Galaga Repair Logs
Repairer: channelmaniac
Symptom: RAM 2L Error
This board was frustrating. RAM 2L bad. Replaced it. RAM 2H bad. Replaced it. RAM 2L bad... RAM 3L bad... RAM 3H bad...
Cleaned the pins on the custom IC at 1H on the video board to fix the error.
Symptom: No characters on screen
The player and enemies were both missing. Cleaned the pins on the custom chips on the video board.
Symptom: Cannot control Player 1. Enemies go in circles then disappear
This was odd. Game self tested fine and coined up. Player 1 controls were non responsive. Couldn't fire and couldn't move. When enemy ships came down the last one tried to peel away and would go in circles for about 10 seconds then disappear. Once it disappeared the next wave came and the last one repeated the circling and disappearing act.
Checked the ROMs on the CPU board. Found EPROM 3300 was for revision 2 of the Galaga software while the rest of the ROMs were for revision 1. Replaced the EPROM with one holding the correct code to fix the game.
Symptom: Random lockups
Board would lock up even during self test. Could tap on the board and cause it to lock up.
Replaced the 3 Z80 CPU sockets. The old TI sockets were making intermittent contact with the chips.
Symptom: Locks up after the initialization explosion sounds
The ROMs on this board were Namco instead of Midway with the exception of the ROM at 3M. Replaced it with one programmed with the Namco version of code.
Symptom: Faint digital artifacts instead of music.
Resistor pack RM2 was broken. Replaced resistor pack and artifacts were gone. Music was still missing.
The 74LS273 chip at 3B had inputs but no outputs. The clear line was being held high, which is correct. The clock line was cycling during explosions but when high when music was supposed to play. This line is controlled by the PROM at 5C.
Replaced the PROM and the music played as it should.
Read the PROM in the EPROM programmer (7611 which crossed to an 82S129) and it read good. The code didn't change when reading multiple times. Ran the code through ROMIDENT - It's a PAC MAN PROM chip for location 3M. Someone stuffed the wrong PROM on the board.
Model: Galaga Symptom: Passes RAM tests then instead of explosion sound & crosshatch it dies
Replaced 3 CPU sockets. Board would then get to the point where it says RAM OK then the screen slowly paints white then the board resets.
Checked the EPROMs. Found 2 were for Galaga set 2, 2 were for the fast shoot hack, the ROM at 3J was bad, and the ROM at 3E was for Bosconian!
Replaced the ROMs and the board came right up.
Symptom: No sounds except explosions
Sound PROM at 5C was incorrect. It had a PROM for 1D in the socket.
Symptom: Reboots constantly
This is different than the bad CPU socket problem in that it wasn't stuck at RAM OK but rather as soon as the crosshatch then the text came on the screen it rebooted. The Reset line on CPU #2 and 3 came active for a VERY short period of time then the game reset.
Swapped out the CPU sockets. Subbed out the custom chips. Nothing. On a whim I turned off each DIP switch and found by playing with them that when DIP switch #3 at 6J was on it would cause the system to lock up when running or to constantly reset when booting.
Replaced the 74LS151 at 4K and the 74LS368 at 5A. Problem still there. Checked the EPROMs again. They were all checking good on the programmer. Swapped out the first 2 program EPROMs. Nothing. On a whim I burned a new set of Williams EPROMs (these were Namco) and swapped out all the EPROM.
Problem fixed.
Tip: Replacing PROMs
The PROMs used on the board are Harris 7611. These are 256 x 4 bit bipolar PROMs.
Other PROMs can be used to replace this one, including:
- 5623
- 6301
- 7052
- 74S287
- 82S129
- 93427
- AM27S21
- F93427PC
- 63S141
I have tubes and tubes of Fairchild's 92446 - a 512 x 4 bit bipolar PROM. If you take the code for the 256 x 4 bit PROM and duplicate in the upper half of the PROM these can successfully be substituted for the smaller PROM.
Symptom: flickering garbage on screen.
Board had been worked on before by a ham fisted technician. Removed their soldering work at 2K, 2L and 2M. Cleaned up the bridged soldering mess under the socket at 2L, installed new sockets, and installed new chips.
Board now had a solid white screen with garbage characters on the top row. During the attract play when the boss ship captures the player's ship that part is visible on the screen under the garbage but the ray is mixed up.
Pin 19 on 2L was stuck high. The input, on pin 18, was missing because of a damaged platethru. Ran a wire jumper and played the game.
Symptom: Multiple graphics problems
Another "attempted repair" by a ham fisted technician. This board had flickering video, incorrect text, bad colors, double characters, and a lot of vertical lines through the screen.
Found a bad trace connecting Pin 1 of 4A. 4A drove address line A9 for 4 of the video RAM chips. Fixed the trace to cure the vertical lines problem.
Found a bad trace connecting pin 1 of 3D. Patched the trace to fix the double characters.
