circuit bending details

I haven't modified this instrument yet, but this is what I found out. The following descriptions relate to the instrument opened and placed keys facing down with view at PCB solder side. The bulky plastic case is mainly empty; there are only 4 small main PCBs in it those would likely also fit into an average PortaSound.

The German language PS-20 service manual mentions that it is based on PAS (Pulse Analog Synthesis) system technology, which was Yamaha's marketing term for digitally controlled semi-analogue sound synthesis (used mainly in Electone organs and synths). The hardware architecture is (unlike digital later Yamaha keyboards) well suited for systematic sound modification. But do not expect to find a CS80 synth hidden inside of it. This is a relative of the single keyboard organ CN-70 of 1979 (often used in music schools), thus the main voice timbres are still only a mixture of 4 digital squarewave drawbars with analogue envelope, routed through fixed analogue filters.

The main units are divided in a horizontal and a vertical PCB section (with panel switches) connected with transparent ribbon cables. They are held on their sheet metal frame by bent tabs those need to be straightened for removal. The below pictures show these PCBs removed and laid out flat.

In the middle is PCB 1 which contains the digital main components; they are the main CPU "YM1011" (IC1, labelled "key assign & rhythm"), the sound chip "YM1101 E" (IC2, labelled "tone generator"), the clock oscillator and some strange DAC hardware. Behind it is the vertical PCB 3, which contains sound preset switches , pots and helper circuits.

To the right is the PCB 2, which contains analogue filters for the main voice; behind it is the vertical PCB 4 which contains rhythm switches, pots and further helper circuits. The VCA IC is "IG02602" in Yamaha PS-30 but "IG02600" in PS-20.

In the PS-30, to the left of these is the solo voice section which consists of PCB 5 and the vertical PCB 6. These PCBs are shielded at the keyboard side and partly on the back with aluminiumized cardboard to protect the sensitive analogue filter circuit against hum. Interesting is that the IC numbering on these PCBs restarts with "IC1", thus it was likely not included by Yamaha in the initial instrument design but came as a later addition to the finished Yamaha PS-20. (The solo section is connected with the 10 pin connector "C3", which is only an unused solder pad row in my PS-20.)

Unlike my ancient Yamaha PS-2, the main ICs in this instrument have no strange ceramic packages but are normal grey plastic with silver print and no embossed pin numbering. But bizarre is that both large ICs are named to use negative(!) supply voltages (e.g. "GND" pin 1= 0V, pin 39= -9V, pin 36= -2V), thus key matrix diodes against "GND" need to be installed "wrongways". I will instead use here positive naming convention (adding 9 to each of them, so pin 1= 9V, pin 39= 0V, pin 36= 7V).

Many functions (e.g. preset sounds) in the behaviour of this early analogue instrument are mainly controlled by the wiring of PCB traces and not hardcoded inside ICs, thus with some skill it can be likely upgraded to a versatile analogue synthesizer. So I finally found an original service manual of Yamaha PS-20 (with German language function description), which explains much about the hardware and data protocols, but unfortunately (of course) lacks the solo voice description. Interesting is that it mentions similarities with Yamaha PS-10, which is basically a fullsize PS-2 with simpler single-chip CPU YM1104.

sound generator

Each of the 10 main voice channels has its own envelope capacitor. With chord or accompaniment 6 of them are re-routed as bass, chord and arpeggio voices. Due to negative logics, by the volume envelope not the "1" blocks move up, but the "0" move down (so the whole principle could be considered inverted). The bass, chord and arpeggio voice are made from squarewave with apparently 3:1 duty cycle and have each an own output.

The semi-analogue percussion is output by the CPU (KAR) and has more individual pins than in analogue PortaSounds. Interesting is that percussion has a separate clock and so stays unaffected by vibrato. The 'base' and the knock part of  'snare' are exactly what they sounds like - namely a short digital squarewave pulse without envelope, making a pop noise like powering on an old transistorized radio. Possibly these were originally planned as 2 trigger pulses for omitted analogue drums.

The 'hihat' and 'snare' (hiss part) are made from each a semi-metallic digital waveform (likely shift-register feedback noise) and have each an envelope capacitor. The 'hi bongo' and 'hi conga' are a digital wannabe sinewave with grainy 4 step (2-bit) resolution; only the 'lo conga' has 7 steps (3-bit). Unlike PortaSound, they all have separate envelope capacitors and thus do not mix into that strange bottle clang when they sound at the same time. The envelope capacitors can be also triggered externally (pulling lo through e.g. an 1k resistor).

solo voice section

The pitch of the solo section can be tuned with the "solo tune" trimmer (which is also accessible through a covered bottom hole). Also the adjustable "master osc." coil changes the pitch. The trimmer "cutoff adj." likely sets the synthesizer filter cutoff frequency. (It is likely well replaceable with a real potentiometer.) 

