Q&A Thread

[channelmaniac]

Follow the outputs of the chips to the connectors leading to the bottom board. Check for continuity from the chips, through the connectors, down to the bottom board. I had one board on the bench once that had junk inside the connector causing a fault.

Also, look at the bottom side of the bottom board for gouged/scratched traces.

Racing means the output is oscillating and outputting junk.

[hairy otter]

I already checked all pin's on all four connectors on the board. All give good continuity top to bottom. Didn't look further on the bottom board so far. The six first an last seem to be all connected together but i think this is OK (maybe GND and +5V?)
To night I check for continuity further on the bottom board.
Is there any way to check the output signal on racing? With a logic probe I can only see if it's busy i think?

[channelmaniac]

When you see a racing signal, you will know it by ear, if your logic probe has audio output. If it doesn't, the only real way to see it is with an oscilloscope.

[hairy otter]

Is it safe/usefull to connect the input pin to the output pin of a suspicious chip?
Or pull the output down to look for change?
My logic probe does not have a audio output, and I dont have a scope.

[channelmaniac]

Is it safe/usefull to connect the input pin to the output pin of a suspicious chip?

Sometimes... Depends on if there's a hard short to power or ground.

Or pull the output down to look for change?

That's the easier way. Do it momentarily.

My logic probe does not have a audio output, and I dont have a scope.

Ouch. This makes it hard. The audio output on the probe makes it much easier to compare the input signal to the output signal on the chip. I *HIGHLY* recommend you get a logic probe with audio capabilities.

[hairy otter]

Still confused. Sorry, but every time I try to look for the faulty chip an other one seems to be bad.
If I flip dip8 to pause and start pulling up input signals a different chip alters the appearance of the line. The chip that seemed faulty in previous tests only add another line. Also if I ad a line by puling up an input those lines look different than the one whose always there?
I try to test continuity of all tracks but it takes quite some time. And I think there isn't a problem as with all 32 27F251's lines change or appear on the screen.
The only thing I suspect is the IC in the SFIX IC-foot.(it says 381000-20, 7C6 BK, 9035 Z01) The SC1 and SC feet are empty? Then the tracks go to the SNK PRO CT0 chip.

[channelmaniac]

The FIX chip is a mask ROM - 23C1000. You can read it with some EPROM programmers and compare the checksum to the checksums found in MAME.

There are 2 empty ROM sockets on the Neo Geo 4/6 slot boards. That is normal. Beyond that, you may have a chipset issue and that's as far as I can help you now.

[hairy otter]

Can you please explain what you do mean by "a chipset issue"? Then I won't take any more of your time for now.

[channelmaniac]

Chipset issue means a problem with one of the custom chips.

[CraftyMech]

Great repair logs, I've enjoyed reading through them!

I did not run across mention of the issue I'm having with a Capcom 1943 board set, so I figured I would ask here.

I have a 1943 with glitchy sprite graphics(scrolling background and character text is fine), but from the service menu the roms all check out ok. Near the object rom chips on the B-board is an IC that looks fried, a Capcom 86S100 28pin dip. Inspection of the board underside reveals a nice scorch area directly under the chip. Being a custom Capcom IC, I have had no luck finding any sources for purchase. Have you had to replace one of these chips before on a pre-CPS1 Capcom board? Side Arms & Tiger Heli also use this particular IC, so I am on the lookout for a donor board, but am hoping out there on the Internets these chips might still be floating around.

Thanks!
Aaron

[channelmaniac]

I have a few parts boards... will check and see if I have the component.

Never seen one of those fry... the problems I've had were with the big PLCC custom so I should have a board in the junk pile with one of those dead PLCCs.

RJ

[CraftyMech]

Thanks RJ, that would be awesome if you had a spare I could purchase from you. The chip is part # 9C if I remember on the Video/Sound B board.

[channelmaniac]

I have some. You have a PM.

[CraftyMech]

I did some probing on the suspect 86S100 chip with my scope, and the chip is still alive but acting erratically compared to the other two chips of the same type on the B-board. Shorting pins 5 & 6 improved the display, but still left "jail bars" through all the enemy/player sprites. 5 & 6 are supposed to connected according to the schematic, but for one reason or another related to the chip fault were not. Looking forward to replacing the chip and seeing if this board is back to 100%.

I removed the existing chip yesterday, not an easy job with the tightly packed footprint of that particular 28pin dip, and the very low profile of the package. I'm just using a solder sucker however and a little patience. Good thing you had one on hand, I'm not sure I could have removed one from a donor board without frying the chip in the process. Easier I'm guessing with a vacuum setup? Expensive tools, but I might have to look at one if I get serious about repair work.

