Konami Rock ‘n Rage

[Supergun]

I'm wondering what did you do to the original eproms with original eprom labels?

No worries there! I saved all of the original mask roms/EPROMs and I fully intend to try and populate the board with them once I get it running.
For the time being, they are set aside as they are quite fragile and in need of repair.

So here are the readings on the 63c09

 

[Monstermug]

have you tried replacing it?

 

[Supergun]

Oh yes, absolutely, and I actually already did replace it a while back.
I’m pretty much going to end up with a board that has every single chip socketed.

 

[Monstermug]

Where do the data line go to? Thats all the Ds. They should be toggling. I suspect a bad buffer of some sort?

 

[Supergun]

Ok, that was easy enough.
All 8 of the D pins from the 63c09 at 13A connect to the 8 pins of the Konami 007324 at 14B.

EDIT: And most of the pins on the opposite side of the Konami 007324 connect to an LS245 (Fujitsu) right next to it at position 15B.

Should I replace that LS245?

 

[stj]

you should be seeing some type of interupts generated by the video circuit
and if i had that board the first thing would have been to replace every chipsocket with decent turned pin ones.
if they looked like that on the outside - imagine what may be trapped under them between the pins!!

 

[stj]

Ok, that was easy enough.
All 8 of the D pins from the 63c09 at 13A connect to the 8 pins of the Konami 007324 at 14B.

EDIT: And most of the pins on the opposite side of the Konami 007324 connect to an LS245 (Fujitsu) right next to it at position 15B.

Should I replace that LS245?

do you see data on one side but not the other?
is it enabled?
is the direction pin pulsing?

 

[Supergun]

Yes, when I put the logic probe into pulse mode, I have pulsing both at the D pins on the 63c09e and at the pins on the Konami 007324.

 

[Monstermug]

Sorry I don't understand. Are the data lines pulsing or not? That konami resistor fails quite often but all data lines being held high is quite odd.

Also I wouldn't touch the sockets for now if you have already tested continuity on all the pins. Lets try fix it first before you end up making more of a mess with the sockets. Usually I only socket cpu ram roms and proms (most of the time). TTL's just go straight in (most of the time) but thats only because I can remove them effortlessly. But it is good practice to add sockets if you cannot. No point in adding more weak points of failure to the pcb.

Can you also probe the LS245 buffer as well for me please? Set your probe right whatever probe you use first.

 

[Supergun]

Sorry I don't understand. Are the data lines pulsing or not? That konami resistor fails quite often but all data lines being held high is quite odd.

Well, now I’m unsure as to wether or not I’m probing it correctly then? When probing a CPU, I set the probe to CMOS and when probing a “TTL family IC” I set the probe to TTL/LS. However, the other setting of “NORMAL/PULSE” is where I am unsure as to how it should be set here?

And with regards to the Konami 007324, what exactly is that? It’s not a CPU or a TTL chip. So how should the probe be set?

Can you also probe the LS245 buffer as well for me please? Set your probe right whatever probe you use first.

Yes, doing this now. Will post results shortly! Thank you again.

Here is the LS245

 

[Monstermug]

its a custom resistor pack used to interface ttl to cmos i think.

Or if you can give me some time I can find my board and see what's going on. But that may take many days if not weeks. Having a hard time finding any spare time these days and besides, my bench still has a friends PC10 on it which needs my attention not to mention a fairly large que after that has been fixed.

 

[Monstermug]

Can you set your probe correctly on the data input lines of the 245s again please?

Sorry had to edit this post as i you didn't label the chip legs again. The 245 does look like its working and also the Konami resistor pack but to be sure can you probe the address lines 2 to 9 again for me please?

While you at it can you probe the the 6264 Ram closest to the CPU and also probe the main program rom as well please I believe that's 15C and 16C ? Might give us some more clues. I'm pretty sure the 6264 is the main program ram but it could also be the 6116 ram as well so probe that too please. I'm just going by your pcb pictures btw.

 

[Supergun]

I will start with redoing the LS245.
Here are the results.

Same as before, but this time I also did PULSE mode. Same results either way.

 

[Supergun]

Now here is the rom at position 15C. This eprom is 512kb in size.

Pin 22 is most curious as the logic probe gives no readings whatsoever. But, it is extremely sensitive. Such as, every time I make contact with it and the probe, it flickers and changes the garbage displayed on screen.

Also, in PULSE mode, pins 1, 26, and 27 make the probe pulse go crazy on and off very quickly.

 

[Supergun]

Here is the rom at position 16C. This eprom is only 256kb. And from what I can see, does nothing strange.

