Wilcox Engineering

When I was teaching at Bucknell University in the mid-1980s, I had the opportunity to consult with a major PCB manufacturer who was building 68000-processor boards for the new IBM CS-9000. The board was an up-scale Motorola answer to the Intel PC/XT that was currently on the market. The 68000 was a great processor, and a shame that it wasn't in the original PCs of the day. At the time, none knew how the Intel/Motorola issue would shake out, so I pressed forward on the 68000, especially because the Apple Macintosh was coming on.

My first book, the so-called red book, was on designing and troubleshooting a 68000 board that would work. One of the problems with the IBM board (with 10 or 12 layers and over a hundred added corrections) was signal timing. Whoever designed it hadn't accounted for various circuit delays that caused rather erratic performance of an otherwise good product. The feature of the red book was an example wire-wrapped 68000 board that plugged into the S-100 computer bus. 

Considering my troubleshooting experience with the CS-9000, my main theme was to design for testability. When a processor acts up, it's virtually impossible to get a handle on it without a logic analyzer; my approach was to provide a way to test and troubleshoot with only a scope and logic probe. So I designed the board with a DIP switch that the uP would read at boot-up. Depending on the switch setting, the uP might test memory, drive a serial port, read in serial data, flash an LED, etc.

Even more fundamental, I provided a way to test the uP with only the reset circuit and clock working: I free-ran the processor. "Freerunning" means, for the 68000, grounding the data bus and resetting. When reset, the 68000 would read its program counter and stack pointer from memory at address 0000. After reading its first instruction at address 0000, it would go on to execute it. It happens that the instruction with op-code 00 does an internal operation. Having completed that, the uP increments the address bus to fetch the next instruction, which is another do-nothing instruction. The result is that each of the address-bus lines has a square wave as the uP goes through its entire address range and back to 0000 for another pass. It's a simple matter to touch a logic probe or a scope to each line and see it oscillating at a slower rate for the high-order address lines. One can also watch address decoders enable as the uP sweeps through their particular address ranges. See N8VEM's video and his 68000 hobby page.

Did it work? Yes, within two weeks I could expect my students to have a 68000 built and successfully freerunning in the lab. From there, they could design and build the address decoders for an EPROM and RAM. With the DIP switch designed in, they could decode for first instructions in EPROM to do simple tests. All with logic probes and scopes. Add in the RAM along with memory-test code in the EPROM, and it wasn't long before we had some serious computing possibilities. 

Micro Resources

While I was at Bucknell, I formed a small at-home business called Micro Resources to design and manufacture 68000 boards. I hired a couple top students to do some of the work creating a unique 68000 board that plugged into a PC. The software we wrote allowed us to program and run the 68000 directly from the PC's screen. It could be used for simulations or data communications. Using the company initials and the “8” because of its 68008 processor, we called the new product a “Mister-8."

Utilizing the design principles in my 68000 book, the new board was easily manufactured and tested. A company in State College built the board and shipped them to me for sale. The product idea, although exciting, could have used additional market research, because I sold only about a half dozen of the Mister-8s. There are several left ...

It's pretty difficult trying to do any work on a board that's plugged into the PC's motherboard. The standard fix to that is the "board extender." It was a standard off-the-shelf product of the time.

The early development boards included a wire-wrap model of the Mister-8 and a clock board.

What you don't see in the photos are the piles of notebooks and design papers. And the printed source code for the EPROMs ... all of it properly recorded on floppys, whatever good that is. I think the 8" versions were tossed; probably some 5-1/4" are left, still safely in a folder somewhere. To be read by equipment that's no longer around!

The Bottom Line

I sold Micro Resources years ago, and this is all I have left. So, for the last twenty-five years, these boards have been waiting for some project to come along. And waiting ... 

All this really needs a good home. It's pretty clear, even to me, that I'm just not going to get to them again. 

Anyone interested?