Tapped the board and found that if the EPROM at 4L was tapped the screen flickered and the text changed. Replaced the bad socket to fix the flickering and incorrect text problem.
Repaired a bad trace on the video output resistor ladder to fix the color problem.
Symptom: Multiple video problems
Another board from a ham fisted repair attempt.
Board had no video. Cleaned the pins on the custom chips at 00xx, 7xxx, and 02xx. Repaired bad pins on those chips and piggybacked an extra socket on the 00xx and 02xx chips to protect the fragile pins.
Board now has video but the video has blue lines through it. Replaced a broken SIP resistor pack at RM5 to fix the blue lines.
Board now has video but the layout is wrong. Ships are doubled, text and player ships stack vertically up the left side of the screen and the crosshatch has many tiny squares down the left side. Patched a trace between pin 11 of the 00xx and pin 21 of the 07xx chips.
Board now has video but colors are wrong. Replaced all the cut and resoldered resistors in the output ladder. Patched 3 damaged traces. Now the video looks great and I set a new personal high score.
Symptom: locks up
2 boards with the same symptoms... Passes RAM/ROM tests, makes the explosion sounds, then locks up randomly. When it freezes up the characters on the screen may or may not disappear. Most of the time the screen will then flip upside down. It might play for a minute or 3 then it will start doing strange things.
Symptoms:
- Starfield starts flickering fast
- Enemies wings move a LOT faster
- Enemies MOVE a lot faster
Tested the ROMs with the EPROM programmer and by substitution. Cleaned the custom chip legs and swapped out the 7xx, 6xx, and 51xx chips. Replaced the ROM sockets and CPU sockets on one of the board sets. The logic probe showed the CPU at 4E was halting. I noticed that when shorting the IRQ pin to D0 (15 to 16) on that CPU the board would run (strangely) and not lock up. Replaced the CPU on each board and played a couple of games.
Symptom: Dead. Stuck in reboot loop.
Removed a 68B09 CPU from socket 4H and installed the proper 51xx custom Namco chip. Replaced the 3 Z80 CPU sockets and the 6 ROM sockets. Board would still either be stuck in a reboot loop or act dead. When it acted dead the CPU was stuck in a reset loop. If the board was slightly flexed it would try to boot.
Replaced a bad socket on the 7xx chip. The VBLANK* signal was not getting to IC6A, pin 1, to keep the watchdog signal from resetting the CPU. Reset the DIP switches to factory settings and tested.
The board was now missing all text on the screen except the very top row, which was green instead of red. When playing game the flags showing the level were the wrong color and missing lines and the line showing the credits was missing.
This is handled by the EPROM at location 4L. It is fed by a 74LS273 at location 2L and 74LS86 at location 2N. These were good, the EPROM tested good by a programmer and by substitution. All the signals were correct on the custom chip at location 4H. The IC at 4H outputs the control signals for those images out pins 11 and 12 to pins 5 and 6 on IC 2N, a PROM. Checked the PROM and found that pin 9 was folded under the chip. This is the D3 output. Straightened the pin and tested the game.
Tip: Stuck in reboot
If the DIP switch for Video Freeze (6J, switch 5) is left on and the game restarted it will be stuck in a reboot loop until the switch is turned OFF. The effect is different than when you have bad CPU or ROM sockets on the CPU board where it shows RAM OK then reboots. In this case, it is showing ROM OK, RAM OK, and the rest of the normal text before suddenly rebooting.
If after replacing those CPU and ROM sockets (which should be done on ANY Galaga game that still have factory sockets) the board is still rebooting then check that switch. You'll save yourself headaches.
Symptom: Multiple problems. Board had been worked on before.
Board would boot but had no explosion sounds and would not accept inputs properly. Sprite graphics were missing every other vertical line and there were vertical blue noise 'tails' on the bottom of the screen.
CPU board: Replaced missing 00xx and 06xx ICs. They were in the box with the board but were not installed and had missing pins. Repaired broken pin 28 on custom IC 55xx to fix the input and sound problem.
Video board: The sprite problem would come & go with flexing the PC board. Replaced a bad socket and trace on pin 3 of 1C. Resoldered the SRAM socket on 2148 SRAM at 6B. The ground pin was not soldered properly. Resistory pack RM7 was bad and causing the blue vertical streaks.
Sympom: Dead. No video output.
Board was previously worked on and had problems on both the CPU board and Video board.
CPU board: Fixed one badly jumpered trace. Replaced a bad EPROM at 3K, the D1 and D2 lines were shorted internally. Replaced bad sockets on the CPU at 4M and the Namco custom chips at 4D, 2L, 2J, 2H, and 2E.
Video board: Cleaned board. The 6MHz clock signal was not getting to pin 1 on the 02XX IC at location 4H. Replaced bad sockets on Namco custom chips at 1N, 1L, 1H, and 4H. Fixed a broken leg on the 07XX chip at 1N and cleaned the legs on the custom ICs at 1N, 1L, 1H, and 4H. Cleaned the legs on the SRAM IC at 1K to finish the repair.
