The Sustain I bug is the exact problem I described on here a while ago, and hardly anyone seemed to have come across it before, which surprised me as I've seen it on just about every CS50/60/80 that I've worked on.
This is where the oscillator voltages are disconnected from the KAS and held by the capacitors on the SH card during the Release cycle. Gradually (or quickly in some cases) the capacitor voltage leaks away and the pitch changes. The 'fix' that I came up with was to directly link the 4066 output, the capacitor and the 3140 input on each voice with wire links away from the PCB. It seems that the surface of the PCB goes slightly conductive with age, possibly due to airborne contaminants.
I have a photo of the 'fix' I used if anyone is interested.
Lucid.
----- Original Message -----
From: Mike
To: yamahacs80@yahoogroups.com
Sent: Monday, July 25, 2011 6:49 AM
Subject: [yamahacs80] Re: KAS bug examples
I'm planning on putting all of the troubleshooting results on my own website in the near future because I need full HTML to reference pictures, audio, and maybe some math. I need a better scope to catch everything related to the bugs on my CS-80 even though I already know temporary solutions for both of them. I can't really justify or afford an amazing deep memory scope until at least sometime next month and I still might not do it because, well, it's gonna cost at least a few thousand US $.
The two bugs that I'm going to take a crack at when I make some posts on my website are the SH pitch drop or raise in Sustain I bug and the Volt/Oct Scaling bug. The SH pitch drop or raise slowly changes the pitch of your release tails when you have the sustain set high. The Volt/Oct Scaling bug doesn't quite scale the YM26700 properly when pin 12 spits out the octave data...I think; this is what I need better test equipment for. Like I said in the previous post, when you hold down one key and then play another in a different octave, you can hear the held down key jump a few hertz.
I'm assuming that you and Fernando have run into the same octave scaling problem? If so, please describe your problems as well as your theory related to the problem. A temporary fix is to slow down the Master Clock by replacing the 100pf cap in the oscillator loop with a 180pf cap. I'm not going say why I think it works yet because all I have is a theory without the proper test equipment. This will slow your clock down to about 70K and also clears up about half of the double triggering. Yes, it slows down the key trigger time slightly which seems like a bad idea at first, but, honestly, my whole machine feels more stable and most of the double triggering is gone.
My machine is now tolerable but not quite right after the clock mod. The held down bass key went from about a 4 hertz drift on octave changes in a higher register to about 3/4 hertz drift.
Sorry, but I've been busy and haven't had the time to mess with my CS-80 or get further with these issues.
-Mike
--- In yamahacs80@yahoogroups.com, "Max Fazio" <faxiomas@...> wrote:
>
> Hi again
> This is for Mike: have you managed to record any example of the supposed KAS' bug?
> Let me know!
> M
>
> [Non-text portions of this message have been removed]
>
[Non-text portions of this message have been removed]