Korg DW8000 built in Chorus effect?

Discussions about anything analog, digital, MIDI, synth technology, techniques, theories and more.
User avatar
madtheory
Supporting Member!
Supporting Member!
Posts: 5155
Joined: Mon May 01, 2006 12:45 pm
Real name: Tomas Mulcahy
Gear: Flangebeast Mk1, Plonkotron, Morovdis Arpeggiator, Maplin My First EQ, Jeff Wayne Thunderchild rack, Thermostat, Buck Owens' Moog.
Band: Minim
Location: Cork, Ireland
Contact:

Re: Korg DW8000 built in Chorus effect?

Post by madtheory » Mon Jul 31, 2017 10:33 pm

OK, I was looking in the wrong place. Service manual is not the place to look. Studied the data sheets properly this time LOL. The MN3209 BBD uses a clock driver chip designed for BBD use, the MN3200. It produces two clock signals, identical except one is 180 degrees out of phase. The MN3209 accepts two clock inputs, and has two outputs. The result is a bit like an autopan but not ever in phase, so the signal doesn't cancel when summed. It gives a pleasant wavering effect. The sawtooth pattern on the goniometer is likely caused by the clock pulse being softened by stray capacitance, maybe :)

The Korg chorus circuit is the same configuration as the app cct in the data sheet, but with different R values. That's where my cct theory fails me, no idea why they used different values lol.

Mooger5
Expert Member
Expert Member
Posts: 1459
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Korg DW8000 built in Chorus effect?

Post by Mooger5 » Tue Aug 01, 2017 11:52 am

The circuit in the service manual is pretty much like the recommended application in the 3209 datasheet, with a compander added for noise reduction.

Image

The dry signal goes through C26 for AC coupling, then it splits through R26 and R27. R26 leads to the summing node in IC2 pin6. The feedback resistor is R24. To determine the gain, we divide R24 by R26. 47/39=1.2. So if we take an abstract value say +5 and multiply it by 1.2, we get 6. Since it´s an inverting input, the +6 turn into -6. After that it goes to the volume control slider VR2. There´s a switch preceded by R6. It´s the muting relay. R6 is only there to prevent the relay from shorting the output to ground. After that it goes to the Left channel output. Pretty simple.
Now for the Right channel dry signal:
R27 goes to the summing node of the other opamp in IC2, whose feedback resistor is R22. This resistor should have an identical value to R24, but it´s 75k instead of 47k. It will have a higher gain, as dividing R22 by R27 is 75/39=1.9. +5*1.9=-9.5, a negative value because it´s an inverting opamp as well. However, the output of the first opamp also goes to the summing node of the second opamp, through R25. The dry signal for the left channel was -6, right? Now we also calculate the gain for that portion of the signal, which be R22/R25=75/100=0.75. -6**0.75=-4.5, but since it again is going through an inverting opamp, this time the negative value turns positive =+4.5. We now sum these +4.5 to the other value at the same opamp stage. -9.5. We get -5. These final -5 also go to the Right channel output section, identical to the Left channel. We have thus -6 on the left and -5 on the right.
Wet signal:
Unlike the dry, the wet signal is not split. We have C20 for AC coupling, then straight to the summing node of the first opamp, via R23. Feedback resistor is the same R24, procedure is the same: R24/R23=47k/47k=1. A gain of 1. If we attribute say a value of +5 to the wet signal, we have -5 at the output of that inverting opamp. After that, it also goes directly to the left output, mixed with the -6 of the dry signal. So we have -11.
Like the dry signal it also goes to the second opamp stage via R25 and the feedback resistor again being R22. Same as usual to determine the gain: R22/R25=75k/100k=0.75. -5*0.75=-3.75. Since it´s inverted at the output the result is +3.75 at the right channel. Summed to the dry signal which is -5, we have -1.25 total wet+dry signal at the right channel.

-11 for L and -1.25 for R. Now when the wet signal by its very nature becomes out of phase and turns from +5 to -5 we have:

-5 at R23, output of the wet siganl at the first opamp stage will be +5 (unity gain, phase reversed). Mixed with the -6 of the dry signal we have -1 total signal on the left out.
+5 entering the summing node of the second opamp stage, through R25 and we have +5*0.75=+3.75. Inverted, -3.75. Mixed with the-5 of the dry signal we have -8.5.

