While writing my recent 8-part EDA history (listed below), I became acutely aware that most of the effort devoted to EDA tool development has been aimed at IC design. A much smaller level of effort was devoted to developing printed circuit board (PCB) tools. The reason is simple, I think. IC design tools sell for far more money. While IC design tools cost upwards of six and seven figures for the most complex tools, PCB design tools sell for hundreds or thousands of dollars. Never mind that the market for PCB design tools is two or three orders of magnitude larger than that for IC design tools, in terms of unit volumes. The big bucks are in the world of IC design. Similarly, the history of semiconductor and IC development has been well documented, starting with the development of the point-contact transistor at Bell Labs in 1947 and continuing through the development of the first ICs at Texas Instruments and Fairchild Semiconductor in 1959 and 1970. But what about the history of PCB development?
I’ve been designing and fabricating PCBs since my high school days in 1970. That’s more than half a century ago. In high school, I took a couple of correspondence courses in electronics from National Technical Schools, which billed PCBs as “the new thing.” Those courses were written in the 1950s and 1960s, so PCBs were still the new thing. vertical heat treat oven
In those days, I applied black crepe tape, adhesive pads, and rub-on pad patterns from Bishop Graphics directly onto copper-clad laminate boards and then etched the boards with a ferric chloride solution poured into my mom’s rectangular Pyrex baking dish (thanks, Mom!). I quickly graduated to photoetching laminate boards that I’d sensitized with liquid photoresist poured from a bottle and baked in our home’s gas oven.
During the late 1970s, I created some of the PCB layouts for products that I designed as a lab engineer at HP’s Desktop Computer Division in Loveland, Colorado. That was under the watchful eye of the PCB layout manager at HP in Loveland, Bev Simpson. During those days at HP, we used colored pencils, mylar sheets, and a Calma digitizer to create photoplots for the in-house PCB shop. I wielded the pencils. Someone else digitized my drawings.
In the early 1980s, I designed engineering workstations at Cadnetix to run our proprietary PCB layout software. Soon we were using our own software and workstations to develop next-generation products. In all this time, I’ve never given much thought to how our industry started using PCBs. Like the history of EDA, the history of PCB development is not well documented, and so I’ve written this article.
Mass production of PCBs started in World War II. In those days, electronic products employed hand-soldered, point-to-point wiring using metal chassis to hold larger electronic components such as transformers, tubes, and larger capacitors. The point-to-point wiring was done with individual wires or axial-leaded components such as resistors and paper or mica capacitors tied to the chassis-mounted components or to terminal strips bolted to the chassis. This was the era of expensive components and cheap labor, so point-to-point wiring made sense.
However, even prior to World War II, there were people developing electronic products that tried fastening interconnect wires to stiff substrates. Initially, these wires were riveted, stapled, glued, or nailed to the substrate. Some inventors started to paint or screen conductive inks on a substrate to create a wiring pattern. The best known of these inventors was Austrian engineer Paul Eisler, who fled to England to escape Hitler’s expanding Third Reich movement in the 1930s. Eisler patented an additive PCB manufacturing process, but he found few takers for his hand-made PCBs.
When World War II arrived, the world’s armorers went into mass production: tanks, planes, jeeps, ships, etc. One pressing need was for a way to improve the effectiveness of cannon and mortar shells. Both the Nazis and the Allies worked on perfecting proximity fuzes, which would explode an artillery shell when it closed on a target. Allied work soon focused on a radar-based proximity fuze. Weapons of war must tolerate harsh conditions, but proximity fuzes endure some of the worst environmental requirements of any weapon. They must work after being shot out of a cannon, which means being subjected to 10,000 or 20,000 Gs of acceleration while spinning at 500 revolutions/sec. Getting a radar transmitter and receiver miniaturized so that it fits inside of an artillery shell and designing such a circuit to survive massive shocks and 20,000 Gs was a real challenge.
Development of the radar proximity fuze is a full story unto itself. Development of miniature vacuum tubes that could operate after being shot from a cannon is another full story. Those stories are for other articles. The PCB used in the radar proximity fuze, developed by Centralab, employed a steatite ceramic substrate with screen-printed conductors made from a silver metallic paint. Baking cured the ink. Centralab then added resistors by screen printing a carbon paste followed by another baking pass. This additive manufacturing process became known as the thick-film ceramic hybrid process, for which Centralab became famous after the war.
After making the bare PCB, subminiature vacuum tubes were soldered in place on the PCB. The tubes were encased in a rubber shock-isolation sleeve developed by physicist James Van Allen – for which the Van Allen radiation belts are named – who worked on the war-time proximity fuze development project. Then the passive electronic components were stacked vertically around the tubes cordwood style, soldered in place on the PCB, and encapsulated in wax to protect the entire assembly from G forces and vibration. By the end of the war, U.S. manufacturers including Crosley, Eastman Kodak, McQuay-Norris, RCA, and Sylvania had produced more than twenty million proximity fuzes. PCBs first went into mass production with the proximity fuze.
The first mass produced PCB was used for the electronic subsystem in World War II proximity fuzes. The loaded PCB resides at the base of the conical section of the fuze. Image credit: United States Navy
After the war, the U.S. military became extremely interested in miniaturized electronics for communications and other applications, like the proximity fuze. The U.S. Army disclosed details of the ceramic PCB technology used in the proximity fuze in February 1946, which seems to have triggered a wave of intense PCB development during the post-war period. It was so intense that the U.S. National Bureau of Standards published two books during the period immediately following the war: the 43-page “Printed Circuit Techniques, National Bureau of Standards Circular 468,” published in 1947, and the 73-page “New Advances in Printed Circuits, National Bureau of Standards, Miscellaneous Publication 192,” published in 1948. Page 1 of the 1948 publication says:
“Over the past 2 years, the National Bureau of Standards has received an unprecedented demand for technical information on the subject from other Government agencies, from industry and scientific institutions.”
The U.S. Army Signal Corps at Fort Monmouth, New Jersey developed the subtractive PCB manufacturing process that resembles what we use today in 1947. The process starts with an insulating substrate board laminated with copper foil, applies a resist (screen printed initially, eventually applied using direct photolithography), etches the traces on the board, and then drills or punches holes in the board to accommodate component leads. This process was adapted from the process that nameplate manufacturers were already using to etch metal nameplates.
Although the initial success in making bare PCBs came in 1947, components were still hand-soldered to the board one at a time. The breakthrough in mass production of PCB assemblies came in 1949 when Moe Abramson and Stanislaus F. Danko, working for the U.S. Army Signal Corps in Fort Monmouth, developed the “auto-sembly” technique, which plugged through-hole components into a blank PCB and passed the assembly over a solder wave to make all the component connections in seconds. That’s the year that mass manufacturing of electronic systems became possible. Danko and Lanzarotti published an article about their auto-sembly methodology in the July 1951 issue of “Electronics” magazine, and the PCB industry has been growing ever since. (Danko discussed many of the PCB techniques developed at Fort Monmouth during the 1940s and 1950s in Chapter 3 of Edward Keonjian’s book, “Microelectronics: Theory, Design, and Fabrication.”)
A tube-based Auto-Sembly circuit board created by the U.S. Army Signal Corps circa 1949. Image credit: Electronics Magazine
Today, there are hundreds if not thousands of bare-board PCB manufacturers around the world vying for your business. I am contacted multiple times each week by vendors, mostly in China, eager for my business. I’ve had to fabricate only one PCB recently. Rather than get my own account, I leveraged my friend Ron Sartore’s JCLPCB account. In fact, I leveraged Ron too. He laid out the board to my specifications. The JCLPCB boards are well made and were finished quickly, but the shipping from China took another week. If you’re fortunate to live in a major electronics center like Silicon Valley, you can get finished boards overnight, stuffed with components and soldered if you need it. Things have certainly changed in PCB technology since the 1940s.
While researching this article, Google took me to several PCB vendors that have added very brief PCB histories to their Web sites. Most of these histories mention the proximity fuze project during World War II, and some mention the US Army Signal Corps. All the histories seemed incomplete and some were clearly inaccurate, which spurred me to dig deeper. If you feel you want even more detail, the references for this article appear below.
“Printed Circuit Techniques, National Bureau of Standards Circular 468 ,” 1947
“New Advances in Printed Circuits, National Bureau of Standards, Miscellaneous Publication 192 ,” 1948
“’Printed’ Radio Circuits ,” Popular Mechanics, July 1946, reproduced by RFCafe.com
S. F. Danko and S. J. Lanzalotti, “Auto−Sembly of Miniature Military Equipment, Electronics Magazine ,” July 1951, reproduced by RFCafe.com and here .
Samuel J. Lanzalotti and Sherman G. Bassler, “The Design and Layout of Printed Circuit Patterns ,” Radio & Television News, November 1952, reproduced by RFCafe.com
“When Milwaukee Went to War, Part 4: Bullets, Fuses, Quintants and Can Openers ,” warmemorialcenter.com
“The Proximity Fuze: How Ipswich women helped win WW II ,” HistoricIpswitch.com
“The Proximity Fuze Part 1 ,” navalgazing.net
“The Proximity Fuze Part 2 ,” navalgazing.net
“Tiny Miracle—The Proximity Fuze ,” U.S. Naval Institute, Naval History Magazine, August 1999
Michael W. Robbins, “The Allies’ Billion-dollar Secret: The Proximity Fuze of World War II ,” historynet.com, October 19, 2020
David D Jackson, “VT Proximity Fuze Manufacturers of World War Two ,” The American Automobile Industry in World War Two, May 6, 2021
Henry H. Porter, “Recollections on the Development of Radio-Controlled Proximity Fuzes ,” Johns Hopkins APL Technical Digest, Volume 4, Number 4, 1983
VT Fuzes For Projectiles and Spin-Stabilized Rockets , U.S. Navy Bureau of Ordinance, OP 1480, 1946
“A Peek at the History of PCBs ,” ACDI, March 18, 2020
Peter Brownlee, “Tracking the story of the PCB – part 2 ,” What’s New in Electronics, April 10, 2007
“The Early Production of Printed Circuit Boards in America ,” Advanced Circuit, Inc, August 14, 2013
Edward Keonjian, “Microelectronics: Theory, Design, and Fabrication,” McGraw-Hill, 1963
You must be logged in to post a comment.
Sponsored by Mouser Electronics and Littelfuse
In this episode of Libby's Lab, Libby and Demo investigate quiet, reliable SRP1 solid state relays from Littelfuse availavble on Mouser.com. These multi-purpose relays give engineers a reliable, high-endurance alternative to mechanical relays that provide silent operation and superior uptime.
Click here for more information about Littelfuse SRP1 High-Endurance Solid-State Relays
dry storage cabinet Implementing a low precision network in FPGA hardware for efficient inferencing provides numerous advantages when it comes to meeting demanding specifications. The increased flexibility allows optimization of throughput, overall power consumption, resource usage, device size, TOPs/watt, and deterministic latency. These are important benefits where scaling and efficiency are inherent requirements of the application.