[sdiy] Korg MS-20 VCO woes

[Simon Brouwer]

Hi Aaron,

To find out why it does not want to oscillate I would disconnect the FET
and provide a simple current source using a resistor to ground. That
leaves you with just the VCO core to debug.

I think this is how the oscillator core works:
The current ramps the voltage on the capacitor down. Assume the voltage on
the base of Q3 ca. 0.7V higher than its emitter. When this voltage goes
below the Q2 emitter voltage minus the Q2 base-emitter zener voltage, Q2
and then Q3 starts to conduct. Q3 and Q4 then keep each other in
conduction. The capacitor voltage will quickly rise until it goes above
the Q2 emitter voltage, at which point Q2 starts to forward conduct and
switches off Q3/Q4 again.

If there is +5.9V between the base and emitter of Q2 there is something
wrong with it; that voltage drop should be not much more than 0.7V.
Are you sure it's not accidentally a PNP instead of an NPN? 5.9V would be
around the base-emitter zener voltage of typical silicium transistors.

Best regards
Simon




lanterma at ece.gatech.edu schreef:
>> If you look here:
>>
>> http://www.korganalogue.net/korgms/images/service/ms20/circ1.gif
>>
>> You'll find the two VCOs of the Korg MS-20. They use different size
>> integration caps, but other than that, they're the same as far as I
>> can tell, and they're also the same as the VCOs in the MS-10 and
>> MS-20.
>>
>> A team of a couple of my students are working on building the Korg
>> MS-X0 VCO and getting it working, and they've yet to get it to
>> oscillate. :(
>>
>> They took some DC voltage measurements with a multimeter at various
>> parts. I'm not sure what we should expect to be seeing at various
>> points - and there's a couple of things they're observing that just
>> don't make sense to me at all.
>>
>> I should mention they're using a 3906 for the PNP (Q4), 2N5457s for
>> the
>> FETs (the Q1 voltage buffer and the Q5 current buffer - many thanks to
>> the folks who helped decode what Q5 was!), and a CA3083 for the
>> remaining NPNs, with the extra-matched pair serving in the expo
>> converter. They have the substrate pin of the CA3083 connected to the
>> negative rail supply (-15 volts)
>>
>> I've uploaded a scan of a rough quick sketch, where one of the
>> students measured some voltages:
>>
>> http://users.ece.gatech.edu/~lanterma/ems11/korg_ms20_stuck_voltages.pdf
>>
>> You will see several voltages written down, stacked on top each other,
>> with a horizontal line between them. That doesn't represent division;
>> it represents a voltage range, as the student twisted the expo-control
>> pot (pot on far left). I should also mention that the Korg has a 22
>> Meg resistor (with additional 100Ks switched in depending on the
>> octave setting) with 15 Volts across it to generate the reference
>> current for the expo converter. It didn't seem to work with 22 Meg or
>> smaller, so I suggested trying 1 Meg to try to make extra sure some
>> current was flowing. So these measurements are with a 1 Meg in the 22
>> Meg spot.
>>
>> Anyway, twisting the expo control pot, the emitter of Q2 (which is at
>> a spot where two 427 ohm resistors split 15 volts and ground) stays at
>> 6.4 volts. The bases of Q2 and Q3 (and the bottom of D4) stay at 12.3
>> volts. The student measured a 0.65 mV drop across D4.
>>
>> The bottom of R15 ranges from 3.37 to 6.2 volts, while the top of R15
>> (bottom of the integration cap) ranges from 3.1 to 6.2 volts.
>>
>> Something strange is happening with the FET buffer - the output of it
>> ranges from 7.11 and 7.27 volts. So something must be wrong with that,
>> at least, but we can't figure out where things are going astray. The
>> FET buffer in the MS-20 is powered from the +18 supply, and we're
>> using +15 - but that shouldn't matter here if the FET gate is ranging
>> from 3.1 volts to 6.2 volts.
>>
>> The thing is I don't have a good feel for how the circuit works, so
>> I'm not sure of what voltages I should expect to see in most of the
>> circuit. The FET buffer is certainly acting weird.
>>
>> As always, any advise of what to look for or things to try is greatly
>> appreciated! :)
>>
>> - Aaron
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>
>


-- 
Vriendelijke groet,

Simon Brouwer
-*- nl.openoffice.org -*- http://www.opentaal.org -*-