Issue link: https://iconnect007.uberflip.com/i/1359517
18 DESIGN007 MAGAZINE I APRIL 2021 started off attending conferences and techni- cal sessions. Happy's "Design for Manufac- turability" was probably one of my first back in 1990. You can learn about the engineering process, but design for manufacturability is very important, and it's not taught in universi- ties. ere are a few courses on signal integrity that are targeted at EEs but not so much PCB designers. I developed the "Advanced Design for SMT" course for the Australian Electron- ics Development Centre back in 1994 to teach high-speed design techniques. Holden: I guess we should qualify that a lot of what we're talking about in terms of problems apply to analog and digital designers, but if people are involved in RF design and doing it correctly then they have learned a lot of these things early on. It's just that the percentage who are really good RF designers is so small. Olney: RF and microwave design is a different world altogether; the circuits are very high frequency with fast rise time and short wave- lengths, so you need to be very careful. Every stub is an antenna. at's why coplanar wave- guides are typically used for RF, because they tend to stabilize the signal. For a microstrip signal, for instance, you've got coplanar wave- guides with ground pours on either side of the signal which tends to dampen the radiation. Shaughnessy: A lot of designers and engineers agree that simulation is a great idea, but many of them don't use it. Why don't more people use simulation? Olney: I'm sure it's mainly to conserve time. Managers always strive to reduce their design cycle. But simulation isn't that expensive or time consuming if you outsource it. For instance, I can analyze an average complex board for about US$5,000 and find the prob- lem. If I were a manager and I had a choice of a small investment in analysis or taking the risk of a re-spin, I know which I'd choose. It's a no- brainer as it verifies that the product will per- form reliably! Dan Feinberg: One of the things that you just raised and Happy just mentioned is that peo- ple oen look at price, but it's really cost, and they're different. Olney: Absolutely. e cost of a re-spin is a lot more than the price of the tools or outsourc- ing analysis. It's not just engineering time you are wasting, but time to market and lost opportunity which can be extremely costly. Shaughnessy: Lee Ritchey says a lot of these conflicts stem from a managerial viewpoint versus the engineering viewpoint. e man- ager thinks, "Great, the design was done quickly," even though there may be a costly re-spin down the road, because for now, the EE can get on to another project. Olney: at's right, you'll always have conflict. Engineering and management require dif- ferent skill sets. e manager is responsible for getting the project completed on sched- ule and to budget. And, he has to look good to his boss. If he schedules a re-spin, then it is expected. It's in the timeline. Whereas the engineer or PCB designer is more interested in the completion of their work and learning new techniques to improve their skills. Holden: Is PCB design growing in Australia? Olney: I guess I am a little out of touch with the Australian electronics industry as I mainly do business overseas now. We have small groups of designers, but there is not a large electron- ics industry except for defense contractors. It was always difficult to sell VeriBest PCB and Mentor (Siemens) tools in Australia because it's the home of Altium, of course. Everyone uses Altium and going against tradition is very difficult. I guess you could say that the Altium Designer's market is growing. ere isn't much