Issue link: https://iconnect007.uberflip.com/i/1398328
AUGUST 2021 I SMT007 MAGAZINE 61 Johnson: Oen, if you go back to where the decision was made that created the situation, it was made in design; it happens in the CAD tool. ose are the issues you're diagnosing. Boguski: It's true. e unfortunate reality is, by the time it reaches us, it's months, if not years away from the original design, and usually, that design has gone through many iterations and it's reached production level, so you can't go back and retroactively update the design. We'll point out certain things, and maybe we will actually talk to the designer. On that rare occasion that we do, their response invari- ably is, "I'm sorry, but it's a done deal. We've moved on. We're two product lines down the road from this, and you and the assemblers and everybody else associated with this, you're just going to have to live with it." It's rarely a coor- dinated effort, which is really a shame because there are great DFT tools available to screen CAD data on the front end and point out many of these problems in their infancy, before they become catastrophes. But these tools are very expensive. A tier one or OEM might invest in them, but is a mom- and-pop shop, much less tier three, tier four, going to spend a hundred grand on a soware suite? I doubt it, so they end up band-aiding it, making the best of it, and trying to pass acces- sibility standards at some minimal level with little or no consideration to long-term reliabil- ity. Okay. It ships, all is good. Electrons are flowing where they're designed to flow; ones and zeros transmit. When it goes out the door and they can certify that they've got the test data to prove it, Industry 4.0, up the stream, laterally, verti- cally, everything that the buzz- word says. Even despite that, seven months from now, boom, it shuts down. Blackout. How do I know? Because that's when we get the call. Johnson: What are some of the busiest areas in your shop? Which testing is the most in demand? I'm asking that question trying to get a feel for where the OEMs are potentially struggling the most. Boguski: Number one, board-level failure anal- ysis. We have a thriving X-ray business for board-level failure analysis. On any given day, a normal workload, we might have two or three jobs in the queue. When I got up this morning, we had 10 with four or five more emails from various customers saying, "I've got incoming for you," so those 10 could expand to 15 by the end of the week. Our poor X-ray technician who's responsible for that part of our shop is probably going to have his weekend ruined for the third consecutive week. at's area num- ber one. Area number two, we also do indus- trial CT scanning as I described; the valve example is a good one. at's very busy. People send us other objects. We don't just do elec- tronics. We do all other objects, medical prod- ucts, mechanical products, castings, materials science/additive manufacturing applications, various things to ascertain what the failure mechanisms are. Weld inspections, joint inspections, mea- surements of various kinds, void checks, porosity checks—we are very busy there for the same reasons. e third area that's really busy is flying probe testing. We operate SPEA 3D image of head-in-pillow defect.