Repair Logs

Galaga (Midway) PCB repairs

Some general notes about repairing Galaga boards: I've heard it recommended that you should only replace the TTL chips with the exact TI (Texas Instruments) 74LS parts. Many of the boards I have seen have been repaired using many other brands of TTL chips and I have seen none of the timing problems that are said to occur if using a different brand of TTL chip.

If possible, it is always a good idea to have a complete and working set around. This will let you narrow the problem 1/2 of the board set down to either the CPU or Video board, by using a known good board and mixing it with the bad board set. It's much easier if you know which board has the problem. :) And even if both board have problems, you can tackle them one at a time this way.

Many problems are the result of dirty contacts and brittle legs on the many custom chips used on Galaga boards. It's a good idea NOT to remove these chips unless you are having problems. If they are removed, be very careful when removing, cleaning and reinserting them into the board. I use a Dremel and a small wire wheel to clean them, but I am very gentle while using it. Cleaning the outside of the legs makes the chip look nice, but the inside of the legs is where they make contact with the stock sockets. If any legs break, I solder on new legs as needed, borrowing them from bad chips I keep from other board repairs, and put the chip in a new dual-wipe socket. That socket is then inserted into socket on the board. This gives the chip a nice new carrier that makes contact with both sides of the legs. The extra socket will hopefully give the custom chip a much longer life. Many times the new legs will be a bit thinner and shorter, requiring you to also change the socket on the board itself with a new dual-wipe socket also.

10/16/2001

Problem: Sound fades in and out at random times.

Solution: Bad electrolytic caps, C20, C21, C22 on the CPU board.

10/16/2001

Problem: Board would hang at the ROM OK during boot.

Solution: Broken pin 11 on the ROM socket at 3J.

10/16/2001

Problem: Fire button doesn't work.

Solution: Bad ROM at 3N on the CPU board.

11/14/2001

Problem: Streaks of graphics trails coming off to the left of many of the characters.

Solution: Pin 12 on the Custom 04xx IC at 4M on the video board was broken and making intermittent contact with the chip. It was repaired by soldering on a new leg, then inserting it into a new socket. That socket was then inserted into the original socket.

11/14/2001

Problem: Players ship and the Galaga character have every other horizontal line either missing or mixed up.

Solution: Replaced the eight 2147 rams with two 2149 (2148 equivalent) rams.

12/19/2001

Problem: Dead board

First step was to clean all the customs. After this I had the self-test starting, but it would constantly reset itself in the middle, or just hang. I replaced the socket on the 07XX and the board would then go thru it's diags and boot into the attract mode.

Solution: Bad Socket on the 07xx custom on the CPU board at 4D, dirty custom chip legs

12/19/2001

Problem: No sound except BOOM.

Solution: Bad resistor pack at RM2 on the CPU board

12/19/2001

Problem: When the enemies entry the playfield, they are all bunched together. Board will reset occasionally.

Solution: Bad custom chip at 2J on the CPU board

12/19/2001

Problem: Board would reset itself. Happened more often as the board was on longer.

I was pretty sure there was a heat problem so I touched the top of the chips on the CPU board with my finger and found the main CPU at 4M was getting really hot.  I started to grab the freeze spray to determine if that was the problem when I noticed the CPU was a standard Z-80 and not a Z-80A.  

Solution: Replace the CPU at 4M with a Z-80A, it had a standard Z-80 instead!!!

12/19/2001

Problem: Dips switches have no effect

Solution: Bad 74LS151 at 4K had stuck outputs on the CPU board

12/19/2001

Problem: Game plays, but star field was erratic. Sprites would streak horizontally and the game would lose video sync, sometimes a row of white or light blue zeros would appear at the top and bottom of the screen.

When I first fired up the board I would have sworn it was a bad 2147 video ram. So I replace all 8 with 2 2149 rams, (2148 equivalents). This changed nothing. Next I looked at the LS161s on the video board seeing that now it was losing the picture altogether now on occasion. They all seemed fine. Then I looked at the star field generator chips making sure it was seated well and the legs were all clean. I knew there was a problem with timing, but not sure why I turned my attention to the chip next to the star field chip. I grabbed a LS368 and piggybacked it on 3N and the display cleaned right up. I replaced the chip and all is well.

Solution: Bad 74LS368 at 3N on the video board

2/14/2002 - From Brian Crowe

Problem: Game plays but enemies and the player's ship is missing.

Solution: bad 74LS20 at location 3A on the video pcb

3/16/2003

Problem: Sprites would kind of fade in an out, like they where only being partially drawn

Solution: Cleaned legs on Custom 07xx chip at 1N on the Video board

3/16/2003

Problem: Split screen, left hand side of screen was displayed twice, with a garbled vertical line of red jumbled letters down the center (the high score display).

Solution: Cleaned dirty legs of the custom 08XX chip at 2E on the CPU board

3/16/2003

Problem: Garbled sound. Boom! was fine, but the rest was short chirping sounds

Solution: Cleaned legs on custom chip 07XX at 4D on the CPU board

3/17/2002

Problem: Dead board, no clock to the CPU.

First thing I checked was the clock to the CPU, none, next checked the crystal and it was pulsing. The LS368 at 5A was fine as was the LS107 at 5B. The clock then goes to pin 1 of the 07xx chip at 5D. It was pulsing and so were a few of the outputs, but most were floating. (nothing on the logic probe). I replaced the 07XX and the same thing. Just for fun I changed the socket. I now had clock to the CPU. Better, but the board now hangs at ROM OK! See the next entry for the log on that problem.

Solution: Bad socket on the 07xx custom at 5D

3/17/2002

Problem: board boots but hangs at ROM OK.

Testing the reset lines on the 3 CPUs shows that they are going high when they are supposed to. Then checking the address and data lines on each individually shows that CPU 3, the Z-80 at 5E was not getting any data back on the data lines. I swapped the socket and reinserted the CPU, the board then fired up.

Solution: Bad socket on the Z-80 at 5E on the CPU board

3/17/2002

Problem: Sound volume really low, you can adjust it with the sound pot, but it's still very faint all the way up.

Solution: Open electrolytic capacitor C20, 22uf @ 10VDC

01/28/2004

Problem: Ram 4L Error

This was a fun one to track down. It was a video board problem for sure. I tried a known working CPU board with it with the same results, 4L error. After replacing the LS245 at 2H, , the RAM and socket at 3H, replaced the socket on the 04xx custom at 1H and the LS08 (chip select circuit), the board sill came up with the same error. I traced out all the connections from the rams to ensure the traces/connections were solid and they were. After that, I hit the lines with a logic probe with power on. Everything seem fine. I then check the connections on the LS08 at 1H and found pin 4 to be floating. Checking the schematics, this is an input from pin 12 of the LS138 at 1E. Pin 12 of the LS138 was pulsing, but nothing on the LS08, pin 4. I jumpered between the two and rebooted the board. This time it passed the RAM test and I had a working board.

Solution: Repaired broken trace between pin 12 of the LS138 at 1E and pin 4 of the LS08 at 1H.