The ShaperPhaser

The waveshaper is a 'wavefolder'-type waveshaper, that in its simplest form with a pure sinewave as input creates a odd harmonics spectrum similar to 'through zero'-type linear FM modulation, but always follows the input pitch exactly and can never go out of tune, also the waveform will always stay the same. In essence it is comparable with a wavetable containing multiple cycles of a sinewave and where the input signal is used as an 'index' to read values around the centerpoint of the table. But as this is a pure analog module an array of diodes and opamps is used to 'create' the table. The carrier signal is 0Hz so it is not present.

The input is a bipolar VCA. The modulation depth depends on the level of the input signal. This can be controlled by the bipolar VCA or when the SHAPER IN input receives a signal directly from a OscHRM VCA OUT it can be set by the VCA in the oscillator.

The bipolar VCA can also double as a ringmodulator when controlled by a bipolar CV signal, no matter when at LFO or at audio rate. If no signal is connected to the VCA MOD CV input then it fades between no signal and full level. The VCA MOD control knob fades between an input signal in the leftmost position to an unmodulated full output level in the rightmost position.

The output connector of the VCA is normalized to the input of the waveshaper, so if the waveshaper input is not connected it will get its input signal from the output of the VCA. This normalization allows the VCA to be used separately from the waveshaper or as the input level control for the waveshaper.

The SYMM knob is sort of a pre-shaper that will add even harmonics when a sinewave is used as input. It makes the input assymetric. Fully opened, this double sthe pitch which creates the second harmonic.

The OD knob adds overdrive distortion to the waveshaped sound, adding fuzzy sounding high harmonics to the sound, but it does not alter the modulation depth.

You can create very complex spectra when using the the VCA as ringmodulator on the input of the waveshaper with its pre-distort and fuzz options...

The switch routes to the left side of the MIX knob.

In its top position it receives the input signal to the VCA and the MIX knob crossfades between the VCA input and the shaper output. Used this way one can fade between an unaltered input signal and the waveshaped signal where the waveshaping depth is set by the VCA MOD knob and can be modulated by a CV signal on the VCA MOD CV input. One would normally want to route the output to the Filter/VCA module.

In the middle position the switch is off and the MIX acts like an output level control. One would have the switch in this position when using the VCA and waveshaper separately.

In the bottom position the MIX knob leftmost position gets its signal from the waveshaper input and this would be a dry/wet type of efx-mix. One would use this mode e.g. when the VCA and waveshaper are used independently and one would like to have a dry-wet mix control for the waveshaper only, or optionally when the bipolar VCA would be used as an enveloped VCA.

Here is the flowchart of the Shaper :

I was not able to upload the video of the Shaper at Novars Masterclass as I cannot compress it under 100Mo so here is the link : https://vimeo.com/240798045. It includes technical information on how it was designed. Very interesting video.

For the Phaser I will only talk about a single Phaser, the DualPhaser is basically 2 Phasers with internal normalization for the inputs and outputs. If you really want to know more about it please let me know.

The Phaser a 8-stage phaser. It has a reasonably accurate 1V/Oct direct control input that can track the keyboard voltage.

Total control range is about 18 octaves. The Freq knob goes over the top 9 octaves of this range. Through the V/Oct and Modulation inputs you can go deeper, but you get into the LFO range and audible phasing effects would disappear.

It is however possible to use the phasing effect on LFO control signals in the 1Hz to 10Hz range by supplying the V/Oct with e.g. a fixed -5V control signal, which can create quite interesting LFO effects on e.g. drones. All inputs and outputs are DC coupled. Only the internal resonance is AC coupled, so resonance drops off below roughly 10Hz.

Additionally the phaser has a modulation input, also at 1V/Oct when the mode is set to sweep. The input is not normalized, in fact if no plug is connected the modulation level knobs receive a fixed voltage so a manual spread value can be set.

Audio input is maximum 12V peak/peak before clipping occurs and there is 6dB attenuation from input to output to enable resonance peaks without clipping. The Phaser can act as a distortion device, where you modulate the cutoff with the audio input.

The MOD input has three functions, determined by a three position toggle switch:
1. SWEEP MODE - Standard frequency sweep, like other phasers, where all peaks sweep up & down as a group.
2. SPREAD MODE - The upper peaks and the lower peaks are spread apart & pulled closer together and cross over each other. This produces vocal timbres/formants and all sorts of complex filter effects and sounds really wonderful.
3. HALF-SPREAD MODE - the upper peaks move further away & closer to the lower peaks and the lower peaks are not affected.

The MIX control is bi-polar which adds even more sonic possibilities. Set straight up at 12 O'Clock you get dry only, no phaser (bypass mode). Turn it clockwise and you increase the phaser mix, and the phased output is positive. The original signal is still present. Turn it counter-clockwise and you increase the phaser mix, but now the phased signal is inverted which sounds very different from positive phasing. The original signal is not present. It then may sounds a bit less louder at higher resonance settings.

The resonance control always provides positive (non-inverted) feedback, regardless if the mix control is set to positive or negative. If the resonance is fully turned, it can produce very nice percussive effects (pings).

There is no flowchart for the Phaser as it is basic and obvious.

Here is a video of the Phaser with great sounds and tips :

Thanks to Todd Barton for providing the picture of the module.

The Rungler

The purpose of the rungler is to create short stepped patterns of variable length and speed. One could categorize the circuit somewhere halfway between a plain S&H and a shiftregister-based pseudorandom generator. It needs two frequency sources to work and basically creates a complex interference pattern that can be fed back into the frequency parameters of the driving oscillators to create an unlimited amount of havoc.

