As a companion to my simple op-amp AR envelope circuit, here’s a discrete version. It has the same basic functionality – gated input, variable attack and release times – but is made with transistors instead of integrated circuits. Power consumption is very low (just a handful of mA), and it runs from a positive supply of your own choosing. Like its op-amp cousin, it could be powered with a battery, or in a Eurorack system, or you could add it into an existing synth like the Moog Werkstatt as a mod.
The main difference between this and the op-amp circuit, aside from it being discrete, is that I have included a very simple way to set the level of the envelope output (see below for details).
RV1 is Release, RV2 is Attack. The Gate input can be anything over a couple of volts. Negative-going inputs (eg., from a bipolar LFO) will be removed by D1. The output goes to nominally 0V when fully off (closer than the op-amp version, in fact).
R1, R2, R6, R7, R10: 100k
R3, R4: 47k
R5, R8: 560 Ohm
RV1, RV2: 1M linear pot
C1: 1µ non-polarized
D1, D2, D3: 1N4148 or equivalent
Q1, Q4, Q5: BC549C or equivalent
Q2, Q3: BC559C or equivalent
How it works
Compare the first pair of transistors with my discrete gate buffer circuit. A positive voltage on the input turns on Q1, taking the base of Q2 low. This turns on Q2, taking its collector high. This is how we drive our envelope.
Now compare the diode and potentiometer arrangement with my op-amp AR. Once you’re past the transistors, it works in basically the same way.
Q3 inverts the output of Q2, so when Q2 is on, Q3 is off, and vice versa. When the collector of Q2 is high, the capacitor charges through diode D3 and pot RV2 (Attack). When the gate input goes low, the transistors Q1-3 switch off, off, and on, respectively. In this state, the capacitor discharges through RV1 (Release) and D2.
Note the two 560 ohm resistors: one on the emitter of Q2, the other on the collector of Q3. When the gate input is high and the capacitor is charging, current flows through Q2’s emitter resistor; when the gate is off and the capacitor is discharging, current flows through Q3’s collector resistor. These two resistors put a lid on the current flow and limit the fastest times for Attack and Release. The value is a trade-off between current and snappiness. With the values shown, maximum current through these resistors is around 16mA and the fastest rise and fall times of the envelope are around 2ms.
The final two transistors in the circuit after the capacitor are the output buffer; notice the two resistors between them, forming a potential divider. With the values shown, if you run this circuit on 12V, the envelope output will be around 8V max.
There are better ways to set the peak level of an envelope, but my aim here is to keep things simple as a base for experiment.
The most obvious things to tweak are the envelope times and the output level.
The values of the two potentiometers affect the attack and release times, but the envelope can be substantially stretched by using a larger capacitor. It would be easy to add a switch that connected, say, a 4.7µF or 10µF capacitor in parallel with the existing one, which would multiply the envelope’s times substantially (use perhaps a 25V electrolytic, with its -ve terminal to ground).
The two resistors between the output buffer transistors can be adjusted to suit your requirements. If you want full-scale output (ie., envelope peak closer to the supply voltage), remove R9 and R10, and connect the emitter of Q4 directly to the base of Q5. In fact, this circuit will also work with just a single NPN as a buffer (miss out Q4 and the divider resistors, connect the cap to the base of Q5), but amongst other things the ‘zero’ value is less close to actual zero; if you want to experiment with a single transistor here, setting the level of the output can be done by replacing the 10k resistor on its emitter with a pair of resistors as a potential divider, or even a 10k trimmer with the output taken from the wiper.
Feel free to experiment with the circuit in Falstad’s handy online simulator.
By default, the Boss DR-55 does not receive any kind of incoming clock. The ‘FS’ footswitch input takes a latching footswitch that starts and stops the existing clock, but that’s it. Although you can clock other equipment from the DR-55, it would be nice to be able to use an external clock to sync the Boss to, which would allow the Boss to trigger yet more devices with its CSQ and DBS outputs (active on Accented steps only and every step, respectively).
My mod as detailed here does exactly that. By replacing the existing FS jack socket, adding a small circuit, and replacing a jumper, we can safely trigger the DR-55 from an external trigger.
A quick internet search will turn up an existing clock input mod which is simpler to do and requires no extra parts; however, it puts the RAM at risk of damage from high triggers, and it does not sync the Boss’ own DBS output. It also requires ‘arming’ by hitting start before external triggering.