SOM pin 13 seems to be filter resonance envelope and pin 17 cutoff envelope out. Pin 12 seems to be a VCF audio input and also related to resonance (high resistance and very hum sensitive). But I haven't fully investigated how it works. The filter definitely can sound good and makes some wicked tekkno noises when controlled properly. With some upgrades the PS-30 can be certainly modded into a synth roughly like Yamaha SY-1 (plus percussion & accompaniment). 

synthesizer features

Because the sound IC YM1101 (DOM) outputs 4 separate squarewave organ drawbar tones, it is likely easy to add 4 potentiometers here to mix your own timbres. By adding a switch with a diode from +Vs to the vibraphone output (sound IC pin 13), the vibraphone vibrato can be enabled in all sounds. This vibrato seems to contain even a mild chorus effect to enrich the timbre. The LFO output to the pitch clock oscillator is DOM pin 14, where certainly a vibrato depth potentiometer or even external LFO can be added. 

As usual with analogue drums, the percussion is also easy to modify. 

The  IC3 "1G02611" of the solo section is the VCF (voltage controlled filter). This is a close relative of the chip in Yamaha CS-01. It is connected with multiple external capacitors and the filter envelope reacts sensitive when the IC is touched by hand, thus it can be likely upgraded easily with additional pots to change the filter behaviour. The SOM seems to output envelope control voltages for cutoff (pin 17) and resonance (pin 13?) and outputs its unfiltered waveform at pin 19. There is also an audio feedback loop(?) at pin 12. Pin 18 is the volume envelope cap and pin 20 a cap for the attack pulse of the filter.

I haven't fully investigated how the VCF works. But it is undoubtedly possibly to install switches to feed it with external audio (e.g. from main voice, chord or percussion) and control the envelopes externally, to make of it something like Yamaha SY-1 or (closer) Casio CT-410V. Richard Atkinson has thoroughly examined this chip in Yamaha CS-01 and wrote how to mod that.

(Attention: I have neither modified nor fully analyzed this instrument yet, thus I can not guarantee that the above instructions are correct. Especially the SOM pinout may contain errors.)

keyboard matrix

The input lines are active-high, i.e. react on +Vs. Any functions can be triggered by a non- locking switch in series to a diode from one "out" to one "in" pin.

The control panel switches are not in a matrix but wired from each a separate pin of "PSC II" ICs to +Vs, those send switch settings as a loop of serial data to the KAR. The unused PSC II pins do nothing, so there are no eastereggs.
 

3 P1	4 P2	5 P3	6 P4	7 P5	9 P6	10 P7	11 P8	12 P9	13 P10	PSC II pin
in 1	in 2	in 3	in 4	in 5	in 6	in 7	in 8	in 9	in 10
-	C. memory	C. fingered	C. single	O. sustain (vibraphone)	O. string accordion	O. vibraphone piano	O. harpsichord	O. organ 2	O. organ 1 trumpet clarinet oboe	IC4
C. multi bass	C. arpeggio var.	R. 8-bar var.	R. rhumba	R. tango	R. waltz	R. march	R. select	R. synchro	R. start	IC3
-	-	-	SV. violin	SV. trumpet	SV. trombone	SV. piccolo	SV. select	-	-	solo voice IC2

The locking preset sound switches select through a diode among only 5 raw preset sounds, but have additional diodes to select through FETs fixed analogue filter settings. The 'vibraphone' switch this way also enables sustain. Their switch contacts are daisychained in a way that if multiple switches are pressed only the rightmost takes effect. But (unlike PS-3 etc.) the voltage control concept through diodes (hi=on) is very straight forward and easy to modify.

On solo voice PCB, the 'solo voice' switch feeds voltage to all related preset sound switches. The orchestra switch goes to its C1 connector pin 16 'EMS'. The solo voice switches set no fixed filters, but only select each a preset sound of the SOM, which changes cutoff and resonance envelopes of the VCF to shape its timbres.
 

legend:
"o"	= keyboard key
underlined	= function needs locking switch (i.e. stays active only so long the switch is closed)
R.	= preset rhythm
O.	= preset sound ('orchestra')
SV.	= solo voice preset sound
C.	= chord
orange background	= easteregg (unconnected feature)

• additional keys

12 additional high note keys can be added by connecting them through diodes from {N2..N7} to KAR pin 21 and 22.
 
