Monday, 24 March 2014

Adapters ready to go

Thanks mostly to my colleague who has been working hard on layout in his spare time, the adapter boards are ready to go out!

As promised, some eye candy, and I'll attempt to explain what they are and how they're going to be used for the project:

MVS Adapter Board (System Adapter configuration)

Both the AES and MVS adapter boards connect to the respective AES and MVS system boards and cartridges. Shown above is a 'system' adapter, with the card-edge connectors on the bottom side of the PCB. In this case a fingerboard (shown below) connects the adapter board to the system board. The adapter has been designed to (just) fit into a horizontal 1-slot MVS motherboard, although the size was ultimately constrained by routing requirements. We'll have to wait and see if it does actually fit when fully assembled.

MVS Fingerboard - system adapter to MVS board


At the cartridge end, the same adapter PCB is loaded with the card-edge and (opposite gender) EURO connectors mounted on opposite sides of the PCB. The AES/MVS cartridge plugs directly into the card-edge connector on this adapter.

Ultimately then, a pair (CHA & PRG) of analyser/programmer PCB's plug between the EURO connectors on the system and cartridge adapter boards. In the analyser configuration, the PCB snoops the cartridge bus accesses for display via the Altera Signal Tap analyser in the FPGA.

The analyser/programmer PCB also has a programmer configuration for the flash cartridge which I won't go into now.

The plan is to send the adapter and fingerboard PCB's out to be manufactured this week, and also order the rather expensive connectors from Digikey. When they're assembled they can be tested by plugging the cartridge and system adapters back-to-back via the EURO connectors, which should act as a completely passive pass-through on the MVS and AES systems.

AES Adapter Board (System Adapter configuration)
While we wait for the manufacture of the adapter boards, I'll press ahead with the design of the analyser/programmer board. I've chosen a candidate FPGA and have started on the HDL design of both analyser (trivial) and programmer functions. Once I've convinced myself that the FPGA is sufficient for the task, I'll start on schematic capture.

Saturday, 15 March 2014

Friday late night update

Happily I can report that progress is being made. My colleague, who has some down time, has volunteered to take on some of the development. This has also spurred me to get the project into a state where he can work on it.

I've finished the schematics for both MVS and AES adapters. And as of this afternoon the AES adapter is almost fully laid out; only a few traces remaining and then the fun bits like the silk-screen overlay. The MVS adapter will be next, and will also leverage off the work done on the AES adapter, which was tackled first in order to gauge the size of the PCB - the MVS is much simpler to route, since the connector pin-outs are closer to 1:1.

At this rate we might have something to go out to manufacture by next Friday!

I've also been re-thinking the analyzer/programmer board and the flash cart board - considering making them both single PCB's to handle both CHA and PRG functionality. I think the flash cart is a definite candidate, the analyzer/programmer not so certain. A lot will depend on the real estate required by the level-shifting logic.

Donkey Kong is always in the back of my mind too.

And I've had a renewed interest in the Coco... need to prototype it on NGPACE first!

Friday, 7 March 2014

Still here...

Not much to post about - work and Real Life have been taking most of my time. I have been picking up a few Neo Geo-related items here and there in the last few weeks, like a couple of (faulty) 2-slot boards, hoping I can fix at least one of them.

Tonight I cleaned up the MVS adapter project, fixing the cartridge connector footprints and getting the Altium Designer libraries sorted - with the assistance of my knowledgeable colleague. Whoever designed that library system needs a good, swift, kick up the bum.

Since the AES adapter is going to be the messier of the two, I really need to lay it out before the MVS so I can match the PCB sizes and, more importantly, the EURO connector spacing. To that end I started on the AES adapter schematic, and would be working on it right now if only the ^%!@$# project files had gotten saved and/or checked-in to SVN before I left the office. Don't know WTF happened there.

Neither of the boards are particularly complex, so I'm hoping I can knock them over in the next few weeks as I'm expecting work to quieten down for a little while at least.

Lastly - Donkey Kong. I still have every intention of picking it up - and almost did a few nights ago - ever mindful of the fact that the longer I leave it, the harder it will be to continue where I left off. I just haven't really had a large enough slice of free time to devote to ramping up again.

My next update should be news about the adapter PCB's going out to manufacture. Not exciting in itself, but a step towards moving the entire project ahead. I did briefly flirt with the idea of putting the analyser/flash cart on hold and just going straight for the NGPACE main board... and I still might...