My own mod, though more complex, overcomes all these issues: the trigger input is protected, both the Boss’ trigger outputs maintain their correct functions, and triggering occurs without ‘arming’. The only two functional disadvantages of my mod are that you must set the Boss’ tempo to Fast, and to reset the pattern when stopped mid-way you need to remove the trigger plug.
I’m going to blog another small mod which will overcome the latter inconvenience [EDIT: No I’m not! I sold both my 55s, thereby halting this particular project].
The Clock Modification in detail
Below is a diagram which shows everything you need to know about building this mod. Below that is a parts list. Key to this is the replacement FS jack socket; it needs to be TRS (ie. a stereo jack), with single pole changeover switches on the tip and ring contacts. I used a Lumberg KLBPSS3 (datasheet here, Farnell UK stock page here).
The additional circuit can be made very small indeed (3 rows * 8 holes on stripboard), and there is plenty of room for it inside the DR-55, particularly towards the right-hand end. The photos below illustrate my own placement.
There is one jumper to be removed, the one immediately to the right of the Variation switch. The replacement connections for the upper and lower point of this removed jumper are shown in the diagram, and you can see in the photos how I wired this up.
In brief: remove that jumper, solder the two points to two jack pins; build the extra circuit, and solder that to the jack and to the main PCB; replicate two of the pre-existing connections from the jack to the PCB. That’s it. I also stuck a small folded piece of card to the PCB to stop the extra circuit from shorting against components.
The image above is a jpeg; click here for a PDF: Boss DR-55 clock input mod revised
1 * TRS 2-pole changeover jack socket – eg. Lumberg KLBPSS3
2 * 47k resistors – I used 1/8W for their smallness
1 * 10nF capacitor – I used a ceramic, again for smallness, but polyester film etc. would be usual
1 * 1N4148 signal diode or equivalent
1 * BC549C transistor or similar standard NPN
Here’s the modified DR-55 (also incorporating my DC supply mod):
And here’s a close-up of the clock mod:
Here are two angles to show the extra circuit more closely:
How to use your new trigger input
The new trigger input will accept any positive pulse over a couple of volts. It’s edge triggered, so the pulse can be any length over a couple of milliseconds. The operating principle is to use the DR-55’s existing clock, but to gate it on for a very short duration; normally when the clock is gated off again, the pattern resets, but the new jack socket enables us to disable that by breaking the reset connection when a jack is inserted.
As I mentioned earlier, the Tempo must be set to Fast (ie. all the way clockwise) for correct function. This is because the DR-55’s clock, once triggered, finishes its pulse cycle. If this is longer than the incoming trigger cycle, it will ignore the new trigger; if we set the speed dial to its fastest, we can clock the DR-55 at any rate up to its natural maximum.
The pattern will cycle round as usual, but if you stop mid-pattern, new triggers will continue where they left off. To reset the pattern at this stage, you need to unplug the trigger jack and hit Stop. This is not ideal, I know
, and I will be making an amendment to correct this later [EDIT: project halted, see above. I have no current plans to do any further work on the DR-55].
For now though, this mod works fine, as shown in the (slightly rubbish) video below:
One of the drawbacks of the DR-55 as it comes unmodded is the power supply. In its original form, the DR-55 takes only batteries, and though this might be good for reducing cable clutter and having to find yet another wall-wart, it does mean you need to keep a regular stock of fresh AAs, and can guarantee that just when you want to use it, your DR-55’s batteries are too drained for the unit to function correctly.
Luckily, it is a relatively simple process to modify the DR-55 so that it takes a commonly-found 9V DC supply instead. I provide instructions for this below. It’s not the only way to do the job, but this is how I did it, and it works just fine. Modding the DR-55 in this way means it no longer accepts batteries, which means two things: 1) you will need access to a 9V adapter, and 2) pattern data will not be retained on power-off. Given that filling the memory of this humble machine can be done in less than five minutes, and I never use this outside my own home studio, I never found memory retention to be an issue. It would be possible to design a DC input that also catered for memory backup via battery, but I’m not going there.
