Motorola 6800 6809 EXORbus-based systems

Most recent revision date of this page, Oct 20 2025 Copyright 2025 Herb Johnson except content as provided by the authors noted. These are discussions and content about 6800 and 6809 systems based on the EXORbus, like the Exorset 30 system I have Content available and correspondence quoted with permission, thanks! Look at all my 6800/6809 work on another Web page of Web links. - Herb.

Stan Ruppert

Stan Ruppert contacted me on July 29-30 2019 as follows:

I'm working to restore two Matrox brand graphics cards ( exo-2480 and exo -512 ) These are the Motorola exorciser system versions ( 86 pin bus ) of the s-100 alt-2480 and alt-512 cards. Hoping to add graphics to my vintage exorciser systems. I'd like to pick up some clues about their circuit design philosophy from the S-100 Matrox ALT-512 manual you have. Google searching led me to your S-100 site. [I supplied him with a ALT-512 S-100 manual. - Herb]

I've sent photos of my EXO 2480 and EXO 512 boards. Looks like 90% of the layout is similar between the S100 and Exorciser versions. At first glance, only the bus interface section and removed voltage regulators looks to be different.

Looks like your Exorset 30 is a terrific and quite capable system. That's a system I haven't actually had the pleasure of using. Love how it is an all in one with screen, keyboard, and cage for cards.

The 6800 Exorciser system I mainly use is indeed the first version Motorola released for the 6800. One heavy beast of a chassis. It's the system documented [at bitsavers.org]. [Later Ruppert said he found documents at archive.org as well.]

The disk system that goes with it is also first version of the EXORdisk - using the Pertec FD400 drives in an FC360-2 enclosure. [Here's a photo of the controller board.] The controller has the boot code in bipolar prom specific for the Pertec drives. The controller uses the 6852 for IO. The MEX68DB card has the MAID monitor code that then calls the boot code on the controller card when booting XDOS or MDOS.

Most other systems I've seen are later versions, using the second rev of the floppy controller 68FDC2 - documented on bitsavers and other vendor floppy drives. Haven't found much info other than brochures for the older controller and had to try and reverse engineer the schematic from the board. [Later he found the manual. - Herb]

I've been using and enjoying exorciser based systems ever since high school in the mid seventies. A relative gave me a 6800D2 kit to experiment with - and it sparked my interest and led to a career in engineering and science. I just recently restored my original D2 system; amazed I could find all the boards and my original notes stashed here and there through many moves. [I discuss a D2 on another Web page. - Herb]

[It became] quite a Frankenstein system in the end - with IO boards enabling use of S100 memory cards and Wintek 12K Basic and Wintek Fantom Monitor and an electronic systems TV typewriter board with custom UART circuit for keyboard. I had also pried the lid off of an early 1K dram chip back then and used it's light sensitivity on the 32x32 die as a crude telescope control / guidance system. Cool proof of concept. Learned an incredible amount amount electronics as a kid (nerd) experimenting with it.

I've restored a variety of Exorciser based boards, SWTPC stuff, and later Motorola dev kits since then - including math coprocessor cards, IO/control cards, A/D cards, even a magnetic core memory card - Love ferrite cores! And now the Matrox cards for graphics.

Have several systems running for my continued 6800/09 code development and I use some of them to digitize signals from various earthquake sensors (my day job is earthquake research and engineering). Fun to combine the profession and hobby. Amazing how at work I think nothing of the petabytes of data we store and analyze when it all began with kilobytes on my 8 bit systems.

Regarding the ALT-256/512 cards, Hugo Holden just posted a draft whitepaper he wrote on restoring his 256 and 512 cards. Great read if you haven't seen it yet.

Your site has been a great reference - I appreciate all the time you've put into documenting and sharing your experiences with other enthusiasts. The historical and technical restoration information has been super informative on many projects.

- [regards] Stan Ruppert

Neil Cherry and various ExorBus boards

Creative Micro Systems CMS

My friend and technical colleage Neil Cherry was working on some Exorbus boards and contacted me on Oct 2019. He found a Web blog about some Creative Micro Systems CMS 6809 hardware. It's linked to a github site of CMS ROM disassemblies and some bits of documentation for the CMS 9639 and 9642 boards. Here's his 9609 CPU board. These are EXORbus compatible cards apparently used for industrial controls.

Neil Cherry: Here's what I have found [from Web search] the debug roms for the CMS boards. Very useful but I've not dug into them deeply yet. That should be useful for running the CMS 9609 board I have.

While reading about the 9639 (I don't have this one but do have the PDF), I found that it has 2K of DAT (6116 RAM used as Data Address Translation). The 9609 has an odd setup for RAM, it has a 6116 (2K) and 2x2114. That makes no sense, but I opine that it is the DAT RAM. It's not quite the same as the CMS 9639 but if I were to engineer something like this. My later designs would be based on earlier ones and many good engineers follow a similar pattern. If I can get this set of boards working I think I'll have a rather unique example of a commercial/industrial computer from the pre-ibm PC days. - Neil

CMS employee Dave Lyons

Dave Lyons contacted me in late 2025 about his time at CMS. Here's what he told me. - Herb.

