Rebuild and repair log: Golden Axe PCB...

ifkz

Well-known member
Joined
Oct 3, 2006
Messages
1,544
Reaction score
45
Location
Texas
BACKGROUND
This Golden Axe PCB has always been like the classic 1964 childrens tale "The Giving Tree" by Shel Silverstein (thanks wikipedia!). I bought it many years ago from ebay in piss poor shape. It taught me the ropes of board repair: replacing sockets, soldering EPROM legs when they broke, hunting down bad EPROMS, and CPU replacement. I enjoyed the game for many years. Recently I started raiding this set for parts to make other games from the same hardware platform work again. In the end this PCB set became nothing more than an incomplete set of EPROMS on a scrap piece of styrofoam. A lot like the stump in the above story.

Last night some parts came in for an unrelated project and I had enough spares to give this game a shot at running again on its own. I'll be rebuilding and repairing this using parts boards and whatever scrap I can find. I made very good progress last night, so time to share!

THE EPROM BOARD
This stripped, scrap board was given to me in a lot of other parts awhile ago. I rebuilt it during troubleshooting of my MVP boardset and I learned that it can run Golden Axe through some pictures online. I moved over every original EPROM I could and spent a lot of time with MAME sets reconstructing the other code. I programed the code to spare chips and had a heck of a time adding and getting the jumpers correct. It is not pretty, but it does work; major accomplishment!

The only thing it needs is a specialized socketed chip marked PLS153N. This may be the chip that holds this entire project up in the end, I still haven't tracked one down!

A picture of the populated board, notice the bad bend on the left. This is actually a crack in the board that I repaired:
HPIM1471.jpg



THE MAIN BOARD
Not much to say on this yet, I have a source for the two missing PALS and I will be looking through my parts boards to find a 10Mhz 68000 CPU and a Z80 processor. This board will have a slight audio issue to work through after everything is populated again (I tried using this board before when I rebuilt my Shinobi PCB). This board has a lot of repair work done to it, but is complete aside from a heatsink and socketed chips. More will be added to this thread as progress is made and more parts come in!

A picture:
HPIM1363.jpg
 
Last edited:
Subscribed.

I've been fighting with an Altered Beast that i replaced my Z80 once and encrypted sound rom twice with 0 change.

my Golden Axe hasn't /wrist yet, great game well worth the time.
 
That's the model of ROM board I have set aside for you. It has 315-5298 in the socket.

RJ
 
I've been fighting with an Altered Beast that i replaced my Z80 once and encrypted sound rom twice with 0 change.

The audio on System16B boards is very susceptible to corner damage where the rear PCB mounts attach, its a major weak point of the board. The data bus for the Z80 runs round the very outside of the board, all the way round the back, and up one of the connectors on the far side of the board to get to the ROMs.

gaxe15.jpg


Its very common for these boards to have slightly bent corners which can often crack a track or two, or in the case of the above photo, take out a load of tracks when the corner snaps off at the PCB mount hole entirely. You should be check for continuity between the sound ROM data pins and the 74ls273 next to the Z80, you should have 8 connections,
 
Last edited:
The audio on System16B boards is very susceptible to corner damage where the rear PCB mounts attach, its a major weak point of the board. The data bus for the Z80 runs round the very outside of the board, all the way round the back, and up one of the connectors on the far side of the board to get to the ROMs.

gaxe15.jpg


Its very common for these boards to have slightly bent corners which can often crack a track or two, or in the case of the above photo, take out a load of tracks when the corner snaps off at the PCB mount hole entirely. You should be check for continuity between the sound ROM data pins and the 74ls273 next to the Z80, you should have 8 connections,


will do! i heard about the traces for the sound running close to the edge/corner of the boards but mine doesn't show any damage there, atleast not that i can remember. I'll have to take another look once i dig it out of my multi jamma cab.
 
Well, I am both surprised and also stumped. I went through my parts boards and I found my last 10Mhz 68000 and a Z80 that will work. I also created a super lame (TM) replacement heat sink using the last remaining foot for my spare switching power supply. I installed them on the board along with some PALS borrowed from other boards. I was greeted by an almost perfect game.
HPIM1474.jpg


The only issue that needs to be addressed is sometimes the speech samples will become corrupt during playback. Other times the samples are 100% correct. I am happy though, the game is fully playable, the BGM is present, all self tests pass, it only has this nagging issue and it will be 100%

I beep tested the damaged area and it all checked out. I looked over the bottom of the board for scratches (all okay) and bent pins (none).

Humm.....Maybe slightly stinky sample ram? Any ideas?
 
Last edited:
It's not a sample RAM... check the 40 pin chip next to the sound CPU and the little blue/orange/yellow/red plastic resonator next to it that is used for the clock for that chip.

The ROM with the samples is A13 if my memory is correct. Check the sockets and check for bad traces or connections on the ROM board.
 
A11 is the sound sample EPROM in this case, a large 1Mbit (128KB) chip. The MIDI tunes are held in A13 (32KB). All MIDI BGMs play perfectly, all sound fx that are MIDI based are perfect. The fault does not follow the EPROM board, so it is somewhere in the sound section of the bottom board.

I'll go around again and wiggle components looking for cold solder joints and other damage.

EDIT: With what I am reading above, I need to beep test each leg of the 1Mbit EPROM in A11 against the 74ls273 and check for 8 points of continuity. If that checks out, can I start to think this may be the point of failure, a bad chip?

Sometimes the samples will address junk and keep playing until another sample interrupts the operation. I can get staticy samples of two death sounds, and the crowd running away all during one sample hardware call, for instance. A lot of the time, it is spot on perfect; which is why I was thinking RAM. But maybe this is an addressing problem, it calls for a spot of memory from the chip but the signal is getting there but interpreted wrong by some 74lsxxxx logic chip?

Even if this is as close as I get it, it is pretty exciting that I had a 95% working game hidden in my parts boards :)
 
Last edited:
My final repair-in-progress thread to close out. Thanks again to channelmanic, who supplied me with my three missing PALs: Two for the bottom board, one for the EPROM board. It still has an audio glitch whenever an elf appears, but no worries (all video and FM music okay). Another game that came to me in terrible shape: six EPROMS from my parts pile!
 
who doesn't love gold and axes?

I've been subscribed to this thread and i'd just like to say awesome job and thanks for sticking it out till the end!
 
Back
Top Bottom