There are two basic stages to this modification:
- Making a 9V DC input: the basic voltage supply circuit
- Installing the Mod: adapt some wire links on the output jack and PCB
Making a 9V DC input
Because the DR-55’s RAM can be killed by voltages higher than around 7V, we take a 9V input and regulate it down to between 5V and 6V. I chose to use a 5V regulator propped up with a diode to give around 5.6V, but you could also use a 6V regulator and omit D2. The input jack I used is a 3.5mm mono minijack of the kind often used for audio and CV interconnects, mainly because I had lots of them and the holes are easier to drill than the larger ones needed for a plastic-bodied insulated barrel connector. Use whatever type you prefer, but note the polarity of your incoming DC, and don’t connect the +ve to the case… with a tip-positive 3.5mm jack, the sleeve of the input jack is connected to the shell of the socket, so it makes sense for that to be the ground. Some barrel connectors do likewise.
Here’s the schematic:
Here’s the final circuit built onto stripboard. It will be panel-mounted using the socket:
Installing the Mod
Now we have a simple DC input, we could just solder the +ve and Gnd outputs to the corresponding solder points on the main board – that is, where the battery clip attaches. Black is ground, red is positive. This works, but you still need to insert an audio cable to turn the DR-55 on. I chose to remove that ‘feature’, as there are no longer any batteries to protect from accidental drain. It’s a simple mod that just means a couple of wiring changes.
The diagram below shows the required re-wiring. The audio output socket is wired by default to both ground and audio signal, as well as having two pins wired to act as a switch when a jack is insterted. We want to retain the audio and ground connections, but not the switch. We remove those wires and instead bridge the corresponding points on the PCB.
Here’s a photograph of the full mod (note the wiring):
I damaged a track while desoldering the battery wires, which is why the red wire goes to the un-numbered hole next to point 9. They’re directly connected, happily.
Below are some photos of the hole I drilled for mounting the new DC input, and the final appearance when mounted and labelled with cheap Dymo (should have gone with black… oh well):
So there you have it. My humble DR-55 now works from a regular 9V DC wall-wart supply, and switches on whether or not its audio is connected. The hardest part is putting the DR-55 back together again…
The DR-55 is easy to get into, but I’ve found that removing the screws on the front panel makes it harder. I recommend the following procedure:
- Unclip the battery holder if present
- Remove the knobs (they should just pull off)
- Unscrew the nuts securing the 1/4” jacks
- Remove the screws on the rear and front edges
- Prise apart the shell, taking care with the jack wiring
- Remove the screws on the rear of the PCB that secure it to the standoffs
The washers are easily misplaced. Those standoffs are better left attached to the body. Note the arrangement of the jack wiring when you gain access, as it’s easy to trap wires on reassembly. The Boss DR-55 Service Manual also recommends avoiding certain wire placement due to possible interference.
Here’s the DR-55 in the nude, complete with a broken wire or two:
The Boss DR-55 is a humble little black box that runs on batteries, and provides the user with 8 memory slots for programmable rhythm patterns. Kick, Snare, and Rimshot can be programmed as desired, as can an Accent control that boosts the volume of the steps on which it appears. The Hi-hats are non-programmable and appear at every step, every other step, or not at all, as governed by a switch. The sounds themselves are generated by analogue circuits, and are simple but punchy, sounding similar to the CR-78. The only other control over the sounds is an overall Tone control, which is kind of a one-knob EQ and emphasises the lows or highs as swept along its rotation.
The patterns are in two halves, labelled A and B. The Variation switch sets whether the pattern plays only its A part, only its B part, or cycles between A and B. Each part is either 12 or 16 steps in length, as governed by the memory slot selected. There are 6 * 16-step memories and 2 * 12-step memories. Generally, each step is assumed to be a semi-quaver or 16th-note. In 12-step patterns, the pattern is simply shortened to 12 steps and therefore at a given tempo will play three quarters the length of a 16-step pattern at the same speed; there is no onboard way to define the pattern as being in 3/4 time, or in triplets, etc. The onboard clock allows running speeds of around 30-300 BPM, if 4/4 time is assumed.
Patterns are programmed by selecting Write Mode and tapping the Beat/Rest buttons for the length of a pattern. The sounds are programmed separately, as governed by the Sound switch, allowing independent writing of the Kick, Snare, Rimshot, and Accent. Each time a button is pressed, the step is written with the appropriate data, and the pattern moves on to the next step. Both the pattern memory and the A/B switch function during programming just as during playback, allowing either or both parts of a pattern to be programmed. Programming is exited by switching back to Play Mode. All sounds are heard during programming.
All changes made to the running of the patterns are made in real time from the panel. If the pattern or the pattern mode is switched during playback, the change is instantaneous.
Rhythms can be started and stopped either from the panel or via a footswitch, connected to a 1/4” jack on the right-hand panel.
