Issue link: https://iconnect007.uberflip.com/i/1140547
18 DESIGN007 MAGAZINE I JULY 2019 with this weak interface microvia reliability issue, and this is a nagging industry problem with multiple causes. We don't think there will be a single silver bullet. Most companies have thought that they alone had a microvia prob- lem with three, four, or five stacked layers. We're in the process of trying to share informa- tion at IPC to see if we can avoid big test pro- grams with hundreds of boards and multiple design iterations. Shaughnessy: What are your plans at this con- ference? Fritz: Here, I'm mostly involved with stacked microvias. We had a good meeting Monday afternoon with about 50 attendees of commit- tee workers trying to sort this into sub-teams. We have seven sub-teams that address mate- rials, laminates, processes, test methods, etc. One big item that needs help is gathering data and trying to organize that anonymously un- der IPC to get the answers as quickly as we can. Shaughnessy: Is this whole thing related to the military board via failures? Fritz: It can be, but it does not have to be. Military is a factor, but we have aerospace, medical, and high-complexity telecommuni- cations people that are involved with weak microvias too. In the U.S., we utilize specific constructions with relatively thick layers that those industries use. Stacking saves space for the thousands of IOs from advanced semicon - ductor devices. The military has been pretty conservative in design. They tend to use more standardized ICs, but microvia stacking is on the horizon. The people who are driving the hardest are the companies who would love to use AI, virtual reality, and IoT. Defense can see that they want the same features in their future weapon sys- tems, needing to use high-density packaging, which includes stacked microvias. Shaughnessy: But it's not happening with the staggered vias. Fritz: If you don't make staggered microvias correctly, conductor breakage can occur. I think everybody has seen some single-layer micro- vias that weren't made well and failed. The thought here is that it's multiple levels of mi- crovias, and the stacked microvias are show- ing the most problems, but it's not exclusive to stacked—predominantly, but not exclusively. We anticipate that IPC will get involved in the specification for test methods to weed this out. I think testing will be a very active topic within IPC for the next year or two until these test methods are capable. The industry sum- mary right now is that companies have mitiga- tions in place that show that they are not ship- ping bad product; however, they don't know why they have to throw some suspicious pan- els away, and the root causes are unknown. And it's not economical to reject panels due to the mitigations that are in place. Shaughnessy: It sounds like you're involved with a lot of things happening right now. Fritz: Certainly enough to keep me busy. Shaughnessy: I know you've tried to retire, but it's like Al Pacino, right? You keep trying to get out… Fritz: But they keep pulling me back. Exactly. Shaughnessy: What would you do if you weren't doing this? Fritz: I live on old family land in Indiana that has been in my mother's family for 150 years and never been sold. I'm not exactly a hobby farmer, but I do rent some land out, raise a big garden, cut firewood, and other things like that. I have community activities, and I have worked half-time now for almost 20 years. I would just flip more into retirement. Shaughnessy: Sounds like a great work-life bal- ance. Thanks for talking to me, Denny. Fritz: Thank you for covering this conference, Andy. DESIGN007