DiagROM stuck at "Detecting if we have a working raster"

[Turrican_3]

Hey there,

I'm trying to diagnose an A500 rev5 green screen issue.

Since I currently do not own an EPROM adapter, I was able to run DiagROM by putting its code into the flash of my ACA500plus.

EDIT 2 there's progress, kinda...

Thanks to the combined use of the aforementioned ACA500plus (to load DiagROM and activate the option to have 1M chip via trapdoor expansion) and a spare 8372 Agnus I was able to properly run DiagROM and...

...its memory tests run flawlessly (did not expect this at all?!), plus I can actually see text on screen because DiagROM doesn't stuck at "checking raster" anymore.

But, the ACA500plus menu is still corrupted when it activates, plus booting the system normally still gives green screen + keyboard led flashes.

=========================

OLD TEXT

I then connected the A500 to my PC via serial link and DiagROM apparently runs its early tests flawlessly, then starts sending an endless stream of "Detecting if we have a working raster" messages.

Any suggestions? I am aware that green screen at power on usually means issues with RAM, and I'm currently waiting for a batch of spare ICs because I currently do not own any.

I also wonder if, meanwhile, I might able to isolate the issue by connecting a trapdoor 512K expansion: I've read about a jumper that swaps internal vs external memory, would that mean I could use the aforementioned trapdoor RAM as if it was the built-in 512k CHIP memory, is that correct? Is there anything else that must be done?

=======

EDIT: wonder if this thread might be relevant, even though I do NOT see a "not detected" part of the message

[dalek]

I just had this same issue on an A2000 Rev 6. It was a faulty chipram DRAM. The piggyback method helped identify the faulty chip.

But it could also be bus buffers or Agnus etc.

[Turrican_3]

Thanks, yes I know it could be many different things, but right now I have to wait for the replacement chips to arrive from East Asia, can't do any piggyback test, that's why I was wondering about the swap "trick". :-\

[Turrican_3]

There's progress, kinda...

Thanks to the combined use of the aforementioned ACA500plus (to load DiagROM and activate the option to have 1M chip via trapdoor expansion) and a spare 8372 Agnus I was able to properly run DiagROM and...

...its memory tests run flawlessly (did not expect this at all?!), plus I can actually see text on screen because DiagROM doesn't stuck at "checking raster" anymore.

But, the ACA500plus menu is still corrupted when it activates, plus booting the system normally still gives green screen + keyboard led flashes.

[Turrican_3]

Problem solved.

There was a bad DRAM IC, and quite likely issues with the Agnus socket. Replaced both (well actually all the DRAM chips are brand new, but I was able to identify the previously installed faulty one so I'm keeping the remaining as spares)

Now system seems to be running flawlessly again. :-)