(Project completed: 1991-06)
During the years of 1989 to 1991 I took a six-semester Electronics Engineering Technology (EET) diploma program at George Brown College in Toronto.
In the final two semesters there was a sort of mini-thesis project where we each had to pick a significant electronics design project, produce a detailed report about it including schematics, component selections and pricings, and then implement it and make it work.
I wanted to make an EPROM programmer of some sort, but I was told that wasn't ambitious enough and wouldn't be worth a full grade. Instead, I was given a project that would put a useful tool in future students' hands.
Earlier in the program, all of the students assembled a microcomputer kit called the µSOKit MK3. I haven't been able to find any documentation about it online, but it was 6802-based, had a generous amount of RAM (2K I think), some I/O ports, and not much else. It was on a circuit board about the size of a regular sheet of paper, and was programmed with toggle switches - you set eight switches to the bit pattern you wanted, then pressed a button to transfer that pattern to either the upper or lower eight bits of the address bus or to the data bus to write to memory. Once you had finished inputting your program this way, you could hand over control of the bus to the CPU to execute your program. LEDs displayed the state of the address and data lines. Every student had to assemble this computer and make it work early in the program, because it was an essential tool for several of the later courses.
My project was to build a flopy disk interface for the MK3, which would enable students to save their programs to disk instead of having to laboriously re-enter them after each power cycle.
The blurry film photo above is the only picture I have of the project. The MK3 computer is the board on the left. My project is on the center breadboard and spilling over onto the right breadboard for some extra debugging circuitry. The floppy disk drive is next to it, and I used a PC power supply, seen in the background.
The oscilloscope and logic analyzer on the left and right edges of the photo were used for debugging, and there were some bugs. I messed up the power connections once and made the quartz window on the EPROM I was storing my program in glow - a lot of current went through one of the gold wires connecting the chip's wafer to its external pins, and it got pretty bright before I pulled the plug. I had to test every chip after that and there were a couple of dead ones.
In the end I didn't quite get the circuit working. Everything seemed correct, all of the signals were doing the right dances on the screens, I had others check my work, but for some reason the disk just wouldn't foom. I ran out of time, unfortunately, but I still got an A+ for the report and for the addendum I added to summarize the results. The project was set aside for someone to pick up in a later year, but I don't know if that ever happened.
Anyway, here for posterity are my reports on this project. These two PDFs are made from scans of the printed reports; there was no digital master copy of the orignal. I used Print Shop for the fancy fonts and most of the text, but there wasn't much software interoperability in those days so I had to do the layout by paste-up and then photocopy that to produce the final results.
I was frustrated and disappointed that I didn't get this project working, but it was a fun challenge and I was happy to get a decent grade for my work.