Issue link: https://iconnect007.uberflip.com/i/1189040
42 SMT007 MAGAZINE I DECEMBER 2019 With the advent of fine- and ultra-fine-pitch, high-pin-count BGAs—0402, 0201, and 01005 resistors and capacitors—as well as the wide- spread use of no-clean flux, yield problems are getting worse, especially when you use large BGAs and BTCs in addition to a smattering of through-hole components on the same board in a lead-free process. It is also worth noting that no matter what our job titles are in various departments—such as purchasing, design, manufacturing, quality, inspection, test, or repair—the key focus of what we do is to make sure no defective prod- ucts are shipped. That means the purchasing manager is responsible for not just focusing on cost but also on the quality of incoming mate- rial, such as components, PCBs, and other materials. And DFM managers are responsible for the overall defects. 1. Choose the Right Defect Let me be clear; we all want zero defects in our final product, but good luck with that. We certainly can ship products with zero defects, but how can you do that if you don't achieve zero defects in manufacturing? The obvious answer most people will have is to inspect and test followed by repair of defective units, but no; you cannot prevent the escape of defects through inspection and test completely, no matter how sophisticated or comprehensive those inspection and test regimes are unless you choose the right kind of defect. Choose the right kind of defect? Yes. Even though there are hundreds of types of defects classified in industry standards, such as IPC 610 and J-STD-001, there are only two types of defects that test methods, including functional test and in-circuit test (ICT), are intended to flag. They are shorts (bridge) and opens. For- tunately, we have a great deal of control over which one we get. By the way, they are both bad, and the assembly will not function if we have either one, but one of them is better than the other: shorts. Which one do you think is the predominant defect in every company? Nearly all compa- nies, including yours, have way more opens than shorts. All you have to do is look into the defect data over the past six months to a year and put the defects in three categories: opens, shorts, and others. Whatever defect does not fit into either the "opens" or "shorts" catego- ries should be put into an "others" category. The defects in the "opens" category will be higher than the other two categories because they're the type of defect that can pass even ICT since partial opens can appear as good joints when vacuum pressure is applied during ICT; opens often end up being discovered by the customer or in the field. And this is why it is not as desirable a defect as a short. If you change your design and assembly process to focus more on solder paste deposit and better stencil design, most of your defect 1560s will revert back to shorts from opens and be caught before shipping. Your cus- tomers will never know your problems, and there's an easy solution to a difficult prob- lem, but it's rarely practiced. You can, how- ever, learn to change that. You are not achiev- ing zero defects, but you are choosing your defect wisely. 2. Develop an In-house, Company-specific DFM and Process Recipe No one gets up in the morning and says, "I am going to screw up three things today at work." We all follow the process and proce- dure, but the problem is that very few com- panies have written processes and proce- dures, such as in-house DFM and assembly processes. Even those who spent considerable resources to produce these documents don't keep them updated, as technologies change, and these documents become obsolete within a year. Thus, it is not only important to develop your own unique DFM and manufacturing pro- cess documents, but you must also maintain The key focus of what we do is to make sure no defective products are shipped.