Amiga 600 black screen, no bus activity

[Eduardo L M]

Well, I did some more testing.

- The bus transceivers U21 and U22 are getting a constant high level on pin 19, so they are kept "open", or not communicating. The high level starts as soon as I turn on the Amiga

- /ROMEN is high all the time, except for a very brief moment, I think when /RST goes high. Then /ROMEN goes high and is kept there forever. This is with diagrom. With KS 2.x, /ROMEN is kept low for a longer time, exactly the time where I see bus activity.

- I desoldered both RAM chips, the symptoms didn't change.

It seems that, with a regular KS, the Amiga starts booting fine, but suddenly the ROM is cut off the bus, and nothing else happens. Should I start worrying about a bad Gayle? Is there a way to test if Gayle is the culprit here?

Another theory: the transceivers should connect the chipset (like Agnus) to the 68000, right? What if I'm facing an issue where the initial ROM runs some code, but when the time comes to command some screen activity, the chipset (Agnus, Denise) don't respond, Gayle never turn transceivers on and the computer looks like it crashed? Does that make any sense?

Many thanks for any clues about this mistery!

Eduardo

[supaduper]

Just take off U17 ram chip by leaving only U16 on the A600 will still boot with 1/2mb, and yes the transceivers also can go bad I have had the odd one go bad like u22, but you are now just getting into swapping chips out and Gayle also can go flakey but they are so expensive nowadays, good luck

[stephen_usher]

Well, I did some more testing.

- The bus transceivers U21 and U22 are getting a constant high level on pin 19, so they are kept "open", or not communicating. The high level starts as soon as I turn on the Amiga

- /ROMEN is high all the time, except for a very brief moment, I think when /RST goes high. Then /ROMEN goes high and is kept there forever. This is with diagrom. With KS 2.x, /ROMEN is kept low for a longer time, exactly the time where I see bus activity.

- I desoldered both RAM chips, the symptoms didn't change.

It seems that, with a regular KS, the Amiga starts booting fine, but suddenly the ROM is cut off the bus, and nothing else happens. Should I start worrying about a bad Gayle? Is there a way to test if Gayle is the culprit here?

Another theory: the transceivers should connect the chipset (like Agnus) to the 68000, right? What if I'm facing an issue where the initial ROM runs some code, but when the time comes to command some screen activity, the chipset (Agnus, Denise) don't respond, Gayle never turn transceivers on and the computer looks like it crashed? Does that make any sense?

Many thanks for any clues about this mistery!

Eduardo

Well, it sounds as if the DiagROM is trying to set up the serial port and is then waiting for something to happen, which never does.

This could mean that U8, one of the VIAs is faulty. That's on the same side of the transceivers as the ROM so they won't have the enable PIN go low. Opening the serial port is one of the very first things that the DiagROM does so that would explain why it stops so quickly. Nothing to do with the RAM and the RAM hasn't been accessed yet.

[Eduardo L M]

Thank you guys once more!

Stephen, this is a great insight! Makes complete sense, and explains very well why diagrom hangs so early. I'll try swapping U8 with the other CIA on the same board. If I remember well, U8 is needed for boot, but the other one is not.

And Supaduper, thanks for the suggestion on checking bus transceivers, I'll remove and test one by one if the CIA swap fails.

Thanks guys! I'll get back here to report my findings.

Eduardo

[Eduardo L M]

I finally had some time to make some more tests. I removed both CIAs, and soldered the CIA that was at U8 position, into U7 position. As far as I know, U7 is essential for boot, but U8 not.

However, nothing changed, the machine showed the same symptoms, with both ROMs.

I then dessoldered the CIA I just soldered at I7. I left both CIAs out, and powered the Amiga on again. Initially, I had the same symptoms, but shouldn't I get a green screen on a healthy Amiga missing the "main" CIA, or U7 in A600? But no, still got a black screen.

However, after some power cycles, I had something different, with KS 2.x ROM: got continuous processor activity at a random data line! And I had some strange activity in composite video output. Since I'm using a LCD monitor, it showed nothing. I'll try with an analog composite monitor to see if it shows anything.

But that was intermittent, it repeated two or three times again after dozens of power cycles.

I think this fix, if possible, will only happen if I test each involved component and PCB track, one by one, including passives. I will do that slowly as time permits, and I'll report back here.

Many thanks!

Eduardo

[Eduardo L M]

I have an update: after checking lots of PCB traces and finding nothing wrong, I decided to measure resistence at various PCB spots, comparing to a working board, of same revision. And I noticed resistence between 5v power rail and ground is about 60 ohms, and a good board shows 120 ohms. So, the defective board shows half the resistence. I don't know exactly what may be causing it, but I'll try to locate the culprit. Perhaps I have a partial short somewhere, or an altered capacitor. Let's see!

Eduardo

[CCRider]

(...)
Yes I have just repaired an A600 where someone had broke a trace under the Rom socket when using a screwdriver to remove it so get a magnifier and have a good look

...and indeed it was! I botched the trace between pin 54 of 68K and pin 52 of Gary.

Now solved. Thank you so much!