First of all, a premise: All the repairs i do are for hobby purpose ONLY. It is NOT a job, and i fix stuff for a selected people group that i consider needs my help. All other people please ask somewhere else to get your items repaired back.

A guy contacted me 3+ months ago (April 2019) asking help to fix his Amiga 4000D (Yellow screen at boot for a couple of seconds, and then black). I agreeded to help him.

This Amiga 4000 was tested with A3640 and a Phase5 Cyberstorm MK3 CPU cards, with which the A4000 does not work with the same issue descripted above.

I immediately noticed a fairly important damage caused by battery and caps acid leakage (especially in the audio area, see photo).

After cleaned everything, i rebuild some interrupted pcb tracks on the motherboard (audio / ram / status led section) and replaced capacitors on both motherboard and 3640 (the fault persisted, of course).

Next, i used the Chucky DIAGROM: the DIAGROM is useful to understand what might be that doesn’t work, and the output via RS232 is also very useful if nothing is displayed on the screen.

DIAGROM reported some issue with all IRQs and CIA Timings, but everything else was working (video output, images, sound, mouse, joystick ..ect..).

The IRQ problem is well known in the Amiga 4000 repair circle, because it could be caused by IRQ _CIPL or _IPL interrupted lines, the pcb tracks start from PAULA and goes to the GAL U701, and from the GAL to the CPU connector.

Well, I checked all the IRQ lines, and they were good. I checked again, everything was ok (the checks must be done from PAULA to the GAL and from the GAL to the CPU connector, amiga pcb solder side and CPU card pcb solder side) at this point i thought that GAL U701 was dead, so i programmed a new GAL and replaced it. Unfortunately nothing changed.

Thinking about what it would cause the GAL not working properly, i pointed the finger to the GAL CLOCK, exactly to the pin 1 of the GAL U701.

The Clock was there but something was not clear to me: the signal i see on the oscilloscope was corrupt, fixed at 5V. At this point i said to myself, what is that supplies clock signal to the GAL? Clock on the pin 1 of GAL U701 is supplied by the IC 74F86 U711, pin 6, via the resistor R702 (obviously resistor was good).

I called my friend Andrea (Andry) and asking him to check on his A4000 the value of that pin (6) of the IC U711 (i could have done that myself on my Amiga 4000, but disassemble it’s a nightmare). My friend does not have an oscilloscope, so i asked him to check the power using a digital or analogue DC tester. On his A4000 the value of U711 pin 6 was was completely different than my value.

Bingo, it is! i run to the local electronics store to buy a SMD 74F86 IC, but obviously they didn’t have it.. so, i purchased it on Ebay, replaced it, and yes, it was dead.

It was not the only faulty component, because replacing the new GAL U701 with the original GAL U701 did not work. So, the original GAL was dead, too.

New IC U711, new GAL U701 and the A4000 is back to life.

Other minor fixes i have been done on this Amiga 4000 are: useless replacement of an 8520 (U300), installation of cooling fan on the CPU 68060 – 50MHZ on Cyberstorm MK3, rebuild of a broken resistor damaged by the caps leak.

No more other to say, except that i am happy i succeeded to repaired this Amiga 4000 for a person that needed my help.

Gallery of the repair:

1234►

Video: