Issue link: https://iconnect007.uberflip.com/i/836472
16 The PCB Design Magazine • June 2017 A DEEP LOOK INTO EMBEDDED TECHNOLOGY of these ultra-smooth coppers are not made here. We have to go to Asia to get some of these super-fine things, and we're competing. The problem we see in the marketplace with getting copper foils is that the lead times are starting to lengthen and what our suppliers tell us is that fine line etching is really using up thin material. But even more so, these energy storage batter- ies, not just for cars but for homes, for storing solar energy or whatever energy require a lot of copper. They're sucking up all the copper in the in- dustry and now we're seeing lead times out 6−8 weeks where before it was a couple weeks. The lead times are going up at the same time that our volumes are going up. I don't know where that's going to end, but every time Tesla builds a new megaplant somewhere, believe me, they're going to use a lot more copper. I don't have an answer to that question except that so far, we're dealing with it and we're keeping up with the demand. We'll see how that goes. Matties: What sort of increase in copper price do you expect we'll see this year, 30%? Brandler: I have no idea. What we've seen so far is an increase in delivery and the lead time is being pushed out. But you're asking the wrong guy, I'm the technical guy. I don't really get in- volved in economics but you asked the right question. I think it's inevitable that's going to happen. But the lead times being pushed out tells me they're struggling with capacity issues which means the price is going to start climb- ing. Matties: Thank you very much. Happy, do you have any thoughts or questions on this? Happy Holden: Well, it's an interesting top- ic that's been around a long time. One of the problems I always felt with the material suppli- er is that they always had the cart in front of the horse. In other words, they were always intro- ducing and pushing new and better materials but had failed to establish any kind of cost trad- eoff. The first step is always providing some cost trade off. In other words, everybody asks how much is this going to save us or what benefit is this? And since they never really produced any kind of model or software to allow people to do a cost trade off, those that experiment with it got burned severely when they found out how expensive it was versus conventional surface mount and even embedded surface mount. So the industry still has fundamental prob- lems that it can't answer the question of how do I benefit from this? There are some com- panies like in the aerospace industry, because of reliability and temperature and/or size or weight, they have to use that and they've been a user for 35 years. The mobile phone people are using it for the same reason and that is size and weight as well as performance advantage. But that's not the North American market too much or the rest of the world, everybody is still standing around waiting saying, "Show me the benefits." And I don't mean a whole bunch of bullets about how great it is, because we know that's only part of the picture. Show me the cost model. Prof. Peter Sandborn at the University of Maryland developed software for MCC and Savantage Inc. that does cost tradeoffs of em- bedding components, either discreet or pro- cess. And it does it iteratively as it redesigns the board to find the sweet spot, where if you do it this way, you have this many resistors or this many capacitors or this size, you can reduce the size so that the material cost drops and you'll make money. But that's not available, or no- body's commercialized that or made it available for free. We're all still stuck with that first ques- tion. Then comes the second question: What are the characteristics available of the material? Then comes, how do I design this, what are the design tools out there, and how do I use them? The fourth question is, how do I fabricate this? Is it a drop-in, or do you have to add special processes and things like that? I've put embedded ICs into active produc- tion and shipped tens of millions of the units. The fabrication of some of these things is a lot more difficult than many times what's adver- tised, especially if you're doing active ICs not just passive devices. Brandler: I definitely have a different point of view from Happy. Ten or 15 years ago, when I