C070523-001 not just layout silly

[Robson]

Not much done today.
Forgot to mention that I'm still running with the alternative ram solution, but all lines seem to be connected (including data between cpu and mmu). I'll put the onboard ram back with sockets and continue testing like that.

The pattern has changed slightly it's running this way on the screen and on serial.

IMG_7688 (1).JPG

IMG_7687.JPG

[Robson]

One step closer.

All drams individually tested, socket connections checked (one repaired) and came back to life but not working without errors yet. After warming up the first thing to go is floppy access (with known good drives and hxc too), then I left at the desktop and after 1 or 2 minutes the screen goes bonkers too.
So there is something still poisoning the bus.

IMG_7706.JPG

IMG_7705.JPG

IMG_7707.JPG

[exxos]

If the screen slowly gets worse, it might be a problem with the ras,cas lines.

[Robson]

karate.jpg

I had some time to get this board back on the my torture desk again and now it's all back to life: needed a new MMU and few more checks.
So in conclusion: corrosion, bad ICs, tracks, somebody else's previous "work" (all HI tos sockets were soldered upside down) all worked together.

The only thing that I haven't yet worked out is the issue that if I leave it off for hours (5-6) and power on, the display is sometimes (not every time) lazy, meaning comes in after a good 2-3 minutes. I have output through serial immediately, when starting up cold. After it's back and warm....all works flawlessly. The issue is the same on TV and Monitor out (rgb scart) , they come back together at the same time.
I'm thinking a capacitive element somewhere but haven't yet started to troubleshoot this, just posted the 99% success. But if you've seen such thing before I'd appreciate the input. Cheers.

[exxos]

By the looks of the PSU its not been recapped and that's always a good place to start.