Here's a head-scratcher - I have a Battlezone AVG (main) board that, when
hooked up to the Fluke 9010A troubleshooter, exhibits an obscure watchdog
related fault. When running the RAM Short (or RAM Long) test sometimes a
random RAM address will be hit that the Fluke decides is bad, but on
putting that address on Loop it then rapidly toggles between OK and the
error. If I then ground the Watchdog Disable test point the problem
immediately ceases and the RAM location is displayed as OK. RAM areas 0-3FF
and 2000-2FFF all pass fine 100% of the time when the Watchdog Disable is
grounded.
There's a similar fault when using the ROM test to check for a signature -
if the Watchdog is enabled then the sig for a particular ROM is always
variable and always wrong. However, if I once again ground the Watchdog
Disable test point then all the sigs are perfect.
The Watchdog circuit seems fine, but I did try swapping out the LS393 to no
avail.
Why on earth should the Watchdog circuit affect the RAM and ROM checks?
Presumably it's slapping something bad onto the address or data bus at
random intervals?
There's also a fault somewhere in the vector state machine, ie by disabling
that (clip pin 6 of the LS244 at R11) the game will run in self test mode
(blind of course - tones are generated as they should when pressing the
relevant buttons, toggling the stick inputs, etc). This works *even though
the Watchdog circuit is enabled*). That said, the board won't play a game
blind, ie with the vector machine disabled. I'm still looking into that but
mention it in case it's relevant.
All relevant LS244's and LS245's near the CPU have been replaced, as have
some of the 2114's.
Any thoughts please?
Thanks,
Phil
List of worldwide arcade repairers: www.arcaderelics.co.uk
---------------------------------------------------------------------------
** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org
** Please direct other questions, comments, or problems to chris@westnet.com
Received on Tue Aug 29 08:54:37 2006
This archive was generated by hypermail 2.1.8 : Tue Aug 29 2006 - 14:50:01 EDT