• strange noise

Pulling KAR test pin 3 lo through a resistor increases pitch of any sounds/percussion and turns their tone scales into disharmonic mess.

pinout YM1011, YM1001 (KAR)

The KAR polls the keyboard matrix and outputs data for the sound IC YM1101 (DOM). Additionally it produces digital waveforms and trigger pulses for semi-analogue percussion, those need external envelope capacitors and are routed through external fixed timbre filters.

The 'hihat' and 'snare' (hiss part) are made from each a semi-metallic digital waveform (likely shift-register feedback noise) and have each an envelope capacitor. The 'hi bongo' and 'hi conga' are a digital wannabe sinewave with grainy 4 step (2-bit) resolution; only the 'lo conga' has 7 steps (3-bit). Unlike analogue PortaSound, they all have separate envelope capacitors and thus do not mix into that strange bottle clang when they sound at the same time.

In CN-70 schematics (not its text description) this IC is depicted as YM10010, which may be an alternative name of an even older version.

caution: This IC uses "negative logic", i.e. technically +9V is its GND while 0V is its -9V supply voltage. So the voltages are not was the pin names suggest. I use the positive voltage naming convention (from 0V to +9V, not -9V to 0V).

This pinout is based on my Yamaha PS-20 service manual.
 

pin	name	comment
1	VSS	supply voltage +9V (measured ca. 8.5V)
2	N2	[connector C1] key matrix out
3	N3	key matrix out
4	N4	key matrix out
5	N5	key matrix out
6	N6	key matrix out
7	N7	key matrix out
8	N1	key matrix out
9	/KC1	[connector C3] key data out
10	/KC2	key data out
11	/KC3	key data out
12	/KC4	key data out
13	/B11	[connector C2] key matrix in
14	/B12	key matrix in
15	/B21	key matrix in
16	/B22	key matrix in
17	/B31	key matrix in
18	/B32	key matrix in
19	/B41	key matrix in
20	/B42	key matrix in
21	/B51	key matrix in (not used)
22	/B52	key matrix in (not used)
23	/SY	[connector C3] synchro data in
24	/Si	serial data in
25	/IC	[connector C3] reset
26	AG1	analogue supply voltage +9V
27	R1	percussion hihat+snare hiss audio out
28	R2	percussion hi bongo+hi conga+lo conga audio out
29	R3	percussion snare drum audio out
30	R4	percussion base audio out
31	C2	percussion hihat envelope cap
32	C1	percussion snare envelope cap
33	C3	percussion lo conga envelope cap
34	C5	percussion hi bongo envelope cap
35	C4	percussion hi conga envelope cap
36	-AV	rhythm supply voltage +7V
37	/TCL	tempo clock in
38	/TL	tempo led out
39	VDD	ground 0V
40	øM	clock in (470 kHz)

Pin 28 outputs a very clear and metallic cowbell clang, which seems to be used only as a component of the bongo(?) sound although it sounds surprisingly realistic (may this have been Yamaha's first and unofficial use of an FM sound on a home keyboard?!?). The rhythm waveform outs respond with continuous tones when connected through a resistor with GND(?); these signals are sent through their envelope capacitors.

The keyboard matrix ins (N1..N7) are notes, while the outs (B...) are octaves (B1x = lowest). So their strange numbering (Bx1, Bx2) was likely done to indicate both halves of the same octave. At the unused pins 21 and 22 are higher note keys addable. The KAR protocol reads 22 time slots of each 4-bit (data pins KC1..KC4) to scan all keys at a rate of 4.5ms, synchronized by a 2us hi pulse on pin 23. Any key press is transmitted as (word 1) a 4-bit note number, followed by (word 2) a 3-bit octave number + 1 bit "note on".

pinout YM1002 (PSC II)

In CN-70 schematics (not its text description) this IC is depicted as YM10020, which may be an alternative name of an older version.

caution: This IC uses "negative logic", i.e. technically +9V is its GND while 0V is its -9V supply voltage. So the voltages are not was the pin names suggest. I use the positive voltage naming convention (from 0V to +9V, not -9V to 0V).

