Exceltronix Multiflex Super System - page 1 of 5

Jeff Avery on the Web

HOME        SITEMAP        ABOUT        CONTACT        INTERESTS


Z80 Exceltronix Multiflex Super System

October 1999: Not having heard from my friend Geoffrey Barkham for several years, it was a surprise to receive an email from him. I wrote back and we corresponded in several more emails, catching up on lost time. Then the subject turned to his Exceltronix Multiflex Super System, a Z80 computer kit that he bought and built in 1982.


The Multiflex Super System consisted of a motherboard designed around the S100-bus with a plug in Z80 CPU board. There were two keypads of 16 keys each that, together with a monitor programme called Z.MON that resided in EPROM, allowed the user to enter data, run programmes, set break-points, and do many other useful things. The computer was sold by a company called Exceltronix based in Toronto, Ontario, and on the right is one of their ads that appeared in the September 1983 issue of Computing Now! magazine.


The Multiflex motherboard provided for four S-100 slots and so the system could be expanded by plugging in other boards that supported the S-100 bus protocol. As well as the Z80 CPU board, Exceltronix offered a floppy disk controller board with 8” or 5-1/4” floppy disk drives and a keyboard/video controller board; adding these boards converted the system into a good personal computer for its time. Along with the drives, Exceltronix provided a floppy disk containing CP/M, a disk operating system from Digital Research. Pressing a designated keypad key started a boot process that read in the disk, started CP/M, and presented the user with an A> prompt.

Geoffrey still had the computer but, in the course of moving house over the years, he had lost all of its floppy disks and thus could no longer boot CP/M. He was interested in resurrecting it and was trying to find replacement operating system disks, so far without success. A problem is that CP/M was obsoleted many years ago by Microsoft’s DOS and so CP/M boot disks are hard to come by. Worse than that, CP/M systems all appeared to use different floppy disk formats and, in order for them to successfully boot, the disk needs to contain the correct BIOS (Basic Input/Output System), the code that fits the operating system to the particular computer in question. In Geoffrey’s case, that meant an Exceltronix CP/M disk and it seems they were not to be found, Exceltronix having gone out of business a long time ago. Meanwhile, he sent me much of the documentation to read at my leisure.

February 2000: Geoffrey emailed that he has still had no success in finding a suitable CP/M boot disk. Looking for another avenue forward, he told me he has a second set of Exceltronix boards; if he sent them to me with full schematics, test procedures, a 5 -1/4” floppy drive and a dozen blank floppy disks, would I be interested in fooling around with them? A caveat is that the disk driver board is missing several chips and he does not have a spare keyboard/video board to send.

I had not, up to that time, done any work on interfacing floppy drives and so it was an area in which I knew practically nothing. However, it sounded like an interesting project to dive into. Essentially, the project would be to learn enough about the system to produce, working from first principles, our own CP/M boot disk. The idea intrigued me. With Geoffrey in Ontario and me in Nova Scotia, I felt it essential for the project that we had identical systems to work with and so the absence of a second keyboard/video board was a problem. However, I thought that we might do away with that board and switch to using a terminal to converse with the computer using the UART (Universal asynchronous receiver/transmitter) that the documentation shows to be on the CPU board. We agreed to go ahead on this basis.

I started to dig in. Of prime importance was to understand the Z.MON monitor programme and the floppy disk driver chip, a Western Digital WD2793. I also started thinking about the programming changes needed in the monitor to switch to terminal input/output. Lastly, I started to search the Web for information about CP/M. I found the source code for version 2.2 and assembled it.

March 2000: In March, the boards arrived from Geoffrey. I plugged the CPU board into the motherboard and supplied power but the system did not work. It took quite a bit of debugging to find that a pin on one of the chips had broken in its socket although it looked O.K. at a glance. I didn’t have a replacement chip, but I was able to “borrow” one from the on-board EPROM programming circuit and then everything worked.

I wrote a change to Z.MON to convert it to use a serial connection when booting CP/M and burned it into an EEPROM rather than an EPROM. To have Z.MON boot CP/M, one pushes the “BT” key. This displays a message on the screen asking you to press the carriage return key. On receipt of the carriage return Z.MON reads CP/M into memory. Although I didn’t have the floppy disk controller plugged in nor any floppy disk drives attached, I pushed the BT key and was able to verify that the expected message appeared on the terminal screen.

Next I turned my attention to the floppy disk controller board.


<-Computers             Z80 Exceltronix Page:     <-Previous             1             2            3            4            5            Next->