One other thing to note. Is that during the proving of the 28c512 eprom, twice now I’ve been “scared” by the game suddenly talking. (Yes, as in typical Konami 80’s speech synthesis like Track & Field has) It caught me off guard and startled me, but, it’s at least a sign of life, I think.

 

[Monstermug]

Think you may have bad sockets then.

Edit:

74LS245 looks okay assuming the address inputs can't be read properly with your probe as the outputs are there.
M27C512 looks okay except for G which is the output enable pin. It's active low. Test resistance on the chip pin and the connection it leads too.
M27C256: Check pin 25 (A and see where this leads from. It's an address input and should be toggling. Worth testing resistance as well on it to see if there is a good connection to where it leads from.

 

[Supergun]

Ok. So pin 22 (G) of the M27c512 connects to pin 8 of a Fujitsu LS32 at 17A

The “connection” is very weak. When I run a jumper wire it improves and affects the on screen garbage.

And pin 25 (A8 of the M27c256 connects to pin 3 of the LS138 at 14A.
(which we replaced earlier)

 

[Monstermug]

1. LS32@17A: Probe pins 9 and 10 and where do connections come from? Also find out why the connection is weak. Is it the socket?
2. That LS138@14A: Double check if you got probe set correctly and confirm if all pins are high except ground on pin 8? You mentioned pins 4 and 5 connects to Konami Custom chip 007842? There isn't such thing. Did you mean the 007327 ? That's a custom dac/pallete rgb converter in a complete package. I'm not sure why a dac should be driving anything to be honest. Did you resistance test this connection?

Can you probe that and see what that's up too? the G inputs should really be pulsing in tandem with the a-c select inputs to make all the outputs on Y0-7 pulse too.

Schematics for the 007327 can be seen here:

https://www.jammarcade.net/images/2018/07/007327_007593_schematics_compare_.jpg[img]

Sorry it's It's turning out to be more complex than I had imagine a remote help would be (hard enough having the pcb in front of you). Even if we don't find the fault this way, at least you would have learnt something from this process and how I am approaching the repair. I am sure we find it unless there isn't any more leads and further investigation would require equipment that you don't have. I mean it could be just a bad clock signal but you don't have a scope to test that or some of the signals isn't correct ttl levels. Due to the errors so far, might be worth just double checking if your records are correct as well. 1 error could lead us on a wild goose chase. Good thing with this hardware is that there is only a handful of ttls with the majority of work handled by customs.

 

[Supergun]

That’s quite a bit to unpack there. It will be awhile before I can respond to every point, but I promise that I will. I’ll start with the first here.

“…1. LS32@17A: Probe pins 9 and 10 and where do connections come from? Also find out why the connection is weak. Is it the socket?…”

The connection was weak due to corrosion on the legs of the chip which were barely making intermittent contact at that pin 8. Some pins were not even giving me readings on the logic probe. I already removed the chip, mapped all the traces, installed a socket, and tested continuity on all the mapped traces which are all confirmed good now.

The connection between pin 8 of this LS32 at 17A and pin 25 (A8 of the 27c256 at 16c is now perfect and steady. So this was CLEARLY a problem connection earlier.

I then installed a new LS32 chip and was able to get better readings.

Lastly, pin 9 of the LS32 connects to pin 8 of an ALS04 at 16B and pin 10 connects via a parts side trace underneath the LS32 chip to pin 4 of that same LS32 and then goes up and to the left all the way along the pcb’s border and it goes through a gauntlet of places and I haven’t found it’s end yet.

 

[Monstermug]

In the meantime can you confirm pins 25 (A of 16C is now toggling? LS32 looks like it's doing its job. Not all functions on TTL's are used hence the n/c If its toggling then that rom is doing its job fine now assuming the data on it is valid and verified with mame indent.

Now just to get the output enable on Rom 15C to work. In the mean time can you probe the ram? I think its the 6264 Ram according to mame source code. Pin 1 is n/c or should be. Doesnt matter what logic level it is. All address and data lines should be toggling. Gnd low, Vcc high. CE OE and WE should be toggling as that controls if the ram is being read or written too, If any is not as it should be investigate where they lead to and why its not doing its job. Thing about ram is sometimes they try to jump back to life clinging on to life so some ram faults may exhibit intermittent faults.

EDIT:

Also I was concerned with the 245 address lines being all high (initially i had assumed your probe could not handle cmos level properly and that the high was actually pulsing) and realised that output enable is active low which means the 245 is effectively disabled. Where does this go? So cannot rule out that konami resistor pack just yet as if my hunch is correct the main cpu data lines have nobody to talk to.

What are the changes on screen now that 15C is working