Connections and Power
Though the DR-55 cannot be clocked from another device, it can act as a master clock for other devices. There are two outputs: DBS, which puts out a 5V pulse on every step, and CSQ, which puts out a 5V pulse on every Accent hit. When the CSQ output is connected, the Accent is disabled from the onboard sounds. The Accent level control has no effect on the output pulse level.
The only other output is a 1/4” mono audio jack, which carries the mix of all sounds. There is no way to independently set their levels.
The DR-55 runs on 4 * 1.5V AA batteries, which are fitted in a holster that attaches to what looks like a regular PP3 clip. However, it must be noted that the memory IC used in the DR-55 is only tolerant of low voltages (no more than around 7V max), and a fresh 9V PP3 will destroy it permanently and irreversibly. The effect of a blown memory seems to be two-fold: patterns cannot be written, and sounds appear arbitrarily/on every step. The only solution is to replace the memory. Of course these ICs are obsolete and hard to find. Look after your DR-55!
While good batteries are kept in the DR-55, memory is retained during power-off. Low batteries affect memory retention and sound. It should also be noted that the DR-55 will not power up without a jack plugged into the audio ouput.
To sum up, the DR-55 is an enjoyable but limited machine. It is a cheap source of a basic set of CR-78 style analogue percussion sounds, and is fun in combination with synthesizer arpeggiators and analogue sequencers.
Here’s a short video of me programming the DR-55. It was made to illustrate a working sale item, but shows effectively how cumbersome by modern standards the programming is. Simple rhythms are relatively painless to input, but anything remotely fancy takes a bit of thought. Best to write them down first, or trust to chance and see what happens…
Apologies for the terrible sound quality. You get the idea, though.
A very simple Attack-Release envelope generator can be built with a dual op amp and just a handful of extra components. The input stage is basically the same as my op amp gate buffer, with only its output resistor changed; the rest is a simple low-pass resistor/capacitor setup with an output buffer. Here’s how it works:
The input acts as a comparator. When the gate input goes high, the comparator output goes high, and the capacitor is charged up via D2 and the Attack pot RV1; when the gate goes low, the comparator goes low, and the capacitor discharges through the Release pot RV2 and D3. The diodes directionalise this process, so the attack time is governed only by the Attack control, etc. The output is a very simple unity-gain follower.
With the values shown, attack and release times range from just a couple of milliseconds to around 5 seconds. Larger values for the pots and/or cap will extend the times proportionally, smaller ones reduce them. The 560 Ohm resistor sets the minimum time against a given capacitance.
With an op amp such as the LM358, the output will swing between 0V and approximately 1.5V below the positive rail. If a lower output level is desired, add a potential divider of resistors in the low-mid single Ks after the output buffer amplifier, taking the overall output from their junction.
Supply voltage is not critical, but as mentioned above, the LM358 op amp will swing to around 1.5V below supply at maximum. It does, however, swing to ground too, which when operated on a single supply is necessary in obtaining a correct ‘gate low’ output. If you cannot find a 358, use another op amp which will swing rail to rail, or ground to near-positive.
A circuit like this makes a nice addition to synths with only one envelope, such as the Moog Werkstatt and Mother 32, or Arturia Microbrute. It will run from a 9V battery and is small enough to build into the Werkstatt itself, or indeed any small external box of your choice. You could easily build one for a Eurorack modular system too, and it will run happily on +12V or +15V.
For details of how to modify the Werkstatt, take a look at my Werkstatt page.
Op Amp AR, parts list:
U1: LM358 or similar
D1-3: 1N4148 or equivalent
C1: 1µ poly non-polarized
R1,2: 100k 1/4W resistor (I use 1% Metal Film types, but 5% Carbon are also fine)
R3: 82k —”—
R4: 18k —”—
R5: 560Ω —”—
R6: 1k —”—
RV1,2: 1M linear pot
Input and output connectors as desired.
One of the simplest DIY utility circuits you can build is a gate buffer: you put a gate signal into one end, and get a gate signal out of the other.
Although this might sound unnecessary, there are several reasons you might want a gate buffer:
- compatibility problems between gate/trigger inputs and outputs on different equipment: see my page on the Arturia Beatstep, for example
- the need to trigger multiple devices from one source: passive splitter cables or mults sometimes result in signal loss and therefore unreliable triggering
- tightening up the edges of gates/triggers: for various technical reasons, some trigger outputs are relatively slow to rise and/or fall; in a worst-case scenario, this can skew the timing of down-line devices. A buffer with multiple outputs can deliver a set of tight, sharp pulses simultaneously.
