Amiga 2000 stopped working... (FIXED)

[stephen_usher]

If /HALT and /RESET are going high at some point and you have a clock then the CPU will attempt to pull /AS low, suggesting that you may have a short to +5V on the /AS line. I don't think that it even looks at the other bus arbitration signals at this point.

Lift the /AS pin on the CPU and look at what it's doing when you bring the CPU out of reset.

P.S. This is what the 68000 does after coming out of reset:

(1) Puts 0 on the address bus.
(2) Lowers /AS and sets R/W to read and sets up a 16 bit memory access.
(3) Waits for DTACK to go low to signal that the memory has put the data on the data bus.
(4) Reads the data on the data bus (16 bits)
(5) Puts /AS high.
(6) increments the address bus by 2 and goes to (2).

It does this four times so as to get the various addresses and state information to start processing, including the address where to start executing code.

[Kepler]

The _AS line does not have a short to VCC.
I lifted the pin (a very unpleasant activity) and checked the voltage in the socket upon reset, it goes almost straight to 5v.

[terriblefire]

The _AS line does not have a short to VCC.
I lifted the pin (a very unpleasant activity) and checked the voltage in the socket upon reset, it goes almost straight to 5v.

That's correct. what i want to see is when it makes its first memory access it will go low.

ROMOE will not go low unless AS is low

A lot of this is time based so we need to know that things are doing at the same time

[stephen_usher]

The _AS line does not have a short to VCC.
I lifted the pin (a very unpleasant activity) and checked the voltage in the socket upon reset, it goes almost straight to 5v.

What does the pin do?

[Kepler]

The pin does the same as the socket, however, I have noticed that when I first power the board after it has been off for some time, there is a distinct step from 0v to 0.4v then to 5v. But if I restart it - or reset it - then I don't see this 0.4v step and it seems to go straight from 0v to 5v.

A side - but related - question. Why are we more concerned about what happens when the board comes out of reset as opposed to when it is powered on from cold?

[stephen_usher]

The reset circuit will (should) hold the CPU in reset until the power and other chips have fully stabilised.

So, why we're asking about reset rather than power-on is that it's taking that confusion about power levels due to the power supply initialising etc. out of the question, trying to refine the problem.

With regards to the signals, if your oscilloscope isn't triggering on the /AS line going low you may miss it as if things are going wrong quickly it may not show on your 'scope's screen when you're looking at it. You'll need the horizontal time base at around 800ns per division as if it's too slow you'll also miss the signal.

[terriblefire]

The pin does the same as the socket, however, I have noticed that when I first power the board after it has been off for some time, there is a distinct step from 0v to 0.4v then to 5v. But if I restart it - or reset it - then I don't see this 0.4v step and it seems to go straight from 0v to 5v.

A side - but related - question. Why are we more concerned about what happens when the board comes out of reset as opposed to when it is powered on from cold?

Because AS will be low early and trigger your scope during the POR (power on reset) phase.

[Kepler]

I think I am making some progress, with the parameters specified earlier, I can see movement on _ROMEN after a reset, but nothing on _AS:

IMG_20240616_160920.jpg

A slowed version:

IMG_20240616_163052.jpg

[stephen_usher]

Well, you can't have ROMEN without /AS but that doesn't matter.

I can see that the CPU is loading the first five values from the ROM, there's a slight pause and then it's doing three further reads and then nothing.

This is good.. the CPU is starting to do things and the ROM is being enabled at address zero.

Is this the Kickstart or DiagROM?

At this point we don't know if the CPU is reading the correct values, but it's a start. :-)

[Kepler]

It's the Diagrom.

Whilst I was waiting for your response, I decided to replace the power switch on my Commodore 1942. Completely unrelated but I felt the needed to start and complete something tangible!

IMG_20240616_165450.jpg

OK, what is next..?