What's new

ekorz

Multi Boyz 4 Pi
Legendary
Multi Boyz
Joined
Aug 21, 2016
Messages
5,187
Reaction score
7,770
Location
Boston, MA
I've been working with Psikyo boards a lot, and recently noticed the audio quality differed greatly between the two PS5-pcb-layout games, Gunbird 2 and Strikers 1945iii/1999. Gunbird 2 suffers from what seems like video interference. This issues has been mentioned before on AP and elsewhere, but I wanted to pull this into its own thread since I've made some progress to improve it.

While recapping both boards I noticed the audio sections were populated differently, Strikers routing to a Yamaha YAC516 D/A converter and Gunbird instead using a YAC513. Both use a JRC 4741 op amp. Datasheets are linked in the previous sentence. Trouble is, I'm not capable of understanding or diagramming this circuit to see what could be fixed. That didn't exactly stop me, but I don't think my results are perfect!

Here's a Strikers 1999, you can see the empty YAC513 on the right hand side.

jMgnDFa.jpg


And a close-up of Gunbird 2, where those are populated (and you can't see it here but the YAC516 is not populated)

vWx8mgf.jpg



Armed with little knowledge, I just took a shot and replaced C22 and C19, previously 222K, with 104K film caps that were used everywhere else. That made a large improvement to me. I also installed a C0805,100pF±5% 50V C0G on C12. Possibly that's helpful, but possibly not -- you see I took off my C12 and damaged it while trying to take its value, so I don’t actually know what should be on there. I saw that value in the op amp datasheet so I went for it.

More photos: https://imgur.com/a/WmpEs9Z

So while I'm happy with the results, it would be awesome for someone to look at the chips involved and see if there's an actual factory error... and if there's something more suitable for a solution. I'm convinced that the YAC513 circuit is misbehaving. Anyone with the pcb want to take a look? Anyone without the pcb want me to get more photos of anything? If helpful I documented the electrolytic caps already too.
 
I finally dug out my Gunbird 2 yesterday but I haven't had a chance to plug it in to see if it has the same audio issues you describe.

if you look at pdf page 4 of the YAC513 datasheet it has the sample circuit which should tell you the value of all the associated resistors and caps that they recommend (the table at the bottom lists the cap/resistor values that aren't on the diagram). the first thing to do is determine if the game PCB uses the same value resistors and caps and is routed the same way as that sample circuit.

For the most part on these audio circuits I see:
electrolytic caps for polar caps
film caps for non polar caps on the output tied to ground
ceramic caps for all other non polar caps

this would ensure that the A/D conversion is done cleanly, it looks like the sample circuit also explains how they expect the OpAmp to be wired in conjunction.

If you're getting any sort of hum or buzz or clipping then there's a chance that you'll want to check out the actual audio amp circuit, maybe there are differences with that as well?
 
Looking at that page4 diagram they look fine then, I think. It shows 2200pf for those caps (CSH, 222K) at 5v, and 360 ohms for the resistance, and 361 is installed. That matches what’s on the board.

I’ll have to see where pins 15 and 16 go since they hit vias in my photos. They should have caps on them too.

Weird that the ones I changed could have such a positive impact then!

The op amp datasheet had a few curves driven with CL (thought that was input capacitor) at either 50pF or 100pF, which is why I used that for C12. I tried both and preferred 100pF.
 
Last edited:
My GB2 is more than 40 days late from Japan, it's out of New Jersey, so when it shows up (if?) I'll test mine as well. Do you think I can hear the buzz via a HAS and OSSC or a HAS and CRT, or should I test in a JAMMA cab?
 
I know nothing about audio circuits, but if you need yet another person to test something, let me know?
 
@xtrasmiley It was easy to hear the buzz in a cab and with a supergun, for me. @rewrite you can test if your audio sounds like crap too ;)
OK, good, I'll test ASAP. Right after I posted I got a text saying my package was finally in IL, so in a day or two. I'll test first thing, and assist in whatever way I can.
 
I spent some time looking at this earlier today.

My Gunbird 2's audio isn't terrible but there's like a high pitched flutter that you can hear in the silent parts, and it seems like feedback from the video circuit as it changes in pitch with the screen brightness.

looking over the circuit for the YAC513 and the JRC 4741, they seem spot on, even the cap values used are exactly as recommended.


One really interesting thing though... is that the YAC513 is a 2-channel A/D converter; meaning stereo. and while it can be run in either single channel or 2-channel mode it's actually setup for 2-channel mode. Like me you're probably thinking "wow is it possible we could do a Stereo Mod on this board?" but no.