Symtpom: No sound and graphics problems
Installed a renew kit (new 24/28/40/42 pin sockets... resistor packs did not need replacing), cleaned all chip legs, and reinstalled the chips into their new sockets. Tested game. Game now works, with sound, but has a strange graphic anomaly: The player and enemy ships are all tilted to 45 degrees, player explosions are divided into four pieces, and enemy explosions are numbers 20 over 31. If the enemies are tilted to the left the numbers are upside down and if they are tilted to the right the numbers are right side up.
Checked the Graphics ROMs on the video board. Pins 1 and 2 were stuck low with very rapid racing pulses. Checked the 74LS153 at 1B and found it had normal inputs but bad outputs. Replaced the IC and tested the game.
Symptom: Missing some blue on motion graphics and random blue streaks on board
This was a very tough one to track down. The board worked fine until an extended burn-in was done. Swapped out the motion RAM (8 x 2147), the resistor packs that hold the data lines high, 1 PAL, and a couple other 74xx chips. Replaced the 74LS298 at location 5B to fix the missing blue. This chip multiplexes the 8 data output lines from the motion RAM into 4 for mixing with the text graphics and later with the starfield.
The random blue streaks was just as tough to track down. Checked the 74LS365 chips at 5A and 6A, swapped out 2 of the replacement Video RAM chips, and checked the 74LS157 at 4A. (Yes, 4A. It says 5A on the schematic, but there's a 74LS365 at that board location. The schematic has an error.) Traced the problem back to pin 10 on the 74LS20 at location 3A which feeds the LS157. Pin 10 is the 6MHz* clock. Swapped the 07xx IC to no change. Replaced the 74LS368 IC at location 3N to finish the repair.
Symptoms: Graphics corruption
Replaced every tall, flat style resistor pack and tested. These packs are sources of a lot of problems on Galaga boards.
Symptom: Dead. Garbage on screen.
Installed all new 24/28/40/42 pin sockets. Cleaned the custom chips. Board would run but would die when hot. Replaced CPU at 4M to finish the repair.
Symptom: No coin inputs and faint garbage lines on screen during self test
Replaced bad resistor pack at location RM10 to fix coin inputs. Replaced EPROM sockets and 42 pin socket on the CPU board. Replaced all 24/28 pin sockets on the video board. Tested. Replaced bad PROM at location 2N to fix the faint line issue and retested board.
Symptom: RAM 0L error
Replaced all resistor packs and 24/28 pin sockets on the video board. Cleaned legs on the custom chips. Board then fired up with a 2L error and random garbage on screen. Garbage would change when pins 1-7 were checked on the SRAM with a logic probe. Replaced bad 04XX chip and tested. Replaced all 24/28/40/42 pin sockets and all of the resistor packs on the CPU board. Tested.
Symptom: White screen. Dead.
Board had been worked on before. Installed a renew kit (new 24/28/40/42 pin sockets & all resistor packs) on the CPU board and found that 2 of the 08xx chips on the CPU board had pins bent under them. Repaired the pins as they will snap off when straightened, cleaned all the custom chips and tested. Game now worked but had no starfield and had blue lines on the screen.
Installed the renew kit on the video board, replaced the starfield generator, cleaned the pins on the custom chips and tested. Replaced the 2147 at 6L to fix the blue line issue and re-tested board.
Symptom: Reboots, crashes
Installed renew kit and replaced 1 cracked 1K resistor. Tested.
Repairer: GameDude
Forum Thread: Galaga (Midway) PCB Repair
Board #1 Constant reseting - Nice easy fix this one... broken trace between cpu's thats it, 4L was the location.
Flex and soak tested the PCB for a day and it works great!
I did notice that on the attract that the ship could not hit the enemy on one of the score examples, turns out the rapid fire chip had been installed on this board at 3J - Not really a fault but a quirk of the reprogramed rom.
Board #2 (SOLD) Constant reseting - This board set had problems on both the CPU and VIDEO board as I tested each against a known working set.
First was the CPU board, no physical damage was evident but it should be noted that many of these midway boards have corrosion / rust on many of the customs.
Carefully removing each and cleaning the legs with a very fine file I found one with a broken leg 2H 5volt pin, after reseating all chips carefully and soldering a new leg I now get boot to the game but with bad graphics.
The crosshatch at the start has big then little squares and the game screen is doubled, cut to the chase after cleaning all the customs the video board now works perfect.
It seems that all these old boards suffer from is usually corrsion / rust, I make it a habit now after cleaning to laquer the PCB to prevent further rusting.
I now have two original galaga PCB
Note:
FLEX = Flexing of PCB to see if there are any bad joints or sockets.
SOAK = Running for a while to see if the PCB crashes, sometimes using various voltage levels.
Repair Logs converted to wiki format by Brad from Aussie Arcade.