Centipede (Atari) PCB repairs
Tools: soldering iron, parts list, eprom burner
Problem: Board is missing chips and will not run
I purchased this board as untested/non-working off eBay. The board obviously
would not run since it was clearly missing the 6502 CPU and the Pokey sound chip.
Further inspection of the board showed both the 12 Mhz crystal and an eprom was
missing also. I had a friend mail me an Atari 7800 Ballblazer cartridge, which
was, to my knowledge, the only Atari 7800 cart that had a pokey in it. I de-soldered
the Pokey and put it in the socket. Then bought an Apple IIe, that was in a very
broken state, at a local thrift shop for $5 and stole the 6502 for the other socket
on the centipede board. :) I had an extra crystal, so I installed that too.
Next, I burned a new eprom for the board and when I went to install it noticed
the traces underneath the hollow socket had a big glob of solder across 3 traces,
shorting them together. With the soldering iron and some solder wick, I removed
the solder glob and installed the freshly burned eprom. The rest of the board
looked very clean with no stray solder drippings, so I plugged it in my Centipede
cab and turned on the switch. The power light lit on the board and nothing started
smoking. So far so good. Then I went around to the front of the cab expecting to
see garbage on the screen and to my surprise, the attract screen was up. I coined
it up and played several games. It seems the only problem this board had was that
someone used it for parts and dropped a few stray bobs of solder on it.
Solution: Replace missing parts, remove solder blob
Problem: Stuck inputs, the archer would just run all the way
right and stay there
Solution: Bad 74LS257 at D9 and E10
Problem: Screen solid blue & was stuck in test mode.
It appeared to be a dead board. I mashed on a few buttons and found that
it was stuck in test mode because the colors would change from blue to black. Following
the test pin lead me to the 74LS257 at K9. Replacing it fix it from being stuck in test
mode, but it still only showed blue on the screen. Looking at the video output section,
the only pins pulsing on the card edge were blue and sync. So I traced back the red and
green signals to a 7407 chip at A9. Replacing that chip gave me back all the colors and
I had a working board.
Solution: Bad 7407 at A9 fixed the colors, and the 74LS257 at
K9 fixed the test mode toggle
Problem: Trackball doesn't move down.
If you moved the trackball slowly in the down direction,
it would sometimes work, but usually it would the archer would move up. After replacing the LS191s, an LS74
and the MC14584 and finally the LS157 in the trackball section, the problem remained. I just happened to
wiggle the capacitor at C99. It's a 1000 uf @ 100V dipped cap. I noticed one leg had a cold solder joint
and made poor contact with the pcb. I soldered it back on and the down didn't work at all. Replacing the
cap fixed the problem. A very long way to a bad cap indeed.
Solution: Bad CAP at C99
Problem: Black screen, appears to be dead
The board appeared to be dead, but when I pressed
on the prom at P4, the playfield appeared. Replaced the socket and we are good to go.
Solution: Bad socket at P4
Problem: Centipede and other motion sprites move up/down with the trackball up/down.
Moving the trackball up/down would move the entire centipede
and spiders up and down the screen. The archer was no where to be seen, but you could see shots coming
from below the playfield when the fire button was pressed. I found the problem by accidentally shorting
pins 10 and 11 of the LS42 at C5 together on a working board. Shorting these pins caused the same
problem I saw on this board. Replacing the IC fixed the problem.
Solution: Replaced 74LS42 at C5
Problem: Picture fades out as the board warms up.
If I hit the 7407 at A9 with freeze spray, the picture
would come back. Replaceing that chip fixed the fading out problem.
Solution: Replaced overheating 7407 at A9
Problem: Picture only shows green pixels.
The screen only displayed the green pixels. Probing the
outputs of the LS175 at A8 showed 2 of them were dead. I checked the inputs against a working board
and they seemed fine. Replacing the LS175 brought back all the colors.
Solution: Replaced faulty 74LS175 at A8
< Prev Page