Does this sound like battery leakage damage?

[ArcticSon]

Greetings!

I have a vintage 1989 A2000 on which I replaced a
moderately leaking battery about 4 years ago.  There
was some discoloration of the PCB in the area, but
no corrosion.  I just installaed a new (refurb) internal CD-ROM drive a few days ago, and the "new" battery still looks fine.

So the issue is, when the Amiga is powered on, the RTC stops.  With it off, it keeps great time.  For example,
if I set the date/time with a synch program just before
powering down on a Monday, I can go back a week later and
find the system date and time spot on at my next power-up.
If I have the system on for two hours, then check the time,
it matches the time I powered on.

Not sure if OS or other details are relevant, but I'm running OS 3.9, and a TekMagic 2060 w/RAM on-board.

Is this how this is supposed to work, or could battery leakage harmed a chip or trace in the RTC circuit with
this seemingly odd outcome?  Any chance the likely chip is still available?  I would like to keep the RTC functionality if possible, but know I could synch time at startup if needed (and remove the battery) too.

Thanks for any ideas!  

[Castellen]

The system only reads the RTC once at power up.  From then on the time is kept via a free running counter.  So it sounds like the RTC is working correctly, but not the counter.

The counter is usually clocked by a 50/60Hz signal from the power supply, so this may be issue.  One suggestion is to move the clock source jumper, J300, to source the counter's clock from the system vertical sync pulse.  This locally generated signal is divided down from one of the onboard oscillators, which have an accuracy of +/-20ppm at best.  Therefore you may find the time drifts if the computer is running for any length of time.  The frequency of the AC mains is extremely stable which is why it's ideal for such timing purposes.

Might fix the problem, though some Amigas won't boot without this timing signal being present.  At least the A3000s don't, not sure about the A2000.  Worth a look.

More detail on the A2000 jumpers here.

[ArcticSon]

Thanks for the info and link to the jumper data, Castellan!
I will be buying and installing a USB board soon so will
have the case open this is definitely worth a look.  My oscilloscope days are long past but trying the jumper change I can still give a whirl.

[rkauer]

 The A2000 will not boot if no tick signal present, too.