The SBC is now rebuilt with GAL chips. Ended up using one GAL20V8 and one GAL16V8 to replace the whole lot of TTL. I also took some advice and tied some signals with the phi2 to get more accurate timing and to avoid some trigger signals being present before other needed signals being ready. Initial tests look hopeful in that I've yet to encounted similar 'garbage' and odd behaviour like I had previously. Currently the system is up and running at 8 MHz clock. I'm yet to do extensive stress-testing, but I shall get there, eventually.
Until that, happy hacking and remember not to spend whole summer inside but rather, to enjoy it while it lasts
This is just to let you guys know, that while the project has indeed slowed down to complete halt since my last post, it is again picking up speed since this project is supposed to be my EE degree Thesis work and my deadlines are approaching So, here is a picture of the routed mainboard for your viewing pleasure to demostrate the point, that we intent to get this done.. eventually
6502.org wrote:
Image no longer available: http://user-200-168.kanetti.fi/janik/stuff/Prometheus_Rev06_Routed_Small.jpg
The IC count is still large but manageable on that boardsize. I will post the schematics here soon for your input and scrutiny so that all remaining bugs and oopses get fixed before I order up the PCBs
Very nice, if I was going to bang round with it, I would have put the two VIAs side by side if you are using the pins 1-20 as the ports, then they can all stay to a single edge and you just have to route the databus and control lines betwene the two instead of routing all the io pins from the inner via. would have also done all chips oriented the same.
tis a sweet monster and you are going to town. party on!
"My biggest dream in life? Building black plywood Habitrails"
I see pads for a 4-pin power connector, of the type used by harddrives on PC Power Supplies. What part number are you using for that power connector? I've been looking all over the place for that style of power connector!
Thanks for the tip Nightmaretony. I had lots of trouble to get the chips placed in a manner, that allowed Eagle's Autorouter to manage to route them all. Each autoroute attempt took some 45mins of CPU time and that with the PCB boardsize of just touch under 28 sqrt.inches. It is big, but it just would not autoroute with smaller PCB size. Too many chips due to feature set getting out of control
kc5tja: Yes, that indeed is a 4pin Molex connector for power that is used in HDDs. I can check from my schematic files for the item number I used if you wish.
That would be great if you could. I searched on Jameco for the part, and it appears OK, but its "original mfr part" number doesn't register anywhere else on the Internet that I could find.
It appears to not be a Molex or AMP connector, because after trying to search on their respective websites, nothing comes up. The source of these connectors are the best-kept secret in the PC and electronics industry, it seems.
It wouldn't surprise me at all if the company that makes these connectors is named "Illuminati Electronics, Inc." or some such. Especially since a connector that is so universal is so damn hard to find!
Janik: it is more a persoal style thing on me. I didnt use an autoroute on mine because ExpressPCB doesnt have that feature. It would seem to make sense to me about the 6522s, thats all.
Still, yours is one hell of a tour de force. Glad you keep us updated on it all.
Now, I simply have to do that for mine, get something up in here and all...
"My biggest dream in life? Building black plywood Habitrails"
I have been very, very unimpressed with the autorouters I've seen. Although they're much faster than doing the job by hand, there's a lot they cannot do at all when conditions aren't apetizing to them. In our company we absolutely do not autoroute anything.
So, here is a picture of the routed mainboard for your viewing pleasure to demostrate the point, that we intent to get this done.. eventually 
