Issue link: https://iconnect007.uberflip.com/i/1210212
54 DESIGN007 MAGAZINE I FEBRUARY 2020 delivered it on time—one day before the dead- line ended. That's how I started working on EMC. Shaughnessy: Are your customers primarily in Europe? Löwer: They're mostly in Germany. Shaughnessy: Are they primarily automotive? Löwer: About one-third of my customers are, but half of the workload is automotive. Shaughnessy: We don't seem to have as many EMC problems now, do we? I don't hear as many horror stories. Löwer: The problems have changed in ap- pearance. Engineers have grown accustomed to handling EMC issues. It is normal to solve EMC issues in more complex projects. There- fore, engineers recognize EMC or EMI prob- lems earlier in the project. They try to handle these issues, but there are a lot of things you can do—some better, some worse. Most people only think in differential pairs and split pow- er planes, but there are more possibilities de- pending on the issues, starting with good lay- out practices and not ending in more elaborate power filtering. I see that many engineers have an understanding of EMC but lack experience in handling it. However, the issues start earlier. There are a lot of non-technical issues that need to be ad- dressed, as well. Look at the number of reg- ulations. In electrical mobility and electric cars, you have a lot of EMC norms and regu- lations to fulfill. Each region of the world has its own regulatory bodies with specific rules. For example, the U.S., Canadian, European— UN ECE—market and Chinese market all have slightly different EMC regulations in place that you need to comply with. There are other challenges, too. Chinese reg- ulations, for example, are similar to the Euro- pean ones, but when you only comply to the European norms, you will not be within the Chinese norms. It is not complicated to com- early design stages, where you can make good and bad decisions. Shaughnessy: You'd rather have the custom- er come to you early in the design before the EMC problems arise. Löwer: Yes. In that stage, it's possible to avoid problems. Unfortunately, most of the custom- ers come to me when decisions are already made, meaning after the first EMC test results come in, and their customer or product man- ager says, "No." I got into EMC when I was with a custom- er. It was one of the typical "we don't have enough engineers for the amount of work" jobs. I didn't have any clue about EMC at that point. The project had already run for more than a year. Their customer said, "We're can- celing this project if you don't fulfill our EMC requirements within the next three months." I had the unfortunate luck of being the only one who could make PowerPoint presentations. The chief engineer of my customers' custom- er visited and demanded to see how we were planning to improve the EMC performance of the product. All my colleagues in the compa- ny were more experienced engineers, but their presentations were strongly influenced by the local dialect. It was hard for non-native Ger- man speakers to understand them. Again, they were all great engineers, but they did not grow up with PowerPoint. I was the youngest, so I was the one who had to give the presentation, but I made a deal. It was clear that their customer would assume that who- ever did the presentation was responsible for the EMC part and would be blamed if it didn't work out in the end, so I agreed to give the pre- sentation under some concessions. I request- ed the local EMC chamber for the next cou- ple of months, an additional engineer for gen- eral support, and a measurement engineer to help me because I had no clue how to measure stuff. My customer agreed. Over the next four months, we had our own EMC chamber and started experimenting a lot. We learned by doing. Looking back, some ideas were stupid, but others were great. We