This pinout is based on my Yamaha PS-20 service manual.
 

pin	name	purpose
1	/Sy	synchro data in (from KAR)
2	/Si	serial data in (from other PSC II)
3	/P1	parallel data in (a switch)
4	/P2	parallel data in (a switch)
5	/P3	parallel data in (a switch)
6	/P4	parallel data in (a switch)
7	/P5	parallel data in (a switch)
8	VSS	supply voltage +9V
9	/P6	parallel data in (a switch)
10	/P7	parallel data in (a switch)
11	/P8	parallel data in (a switch)
12	/P9	parallel data in (a switch)
13	/P10	parallel data in (a switch)
14	/SO	serial data out (to KAR, DOM, other PSC II)
15	VDD	ground 0V
16	øM	clock in (470 kHz)

Pin 2 is described "serial data IN <= PSC II"; in PS-20 schematics the last one is wired to VDD (grounded), so this is likely a daisychaining mechanism to connect multiple YM1002 for handling a large control panel.

The protocol is simply a loop of bits (each 2us), those are hi for every enabled switch. The bits are shifted out in reverse pin order (P10, P9, P8,..., P1 followed by P10 etc. of other PSC II at its /Si pin). The sequence repeats every 47us and is synchronized by a 2us hi pulse on pin 1. If anybody wants to control the sound synthesis by a microcontroller, this port would be certainly the best place to add it. Of course voltages need to be converted to the strange 9V level.

pinout YM1101 (DOM)

Unusual for a Yamaha keyboard is that this sound IC has no GE# number (e.g. PS-1 has GE1, MK-100 has GE7 etc.), which suggests that it is so old that it was originally designed for Electone organs rather than home keyboards. Despite PS-20, PS-30 were released one year after 1st generation PortaSound (PS-1, PS-2, PS-3), not least the related Electone CN-70 organ of 1979 (seen in CN-70 service manual) proves that the PS-30 hardware platform was developed first and later simplified to create those PortaSounds.

In CN-70 schematics (not its text description) this IC is depicted as YM11010, which may be an alternative name of an older version.

caution: This IC uses "negative logic", i.e. technically +9V is its GND while 0V is its -9V supply voltage. So the voltages are not was the pin names suggest. I use the positive voltage naming convention (from 0V to +9V, not -9V to 0V).

This pinout is based on my Yamaha PS-20 service manual. The word "data" in its pin description often seems to mean simply a switch or voltage output for analogue functions (e.g. vibrato). I instead have renamed these pin functions "control" where no actual digital data (binary numbers) are transmitted.
 

pin	name	comment
1	VSS	supply voltage +9V
2	TEST	test pin (not used)
3	TEST	test pin (not used)
4	/IC	reset
5	/KC1	key code data in (from KAR)
6	/KC2	key code data in (from KAR)
7	/KC3	key code data in (from KAR)
8	/KC4	key code data in (from KAR)
9	NGL	normal gate control out
10	NGP	(not used)
11	/Si	serial data in (from PSC II)
12	/KON	key-on signal out
13	/VIB	vibraphone-on control out
14	DV	delay vibrato lfo out
15	-AV	sound source supply voltage +7V
16	øOSC	sound source clock in (588 kHz)
17	NENC	(not used)
18	M2'	melody footage 2' audio out
19	M4'	melody footage 4' audio out
20	M8'	melody footage 8' audio out
21	M16'	melody footage 16' audio out
22	L4'	chord audio out
23	A4'	arpeggio audio out'
24	P8'	bass audio out'
25	MAG	main voice analogue supply voltage +9V
26	CM1	envelope cap (melody)
27	CM2	envelope cap (melody)
28	CM3	envelope cap (melody)
29	CM4	envelope cap (melody)
30	CC1	envelope cap (melody/chord)
31	CC2	envelope cap (melody/chord)
32	CC3	envelope cap (melody/chord)
33	CC4	envelope cap (melody/chord)
34	CA	envelope cap (melody/arpeggio)
35	CP	envelope cap (melody/bass)
36	AAG1	bass analogue supply voltage +9V
37	-AV	sound source supply voltage +7V
38	AG1	analogue supply voltage +9V
39	VDD	ground 0V
40	øM	clock in (470 kHz)

The ciphers in audio out pin names indicate organ footages. The volume envelope levels vary among preset sounds to match the gain of their corresponding filter settings.

Pulling test pin 2 lo through a resistor (reacts even on finger) increases pitch of any sounds/percussion and turns their tone scales into disharmonic mess. Test pin 3 seems to do nothing.

