Greetings earthlings! This post is a continuation of this mess. Basically, when I first got my Moon Patrol cabinet, I got a PCB along with it. That PCB had some major issues which I tried to fix. I got nowhere with it, and decided to just buy a working one. I’ve had the old dead board sitting here and figured I would take another crack at fixing it for fun.

Before this round of screwing with it, one of the things that was most noticeably wrong with it can be seen here:

Moon Patrol Problems

That should read BEGINNER COURSE GO! Half the letters are incorrect. In diagnostics mode, there was a RAM error, but it was difficult to read because of the character problem. I’ll try not to bore you too much, but here’s what I did for this round…

Moon Patrol is a 4 PCB stack. Now that I have a working set, I figured I could use it to isolate the problem to a particular board in the stack. I swapped each board from the bad stack into the good stack one by one, powering it on in between to see if the problem would follow. Doing this lead me to figuring out that the main CPU board, #2 in the stack, was the one with the problem.

With that out of the way, I noticed in the manual that if you don’t have any characters on screen at all, the ROM in  3F on the scrolling video board is bad. I swapped it with a good one, and the problem was still there. I guessed that maybe something connected to 3F was bad, and if I remember correctly, I traced 5M on the CPU board back to that chip. It’s a logic chip - 74LS245. I searched the RGVAC newsgroup for this chip, and many people were reporting that they fail pretty often in a lot of games. I then did a test I learned from Randy Fromm on that chip, checking each pin to see if any were out of wack. The basic idea is you put your meter on the diode test setting, put the red lead on ground, and probe each pin on the IC. All pins on the chip except for ground and Vcc (+5v) should read around .5 to .9. Anything way lower (like .004, for example) or way higher is probably bad. You can then compare any suspect pins to another of the same chip on the board if there is one. I did my comparison to the same chip on the working board. I found that pin 18 read .002, and was shorted to ground. I thought this was pretty cool, because I really am clueless with this stuff. :)

So, I removed 5M, put a socket in, and replaced the chip. With the old chip out, I tested it again. While pin 18 still read wrong, the short was not there. The short was actually on the board somewhere else. I went over every single trace on the board visually looking for problems, hoping to find the short that way. While doing this, I spotted this on the chip in 1K!

Moon Patrol - 1K bent pin #1

That is definitely wrong. I checked the same chip on the working board and it wasn’t intentionally not soldered. So, I removed that chip, bent the pin back, put a socket in and popped it back in. I still had that pesky short however. I read through the schematics, and wrote down every thing that was in circuit with 5M pin 18. That gave me a list of about 6 or 7 chips. I figured I could test them one by one, but the fastest way was to clip pins on the chips in question. I had replacements for them, so I started clipping hoping I would get lucky. I found the suspect chip at 1D. Pin 8 was shorted to ground. I replaced that chip and the short was gone. In the process of finding the short, I clipped pins on 3 other chips so I changed those as well.

BOY this is long and boring. :)

Anyways, after all that, I hooked it back up and the character problem was solved. The board still has issues however. It keeps resetting, the tank is missing sometimes, and the background looks a bit weird. In diagnostics mode it says RAM NG 8000, which according to the manual means RAM 3C on the scrolling video. board is bad. I know that’s not the case because I had already changed that RAM prior to this, so I have no clue what’s causing it. I’m taking another break from this board but maybe I’ll get back to it in the future.