[channelmaniac]

Night and day difference.

Having a good desoldering station will just flat out amaze you.

[CraftyMech]

Night and day difference.

Having a good desoldering station will just flat out amaze you.

What brands would you recommend? I've had good luck with sub-$100 Xytronic and Weller solder stations, but am not familiar with their more specialized products.

[channelmaniac]

I use Pace gear. Hakko is good too.

[Artemio]

I have an interesting Issue with a couple of boards and am looking for pointers.

One is a Kuri Kinton, it looks pristine but when playing the game, or in demo mode, there is no collision detection that produces damage. The enemies can't hit the main character neither the enemies can be hit. They cannot go through each other though. The same happens in the demo mode. I've already checked the program roms, and burnt all alternate sets form mame with no changes, and also played with all the dip switches.

I've scoped around it and have found no evident cause.


The other is a Side Arms. Some sprites are shown incorrectly, they are made up of the same half twice. I have been testing it slowly, and noticed that one of the data lines had noise, specially after some time and one mask rom was heating fast. Replaced three mask roms that had that pin apparently burnt... they could not be read, and they caused noise in the line. They are normal now after replacing them with suitable eproms However, I am still looking for the half sprite issue.

My guess is that a it must be wrong that causes some multiplexing for the sprite data... any other ideas? will continue checking it.

Thanks!

[channelmaniac]

Wow... not sure on the collision detection. That's something that is different on every boardset. If it's a surface mount boardset then I'd start with reflowing any custom chips.

On the double graphics issue, check the address lines for missing or stuck signals on the graphics ROMs and graphics RAMs.

[misschivous]

I just received a Neo Geo 6 slot and it doesn't boot to the crosshatch screen. My monitor just displays red blocks.Please Help.
Photos:

http://i1272.photobucket.com/albums/y397/misschivousyoo/IMAG0063_zpsba6be8a5.jpg
http://i1272.photobucket.com/albums/y397/misschivousyoo/IMAG0058_zps9c7ea632.jpg
http://i1272.photobucket.com/albums/y397/misschivousyoo/IMAG0060_zps60dafe93.jpg
http://i1272.photobucket.com/albums/y397/misschivousyoo/IMAG0065_zpsa779516a.jpg
http://i1272.photobucket.com/albums/y397/misschivousyoo/IMAG0066_zpsc2a442cb.jpg
http://i1272.photobucket.com/albums/y397/misschivousyoo/IMAG0062_zps542d7de0.jpg
http://i1272.photobucket.com/albums/y397/misschivousyoo/IMAG0056_zpse7531ebf.jpg

[channelmaniac]

Hmmm... Looks like the CPU isn't starting at all. Check the power then the /CE and /OE lines on the backup RAM for missing signals. (These are the RAMs connected to the CPU, but are powered by the backup battery circuitry.)

Most likely culprits: Bad BIOS, bad Backup or Work RAM, missing control signals on the Backup RAM (see above), corrosion damage from a leaking battery (causes control signals to be missing on Backup RAM), or a gouged trace somewhere.

Raymond

[CraftyMech]

I picked up a Capcom CPS1 Mercs board for cheap as "untested". Powered up fine, but there appeared to be palette issues, as the backgrounds were two tone (just black and white) and the game sprites had weird colors. Checking the color bar test in the service menu showed only muted beige and grays instead of the usual r,g,b.

The graphics all seemed intact, just with palette problems, so I assumed the ROMS might be fine and swapped out the A board with one from a working Magic Sword set. Game powered back up fine! So the issue is somewhere on the A board, but I'm not sure where to start. Any recomendations for graphic related A board chips to take a look at?

[channelmaniac]

That's a difficult boardset to work on. I've found that generally it's the fault of the large surface mount custom logic chip.

You can check the pallette RAM by replacement... but that's really about it.

[CraftyMech]

Thanks for the reply, it does not sound very promising for salvaging the A board.

Schematics appear to be very hard to come by for the CPS1 board set, so I have not been able to locate the palette ram. I did switch out the 5 socketed 20pin dips on the A board (they appear to be PLA chips), but no luck. The board had some rust and corrosion on a number of proms on the top layer (A board looked better), I wonder if reflowing the graphics chip might catch a bad joint that I'm not seeing with my magnifying loop. Can't do any harm I guess, so I might give that a try after I've located the palette ram.

[channelmaniac]

Who's yer daddy?

http://arcarc.xmission.com/PDF_Arcade_Manuals_and_Schematics/Forgotten%20Worlds%20%28Capcom%29%20CPS%20Schematics.pdf