What's new
By:

Another Stumper CPS1 A Board Repair

YesAffinity

YesAffinity

Enthusiast
Joined
May 20, 2016
Messages
242
Reaction score
150
Location
Escondido, CA, USA
Been probing this board for many hours over the course of a couple weeks (hard to get a single window of time to sit and dedicate to it...).

I have tracked down "a" problem, which I'm hoping is "the" problem.

Video of the issue here:

I found that pin 20 of video rams 11L and 11N are pulsing at a spasmodic rate. Way too fast, compared to a known good board. This is the Chip Enable pin, CE*.

I tracked it back through a 74F32 at 11K. Pins 1 and 2 are inputs, pin 3 is the output of pins 1 and 2, and pin 3 output to both pin 20's at 11L and 11N.

I tracked it back further to the PAL BUF1 at 13K. BUF1 pin 19 outputs to 11K pin 1. Pin 19 is outputting at a spasmodic rate.

I have swapped BUF1 with known good, replaced 11K, 11L and 11N, with no change.

I've checked BUF1 inputs against a working board, and they appear to be operating the same. I've checked the rest of 11K's (74F32's) inputs and outputs, and they seem to be operating the same as a working board.

11L and 11N WE* and OE* pins are shared with other rams, and I would expect if something was wrong on those lines, they would lend to similar problems on the other rams.

I've also probed throughout the circuit feeding pin 2 of 11K. Everything looks okay there as well, compared to a working board.

I believe the problem is isolated to BUF1, 11K, 11L and 11N, but possibly I'm overlooking something.

Also worth noting: the problems aren't quite as body when the board starts from cold. The title screen is perfectly normal. The visible sprites, health bars, etc. are normal. As the game warms up, the problems progressively worsen.

Anybody have any good insight, or see something I'm missing in the schematics, in this section of the circuit?
 
The A custom has died, all the graphics data goes through that custom. If it's bad, you get graphics errors. The spasmodic data you're seeing will be coming from the custom.
 
xodaraP said:
The A custom has died, all the graphics data goes through that custom. If it's bad, you get graphics errors. The spasmodic data you're seeing will be coming from the custom.
Click to expand...
Thanks for confirming xodara. That was my fear, but it's an inevitability with these boards. Let's see...do I have any scrappers lying around here that might have a salvageable custom...?
 
Sadly it's the most common fault with CPS1 :(

I have seen and fixed a few with mask ROM faults, but if you've isolated it to the A board, 99% chance it's the custom.
 
If the board boot ok and fail some time later, maybe there's a logic problem that's cause a instability on cycles, not a custom burned. First you have to test/replace ICs on the BUS signal lines.
 
i manged to fix one a board, i found a guy selling a dead cps2 board for parts. Took the A01 and it worked again.

I do not recommend killing a working cps2, be patient and a parts board will show up
 
Elfito69 said:
i manged to fix one a board, i found a guy selling a dead cps2 board for parts. Took the A01 and it worked again.

I do not recommend killing a working cps2, be patient and a parts board will show up
Click to expand...
Exactly what I did. Picked up 2 busted CPS2 A boards last week. CPS-A-01 fixed it right up....well the video at least. Now I have to track down the problem causing missing audio. :P ;(

Thanks again, xodaraP. I will report back on the audio once I get that sorted, as well.




 
I wish...

Someone had done a terrible job socketing the z80. Looks like they solder it in with dirt.

Repaired a couple floating lines, one of which was IRQ* between YM2151 and z80. YM2151 is supposed to issue the requests. It is stuck high (YM2151 pin 2).

Most lines that are supposed to have activity don't:
SOU1 pins 16, 17 and 18 are not behaving normally (stuck), corresponding to YM2141 pins 5, 6 and 11H pins 4, 5 which in turn
Address lines at 11H - pins 1, 2 and 3 - do not have activity
z80 pins 19, 21, 22 (MREQ*, RD*, WR*) are not behaving normally (stuck)

z80 pins 17, 24 and 25 are all tied to Vcc (24 and 25 were floating previously)

Clock lines all have normal activity.

The z80, MSM6295 and YM2151 reset lines behave normally.

I confirmed the z80 is good, have swapped SOU1 with known good; replaced 11H, YM2151 and MSM6295.

Going further downstream, YM3012 does not have normal audio on its outputs, but I expect that it wouldn't.

What am I overlooking?
 
Jesus H....

I soldered the jumper for z80 pins 24 and 25 to the wrong pin. They are supposed to be connected to z80 pin 17 and pin 11. I had it on pin 10.

Well, that's 5 hours of my life I'll never get back. :S

But, in the process of all the unnecessary probing, I did find a bad cap at CC12. Had some brand new nichicon spares and replaced it, so I guess that's a plus.

Thanks for all the help, fellas. On to the next one.
 
I have 8 cps1 A parts boards in various states. Almost all are complete. Too much shit in boxes, but want to get rid of them since they are beyond my knowledge and skill and I will never find the time.
 
You must log in or register to reply here.
Back
Top