Hey, Herb. I was looking around [the Web] to see if there were any pages of history on CMS stuff, and I came across your site. I worked at CMS from ‘78 to ‘86.

I started at CMS in the spring of ‘78 [when I was 16]. Pete Prossen lived across the street from me and, at that time, CMS was being run out of his and Elaine’s house in Westminster, CA. I worked in the garage “stuffing” boards [with components] before they were taken to a place to be run through the wave solder machine. As best I remember, at this point the boards being made were the 9600 (6802 processor board), a board with 6821s PIA’s on it, another with 6850 ACIA parts on it, an 8K static memory board, an extender board, and then the card cage. [The extender had] pins on it to attach scope probes to any of the lines coming from the backplane.

A few months later, around June of ‘78, CMS moved into an industrial park up in Garden Grove, CA. This building had a large warehouse, a room with a raised floor where we kept the burn-in racks and test equipment, and then the front office. Around this time, two other guys my age (16 at the time) were hired and, along with stuffing the boards, we started to run them through the newly purchased wave solder machine and [then on] to test and trouble shoot. The 6802 processor had a special instruction known as the “halt-and-catch-fire” instruction that we would have the processor execute (I think the opcode was 0x3F). That put it in a mode where the address lines would constantly increment so that each made a square wave half the frequency of the previous line. We could detect an address line short by using a scope to see if there was a line that had a combination of two of those signals.

[While working at CMS,] the three of us went into college in 1980 all studying electrical engineering. Two of the guys started working on hardware design for new CMS boards. I’m fuzzy on what got built when, but we had multiple 6809-based processor cards including one that allowed paged memory management so that individual OS process could have up to 64KB of memory. There was also a floppy disk controller (the 9671) and an I/O processor which had multiple serial ports and its own 6809 to offload some of the user communications from the operating system processor.

[Meanwhile I started working] on the software side. In early 1981, we started supporting a real-time operating system called OS-9 that was developed by Microware in Des Moines, IA. I worked on device drivers that adapted our hardware to their OS. Around 1982, CMS moved from Garden Grove to a building in Los Alamitos that had more room.

Over the next few years, besides the embedded hardware business, CMS branched out into developing a payroll system called PayLink that included time card reading hardware and all the software needed to track employee hours. There were around 15 - 20 engineers working on it (mostly on the software side). The system was successful and used by some relatively large customers.

Toward the end of my time [at CMS] I remember there was a processor board developed that used a 68008, similar to the 68000 processor used in the first [Apple Macintosh] systems. After I graduated college in 1986, I moved to Iowa to work for Microware. While I kept in touch with people from CMS, I don’t remember many details from then on. The company was eventually sold and its three founders retired.

Let me know if there are any questions I can try to remember the answers to :-) - Dave Lyons

Motorola Micromodule

[Photo of the Motorola M68MM19 6809 Micromodule]

Neil: I have spent so time putting my parts orders in for my 2 Corsham Tech SWTPC 6809 replica [CPU boards]. I'll order more parts next month and work on adding what I have over the next few weeks.

I did get some information on the Exor bus. I'll compare it to the schematics you provided. I'll then post that up on my web site. - Neil

Mikul Exorbus products

Neil Cherry has a Mikul 6809-5 board, that Neil is working on. He says: "The MIKUL board is a terminal board. It has a 6809 (MPU), a 6847 (Video), a 6116 (SRAM), 6821 (PIA) and a Z80 SIO. No space for ROM or extra RAM so it expects that to be on the bus. Not sure how I can run both the terminal and the MPU boards in the same bus (probably can't) but I'm still thinking about it. Probably won't get further than that. ;-)"

From some industrial sources, I extracted the following information. - Herb

The 6809-5 was a "standard product board" that has an operator interface consisting of an encoder for a 20 key keyboard matrix, and a 16x32 character monochrome composite video CRT display driven by the 6847 and the 2K static ram display buffer. It also has a proprietary transformer-isolated industrial network interface using the SIO. The 6809-5 has no on-board ram or prom memory.

The Mikul 600 Series boards were designed for industrial applications, and had 6800 or 6809 processors. Motorola ExorBus compatible cards, the Mikul 6000 series, followed, mostly with 6809 processors. Eventually there were some 68000 VME cards, the Mikul 1600 series.

The Mikul 6809-5 board was used in a heat treating furnace control for a furnace OEM. The control managed the temperature and atmosphere in the furnace and some of the automation related to moving product in and out of the furnace. The operator interface consisted of a keyboard matrix interfaced to the PIA,  and a CRT display (driven by the 6847 and the static ram). It had TL's proprietary transformer-isolated industrial network interface, (SIO, some logic gates and the 2 red transformers at the top of the board) to support statistical process control.

There were a couple other boards in the system with analog I/O (A/Ds, D/As, 6840 timers), and memory (ram, prom and EEROM). The control cabinet was beside the furnace, a pretty hot, dirty place.

AMF brand boards

In June 2022, my colleague Neil Cherry obtained a set of AMF branded boards which appear to be ExorBus format, with 6800 processor-family chips on them. He and I sorted out what these are and where they were produced. I cobbled up a Web page, to gather and solicit information about them. - Herb

Copyright © 2025 Herb Johnson