Some years ago I became interested in trying to recreate the almost mythical 1090 XL Expansion Box. They were never actually released by Atari due to numerous reasons. Still some fortunate few had managed to get their hands on mostly complete proto examples. One individual down under had too much time on his hands and volunteered to trace out his proto board. With his hard work and my still learning experience with Eagle and Diptrace EDA software, a schematic of the original protoboard was born. The foundation was laid to support further development.
Scouring various Atari document dumps and old magazine articles I began to amass almost all the remaining info on the PBI/ECI bus of the Atari 800XL/130XE computers and how that bus or the 1090 had been used or planned to be used by Atari. Most of it was old and newer information had not been released by developers who had currently active projects. Hopefully that information will become available soon. Some of the original Atari documentation was contradictory depending on the date of the documentation, indicating that the feature set of the expansion box was in flux right up to the time the project was dropped by Atari.
This much was clear. Atari intended to allow the 800XL model to have expansion capabilities that would hopefully extend the life of the computer. Those expansion capabilities planned were breath taking in that day and age. They had even provided rudimentary plug and play in the operating system of the computer itself. A decade or more ahead of the Microsoft Windows 2000 operating system that finally began to natively support major manufacture’s devices in the late 90s.
The original 1090 prototype box connected to the 800XL PBI bus via ribbon cable and buffered and distributed the signals to five card edge slots inside. Those slots also contained additional connections carrying more power supply values as well as selection pins for each slot. The pinout of the slots was never an exact match to the 800XL PBI bus by design. As technology has progressed and some signals became obsolete (ie ram chips no longer needed refresh signals, so some signals on the PBI bus were no longer needed.) I reasoned that I could drop some signals and provide additional internal communication with newer technology. This also helped in combining the slightly different 130XE ECI bus to connect to the expansion box through a simple physical adapter board.
Finally a few months later after a few false starts and mistakes I had what I believed to be a functioning expansion box prototype. It is not a replica, nor is it a complete new implantation. It’s somewhere in the middle. The new expansion box could be powered externally and the 800XL or 130XE would boot normally and function. The next step was in developing plug in boards that would prove the concept. I envisioned ultimately builtin parallel and RS232 serial interface ports. Why not start with that as a plugin card? That would provide the building blocks for other plugin card development as well.
Looking through the old documentation I had collected to find out if anyone else had done something like that already, I found a fairly extensive amount of PBI work done by Robert Scholz and published by MyAtari.net in 2001. I located an email address for him and contacted him. I told him about the project and what I currently had designed and asked him if he would be interested in modifying some of his work and design an IO card for the new prototype. He seemed excited and eager to do some work with the Atari again.
Within a couple of months or so we had hammered out the details of what the plugin card feature set would be and Mr Scholz had a prototype up and running on his 800XL PBI bus. During this same time I had also been in colaboration with Jurgen(tf-hh) from AtariAge as he worked on a 1090XLR version of his Syscheck board. I was working on converting the RomBoard as a plugin card and converting the IO prototype to a plugin board.
Everything looked promising for awhile. I had put up a website chronicling the progress and did solicit for donations. It was getting more expensive than I could afford on my own. I had several donors contribute a significant amount, but it just never seemed to be enough. I’m just stating a fact, not complaining. Then tf_hh ran into problems getting the Syscheck to function, and life also began to demand more of his time. I found my own finances or lack there of interfering with my progress as well. Finally I had to put the project on hold when my own life interfered.
That’s where it’s been for the last nearly two years. With little hope that I can afford to move the project along any further I’ve decided to release everything I have to public domain. Perhaps someone else can pickup where I left off. The major areas still remaining is the enclosure and taking the prototype design to a final release version. That in itself is not a cheap task.
I also will be putting up extra bare boards for sale with BOMS. I have 5 complete sets of bare boards that will be sold as a set and then the remainder will be sold as individual boards. Once the inventory is exhausted I won’t be doing another run. Understand these boards are not necessarily ready for prime time. They are proto boards for testing and improvement.
Some eye candy of the 1090XLR project.
Known 1090XL documentation