A4000 will not boot. Diagrom output attached

[Firestone]

Hi everyone, and happy new Amiga year

Today I had some time to give a few old a4000 boards a try.
Tonight I’m facing a strange problem where Diagrom starts to output data over the serial connection to my a1200, and then it just stops. Not able to control anything past this screen.

So far I have tested that these components are known good in another a4000:
Chipram module
Fast ram module(s)
CPU-board (standard 030 board)
PSU

Right now nothing extra is connected. No harddrive, floppy, daughterboard.

Diagrom outputs some strange characters between some stages as you can see in this video:

[ Show youtube player ]

Any clues?

[Firestone]

New clues revealed today:
- Swapped with an A3640-board (J100 and J104 set to external)
Issue still persists.

So i tried to put the diagrom into a working A4000 to see if it prints garbage between the lines on the serialoutput. It does not. Text seems fine, and the remove ansi-menu is displayed fine and functions as intented.


I then noticed that on the bad board, Diagrom seems to detect that the right mouse button is pressed all the time:

Checking status of mousebuttons for different startups:
P1RMB

This happens all the time on the faulty board, and the system will crash at random places during boot.

I have the best chance getting to the DiagROM Menu if I only insert simms in the fastram slots (diagrom successfully detects if there is no chipram installed).
But I'm never able to do anything in the menu. I can see it on the remote machine, but I guess the system must have crashed after it's displayed.
On the statusbar below the menu, it lists the CPU as a 68020 on the faulty board. 68030 on the working one.

Could it be a CIA-problem, or maybe faulty Paula since Paula is responsible for the buttons, and not the axis-handling (which I believe is handeled by Lisa?)

So as a conclusion so far:
- Right mousebutton is always (?) down, or at least detected as down during power on
- Garbage output from the serialport between the lines

= Paula......?

[cpiac64]

I suppose the problem is around the battery
verify the connection and change both 74hct166

[Firestone]

Yeah you might be right cpiac64.... I'll take a look at them.
Will post back with the results. Thanks for the tip!

[DigitalKeeper]

possible -> mdat is low coz of battery leakage -> Lisa prevent to boot

[Firestone]

Okay, got some time to get around checking this.
I have replaced U177, U976 and U975.

Seems like there is no difference as far as I can tell.
The computer still thinks the RMB is held down during boot, and there is some garbage printed along with the serial output.

Booting with only fastram gives me the Digarom diagnostic menu on the remote computer, but the system seems to have crashed, or does not receive commands from the remote computer.

Booting with chipram only does not give me the diagnostic menu. It stops a little earlier.

MDAT on U975 is high.

[DigitalKeeper]

Do you have checked ALL the Traces and the VIA's?

Have a look on Top and Bottom at the R 4,7K Resistors....seems like a cut somewere or a short....



Often the Battery Acid get some shorts and - the capacitors are leaking and do a short between the traces and contacts.

Make sure, that there is no leakage and no corrosion by the 4,7k resistors on top and bottom.

This is a full reworked Board by me, with similar Problems.....

Top:



Bottom:

[Firestone]

Got some spare time tonight checking for broken traces. I have checked every trace and every via for U975, U976 and U177. All traces are found ok and connected to the components listed in amigapcb.org for each pin.

Checked all the resistors on the left side of these components, and checked the 5V with a scope. Stable 4,9x V on every one.

I tried a normal 3.0 ROM tonight. This gives me a dark red screen at first, and when I reset the computer with the keyboard it goes black and nothing more happens.

I have to mention that the board have a Super Buster rev 7 on board. I don't know if this matters? (just because it was the only one I had spare...)

Tonight I attached some pictures from various areas of the board if it's of any interest.

Thanks for trying to help btw!

[cpiac64]

in the photo, chipram socket are empty

[Firestone]

Originally Posted by cpiac64

in the photo, chipram socket are empty

Yes, I know I took the pictures before I moved it into bank 0 and turned on.
Got a green flashing screen with the simm in the wrong socket.

[DigitalKeeper]

Buster 7 -> please get a Buster 11, and solder directly on Board! Coz - the Buster Socket make Problem's

Buster 7 = Buggy
Buster 9 = DMA Problem's, i.e. with Cybervision 64 or similar Card's
Buster 11 = last Version, DMA Bug fixed

(Spoiler -> Dave and some other User's are working on a New Buster Version! NOW!)

Remember -> ChipRAM with Double-Sided Simm's Only!!!

Have a look at this Traces - there are often broken, coz of leakage from Capacitors and Acid from Akku-Cell:

[cpiac64]

i am not sure, but i suppose the board start without Buster, only for test

buster 11 need version 2 of gal u213

[hese]

Originally Posted by DigitalKeeper

Remember -> ChipRAM with Double-Sided Simm's Only!!!

Not true. Single-sided 4MB SIMM works perfectly fine as 2MB Chip RAM when J213 is set to 8M and J214 to 8M. Actually sideness does not matter, what matters is how the memory is organized on the SIMM.