Board #1 – Friend asked me to look at their board

This is one of the Atari repro boards that was built up.. This is my second repro board repair, my first was this Star Wars set that needed a LOT of work to get running. This particular board ran, but the text during the game had some jitters and some vectors were jumpy.

Fortunately – the board is nearly identical to Gravitar. The PCB is the same, the controls are different. I added 4 wires to my harness adapter and made a ‘Y’ adapter to connect to my dual joystick controller.

Connecting to the board, I did RAM and ROM checks. RAM was good, the ROMs @KL/1 and @M1 read bad. I removed them and determined each had a single bit flip. They are 2532’s and those are sometimes tricky to write. I erased and reimaged all of the EPROMS. One of them would not reimage properly – it got replaced.

At this point the board seemed to run correctly in game mode but in TEST mode it would sometimes get stuck watchdogging however mostly it showed bad Pokey’s and often a bad EAROM. I spent some time here looking at the power on reset circuit. It all seemed to be correct.. Rather annoying. I had done some work for a large repair house and got 2 Black Widow boards in trade (great deal for me) I had been using my Gravitar as a reference board, but I decided I needed a BW as a reference to verify the watchdog reset frequency to see if it was off for some reason.. Black Widow #2 below I was able to get repaired quickly..

During all of this testing, I determined that if I powered up in game mode – it ran correctly. If I powered up in test mode, got errors and switched to game mode, the sound would be messed up.

After a fair amount of testing and cross checking – it really seemed like a software problem. I finally did a quick search on KLOV and found out – this is a true software bug Atari missed.. Can’t start Black Widow or Gravitar in test mode.. They both do not initialize the test process correctly from power up – only when the game it put into test from game mode.. Worst part, I read the thread back when it was written and forgot about it. Lesson learned 🙂

Board works!

Board #2 – My board – quick repair and reference board

Mine was missing the AVG chip, EAROM and the CPU. When I first powered up RAM and ROM checked good, but the vector generator was dead..

The 12MHz test point had been ripped up and broke the clock into the State Machine. After fixing that..

Board came to life, no errors except EAROM which I didn’t install. More about this on Board 1 above.. Note: I had not made my adapter for my dual joystick controller just yet. The single stick had enough buttons to do basic things.

Board works!

Board #3 – Board in for repair

Board came in not working.

It has been in the shop a fair amount over the years. Testing revealed a bad RAM and oddly a torn up trace for the CPU test connector.. Doesn’t effect anything – except when you are trying totest the board! Owner asked that I perform all standard board maintenance and burn in testing. Cabinet is new to him with all matching serial numbers. Completed 30+ hours of testing over 7 days.

Board works!

Board #4 – Board in for repair

This one came in from an arcade and I was told it wasn’t playing correctly. Not much detail.

Testing showed that one of the Pokey’s was bad.. Replaced it and cleaned chip legs while I had it.

Board works!