CoolFox
Grand Master
Okay, so around in November of last year I saw a non-working Kaneko Jackie Chan PCB for sale on Ebay.
The problems were:
1) It wouldn't boot
2) There were signs of someone trying to reflow the QFPs with heat and there were actual burn marks on the board. The interesting thing is the picture the seller provided didn't have any of those burn marks or reflow attempts, so I'm guessing he tried one more time to fix it before he sent it to me...whatever
3) The I/O chip was cracked in half
4) Some of the ram wasn't soldered in correctly.
I checked all of the ROMs/Mask roms and found 1 to be bad, but it was a graphic rom, so that wouldn't have any effect on the game booting. I also fixed the ram that wasn't soldered in correctly.
After a while I requested the help of Phil and we worked together to try and get this working.
I desoldered and socketed all of the GALs, so we could compare them to the dumped ones and they all matched.
He noted that the primary 68000 wasn't doing anything, so it could be a bad 68000 or even worse, one of the 2 CPLDs on the board that had no dumps. The two CPLDs were in charge of address decoding and control signal generation for each CPU.
Being a rather rare game,I had to enlist the help of my friend who had the board and luckily his CPLDs were socketed. Even more lucky was the CPLDs were not secured, so we dumped them and preserved them on Jammarcade.
CPLDS in question:
Both of the CPLDs in my eprom burner complained of bad pins, so we burned new ones, socketed them, and the board did this:
Life! According to Phil that check was for the main processors/ram and it was failing it. I noticed if I flexed the board in a certain way, it would get past other checks, but it would not go into the game. I checked all of the ram that was connected to the main processor and they were fine. I had an odd feeling that the PLCC sockets weren't in correctly, so I took the board to another local friend and he professionally put in new sockets.
Nice! It's booting past all of the checks. Next was the cracked I/O chip. After the checks, it would go into test mode no matter what and would not exit it until the I/O chip was fixed. I heard Caius made a replacement for that chip, so I ordered a kit and put it together.
Goody, now lets see what it does. Two videos, one easier to see.
Well great, Jackie doesn't show up in the intro and there are missing layers. We initially suspected the QFP chips CG24143/CG24173 on the left side of the board which were Fujitsu custom graphics generators and possibly the View2 chip which does tilemaps. The Fujitsu chips were burnt on one side because the previous owner didn't know what he was doing when he was reflowing it, but it seemed like the chip was outputting good signals. There are 2 sets of the CG24143/CG24173 chips, one on the right side of the board and one on the left. The burnt ones were on the right side of the board, but the ones we suspected were on the left. I had replacement CG24143/CG24173 chips, so we replaced them to no change. I was hoping it wasn't the view2 chip as it would mean I needed to buy another parts board.
The area in question:
I gave the board back to Phil and he looked around and noticed one gal pin wasn't connected. A trace must have been lifted when the gals were taken off as the solder on this was really hard to get off the board.
That was indeed the cause of the lack of Jackie and the missing layers!
I then noticed the AMP was in horrible condition (other people looked at it and told me as well) and it had no heatsink, so I sacced the amp and heatsink from a broken board and Jackie is now complete and working! Huge thanks to Phil for helping me with this and Caius for the replacement I/O chip.
On a side note: Phil said he won't work on Kaneko boards ever again as they are crazily built.
The problems were:
1) It wouldn't boot
2) There were signs of someone trying to reflow the QFPs with heat and there were actual burn marks on the board. The interesting thing is the picture the seller provided didn't have any of those burn marks or reflow attempts, so I'm guessing he tried one more time to fix it before he sent it to me...whatever
3) The I/O chip was cracked in half
4) Some of the ram wasn't soldered in correctly.
I checked all of the ROMs/Mask roms and found 1 to be bad, but it was a graphic rom, so that wouldn't have any effect on the game booting. I also fixed the ram that wasn't soldered in correctly.
After a while I requested the help of Phil and we worked together to try and get this working.
I desoldered and socketed all of the GALs, so we could compare them to the dumped ones and they all matched.
He noted that the primary 68000 wasn't doing anything, so it could be a bad 68000 or even worse, one of the 2 CPLDs on the board that had no dumps. The two CPLDs were in charge of address decoding and control signal generation for each CPU.
Being a rather rare game,I had to enlist the help of my friend who had the board and luckily his CPLDs were socketed. Even more lucky was the CPLDs were not secured, so we dumped them and preserved them on Jammarcade.
CPLDS in question:
Both of the CPLDs in my eprom burner complained of bad pins, so we burned new ones, socketed them, and the board did this:
Life! According to Phil that check was for the main processors/ram and it was failing it. I noticed if I flexed the board in a certain way, it would get past other checks, but it would not go into the game. I checked all of the ram that was connected to the main processor and they were fine. I had an odd feeling that the PLCC sockets weren't in correctly, so I took the board to another local friend and he professionally put in new sockets.
Nice! It's booting past all of the checks. Next was the cracked I/O chip. After the checks, it would go into test mode no matter what and would not exit it until the I/O chip was fixed. I heard Caius made a replacement for that chip, so I ordered a kit and put it together.
Goody, now lets see what it does. Two videos, one easier to see.
Well great, Jackie doesn't show up in the intro and there are missing layers. We initially suspected the QFP chips CG24143/CG24173 on the left side of the board which were Fujitsu custom graphics generators and possibly the View2 chip which does tilemaps. The Fujitsu chips were burnt on one side because the previous owner didn't know what he was doing when he was reflowing it, but it seemed like the chip was outputting good signals. There are 2 sets of the CG24143/CG24173 chips, one on the right side of the board and one on the left. The burnt ones were on the right side of the board, but the ones we suspected were on the left. I had replacement CG24143/CG24173 chips, so we replaced them to no change. I was hoping it wasn't the view2 chip as it would mean I needed to buy another parts board.
The area in question:
I gave the board back to Phil and he looked around and noticed one gal pin wasn't connected. A trace must have been lifted when the gals were taken off as the solder on this was really hard to get off the board.
That was indeed the cause of the lack of Jackie and the missing layers!
I then noticed the AMP was in horrible condition (other people looked at it and told me as well) and it had no heatsink, so I sacced the amp and heatsink from a broken board and Jackie is now complete and working! Huge thanks to Phil for helping me with this and Caius for the replacement I/O chip.
On a side note: Phil said he won't work on Kaneko boards ever again as they are crazily built.
Last edited:
