ROM and power LED

[orange]

If there is no ROM in A2000 or its pins are corroded, will the power LED still change from dim to bright on power up?

[Matt_H]

Good question. There a lot of things that can go wrong that prevent the LED from brightening (processor card issues come to mind). A bad ROM will produce a red screen and send the machine into a crash loop, which will cause the LED to flash. Don't know about the absence of a ROM, though.

Trying to diagnose trouble without opening the case? ;-)

[orange]

err.. no, I actually opened the case, removed all cards, then removed floppy, sprayed CPU slot with contact cleaner and corrosion remover to get rid off green stuff on pins (kontakt60+kontaktWL+some lubricant), removed motherboard from case and then DESOLDERED 64 pins and soldered new 64pin socket for 68000, but the bloody LED still wont light up..

oh, someone please help me.. :-(  :-(  :-(

[Doobrey]

Nope, the code that changes the power LED is inside the ROM.

Also, just because the screen hasn't gone red doesn't mean the ROM is OK. It's the same as the LED, the code that puts up the  coloured error screens is in ROM too, so if the ROM is dead or  there's data corruption before the checksum routine then it can't put up the error or flash the LED

[orange]

oh well.. ROM socket is next for desoldering, only 40 pins :roll:

[Doobrey]

What happened, major battery leak or something ?
Have you tried doing a continuity test between the address and datalines of the ROM and CPU sockets?
BTW, you might want to inspect the Gary chip.. it's responsible for decoding the ROM address space and generating its chip enable signal.
Also, one of the CIAs has a pin that enables/disables the rom overlay (ie tells Gary to decode ROM at 0x0000000 to 0x00800000 at bootup)

Edit..
 Found it, it's the CIA at U300. You could try swapping the CIAs around, if you're lucky and it boots it'll save you a lot of soldering  :-)

[Chain]

Oh man! You have iron nerves to desolder 68pin cpu base  :-o

[Noster]

Hi,

I have a simular problem with an A3000T mobo. The LED also doesn't gets bright :-(
I have checked every socketed chip and placed an accelerator card into the cpu slot to bypass the cpu -- without effort.
I have changed the 28.xx MHz oscillator and removed every ram accept of the soldered first MB - no result.

Now I've got another A3000 mobo (also dead) and started desoldering the soldered IC's searching for the one that is dead.
(The board is not dead because of a leaking battery, it has just died. I don't think it is a broken trace.).

Noster

[orange]

OK, quit, I give up..
I've decided to store this dead A2000 that seems to be killed by CHIP RESEATING that everyone reccomended!
It will be in a safe place until someone finds a cure for cancer or similar.. :crazy:

Isnt there some guide to Amiga repair that uses logic probes and stuff and not just "check if chip is overheating, clean contacts, remove dust from floppy" :-?

[patrik]

@orange:

I guess you are looking for the A2000 service manual.


/Patrik

[Chain]

okay okay, i will scan the whole bible for ya it seems psu schematic is not enough for your repairments  :-)
but gimme a little time plz its about 200 pages i guess

[orange]

 :banana:  
great, thanks Chain.  :-)  :-)  :-)

[Chain]

Doues anybody know something about 'amiga tester' (i dont know the full name) ?
its a big box with cables for all ports, some LEDs and swithes/buttons

[orange]

I have seen something on ebay but it was sold for more than 200$ !!!.
There is also some   Advanced Amiga Analyzer . I have software and schematics for it. It should check serial/parallel ports and some custom chips. There is a "dongle" with LEDs that should be connected to parallel port.