Issue link: https://iconnect007.uberflip.com/i/1284035
58 SMT007 MAGAZINE I SEPTEMBER 2020 these products, changing? Are they putting more or different demands upon box build qualification than in the past? Malmrose: In the olden days, a lot of products had to be NEBS compliant. There are also UL standards. We had a stretch where we had to be more competitive with China, so we said, "Let's not use the NEBS-compliant box. Let's do some- thing else." Cheapening it or dumbing down the performance of the enclosure is mostly cost- driven. I haven't heard, "I want a NEBS-compli- ant box," from a customer for a long time. Feinberg: You mentioned testing reliability by dropping something three meters. Having to do a drop shock is one of the things that I hear a lot about. A lot of that has come with lead- free solder being more brittle. Malmrose: Yes, we still do a lot of leaded solder. We do more lead-free solder, and we do a ton of it. Not only do you have the shock aspect, but the CTE difference is markedly different between lead-free and leaded. If you're try- ing to pass REACH and RoHS, it becomes more important if you're looking for higher reliabil- ity for your product. Johnson: Especially if you're concentrating on mobile, in-the-field, portable-type of func- tionality, the enclosure is there to protect the electronics inside from drop shock. That must change some of the design concerns going for- ward because mobile and in-the-pocket type usage is becoming the standard for most of our consumer electronics. Malmrose: And in the enclosure space, for example, you have compression set issues. If you're a mechanical guy and you're going cheap, you may grab some kind of foam prod- uct that's going to be your gasketing. You'll say, "I'm going to do my drop test today." If it passes, you still may have a compression set problem in the future with certain types of polyurethane or other materials used on gas- keting. You may find that what was good in your product today, two years from today, your plastic or your rubber gaskets are more brit- tle, you have compression set issues, or you have adhesive fatigue issues. Suddenly, you're going to have field failures. Johnson: Lilia, are there some common issues that you could point to for the interface to the board itself? Castro: We always use the IPC standard for sol- derability and the J-standard all the time. Based on that, what is the reliability of the board? If this is medical, we're using Class 3 for the standard build and the cable. We always use Class 3 for medical devices and aerospace. It depends on the customers' requirements. Johnson: As long as they're adhering to the appropriate IPC standards, they're going to be in good shape. Castro: Those are standards for our contract manufacturing; we always use the IPC stan- dard. Sometimes, a customer says, "I don't want to use IPC. I want you to do this per- fectly." It also depends on the requirements of our customers. As a contract manufacturer, we need to follow what the customer requires. The book is always in front of people. We give it to them to show we're following that kind of stuff. If you don't follow it, all your compo- nents end in failures. Johnson: Let's pivot and talk about reliability during the assembly process and for manufac- turing when assembling the PCBA. What are some of the current trends or issues that you face in the designs you get from customers, espe- cially about when you have to work with them to make fixes before you can be really reliable? Castro: One issue is when we build assem- blies. We run the first article and find out there is a design issue on the boards. For example, when a location designator is indicated on the drawing and the BOM, but the actual design of the board does not include that location, there may be a design problem on a differ- ent drawing where there's an issue with the