I offer two simple designs here, one using discrete components, the other using an op amp. Both require just a handful of parts, both will run off a wide range of DC supply, including a 9V battery, and both can be made very compact if you ever want to include them inside another piece of equipment as part of a build or mod.
Discrete (transistor) Buffer
The transistor buffer is a two-stage circuit, with each stage inverting the incoming signal.
Think of a gate signal as a logic on, or a logic off. When there is no gate present, the first transistor is held off by its base resistor. The base of the second transistor is therefore tied to +V by the two 47k resistors; as it is a PNP type, it is therefore off, and the output is held low.
Conversely, when the input is high, the first transistor is switched on, and the base of the second transistor is taken low. This pushes the second transistor into conduction, and the output is taken high.
Precise voltage levels depend upon the level of the gate signal going in, and the positive supply rail. The circuit will operate on a wide range of positive DC supply: in a 5V logic circuit, from a 9V battery, a 12V or 15V rail in a Eurorack system, etc. The input resistors and diode provide input protection; so, for example, you can send a bipolar square LFO into it with no ill effects, or use it to make a reliable 9V gate from a 15V one without the impedance issues of a simple passive potential divider. It will also allow you to increase a low gate to a high one, so you could (for example) run a 5V signal into this, powered on an existing 15V rail, and get a 15V gate out. With a standard signal diode and two normal low-power transistors, you can trigger this circuit with just a couple of volts.
Op Amp Buffer
The op amp version of this gate buffer circuit consists of a single op amp stage set up as a comparator: one voltage is compared to another, and the output goes high or low depending which input is the higher.
The potential divider at the inverting input provides our reference voltage. The non-inverting input takes the external gate signal we want to buffer.
When there is no gate signal, or it is low, the inverting input is higher, and the output is therefore low. When the gate signal is high, the non-inverting input is higher, and the output is high.
The circuit is designed to run from a single-sided supply, ie. ground and positive. For this purpose, an op amp such as the LM158/358/324 (single, dual, and quad versions respectively) is suitable as the low output state goes to the 0V rail. Their high output state is around 1.5V below positive supply.
The voltage reference provided by the potential divider at the inverting input should be adjusted for purpose: using a 9V supply, the values given will trigger the comparator at around 1.6V; even with a low battery, this circuit should trigger around 1.2V. With a 12V or 15V supply, replace the 18k resistor with something in the region of 10k-15k. This would keep the trigger level around 2V or a little lower, which is high enough to be a clear ‘on’ signal, but not so low as to be confused with a slightly high ‘off’ signal (the Arturia Beatstep ‘off’ gate signal hovers around 0.6V, for example).
It would be possible to use a dual-rail op amp just as well, which would require the addition of a diode on the output to clip the negative-going signal.
I have used an op amp here rather than a dedicated comparator; devices such as the 311 cannot be directly substituted in this circuit.
Note: I make reference to the Moog Werkstatt schematics throughout. Copyright prevents me reposting them here; they can be found on Moog’s website.
After the VCO Frequency CV and Gate inputs, perhaps the next most useful control we can modify is the VCF cutoff frequency. The Werkstatt already has switches to select either LFO or EG filter modulation in positive or negative amounts. Many synthesizers also have a Keyboard Tracking control which routes the CV generated by the pitch control source to the filter cutoff, allowing the filter to open up as higher notes are played. The amount of this modulation is often governed by a pot — giving continuous variable control — but is also often implemented with a switch — giving either preset amounts of modulation, or at its most basic just on/off (that is, 100% or nothing). At 100% Keyboard Tracking, a self-resonant filter can be used as a sine wave oscillator, the pitch of which will follow the keyboard.
The Werkstatt’s filter has a CV input on the header, which is fine for simple self-patching, but two problems show themselves when you want to control this parameter from an external source: firstly, the necessity of hacking a cable together as described previously; secondly, the accuracy of tracking. The Werkstatt’s existing filter CV input point does not, in my experience, give accurate 100% tracking from an external V/oct CV, which spoils sounds that require the resonance to boost harmonics that are locked to note pitch.
The mod below overcomes these problems by giving the Werkstatt a separate, tunable Filter CV Input that can be trimmed to give suitably accurate pitch tracking.
