Issue link: https://iconnect007.uberflip.com/i/1529411
40 PCB007 MAGAZINE I NOVEMBER 2024 Feature Article by Happy Holden I-CONNECT007 ere are three things I know mathemati- cally about PCB yields: Defects that produce low yields may be systemic and repeatable, while others may be random (but you can test for this); you can predict yields if you take the time to collect the data; and yields are such wonderful things, so we should write about them more. Yields are the dividing phenomena that oen decide whether we make a profit. ey certainly control our efficiency and through- put. Most customers acknowledge high yields with competence, control, and capability for the fabricator. So, why don't you read more about them? I think it's because we have been Calculating PCB Complexity and First-pass Yields ver y poorly schooled in statistics. Not boring statistics like the kind I had in college with endless equations and proofs, where only the professor seemed to care about the endless stream of numbers. I'm talking about statis- tics that are fun and revealing, and a tool that ever y engineer must embrace. ese kinds of statistics allow the numbers (yields) to reveal certain truths. My love of the numbers (and statistics) came from four sources: 1. I was introduced to "problem solving and design of experiments" in an advanced chemical engineering course in college. 2. I took one of the E.I. DuPont de Nemours Professional Development Courses in "Design of Experiments" at work.