Reading time ( words)
I recently reflected upon the notion that this year marks my 50th year in the printed circuit industry. It was a bit of a shock when I looked at the calendar and realized that I have been kicking around this industry for a half-century. I was fortunate enough to find my way into the PCB industry through the analytical lab of a PCB company in Mountain View, California in the early days of Silicon Valley. The name of the company was Printex and it was one of the premier PCB fabricators in the United States.
Just prior to signing on with Printex, I had been working for Data Lab in Santa Clara, where we served as an offsite analytical lab checking the plating chemistry for many of the dozens of printed circuit fabricators that were operating in the Bay Area at the time. At Printex, my job was to analyze and maintain the numerous plating and processing solutions used in the manufacture of printed circuits. I was also tasked with preparing and evaluating microsections of the plated through-holes of those printed circuits.
I was young, ambitious, and energetic enough to complete those tasks quickly, leaving me time to go into the manufacturing areas of the facility to learn firsthand the details of every area of processing. It was of immeasurable value to me—for the rest of my career I had learned to troubleshoot, diagnose, and correct problems in processing. I also came to appreciate that there are few products that draw on such a varied palate of technologies to create them—composite material lamination, computer-controlled machining (i.e., drilling and routing), electroless and electrolytic plating processes, screen printing, wet and dry film imaging, developing and stripping, chemical etching, and several others. To me, it has always been one of the most attractive aspects of PCB manufacture and what makes PCB manufacturing tirelessly interesting.
What Has Changed
Today, many of those same processes are still used, though they have been greatly improved in terms of machines, materials, and processes available. Still, they remain fundamentally unchanged except that the circuit features are now approaching or in some cases equal to those produced on semiconductor integrated circuits of that same era. One big difference is that back then, the substrates for early semiconductors were 50- and 75-mm silicon wafers and today those near-same-size features are being produced on 450 x 600 mm FR-4 panels to make printed circuits. This has been a remarkable achievement.
Much has also changed in the realm of PCB design. PCB designers of the early days were largely mechanical draftsmen charged with “connecting the dots” on the schematic provided by the circuit designer. Artwork was often created by taping circuit traces and pads at one to four or more times the size of the final circuit and using a large format (near room size) cameras to “shoot down” the artwork to the size needed for contact printing the circuit image with working film. During that era, Gerber vector photoplotters were coming into use and light pens were mechanically driven using pre-programmed vector-driven information to create the artwork. In that early era, double sided printed circuits were most common and four-layer multilayer circuits were essentially state of the art, and the term “controlled impedance” was basically unheard of. Today’s PCB designers smile at such simplicity.
One unfortunate thing that has happened over the years is that, while semiconductors have grown in respect, appreciation, and even adulation, the PCB has too often remained under-appreciated and undervalued. I have, over the few decades of teaching PCB seminars and workshops, likened the semiconductor to a magician or illusionist and the PCB as the stage upon which the magician/illusionist works. My simple evaluation and statement of fact in this regard is that, without a suitable stage, the potential of magic or illusion simply will not happen. The PCB and the semiconductor must work together for the show to be a success.
In more recent years, I have personally become ever more appreciative of the importance of PCB designers and their work. They are clearly more knowledgeable than their predecessors. Today, the choices they make are of the utmost importance to the end-product in terms of its functionality, performance, manufacturability, and ultimate reliability. Today, designers must become increasingly knowledgeable of many different design attributes to make their designs suitable for the applications intended: DFR (design for reliability), DFT (design for test), DFE (design for environment), and DFA (design for assembly) are checklists alongside perhaps the most important, which is DFM (design for manufacturing). Several months ago, I suggested in this column, that a better approach might be to design with manufacturing or DWM. This was arguably common practice in the early days of the industry when vertically integrated manufacturers built everything “under one roof” and ties between design and manufacturing were much closer and stronger.
Today’s IC packages and PCB substrates must work flawlessly together to meet requirements, and designers must become increasingly attentive to mechanical concerns alongside the electrical concern. Matters such as CTE (coefficient of thermal expansion) and Tg (glass transition temperature) need to be part of their design calculus. So, also, will use of predictive modeling to look for prospective failures in advance and address them before they happen. Such analytical software is becoming more common as electronic products find their way into products which must perform in harsh environments.
In summary, the electronics industry at its core is a partnership between semiconductors and printed circuits and the importance of the PCB designers’ work cannot be overstated. They are the “drum majors” of the printed circuit industry in many ways and it is incumbent upon them to be continuously learning to make certain they keep current on the latest developments in PCB technology, to keep pressing the industry forward. I hope to be around for another 50 years to see what has changed. One thing that seems certain is that the lines between semiconductors and printed circuits will continue to blur as their domains seem destined to continue to merge into the future.
This column originally appeared in the November 2021 issue of Design007 Magazine.