Pin 9 turns hi while manual chord notes in chord section are played. It turns lo again when chord mode is switched off or accompaniment is sounding. This pin is used for 'level control' of a preamp that likely changes the volume of chord (organ tones) in ratio to accomp notes (with piano decay). The German service manual text tells that it can be also used to reliably sync an oscilloscope for examining data an pins 5..8. The unused pin 10 behaves similar, but turns lo while 9 is hi, and else is high resistance.

Pin 12 (key-on) pulls lo while one or more melody voice keys are pressed. Else it stays high resistance. (The key-on pin of analogue PortaSounds behaves different; it outpute a short lo pulse during each additional melody voice key press.)

Pin 13 turns hi while the 'vibraphone' preset sound is selected, which is used to set the analogue vibrato circuit.

Pin 14 is the vibrato control voltage out, which outputs a digital sinewave LFO to modulate the sound source clock oscillator. Depending on the selected raw sound, vibrato can be delayed, i.e. amplitude stays 0 during attack and slowly grows to the final level (where it also stays after note-off). According to the service manual, during 'vibraphone' and 'accordion' the vibrato is routed to FET 14 to make the LFO perform amplitude instead of frequency modulation.

(Initially I had written that connecting the audio pins 18..21 directly with the amp produced no tones. Possibly the signal has very high frequencies (the instrument makes no audible aliasing) - these may be simply PWM bit streams decoded by op-amps and capacitors, decoded by a strange kind of external DAC consisting of a few capacitors and op-amps IC6 and IC7 (both "Texas Instruments TL4558P, J104, LTJ", 8 pin DIL). According to the service manual these should be 4 normal analogue footage outs and not digital data. So this may have been an error by me.)

pinout YM1102 (SOM)

It outputs monophonic waveforms with volume envelope for the solo voice and control voltages for an analogue VCF. As a real sound IC it needs to be controlled by a host CPU. Additionally it has a serial data input for an YM1002 (PSC II) for solo voice panel switches.

Depending on selected preset sound, the raw waveform can be either a 4-bit falling sawtooth wave or a squarewave with 7:1 duty cycle.

I had examined this pinout by myself, and pin names later corrected from Yamaha PS-30 service manual.

caution: This IC uses "negative logic", i.e. technically +9V is its GND while 0V is its -9V supply voltage. So the voltages are not was the pin names suggest. I use the positive voltage naming convention (from 0V to +9V, not -9V to 0V).
 

pin	name	purpose
1	VSS	supply voltage +9V
2	/IC	reset
3	TEST	(not used)
4	/KC1	key code data in (from KAR)
5	/KC2	key code data in (from KAR)
6	/KC3	key code data in (from KAR)
7	/KC4	key code data in (from KAR)
8	/Si	serial data in (from PSC II)
9	/SKO	key-on signal out (not used)
10	AG2	analogue supply voltage +9V (C2 pin E)
11	DV	delay vibrato lfo out
12	Q	Q setting voltage out (cap to vcf pin 11)
13	FEG	vcf resonance envelope out (capacitor for VCF-EG)
14	LO	audio out (cap+resistor to IC4 pin 6)
15	LI	level control in (cap+resistor to vcf pin 6)
16	-1V	+7.5V
17	KV	key voltage out (vcf cutoff envelope?)
18	AEG	envelope cap (for VCA EG)
19	AO	raw wave out (cap+resistor to vcf pin 4)
20	-2V	envelope attack cap
21	AG1	analogue supply voltage +9V
22	øOSC	sound source clock in (1 MHz?)
23	VDD	ground 0V
24	øM	clock in (470 kHz)

The unused pin 3 outputs a multistep spiky signal at about 1MHz (like the pitch clock). It is likely a test pin revealing internal data.

Pin 9 turns hi while one or more keys are held. It is interrupted by a short lo pulse when an additional key right to it (plays new note) is pressed.

Pin 11 (cap to AG1...) outputs a delayed vibrato control voltage (similar like in DOM YM1101). It outputs a digital sinewave(?) LFO to modulate the sound source clock oscillator. Depending on the selected raw sound, vibrato can be delayed, i.e. amplitude stays 0 during attack and slowly grows to the final level (where it also stays after note-off).

Pin 12 (cap to VCF pin 11) seems to be an audio input or feedback line of the VCF. It is very hum sensitive when touched with fingers. It seems to be related to filter resonance and may be its coarse control. Connecting a 6 megohm potentiometer (for safety with 1k in series) to +Vs or to pin 13 made a quite rough and noisy resonance control, that can even drive it to self-oscillation.

