Tribal knowledge and design data formats
After more than six decades in the specialty chemicals, electronics interconnect, and electronic packaging industries, I remain perplexed by the resistance to change. Even higher costs, substandard quality, or impending failure don’t seem to be enough to motivate many of them to act, even for their own good.
One might wonder why I am writing about change today. We have seen dramatic changes in our lifetime. Half a century ago, Alvin Toffler reminded us of the inevitable change in his classic book, Future shock.
I recently started investigating “tribal knowledge” in PCB and PCBA operations. My goal was to determine if anything of value would be lost if it was not recorded or kept for future technologists in the industry. In the face of modern standards in materials science and industry, the descriptive title “tribal knowledge” suggests that one must determine what is fact, what is fiction and what can be just one. simple embellishment, coincidence or even magic. In many cases, tribal knowledge seems to consist of memories of problems solved by trial and error, coupled with observations of the resulting causes and effects.
During my search for information, I read the August 2021 problem of Charger PCB007 which focused on DFM from a manufacturer’s point of view. Several of the articles discussed the challenges involved in DFM and design data transfer. What seemed to come off the pages is that much of today’s design and data transfer uses a 60-plus-year-old format called Gerber.
I remember visiting Gerber Systems Corp. (which no longer exists) in Connecticut many years ago. The system is now supported by continuous developments by an entirely different entity. This system does not capture and store data like modern smart systems do. Important DFM notes are not retained or transferred by Gerber. Those notes are lost with the deaths of those who fixed the manufacturing issues. This important (tribal) knowledge may or may not have been recorded.
It brought back memories of my time as a tech student at MIT Lincoln Labs in the mid-1950s. The designers laid out a design on a large sheet of plastic. We then applied black tape to the pattern, cutting it with an X-Acto knife if necessary. This was then placed in a large vacuum frame, photographed with a large brown camera on a track, and photo-reduced by ten to produce the artwork (negatives) used to image our circuit patterns with KPR (Kodak photoresist ). In 1957, my supervisor, EA (Al) Guditz tackled this torturous procedure and created what I firmly believe to be the predecessor of today’s imaging systems. He converted a programmable head milling machine to directly write a pattern on Kodak film in a “darkroom” using a xenon light source collimated through a hypodermic needle to expose the film.
To read the full article, which appeared in the October 2021 issue of Design007 magazine, Click here.