My 2003 Mazda Tribute doesn’t look very cool; it’s classified as a “cute ute.” This car was designed to spend a lot of time in an elementary school carpool lane, but it is about as reliable as any car can be. It just goes and goes, and it doesn’t break down very often—especially for a car with over 200,000 miles on it (it helps that I telecommute and don’t spend two hours each day driving back and forth to work).
Everything still looks new, inside and out. It’s fun to drive, and it can haul four guitars and a pair of PA speakers with room to spare. It’s been paid off for so long that I’ve been able to put more money away for my rapidly approaching golden years. This was no accident; as the saying goes, “Reliability isn’t just an added feature.” Every person involved in designing and building my car contributed to the car’s long-term reliability.
Reliability has been in the news quite a bit lately, and as always, it’s when something has proven to be unreliable. In 2018, IPC formed the IPC V-TSL-MVIA Weak Interface Microvia Failures Technology Solutions Committee to try to find the root causes of microvia interface failures that have been affecting the defense and aerospace segments. Members of that subcommittee spoke at the IPC High-Reliability Forum and Microvia Summit in Baltimore, Maryland, and I was fortunate enough to cover that conference.
Now in its third year, this conference has continued to grow with over 100 attendees. No one at the conference expressly blamed the designer for these failures, in case you were wondering. But there are steps that designers can take early in the design cycle to head off trouble later. Motorola’s Jerry Magera and J.R. Strickland have been looking into these via failures for years. In their presentation, they offered a few tips for designers seeking to create the most reliable board; namely, avoid using stacking microvias. Try not to use stacked and staggered microvias in the same board. Keep the vias as large as possible, and the aspect ratios as small as possible. Try to avoid mixing processes on the same board. Variation is the enemy.
This is a problem that will likely require competitors to share data and work together for the greater good. As Magera said, “OEMs can’t let their fabricators go out of business.” Incidentally, the IPC V-TSL-MVIA subcommittee needs more members who can share their company’s data on microvia failures and put in the hard work needed to solve this problem. Talk about working as a team. I believe our industry will come together and work with each other on this problem, even if means a few uncomfortable moments of working with rivals.
To read this entire column, which appeared in the July 2019 issue of Design007 Magazine, click here.