A500 plus agnus _WE stuck high.

[solidcore]

Firstly let me brief you with the condition of the a500 plus .

The a500 plus has had its leak and was donated to me by indigolemon who told me that the a500 plus looked as if it had been thrown down some stairs .

Secondly the damage of the leak had spread locally about a few inches from battery and with light spotting across a foot in length . Cleaned up and found a trace has been eaten away to the left of battery , fixed that and what was once
A black output become a flashing green.

Anyway green means memory so I pushed on to replace ram and sockets for Paula, Gary and local u* chips below Gary and right of ram .

Hooked up diagrom over serial and saw that the memory was random each boot and didn't reflect the input data .

Using a logic analyser I discovered _WE (RRW) is stuck high , all other signals to ram is fine .

Followed to u35 pin 3 and 17 , still nothing , back to agnus pin 21 still nothing still stuck high ! Tried without u35 same deal!

Replaced agnus 8375 pal for a500+ ,

Replaced u35 . Booted up , same green screen same situation with diagrom.

Any ideas ? Suggestions welcome.

[amiman99]

When my A2000 leaked battery and corroded bunch of traces and pins I cleaned the area with vinegar, I used solder sucker to remove the corroded solder as much as I could within the damage area. I then re-flowed the solder in those places and replaced the 68000 socket.
Booted fine after that.

Maybe you could do that.

[solidcore]

It's all been done except the 68 socket and agnus socket . They'll be on the hit list .

[amiman99]

If you are getting green error code, that probably means 68000 is fine, maybe one of the ram chip is bad, or traces to it. You are close to fix it.

[solidcore]

Originally Posted by amiman99

If you are getting green error code, that probably means 68000 is fine, maybe one of the ram chip is bad, or traces to it. You are close to fix it.

replaced all ram chips, _WE is stuck high from agnus so ram wont fire up anyway . I think gary/agnus had a divorce and signals between the two are buggered, perhaps anything to do with DMA , maybe _blit and other signals.

[solarmon]

@solidcore

Did you ever identify the root cause of this issue?

I'm having the same issue with my A500+ board - /WE signal is held high, and DiagROM is indicating that all the read back memory bits are stuck high.

[admiral]

Originally Posted by solidcore

Any ideas ? Suggestions welcome.

Is it driven high or pulled high? Later is most likely. Don't just check levels, check continuity and resistance.

It would not surprise me if a trace is cut between, say, the CPU and the Agnus, and thus the CPU can't effectively drive it low.

[solarmon]

On my faulty A500+ board, there is an issue where the /WE socket pin is bent and does not make contact with the corresponding pin on the Agnus chip.

In this situation I can test the /WE signal on the Agnus side and on the U35 side. I find that on the Agnus side /WE is driven high, and on the U35 side it is floating (i.e. not connected), as expected.

I can temporarily fix the continuity issue by using a bit of wire to bridge between the socket pin and the Agnus chip pin. But even doing that, it is the same - /WE is held high.

[dalek]

I'd be expecting maybe it's more than just that WE not making contact. I've had success in the past removing Agnus and with a pin or needle gently pulling out at each socket pin so that it makes more tension when reinserted. That and give the Agnus legs and socket pins a good scrub with contact cleaner.

If WE is stuck high but floating at U35 it could mean:
* other Agnus pins not making proper contact
* dead Agnus

Pretty much rules out a short if it's floating at U35 :/

If you can get a hold of another Agnus to test that would help rule it out, otherwise you might get lucky just cleaning or if not replacing the plcc socket.

[solarmon]

@dalek,

If your comments were directed at me, then I think you are misunderstanding slightly what I'm saying.

I'm using the fact that the _WE pin 21 on the Agnus chip is not connecting to pin 21 on the socket to measure whether it is Agnus that is driving/holding /WE high, or if it is something else is.

With the _WE pin 21 on the Agnus chip isolated from pin 21 on the Agnus socket, it is still measuring high - so it seems that it is Agnus that is driving it high.

Even when I do use a piece of solid wire to bridge the gap between Agnus chip pin 21 and Agnus socket pin 21 and continuity is good to pin 17 on U35, the _WE signal is still driven high (by Agnus, as found in the test above).

The Agnus chip has been tested in a working A500+ and that boots up to the Kickstart screen OK. All the other chips have also been tested in the working A500+ board, along with U10/U11/U12/U13 and U34/U35 chips. I've check all the pins/traces for the Agnus pins and they all seem OK.

But I think _WE being held/driven high is more of a symptom rather than the cause of why this A500+ board cannot boot. I just don't know what else is wrong that is causing _WE to be driven/held high. I'll probably end up checking all memory related traces yet again, as I can't find anything else yet that is giving me a clue as to what the problem is.

[dalek]

Originally Posted by solarmon

@dalek,

If your comments were directed at me, then I think you are misunderstanding slightly what I'm saying.
...

Ahh yes sorry that explanation clears it up for me. It's not the Agnus chip then :/



Sounds like something else related is stuck. What does pin 22 PRW look like? Is it stuck low by any chance?

[solarmon]

Using the logic probe, Agnus pin 22 (RW) does seem to be mainly high but it does pulse low at a high frequency. The red high LED is constantly on, but the green low LED is dimmed but pusling, and I can hear the pulsing tone.

Checking the Agnus pin 21 (/WE) again and more closely, I can see that it is similar to the above, but this time the green low LED is much dimmer and not pulsing.

I don't know then whether a logic probe is a good test for this situation, but I don't have an oscilliscope.