What's new

ShootTheCore

Legendary
Joined
Jan 20, 2016
Messages
3,153
Reaction score
6,169
Location
Utah USA
I have a Konami Aliens PCB that works fine, except that none of the sound samples will play (like the machine gun sound effects).

@caius documented a repair he made on an Aliens PCB with the same problem, here:
http://www.jammarcade.net/aliens-repair-log/

Unfortunately, his solution of replacing the mask ROM with a 27C400 with the data doubled didn't work for me. I tried both with the ROM data written to the chip directly, and with the data doubled just to be sure. The written EPROM passes verification correctly on the chip programmer.

The Konami 007232 PCM controller is showing activity on the CPU lines, it has power, and it's getting a clock signal, but the ROM address and data lines are stuck low with no activity (the ROM lines were all stuck high on @'caius''s board). ROM pins 14,16,18,20,23,25,27 all read as floating on my logic probe, but the schematics show that those lines are unused.

I've also tried pulling the reset line low a couple of times on the 007232 in case the controller had crashed, but that made no difference either.

At this point, I'm thinking the 007232 custom itself has failed, but I want to get a second opinion before going through all the trouble to find another one and performing the soldering work to replace it (64 pins!).

Any ideas?

Photo Jan 13, 1 34 25 PM.jpg
 

Attachments

  • Capture.PNG
    Capture.PNG
    375.5 KB · Views: 344
On the dip bank that only has 4 dips, you can flip dip 3 and it’ll do a more robust test mode on the roms. Not that it’ll solve anything, but I’m curious what shows up when a board is in your condition, where you have replaced the rom already.
 
OK, I desoldered the 74LS273 at E9 and installed a socket along with a replacement chip. No change - still no sound samples.

I also had the board run it's comprehensive self-check and everything passed.

Any other suggestions? Sounds like I'd better start looking for a 007232 donor...

Photo Jan 13, 3 20 29 PM.jpg
 

Attachments

  • Photo Jan 12, 11 11 43 PM.jpg
    Photo Jan 12, 11 11 43 PM.jpg
    947.4 KB · Views: 199
Don't take what I tell you as gold cast, I can't be of great help remotely, you should do some troubeshooting by yoursef too.
Anyway, try to replace (but install first a 64 pin S-DIP socket) the '007232', a good cheap donor board is Main Event.
 
Isn't there a 640kHz crystal on those board for samples? More than often it gets ripped off.

Edit: not on those boards. Check clock anyway.
 
Last edited by a moderator:
I did confirm that the 007232 is getting a clock signal.

I’ve got a lead on a cheap Main Event board to use as a donor, so I’ll swap the 007232 next and update this thread. Thanks for the advice everyone!
 
I did confirm that the 007232 is getting a clock signal.

I’ve got a lead on a cheap Main Event board to use as a donor, so I’ll swap the 007232 next and update this thread. Thanks for the advice everyone!
I had few for sale here for $30 a pop but they didn't sell...
 
OK, I tried piggybacking both the F8 and F9 ICs with no change in behavior.

I probed out E9, F8, F9, E12, and F12 since they each individually contribute a few signals to the 007232 custom. All of them exhibited pulsing data lines and proper behavior. I finally probed out all 64 pins of the 007232 custom itself and cross-referenced it with both the Konami schematic and a pinout writeup I found here:

http://newlifegames.net/nlg/index.php?topic=47.35;imode

I've concluded that the 007232 must be faulty. All of the input signals and the clock line show pulsing activity (the input lines are even pulsing in time with the machine gun fire animation in the game), but the ROM addressing lines, ROM data lines, and output channel lines are all held low.

I'll swap out the custom for a socketed donor in a week or so and update the thread. Thanks everyone for the suggestions and help!
 
Back
Top