-1 for L and -8.5 for R.

Hence the auto-panning and the imbalance between channels as well.
Herrare umanum est.

Mooger5
Expert Member
Expert Member
Posts: 1459
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Korg DW8000 built in Chorus effect?

Post by Mooger5 » Tue Aug 01, 2017 12:27 pm

Another goofup when I tried to simulate the "monofication" with a differential amplifier. Of course the outputs must be added, not subtracted.
So for the mono dry signal which is phase coherent on both outputs we have -6 (L) + -5 (R) = -11.
For the wet signal: -5 (L) + 3.75 (R) = -1.25. Or, depending on its phase: +5 (L) + -3.75 (R) = +1.25
Herrare umanum est.

User avatar
madtheory
Supporting Member!
Supporting Member!
Posts: 5155
Joined: Mon May 01, 2006 12:45 pm
Real name: Tomas Mulcahy
Gear: Flangebeast Mk1, Plonkotron, Morovdis Arpeggiator, Maplin My First EQ, Jeff Wayne Thunderchild rack, Thermostat, Buck Owens' Moog.
Band: Minim
Location: Cork, Ireland
Contact:

Re: Korg DW8000 built in Chorus effect?

Post by madtheory » Thu Aug 03, 2017 12:48 pm

So, to clear it all up:
Korg DW6000: BBD chorus with phase shifting stereo effect. Effect is summed in mono.
Korg DW8000 and SDD2000: DDL chorus with no phase shifting, only a simple polarity flip. Effect is lost in mono.

While we might assume the BBD is nicer because it is analogue, the digital SDD2000 is highly regarded by guitarists. Both are nice... if you want both, I'd suggest SoundToys EchoBoy, superb modelling of a wide range of echo and chorus devices.

Mooger5
Expert Member
Expert Member
Posts: 1459
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Korg DW8000 built in Chorus effect?

Post by Mooger5 » Wed Aug 09, 2017 10:45 pm

Mooger5 wrote:
madtheory wrote: The Juno chorus gets the effect in a much better way, it's basically two slightly different modulated delays, right? Sum that to mono and it's still there :)

I've been trying to find what the delay time (s) and LFO rates are for the Juno, no luck yet.
More on this later, sorry.
Just checked. The Juno chorus is two identical, phase coherent delays and a single LFO, whose split output modulates one delay in phase and the other out of phase. So when one delay time gets increasingly shorter, the other gets increasingly longer.

The LFO rate I think is determined by the time constant of C3 (100n) and R8 (2.2M) around IC1b. According to the ElectroDroid app it´s 723.432 miliHerz. Or one cycle every 1.382 seconds...
Herrare umanum est.

User avatar
madtheory
Supporting Member!
Supporting Member!
Posts: 5155
Joined: Mon May 01, 2006 12:45 pm
Real name: Tomas Mulcahy
Gear: Flangebeast Mk1, Plonkotron, Morovdis Arpeggiator, Maplin My First EQ, Jeff Wayne Thunderchild rack, Thermostat, Buck Owens' Moog.
Band: Minim
Location: Cork, Ireland
Contact:

Re: Korg DW8000 built in Chorus effect?

Post by madtheory » Thu Aug 10, 2017 7:58 am

Awesome! Thank you. So 0.72Hz in SI units :) How is the delay time set?

Mooger5
Expert Member
Expert Member
Posts: 1459
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Korg DW8000 built in Chorus effect?

Post by Mooger5 » Thu Aug 10, 2017 11:53 am

The calcs are wrong, as I had forgotten about the Chorus I and Chorus II different timings.
The selecting is done by the FET bypassing the 2M2 resistor and shorting the 680k to ground.
As I´ve already burned a few neurons trying to calculate the values without success :( I´ll have to cheat here by measuring the timings with the scope. Later.

What sets the delay time, I don´t know. Will have to look at the MN3001 datasheet.

How important is it to you, BTW?

I´m more concerned about the parasitic noise this chorus generates...
Herrare umanum est.

User avatar
gs
Active Member
Active Member
Posts: 350
Joined: Sun Sep 19, 2004 6:10 pm
Location: USA

Re: Korg DW8000 built in Chorus effect?