How it Works
As with the Pitch CV Input mod, we’re going to simply duplicate the existing control input and make a slight alteration. The existing header input for cutoff control mixes its CV via a 47.5k resistor. In order to be able to give tunable tracking, this mod is going to use a 43k resistor and a 10k variable trimmer in series. 100% tracking should be somewhere towards one end of the trimmer’s range.
Solder the two extra components to the board, take the third leg of the trimmer to TP17 (purple wire in the photos below), and take the outer leg of the resistor to the input jack, which is mounted and grounded exactly as for the CV and Gate jacks.
Tuning the tracking is similar to tuning the pitch (a process described in the manual) — with the Werkstatt open, connect the external CV and play as normal, using full resonance on the filter, with the cutoff tuned so you can hear the pitch of its self-oscillation. Adjust the trimmer so the filter’s resonant frequency scales up the keyboard at the same rate as the note pitch — that is, two notes played an octave apart should give a resonant filter peak an octave apart.
Because the Werkstatt’s VCO cannot be silenced without modification, it might be easier to disconnect the pitch CV control while tuning the filter; alternatively, if you have a way of multing the pitch CV, connect it to both pitch and filter. You might try setting the filter to resonate at an obvious harmonic such as a 5th above the pitch, as any deviation in the tracking will result in some noticeable sonic artefacts.
Of course, you might not want a simple fixed 100% tracking filter. It would be possible to add a pot to allow the user to vary the tracking amount; you could install a switch to select between different resistors to give preset fixed trackings; you could route the pitch CV to a break-contact on the filter CV input jack so that it tracks by default unless a jack in insterted to over-ride it. My own mod is simple and quick and functional, and hopefully will provide a point of departure for your own experimentation.
43k 1/4W 1% MF resistor
1/8″ panel mount jack socket
Note: I make reference to the Moog Werkstatt schematics throughout. Copyright prevents me reposting them here; they can be found on Moog’s website.
In its original form, the Werkstatt’s own keyboard generates the Gate signal to trigger the envelope, and there is no obvious ‘Gate Input’ on the header. The existing Gate Out can be (ab?)used as a Gate In, but it’s not ideal, because as with most of these header points, anything coming in here isn’t buffered from the internal signal.
Adding a proper Gate In to the Werkstatt is straightforward enough, though a little more involved than the CV input; my approach doesn’t require the cutting of any traces, the only hack-work being the hole in the enclosure for a jack socket. It does require the end of one wire to be soldered to rather small SMT (surface-mount) components though, so you’ll need a suitably fine tip for your iron and a steady hand.
How it Works
The Werkstatt’s keyboard scanner outputs a logic high at U19 pin 3 when it detects a key press. As well as stopping the scan and loading the current key value into a latch (which feeds the VCO CV), this signal is buffered to provide a Gate, and differentiated to provide a Trigger. The Key On signal is buffered inversely by the Schmitt trigger of U14-F before being flipped back positive by U14-D. In order to add our external gate without affecting any other part of the keyboard circuit, we only need to bring the input of U14-D low. In this way, we can use both the Werkstatt’s own keyboard and an external Gate without having to switch between control sources.
The solution is to use a simple NPN in saturation to take U14 pin 9 to ground when its base is taken high. In other words, a positive external Gate will take the gate inverter input low, just as does the keyboard gate detector. Because there is a diode in the way (D14), our added transistor is isolated from the keyboard scanner clock and data-bus, so there won’t be any accidental mis-readings of the keyboard CV.
Another advantage of this solution is that the Werkstatt’s own envelope retains its Gate/Trigger operating modes, as our external Gate also gets differentiated; we are activating the Werkstatt’s envelope, not over-riding it.
The modification takes just four components and a socket, and fits easily on the PCB. The hardest part is soldering the wire from the collector of the transistor to the appropriate point on the Werkstatt’s circuit – I chose to solder it across the connection between R89 and C64, as the two solder points make a convenient place to lay a thin wire and give it a firmer purchase.
I presume you’ll be doing both CV and Gate input mods; the socket ground can be wired to the CV In socket ground, which I wired to a solder tag around the nearby PCB mounting screw (see also the CV Input page).
33k 1/4W 1% MF resistor
100k 1/4W 1% MF resistor
1N4148 signal diode
BC549C NPN transistor
1/8” panel mount socket
These parts are what I had handy. Pretty much any NPN with reasonable gain can be used here, and the signal diode is a generic one.