Issue link: https://iconnect007.uberflip.com/i/1307491
30 DESIGN007 MAGAZINE I NOVEMBER 2020 I also do seminars for customers, where we talk about how PCBs are manufactured because, believe it or not, a lot of layout peo- ple and EEs have no clue how we build circuit boards—none whatsoever. My seminars are not nearly as good as Happy's, but I do one on how to properly stack microvias, and I've included a lot of what I've learned from you in that presentation as well. Shaughnessy: Do designers ever ask to take a tour of your shop? Hofer: Not unless they become a customer or we're talking about a project. With COVID-19, not a lot of people want to visit these places. Shaughnessy: Related to vias, there have been failures with military boards. They found out later that they were opening during the reflow process, but then they would close back up, and it would pass. After the fact, they found out it was shorting during reflow, and then it would heal itself at room temperature. IPC has a task group looking for the cause, and Happy is a member. They're saying to stagger the vias until they can figure out what's going on. Holden: The more you stack, the bigger the stress on the landing pad. The electroless cop- per is cracking during the reflow but resealing so that you can't detect it any other way. You have to go up to 240–260‐C to get the stress to cause the open. Once the mechanical contact heals, it can take months or years for every- thing to relax so that you start getting an inter- mittent open. Unfortunately, these are military missiles and things like that. Some of them are warheads, and the military doesn't like intermittent opens on warheads when you're underneath the polar icecap. They get really angry when they start seeing red indicators showing that something has opened up. Some people probably have already solved the problem, but they don't want to share it with the industry, so we're still back to look- ing for money and test vehicles where we can get data that can be shared with everybody. Hopefully, we'll have it next year. We have the data from the European Space Agency where they've gone two buildup layers, and their second phase is three buildup layers. They've built test vehicles for our group that are stacked and staggered that are four high. We have two, three, and four high. We're using electroless copper and direct metallization, and a couple of other variables, but a standard material that's used a lot in mil- itary aerospace. The European Space Agency is using materials not commonly used in North America. Although we can report their results, the boards are fabricated in Europe with mate- rials that we typically don't use, so we don't know how to interpret their data so much into our situation. Shaughnessy: Do you see any new technology that looks promising as far as doing vias, any- thing that piqued your interest? Hofer: No. I recently returned from a trip to Asia and watched them using the direct met- allization that we had tried here years ago, and kind of phased out. I found it interesting that the European Space Agency will be using direct metallization in their testing. Holden: We have two PCB shops with electro- less copper and direct metallization with the same process at each shop. As they build the test vehicles, they'll build them both ways. Hofer: That will be very interesting. But currently, there's no new technology to tackle this ongoing challenge. You're going to have an uphill battle tearing us away from a standard electroless and electrolytic copper plate methodology. Shaughnessy: Tell us about Accuwrap. Hofer: Accuwrap is a way to meet the IPC Class 3 wrap plating specification without adding a significant amount of copper plating to the surface. It is a proprietary process that allows you to meet both IPC-6013 and IPC-6018 wrap requirements without adding too much copper to the surface of the PCB. IPC-6018 is 0.002" on all classes, and IPC-6013 is half a mil for