Page 2 of 2

Re: Korg Poly-800 Chorus mod

Posted: Wed Jun 05, 2019 3:18 pm
by Mooger5
Your questions are pertinent. Would like to know the source of those mods.
Q8 is a constant current source. D6 is there for thermal compensation. Removing it from the circuit would make the current fluctuate, affecting the saw wave.

I *think* I got it (someone please correct me otherwise):
bbd.jpg
Q9 is always either on because the base is positively biased by R70, or off because of the comparator IC8b.
When Q9 is on it shunts C46 and a current flows immediately from -D to the non-inverting signal input of IC8b at pin 5. At the same time a constant positive current also flows from Q8 to the same input, therefore what arrives is close to zero.
Now at pin 6 we have the reference voltage for the comparator. It is continually changing, but is kept above zero volts. Since pin 6 is the inverting input, the difference in potential between pin 6 and 5 makes the comparator go into negative saturation. This negative output is nulled by R70, opening Q9. This in turn stops the current at -D from flowing. It will have to build up through C46 that starts charging. This means the positive current from Q8 is at full blast. A positive pin 5 and 6 stops the comparator, triggering Q9 again, until C46 is fully charged and pin 5 returns to zero, and so on.
As the reference voltage is modulated by the LFO, the two positive charges at pins 5 and 6 coincide sooner or later, affecting the rate at which the comparator stops saturating and therefore the ramp time. If we offset the LFO output to a higher voltage, pins 5 and 6 will coincide earlier. R57 and R58 form one of the voltage dividers mentioned earlier and a small value pot in their place could probably do it (along with limiting resistors). Are you sure R67 and R68 you read elsewhere were not a typo?

Re: Korg Poly-800 Chorus mod

Posted: Thu Jun 06, 2019 4:21 pm
by crafter57
Mooger, your analogy makes total sense for that sawtooth wave generator circuit. Those infos were the only thing I found while digging the net and yes there could be typos it could have been R58/R57 instead who knows? I guess Korg figured out the fixed resistors for that constant thick chorus when turned on, but I'm looking at many different guitar chorus/flanger circuit diagrams/schematics and it seems that you can vary the rate/depth either at the LFO or at the MN3102
I have a spare chorus board, I'm thinking may to breadboard the circuit and do some testing. It would be nice to have a circuit simulator app/pgm to try this out tho'

Re: Korg Poly-800 Chorus mod

Posted: Thu Jun 06, 2019 5:29 pm
by Mooger5
It's just that the LFO time and the 'nominal' delay time are two separate things. No matter how much the LFO is tweaked, the chorus effect will never sound like a flanger unless the nominal delay time is adjusted to the adequate value. The one in the Poly800 chorus is fixed and is not in the flanger range. It has to be modded.

Re: Korg Poly-800 Chorus mod

Posted: Thu Jun 06, 2019 5:47 pm
by Mooger5
To shift the master delay time down to flanger range, the MN3102 datasheet shows the minimum and maximum values of resistance 'allowed' which could be adjusted with a potentiometer. A capacitor of a suitable value is also added. Pin 7 though, is tricky... anyway, this involves soldering to the pins of MN chip.

I think replacing R57/R58 divider with a pot is much more practical. A pot in place of R69 could also do it.

Re: Korg Poly-800 Chorus mod

Posted: Fri Jun 07, 2019 12:14 am
by Mooger5
And I got it all wrong hahaha.
See IC8a, the final comparator: the reference voltage set by the divider at pin 2 is -4.5V! There's no way "my" saw wave reached that low, it was +4.5 to zero. So my calcs were nonsense LOL.
If the ref voltage is -4.5 at pin 2, the signal voltage at pin 3 must at one time be identical for the comparator to produce zero.
Q8 - need to know its internal resistance. Because R69 has the same value 100k as R72 and since there's +D and -D, if Q8 is 4.7k when on we have a voltage divider that mimics R72 and R71. So there we'd have at one point in the cycle -4.5V...

More later.

Re: Korg Poly-800 Chorus mod

Posted: Sun Jun 09, 2019 9:20 am
by Mooger5
Yep the saw wave's amplitude varies from about -4.5 to a positive value. R69 100k limits the flow of the positive current. So, a value less than 100k will make the ramp rise earlier, shortening the wave period which means of course increasing the frequency.

Re: Korg Poly-800 Chorus mod

Posted: Tue Jun 18, 2019 1:05 am
by crafter57
Hi Mooger, sorry for delay.. I tried a 10k limiting resistor + 100K pot in place of R69 and it made no difference.