Post by gs » Thu Aug 10, 2017 2:37 pm

Jabberwalky wrote:The Korg DDL is creamy, and amazing. I'd almost go as far to say it's half of the Dw8k sound. The same delay was also in the much maligned Korg Ds8. As far as I know, the same delay is a rack unit, the Korg SDD2000
It is also found in the DSS-1, with an added bonus: you can run the two DDLs in either parallel mode (which the DW-8000 does by default), or in serial mode (unique to the DSS-1, and maybe the DS-8 as well; I don't know for sure). In serial mode you get a mono signal, but with some truly amazing DDL effects not possible on the DW-8000 (including more "chorusing" options). I believe there are enhanced modulation options as well.
Kurzweil PC3, Yamaha MOX8, Korg Z1, Alesis Ion, Alesis QS8.2, Kawai K3M

User avatar
madtheory
Supporting Member!
Supporting Member!
Posts: 5155
Joined: Mon May 01, 2006 12:45 pm
Real name: Tomas Mulcahy
Gear: Flangebeast Mk1, Plonkotron, Morovdis Arpeggiator, Maplin My First EQ, Jeff Wayne Thunderchild rack, Thermostat, Buck Owens' Moog.
Band: Minim
Location: Cork, Ireland
Contact:

Re: Korg DW8000 built in Chorus effect?

Post by madtheory » Wed Nov 08, 2017 6:53 pm

Mooger5 wrote:The calcs are wrong, as I had forgotten about the Chorus I and Chorus II different timings.
The selecting is done by the FET bypassing the 2M2 resistor and shorting the 680k to ground.
As I´ve already burned a few neurons trying to calculate the values without success :( I´ll have to cheat here by measuring the timings with the scope. Later.

What sets the delay time, I don´t know. Will have to look at the MN3001 datasheet.

How important is it to you, BTW?
Somehow missed this post sorry. Thanks again for burning neurons on this!

How important? Very. I want to try simulating this chorus with other delay lines. And I want to see if the TAL Chorus plugin has it right.

User avatar
iphoenix
Junior Member
Junior Member
Posts: 170
Joined: Thu Aug 27, 2009 9:49 am
Real name: Paul
Location: Ireland

Re: Korg DW8000 built in Chorus effect?

Post by iphoenix » Wed Dec 27, 2017 5:32 am

Ok , the digital delay in the DW8000 is much closer to the Korg SDD1000 than any other SDD delay as far as I m aware.

I have had the SDD2000 since the late 80s & know it very well.

I also have a DW8000 & know the delay within bears no resemblance whatsoever to the SDD2000.( apart from the fact that it is a DDL & hence obviously can do echo,chorusing & flanging fx )
The DDL in the DW8000 sounds a lot more 'vintage' & more like an analogue delay in character than the cleaner more 80s sounding SDD2000 .
This fits with the SDD1000, which sounds more late 70s to me , but obviously it is one of the earliest Digital Delays . I think the SDD1000 came out just before the DW8000 & the SDD2000 just after it ,with perhaps a year in between each release.

I think that the SDD1000 is more or less identical in its electronics/ setup & specifications /parameters to the one in the DW8000 (apart from the sampling option).

I also have an SDD1200.

The SDD delays are actually all very different in character.

I love the Korg SDD delays. ( & the DW8000)
Last edited by iphoenix on Tue Jan 23, 2018 9:42 pm, edited 1 time in total.

User avatar
Jabberwalky
Synth Explorer
Synth Explorer
Posts: 2143
Joined: Fri Apr 01, 2005 6:03 pm
Gear: A hybrid of vintage and modern junk
Band: Variar
Location: Pgh, PA
Contact:

Re: Korg DW8000 built in Chorus effect?

Post by Jabberwalky » Thu Dec 28, 2017 12:04 am

Thanks for clarifying that! I'm a big fan of those delays and the dw8k as well. If that's the case the Korg Ds8 may also have the sdd1000 in it.

grenert
Newbie
Newbie
Posts: 15
Joined: Wed Jun 25, 2014 6:43 pm

Re: Korg DW8000 built in Chorus effect?

Post by grenert » Thu Dec 28, 2017 2:55 am

Does anyone have the SDD-1000 service manual who could send it to me? I have the DS-8 and schematics and can compare the two. The DS-8 is pretty primitive (25 kHz sampling rate, 720 msec max delay), so it could well be the same. The overall design, with a compander, is typical for early delays and is very similar to the supporting circuits of analog delays.

Post Reply