1 output channel feeds to pin 11 on the JAMMA edge and the other feeds into the amplifier circuit. Pin 11 is a rarely used feature of the JAMMA protocol for line-level audio


I don't really see any obvious faults in how the circuit is built or the component used.


interestingly the YAC516 is also setup as a 2-channel AD converter, they're both fed from the same pins on the YAMAHA Audio CPU, it's entirely possible that gunbird2 would work using the YAC516 circuit.
 
Great thread and findings @ekorz! I'm inclined to think the Strikers circuit might work if copied/moved over to Gunbird 2 as well.

While I would like to fully resolve the issue and be able to get fixed amped audio from the PCB itself, it seems to me that doing a line level audio mod may be a decent way to further isolate / pinpoint the problem further.

Looking at the YAC513 datasheet, it should simply be a matter of tapping pin 16 (AOUT) and pin 2 (GND) and writing to an rca/phono jack to do a line level mod right?

Or maybe the audio out pin should be pin 10 (CH1)?

Edit, just noticed what @twistedsymphony wrote:

1 output channel feeds to pin 11 on the JAMMA edge and the other feeds into the amplifier circuit. Pin 11 is a rarely used feature of the JAMMA protocol for line-level audio
Cool! Never heard of that JAMMA feature before.
 
Last edited:
Cool! Never heard of that JAMMA feature before.
I learned about it years ago when I found a copy of the official JAMMA spec and noticed that a lot of the "N/C" pins you find on JAMMA pinouts actually have official functions that mainly just go unused.

I think I've only noticed 1 or 2 other PCBs that actually use the line level audio pin, it's a shame really.
 
When I saw connections to that pin I thought it was a factory error of some kind for sure. Like they expected it to ground or something.

Anyway it sounds like we're saying we can rip off the Gunbird 2 components and just install the circuit the same as Strikers. I may just give that a try for fun since I have almost all the components needed already. Just have to decide to buy a YAC or see if I have one on a scrap board or something. Or just borrow one from a working Strikers for a little bit...
 
I'm interested too. My board even misses one 100uf 16v cap from the upper right corner but still plays just fine. I was going to re-cap it anyways but perhaps I wait your results! Not sure if I remember correctly, but Tengai might have this same static silent buzz problem too. Should plug it in and listen.
 
I'm interested too. My board even misses one 100uf 16v cap from the upper right corner but still plays just fine. I was going to re-cap it anyways but perhaps I wait your results! Not sure if I remember correctly,
that's just a general cap to help cut down noise in the power around the CPU and ROMs, it's unrelated to audio. I swear half of my SH2 PCBs are missing this cap since its 3x the height of any of the surrounding components.


Not sure if I remember correctly, but Tengai might have this same static silent buzz problem too. Should plug it in and listen.
Good memory, now that you mention it, if I recall my Tengai sounds like a buzz-saw as well.
 
OK, here are my results. Using a HAS 4.0 to RGB2COMP to Sony Wega CRT.


Gunbird 2, perfect sound, like it's an emulator. Even with the TV cranked up, or the PCB cranked up. No issues at all.

Tengai, fucking awful. I can't believe I didn't notice it before. Horrible extra static. I think maybe I noticed it and thought it needed a cap kit. Bought from someone here, and don't think he mentioned it, but I didn't notice it, so maybe it's thanks to this fucking thread I can't un-hear it. It's terrible.

Pics of both PCBs and up close of the sound area. Note that my GB2 had some silver thing replaced and has a piece of paper stuck on top of a silver part (crystal?) that I have no idea what it means.


Tengai (thanks to you I can't play this shit now, thank you. Now find a fix, asshole!)
IMG_8022.jpeg




Gunbird 2 (part in center with paper attached)
IMG_8021.jpeg


Edited to say I don't know if they silver part (a timing crystal?) was replaced for sure, just that there is a paper stuck to it).
 

Attachments

  • IMG_8024.jpeg
    IMG_8024.jpeg
    702.9 KB · Views: 198
  • IMG_8020.jpeg
    IMG_8020.jpeg
    777.5 KB · Views: 221
Last edited:
Tengai, fucking awful. I can't believe I didn't notice it before. Horrible extra static. I think maybe I noticed it and thought it needed a cap kit. Bought from someone here, and don't think he mentioned it, but I didn't notice it, so maybe it's thanks to this fucking thread I can't un-hear it. It's terrible.
You're welcome :thumbsup:
 
For funzies I replaced my op amp and Yamaha, and returned my capacitors to stock, but I still hear the high pitch video/ram-access noise chirping away. I think overall the audio is a bit more clear though, so maybe old components were contributing to my issues. Still not perfect though.
 
Kind of an off-topic, but how do you get to "dip switch" menu on Gunbird 2? There is no physical buttons or switches on the pcb?
 
Back
Top