Amiga 4000D with black screen

[turrican9]

So I am looking at a dead A4000D.

Owner said it worked the last time he tried it back in 2004.

Comes up with a video signal with blank screen. No disk or floppy activity. HDs will power up. As they get power from the PSU of course. Floppy will not seek or tick. The power LED will not dim up after the first second. Indicating that the 030 CPU card is not starting up. CPU card looks perfect, as do the connector. I have cleaned them and used contact spray, even though it looked perfect. I also made sure it was firmly seated. I do not have a keyboard for this A4000. I know there is a trick using the caps lock on the keyboard to see if the CPU card is triggering. But the indication I get with the power LED is confirming the same thing?

The battery had leaked and U177 was shot. I removed this chip as it is not neccessary for bootup, but in theory it can prevent startup if badly damaged. Many of the legs on the battery side of it just fell off as I desoldered it. The pads are still fine. U891 was also badly corroded on the lower side. As far as I understand this one only has to do with fast RAM, and I'm trying with just the chip RAM for now. I also tried without the chip RAM, and this should give a green screen if the CPU was activating at boot up. I just get the same blank screen with a video signal.

As for U975 and U976 I have probed all the connections to them and the connections looks fine. Only U976 had quite a bit of corrosion on it's legs. As far as I understand, only U975 can prevent boot out of these two. It has a data signal to trigger Lisa as far as I understand. I do not have extra chips to swap with at the moment.

U178 looks fine and does not overheat. I've heard this one could also prevent boot if it's badly damaged.

Other than that the only traces I could find that was broken was two of the traces going from U177 to the resistor pack closest to the battery. Which should not prevent boot. Of course I have removed the battery and done some cleanup.

I do not have one of those signal probes to look at hi or low data signals.

I do not have extra kickstarts to try. I know the A1200 kickstarts are not compatible due to small differences, but would you at least get a boot screen if one tried A1200 kickstarts in a working A4000D?

Also, is there a chance U891 could prevent a boot, even without fast RAM, if it's badly damaged inside?

Oh, and voltages checks out fine. Power supply looks fine. No indication of anything shorting to ground. Board is not recapped yet as the caps should not prevent a bootup. And they mostly look fine. C912 had some corrosion on the legs due to the battery leakage, but it's not making a short.

I do not have another A4000 to try the CPU card. CPU card looks in perfect condition and has very shiny solder joints on the caps.

Lisa is luke warm, I think Alice was luke warm, Denise luke warm, CIAs quite cold and buster stone cold. I think I read somewhere that buster should be quite cold? I know the CIAs runs quite cold usually. Not found any unusually hot chips yet. Oh, and I also reflowed the kickstart sockets. Sockets looks fine and I reseated the chips.

[turrican9]

I found the problem! The grounding pin on U891 was not making proper contact with the pad because of the corrosion. I was probing and had huge resistance from pin to groundplane. The computer booted up to kickstart screen when I put preassure (Or more likely making contact between the pad and the pin using the probe as the conductor) on the pin with the probe as I was booting.

I read somewhere that you don't even need U891 to boot as long as you only have the chip RAM installed and not fast RAM. This is false in this case. At least when the ground pin was "floating" at the same time the other pins on U891 made contact with the pads. It would not boot with only chip RAM until I made the ground pin connection. If I let go of the probe the machine will crash.