Toward the end of 2009, we started entertaining the idea of building some kind of logic demonstrator out of relays for use as a museum exhibit. The idea was to bring the concept of computer logic down to a more demonstrable level that people could see and hear. I browsed through parts catalogs for days, looking at relays, and finally selected one that I thought suitable for the purpose. Once they arrived, I started experimenting with various circuits, trying to construct a J-K flip-flop, but was not successful because of the difficulty in edge-triggering with relays.
About that time we discovered an article in Make: magazine recounting how computer science professor Dr. Harry Porter, at Portland State University, had constructed an entire working computer out of relays. By some chance alignment of the cosmos, the relay Harry had picked to construct his relay computer was exactly the same relay I had picked for our logic demonstrator. That got us to thinking that if we could duplicate his feat the result would be way cool.
I contacted Harry Porter and asked if he would have any objection to our building a similar computer based on his design, with proper attribution of course. He replied that he would be honored and thrilled if we built a machine using his design. We discussed what changes he would make if he were doing it over again, and I suggested a few modifications and additions, as well.
That probably was about as far as the project would have gone had I not discovered a skill I didn’t know I had: begging. I contacted the relay manufacturer and asked if they might have a charitable contributions program that would be willing to donate product to the museum for this project, requesting 450 relays, with sockets and mounting rails. To our surprise and amazement, they replied that our request for relays had been approved, but they could not provide the sockets and mounting rails, which were in short supply. A week later we received a large box of relays and things were underway.
Harry Porter constructed his machine entirely by himself, using hand sketches of schematics and wiring diagrams, often doing the step-and-repeat of repetitive circuits in his head. We planned on using a volunteer crew to do the construction of our machine, so complete documentation was a requirement. I spent the next nine months creating a full set of schematics, timing diagrams, cabling plans, and a bill of materials.
Story continues in Electrical Design.