Archive of the former Yahoo!Groups mailing list: Discussion about the Korg PolySix synthesizer

previous by date index next by date
previous in topic topic list next in topic

Subject: Re: [PolySix] Re: Polysix mods

From: w.james.meagher@gmail.com
Date: 2013-07-10

Awesome.

Sent from my iPhone

On 2013-07-10, at 9:26, "chipaudette" <chipaudette@yahoo.com> wrote:

> In my searching of the archives for different purposes, I came across this sad and lonely post from 2005 (see below)...toorglick was looking for a mod to "open up" the sound of the Polysix. He was looking for something that increased the high-frequency response and that made the Polysix feel snappier.
>
> In my opinion, if you listen to the raw output of the Polysix voice board (ie, if you listen to TP1 on KLM-366), the voices have plenty of high frequencies and are plenty snappy. The main output of the synth doesn't sound quite as good as TP1 with regards to the openness of the raw waveforms. The problem (if you think that there's a problem) are the side-effects of the "noise reduction" VCF that was added after the effects section on the KLM-368 board. If you don't know what I'm talking about, this post talks about the circuit:
>
> http://synthhacker.blogspot.com/2013/02/mystery-circuit-polysix-post-effects.html
>
> In my opinion, I would describe this circuit is a VCF that opens and closes based on the amplitude of the audio signal coming from the voice board. When the audio signal is small (or silent), the VCF is closed, which cuts away the hissy noise coming from the effects board. When the audio signal is strong, the filter opens up to allow the audio signal (and the noise) to pass. In effect, it's trying to work like a noise gate.
>
> The problem is in how fast the amplitude detection circuit can tell that a signal is present (ie, how fast can it respond to you pressing a key). It turns out that it can take a few milliseconds. This means that it takes a few milliseconds before the filter opens up to let the high-frequencies through. As a result, the attack envelope feels slower. The actual envelope on the voice board might be fast, but this post-effects VCF can attentuate the start of the signal making it feel slower.
>
> A second side-effect of this VCF is that, it sometimes isn't stimulated hard enough to open up all the way. So the high frequencies can be attenuated a little bit.
>
> If you'd like to remove this effect, one "cure" was discussed at the end of the blog post above. It is to add a jumper (or solder a wire) to apply +15V to the base pin of Q14 on KLM-368 (personally, I jumper from the +15V side of R125 and connect to the Q14 side of R89). By fixing Q14 at +15V, it basically removes the effect of the amplitude sensing circuit and keeps the "noise reduction" VCF open all the way all the time.
>
> The benefits of this mod is that (1) the synth feels a good bit snappier and that (2) the high-frequencies are always present. The downside of this mod is that you hear the hissing noise from the effects section ∗all the time∗.
>
> If you can live with the hiss (I can), you may like this mod. Plus, if you use clip-leads to make the connection, it is completely and easily reversible.
>
> Chip
>
> --- In PolySix@yahoogroups.com, "toorglick" <toorglick@...> wrote:
> >
> > Greetings again!
> >
> > It's been a year or longer, but I finally got around to purchasing
> > another Polysix.
> >
> > I'm interested in some mods for the thing, especially...
> >
> > "there's a mod to open the rather dark sound, changing a few caps at
> > filter/vca, a mod to make it louder w better S/N. and finally, a
> > resistor replacement to make the envelope much faster."
> >
> > I'm planning on shielding the effects board, but I haven't seen these
> > mods mentioned elsewhere other than in this quote (which I am
> > following up on with the person quoted).
> >
> > It'd be great to "open up" the sound a little. I'm also interested
> > if there's a way to make the ATTACK of the VCA ADSR envelope snappy -
> > leaving the rest of the envelope times alone.
> >
> > Any input would be appreciated!
> >
> > Thanks,
> > Rich
> >
>
>


[Non-text portions of this message have been removed]