|
|
The game was running fine for the first few days but by Sunday the game had developed a graphics problem and wouldn't play correctly. With no spare PCB to put in, it was left as is until closing.
|
Since the game wasn't playing correctly I suspected a RAM issue and setup the Arduino ICT to test it out. "RAM Check All" reported failure "E: 3H d000 00 0f" and a RAM Check on the other nibble of that bank reported "E: 3F d000 90 f0". RAMs 3H/3F seemed not properly accessible. Given the effect on the display during the RAM test it appeared as if the writes were not working properly. Using a scope found IC 3H (2114) pin 10 (~WE) was active during writes. IC 5J (LS245) pin 19 (~VRAMCS) was also active on writes but the signal was very poor, almost noise when active. IC 2R (LS08) output pin 3 per both the Midway and SWR schematics didn't look like the right signal on this PCB. Using a meter to trace the wiring found the right signal at IC 2N (LS08) output pin 8, poor & noisy. Input pin 9 was high pulsing low and input pin 10 was high pulsing to a poor low. On "RAM Write All Lo" ~1V low pulses were seen on pins 9 & 10 when inactive and full low pulses on active. "RAM Read All" was the same except 2V low on pin 10 when active. Tracing back to IC 8M (LS138) awkwardly located under the daugher PCB found on "RAM Read All" output pin 13 around ~1V when low. All the other outputs seemed OK. IC 8N (LS138) with "RAM Write All Lo" similarly found ~1V then ~2V low. I suspected that IC 2N (LS08) was internally shorting signals in some way. Replacing IC 2N fixed the game.
|
The shorting LS08 was the only issue with the game and after clearing the battery RAM the game ran fine on the bench for several hours. The battery voltage was still good - 2.96V.