Time Soldiers arcade PCB repair

Bought as a non-working board for $9.99 – the most obvious problem is that no CPU was present so clearly the game wasn’t going to do much without that!  With a replacement 68000 in hand (again from Ebay for only a few $) the game booted to garbage.

A logic probe on the CPU address and data lines showed zero activity, so the CPU clearly wasn’t doing anything.  Power and ground tested as good.  The reset and halt lines on the CPU were pulsing, so it seemed something on the board was trying to kick the CPU into life.  Testing the CLK pin on the 68000 showed it to be stuck low.  I traced out where this pin went and it connects to the output (Q) of a LS74 next to the oscillator.   The logic probe seemed to show this chip had failed, as a CLK was pulsing on the LS74 input, but the outputs remained stuck (one output feeds back into a LS74 input, so the output should toggle hi/lo every clock in order to form the 68K clock).

Tested off-board – failed – replacement from Dragonninja parts board soldered in – board boots right up!  Quickest TTL level fix I’ve ever done.

At first I thought there was a graphics problem as the title screen was corrupt – though it’s strange all of the other screens were perfect.  Then I remembered I’d seen this a long time ago in MAME..  If you boot the English language ROMs with the Japanese dipswitch enabled the title screen is corrupt (because it expects the Japanese character ROMs).

Unfortunately this board is stuck in Japanese – because the dip-switch isn’t being read by the software.  A previous owner has removed resistor packs for the dip and the jamma inputs, presumably to repair something else!

That shouldn’t be too hard a fix – to be continued.

Pacman arcade PCB repair

Pacman did not appear, and two of the ghosts did not appear.  Additionally the text at the top of the screen (Hi Score) and the bottom (Credit) were corrupt.  A tip on the KLOV forum suggested this was likely to be the VRAM addresser daughterboard.  Nothing looked wrong with a logic probe so I decided to remove the LS157 and LS86 chips as I have a plentiful supply from a Dragonninja parts board.

As it turns out I didn’t even get to the LS86′s as one of the 157′s tested bad when removed, and replacing it solved the problem!

Viper Phase 1 / Seibu SPI arcade pcb repair

Sound would go quiet after a few minutes, and it seemed like different elements of the sound would have incorrect volumes (drum track, music, explosions).  I think that’s because the sound chip outputs multiple channels which are mixed externally before the stereo amp but didn’t look too much into it.

Instead, I just swapped out all the capacitors in the audio section – I think all are 100uF, 25V.  This seemed to fix all the audio problems.

Gauntlet Legends / Midway Vegas battery

The battery is barely mentioned in the manual, but test mode will tell you if it has failed.  Just replace the IC in the picture which is on the board with the jamma connector – it’s in a socket even though it doesn’t look like it is.  Part number is M4T28-BR12SH1.