Issue link: https://iconnect007.uberflip.com/i/1078362
20 SMT007 MAGAZINE I FEBRUARY 2019 1970s assembly procedures and materials, and may or may not have even been fully effec- tive for that time. As components continued to shrink over the last 40 years, the acceptance criteria didn't adjust as it should have. One reason for this is the group of people with the mindset that, "This is how we have always do- ne it, and change be damned, it's how we are going to keep doing it." Guess what? People used to die all the time from polio and tuber- culosis, but as medical science advanced and people embraced vaccinations (except for Jen- ny McCarthy of course), those diseases are a thing of the past. Now, I am not saying that a dirty PCBA is as dangerous as TB, but my larger point is that scientific advances are allowing us to make smarter choices in terms of adopting a qual- ity strategy. Too many times I see an assembly print without any reference to cleanliness or only the default of IPC, which is super vague as there are 200+ test methods and only a few re- lated to cleanliness. The saying "you get what you ask for" comes to mind. It is so important to demand and clearly document cleanliness requirements and verifications on a regular ba- sis to help mitigate cleanliness related issues in the field. So, what is a good strategy to maintain qual- ity related to cleanliness? Here in the lab, we love the customers planning new product to test bare fabrication, components, as well as first-run production with ion chromatography. That is the first and easiest way to see just how clean each of those sample sets are before you go into full production. IC analysis gives you a good picture of the type of ionic residues pres- ent on each sample, and each of those ions can be traced back to a specific part of the process. With that information in hand, and when necessary, you can go back to the supplier, re- view their process, and see where it can be op- timized to reduce residual ionic content. After optimization, a second round of testing is in order to see the effectiveness, and then further optimize from there. Component manufactur- ing and bare boards share a lot of the same risk due to the plating processes, which use highly active chemistries. Those chemicals need to be fully removed or effectively neutralized to a low enough level that when they arrive at the CM, they won't be the root cause of assembly failure down the road. Many PC fabrication shops aren't using the best quality DI water for rinsing after plating because good DI isn't cheap. The higher-qual- ity wash or rinse water uses lower the surface tension. Lower surface tension allows the DI water to penetrate into vias and remove pro- cess residues better. In past studies, we have found that tap water is incapable of penetrat- ing vias smaller than eight milliliters, which are ever present on today's PCBAs. We found one supplier that was using water directly out of a nearby river. I can't make this stuff up. The print didn't have any cleanliness specification or direction for proper washing after plating, and the customer was more than thrilled to save a few cents per board with no questions asked. When the boards were used for actual produc - tion, the customer had massive amounts of failures, and the cost of replacements couldn't be pushed back on the supplier because they got what they asked for—nothing. This company was an obvious outlier in the big picture, but it goes to show that you can't rule out anything if you don't have a rule in place. The strategy here for the PC fabrication shop would be to pay a little more for a good quality water system that can maintain at least 10-megohm resistivity DI water for more ef- Now, I am not saying that a dirty PCBA is as dangerous as TB, but my larger point is that scientific advances are allowing us to make smarter choices in terms of adopting a quality strategy.