The VCF has a cap parralel with resistor from pin 11 to 12, from 12 another cap to 3 and another to 5, which appears to be a feedback loop for resonance.

Pin 13 (cap to AG1...) seems to be filter resonance envelope out. The tone grows brighter when the control voltage falls. Pulling it too low (1k resistor to ground 0V) however makes it dull.

Pin 16 (cap to AG1) is likely an analogue supply voltage.

Pin 17 (cap to AG1...) seems to be filter cutoff envelope out. The tone grows brighter when the control voltage falls. Pulling it too low (1k resistor to ground 0V) however makes it darker. Pulling it high makes it brighter.

I still haven't fully figured out how the filter control really works and what pin 12 exactly is. These are certainly no standardized CV voltages. But it likely wants to see a reasonably low resistance voltage (pot wired as voltage divider between +Vs and 0V or a fraction of that) rather than pulling with a very high resistor into only one direction. The filter definitely can sound good when controlled properly.

Pin 18 (cap to AG1) is the volume envelope capacitor. Pulling it lo (1k resistor to ground 0V) makes the tone loud with plenty of nice analogue distortion like an overdriven e-guitar.

Pin 19 outputs the raw audio (raw digital waveform with volume envelope and vibrato) to the VCF.

Pin 14 and 15 seem to be audio (amplified solo voice including filter). 15 goes to the VCF, and 14 to IC4 (likely VCA).

Pin 20 (cap to AG1) seems to be a filter cap for the attack phase of the filter resonance envelope. Pulling it with 1k resistor to ground 0V makes sounds attack harder.

Pin 22 (resistor to Tr3, Tr4) seems to be the pitch clock (on oscilloscope 3.6*0.3us = roughly 1MHz, modulated with vibrato).

pinout IG02590, IG02592, IG02595,IG02600, IG02602 (VCA)

This pinout is from my Yamaha PS-20 service manual. Also IG02600, IG02590 and IG02592 have that pinout (seen in those of CS-01 and C55-N), and also the IG02595 in my Yamaha PS-3 looks like yet another variant. Pin names are from IG02600 datasheet. But this thing is fairly trivial to understand anyway.

caution: This IC uses "negative logic", i.e. technically +9V is its GND while 0V is its -9V supply voltage. So the voltages are not was the pin names suggest. I use the positive voltage naming convention (from 0V to +9V, not -9V to 0V).
 

pin	name	purpose
1	IN	audio in
2	Icont	amplitude cv in
3	Bias	Bias (resistor to 6)
4	VEE	ground 0V
5	Out	audio out
6	Ein	emitter follower in
7	GND	supply voltage +9V

pinout IG02610, IG02611 (VCF)

It is hard to find any specs about this analogue part, but according to the service manuals of Yamaha CS-01 and CN55-N this is the pinout. Pin functions were concluded from those schematics and may be wrong. I don't know if this is only a quad op-amp in fancy camo to fool competitors or really something special.

caution: This IC uses "negative logic", i.e. technically +9V is its GND while 0V is its -9V supply voltage. So the voltages are not was the pin names suggest. I use the positive voltage naming convention (from 0V to +9V, not -9V to 0V).
 

pin	name C-55N	name CS-01	purpose
1	IC		cutoff cv in
2	ICD		cutoff frequency control (used in C55-N)
3	C1		feedback (cap 22nF to 12)
4	IN		audio in
5	C1'		feedback (cap 22nF to 12)
6	C2		(CS01: resistor 10M to analogue ground 0V?, cap to +Vs | C55-N: with voltage trimmer, other only cap to +Vs)
7	NC		(not used)
8	-15	VEE	ground 0V
9	C3, NF	GN	(CS01: cap 10nF to +Vs)
10	OUT	GO	audio out
11	F2	KN	resonance control (CS01: switch to 12, 10k resistor to 12)
12	F1	KO	resonance control
13	GND		supply voltage +Vs = +9V
14	C3		feedback (CS01: 1k resistor to 4, 'Q ADJ.' trimmer to 11, 100uF cap to +Vs)

The listed connections are from CS-01 and CN55-N and only examples to figure out what it does. Resonance is apparently a feedback loop from pins 11 and 12 back to 4, so this has no simple single control voltage pin. Possibly the slope (dB/octave) or mode of the filter can be even changed by different wirings.

Richard Atkinson has examined this chip closer and thoroughly explained how to mod the Yamaha CS-01 synth. Likely similar principles can be applied to the PS-30 and PS-3.