Centipede Repair Log

mjansma

Active member

Donor 3 years: 2023-2025
Joined
Sep 12, 2023
Messages
208
Reaction score
239
Location
Antioch, Illinois
I'm writing and posting this mainly for my own record as I'm trying to teach myself how to diagnose and repair basic board issues, but also maybe this will help someone down the line. Full disclosure, I'm going at this with only a DMM and logic probe. I know I know I know, I need a scope, especially since I own both an Asteroids and Tempest. But I wouldn't know what to do with a scope if I had one. That will be next on my list of things to learn once I get better at simple fixes.


Background: I picked up a non-working Centipede -01 PCB for $50 on ebay. The board is pretty clean overall. There is no scorching on the edge connector. It is missing all 6 ROMs, the EAROM, both 8-pin dip switches, and a LS42 @ H3.

The board had three broken traces on the backside near the edge connector that were jumpered. After cleaning all the socketed chip legs and washing/drying the entire board overnight the first step was to get rid of the jumpers and repair the broken traces. It was a fairly easy fix. I tested for continuity afterwards and confirmed it was good.

I burned new ROMs, socketed/replaced the LS42 @ H3, the missing dip switches, and EAROM. Time to plug it in to my Centipede cab..

+5 voltage was good (4.998), TIP102/6044s all tested good. -15 was good. +15 was dead. Replaced it with a new 7815. The screen was filled with garbage at startup. Sometimes it's a black screen with red garbage, sometimes a yellow screen with red garbage, sometimes it's just a white screen with nothing at all. I was unable to get a test screen or boot to test. Just solid white screens or garbage screens.

Pin 37 on the 6502 is showing a clock. I'm not sure if it's a good clock, but a clock none the less. The clock has been the only constant while probing around the last two days. Pressing the reset button has no effect on the clock while probing. Pin 40 was stuck high (which should be normal operating procedure?) After I took a break and turned it back on, it was stuck low. Reset button has no effect on it while probing

I probed all of the address and data lines on the 6502. Every time I turn the game on and probe them I would get something totally different.

First check of the data lines showed everything stuck low. Based on the schematics, I immediately thought the LS245 @ E2 is bad. However, after turning the game off and then back on, the data lines were almost all pulsing, with only one or two stuck low. Same issue with the address lines; sometimes high, sometimes low, sometimes pulsing. What are the odds that the 244's AND 245 are bad?

I placed a DigiKey order for a bunch various chips that I might need for repairs, but I'm in a holding pattern as my order wont be here until after the New Year. Don't worry, I'm not shotgunning anything, I just want to be prepared.

In the meantime, more reading and watching videos. Happy New Year...

EDIT 12/31/24: After actually thinking about this a little more, it appears that there is probably something wrong within the reset circuit. For the most part, pin 40 on the 6502 is stuck low. Sometimes when turning the game on it was stuck high and would then get stuck low after hitting the reset. Regardless, there was no immediate low to high shift when powering up the board. I do have a functioning Centipede board here to compare with while I wait.
 

Attachments

  • IMG_6502.jpeg
    IMG_6502.jpeg
    958.6 KB · Views: 9
Last edited:
Since my last post on this I picked up a scope and a few probes on marketplace from a retired audio tech for less than $100, which was a nice score. There's quite a bit of a learning curve, but I've figured it out enough to actually see things.

After probing around and comparing the dead board to a good working board I was able to determine that the LS90 at L2 was shot. After socketing and replacing it I now have a reset signal that is watchdogging. It's a step in the right direction, but now I need to find out why it's resetting.

I still have a screen full of junk and can't get a test screen, but I'll celebrate a small victory.
 

Attachments

  • thumbnail_IMG_6595.jpg
    thumbnail_IMG_6595.jpg
    95.7 KB · Views: 12
  • thumbnail_IMG_6594.jpg
    thumbnail_IMG_6594.jpg
    83 KB · Views: 12
  • thumbnail_IMG_6593.jpg
    thumbnail_IMG_6593.jpg
    72.3 KB · Views: 12
It lives!

I guess the lesson for today is never assume that your processor is 100% working. Even if you are getting some type of clock and a reset line that's watchdogging. After seeing that AB0 through AB9 on the 2114s at H2 & F2 were dead, I followed the schematics backwards to C1 and C2 which were all stuck high. This left only the 6502. I popped in a new one and it's back. Lesson learned.

Now the only issue is these little blips that blink around the high scores and occasionally around other parts of the field. I don't even know where to start looking. If anyone has had this issue before or could help point me in the right direction, I would appreciate it.
 

Attachments

  • thumbnail_IMG_6606.jpg
    thumbnail_IMG_6606.jpg
    43.3 KB · Views: 9
My guess is the cap on the LS42 or you put a 6502A vs a 6502 - it will be some combination of those..
I definitely put a 6502A in it. The LS42 was missing when I got the board so I socketed/replaced it and have no idea if there was a cap was soldered to it or not, but I bet there was!
 
Last edited:
Just following up and closing this post. I soldered on a 330 pF cap to the LS42 and still had the same issues. Once I got ahold of a 6502 instead of a 6502A and installed it, the problem was fixed! I ran the board 2 hours this afternoon without issue. Thanks for your help @lilypad19 I think this one is now considered case closed.
 
Last edited:
Back
Top Bottom