flex circuits? Schulte: Well, for me it was the initial invest- ment in time spent reading up on flex design. I started contacting flex manufacturers and asked them if they had any design guides. Almost every one of them sent me a design guide say- ing, "If you design to these specifications, that's really going to help us," and they gave us tips on making them reliable and more flexible. Once I found a shop that I liked talking to and that was responsive, then I could get a little more specific with my questions as the designs got more refined. But the biggest chal- lenge was just getting up to speed with the ter- minology needed to use for designing flex. It's not a whole lot different than doing a simple rigid board. You just have to keep a different rule set in mind when you're doing flex. Shaughnessy: What do you think about the EDA tools' ability to do flex? Do you think EDA has caught up with flex? Schulte: Well, with Cadstar, it's not ideal. It's definitely geared more for rigid designs, and it does a great job with rigid. However, there's really no tools that I'm aware of in our pack- age that are geared for doing flex boards. We recently made a switch to Altium, I'm looking forward to learning the flex specific tools that are associated with their package. We use SOLIDWORKS for all our mechanical design, Altium will also plug right into that. You can design the flex in that environment and push it to the mechanical engineer, who can then put it into his assembly and make sure that everything fits prior to ordering product. On the subject of challenges, that's some- thing that's actually been surprisingly difficult. It seemed like it would be pretty simple to fig- ure out how long a flex board should be, but when you design things in a 2D space that are going to live and move in a 3D world, some- times it's tough to figure out exactly where those bends are going to happen, and what forces that might put on the connectors once it's plugged in. That's been the only modifica- tion that would have had to happen to any of the boards, either adding a little or take a little length out of them. Other than that, the fab shop does a good job of answering questions and making sure that I can get them a design that's buildable and reliable. Shaughnessy: Do you do paper dolls every time? Schulte: Well, not every time. In fact, in one instance that I am thinking of, all the boards, even the rigid boards, were all in the process of being manufactured for the first time, so we didn't have anything physical that we could put our hands on to see where these were going to line up. It was all conceptual models and our best guess was to draw curves into the models and say, "It looks like it'll probably go right about there." We didn't have anything that we could put a paper doll up against to see if it was going to fit. But once we get stuff in hand, that's actu- ally how I would nail down the length. Just draw up a paper doll, cut it out, and compare it to what we've got, and we can adjust accord- ingly if needed. Shaughnessy: What do you think about the IPC flex standards? Schulte: I read through them to see if there's anything pertinent to what I'm doing. The main use I get out of them is to make sure that I can call out all the UL ratings that I need. The flammability standards are a little different with bend materials than it is for rigid materi- als. I had to pull that out of IPC-2223, I believe. A lot of the design guides that I got from the fab shops were regurgitating a lot of that same IPC information, but boiled down and condensed into a quickly-readable package. I like to look up the standards whenever I can. If I've got a question on how big a hole should be or if I can do something in particular, my first go-to is to try to find the IPC standard that