The rungler is basically a CMOS shift register clocked by one oscillator and receiving its data input from the other oscillator. The output bits of the shiftregister are used as a binary code 'to do something with'. The bits are then fed into a DA converter. This DA level output voltage is fed back to the oscillator frequency control inputs. The output of the DA is the 'rungler CV signal'. To describe the rungler waveform in similar terms as like a sine wave or pulse wave I call it a 'stepped havoc wave'.

When the rungler signal is fed back to the frequency parameters of the oscillators it will change the triangle waveforms and pulse widths of the oscillator outputs, making other types of havoc waves, like a 'pulsed havoc wave' and a 'sloped havoc wave'.

The rungler will try to find a balanced state. In this way it behaves according to principle from Chaos Theory. There seems to be an unlimited amount of possible balanced states and when a balanced state is just slightly disturbed it can be noted that it takes a little time to find the next balanced state, with noticeable bifurcations, etc. Note that a new balanced state is defined by the exact position of the control knobs plus the previous state it was in.

The two wide range oscillators (not compliant to the 1V/Oct standard response) can cross-modulate and/or can be modulated by external signals. Oscillator A produces a Sine wave and provides clocking for the shift register. It has fluct modulation with rate controlled by Osc B Rate. As Osc B can go in audio range, fluct can cause sort of soft sync effect. Oscillator B produces a Triangle wave and provides the material for the shift register. Oscillators A & B are available as direct outputs. The output of the shift register can be applied to oscillator A in Stepped, Smoothed and Pulsed modes simultaneously. These Stepped, Smoothed and Pulsed signals are also available as direct outputs. Smoothed and pulsed are derived from stepped signal. Smoothed is sort of filtered stepped signal. Pulsed has 2 states : On and Off.

Clock In & Out makes syncing to other modules possible.

The Rungler has different operational modes that can be selected with a switch.

Random - Constant change when VCO B is lower freq VCO A, locked when higher. At audio rate, if Osc B is tuned higher than Osc A there is a sort of noise pattern but that’s loop quicker than the normal digital noise. By lowering VCO B Rate there is noise & every time you increase it, it has different kind of timbre.

Sparse - VCO A Rate can be very slow, can have sort of stepped signal. When modulating pitch, you have a short sequence of 32 notes. VCO B is locked when higher than VCO B.

Dense - Chaos increases with VCO B frequency. It never locks.


Here is the flowchart of the Rungler :

Check these videos for sounds and more explanations :

I wasn't able to upload the video of the Rungler at Novars Workshop so here's the link : https://vimeo.com/237385725

Thanks to Todd Barton who provided the picture of the module.

Triple LF-VCO

This module has three independent low frequency voltage controlled oscillators or LF-VCO's that can also be synced and cross-modulate. Three LEDs give visual feedback of the activity of the three LFOs. The modulation inputs of all VC-LFO's are normalized at the input connectors in such a way that everything can crossmodulate and sync.

LF-VCO A provides a Sine wave wich has no corners, giving very smooth modulation. It has a wide frequency range going from quite slow to well into the audio range, so it can be used as a sound source as well as a modulator. This LFO is tied internally to LF-VCO C. The latter will modulate the former in FM and in combined FM-AM called "fluctuation". This "fluctuation" modulation soft-syncs to the harmonics of the modulating signal. Therefore the modulation depth 'feels' more natural even when the frequency of the modulated waveform is increased. This results in an extremely natural sounding vibrato, much gentler than through a normal LFO. Pitch is controlled manually. By mixing LF-VCO A & LF-VCO A Fluct, at a certain point, one can only hear the amplitude modulation which is interesting when modulating the pitch of an oscillator.

LF-VCO B is useful for clocking a sequencer or envelopes with the pulse. It has a slower range than LFO A. There is a triangle output and a pulse output. These outputs are unipolar. The wave shaping control is set manually. The triangle output can be modulated from downsloping saw to a triangle to an upsloping saw. The triangle output can then be used as envelope. The pulsewidth also varies with the waveshaping control.

The frequency of LF-VCO B can be modulated internally by LF-VCO C or by an external signal.

LF-VCO B has three operating modes, selectable with a switch:
- MODE : LF-VCO B is free running
- SYNC : LF-VCO B is synced to LF-VCO C or to an external signal
- HALT : LF-VCO B oscillates at a frequency set by its Rate control during the Rise of the Triangle wave of LF-VCO C and will "freeze" during the Fall of the Triangle wave of LF-VCO C. LF-VCO B will restart with the next cycle of LF-VCO C. When "frozen", LF-VCO B will output a fixed, momentary voltage. Halt mode works best on pitch.

LF-VCO C has a triangle and an pulse output. Rate goes from several minutes to around 100Hz. The LF-VCO C Mod slows down the rate of the downgoing slope of the triangle. The rate of the upgoing slope remains the same. So, when e.g. speeding up the LFO and opening the C MOD knob the waveform will take on the form of a fast attack and slow decay envelope like a sawtooth wave. The pulse output is high when the slope is going up and low when it is going down, so the pulse length will remain the same but the time between the pulses will
increase.

Regrettably this selfmod feature is obsolete as Rob developed a S&H instead.
Here is Rob's description about it :

The modulation input of LFO C has a S&H right after the input connector, but just before the input level knob. LFO C triggers the S&H circuit. 
On every positive peak and on every negative peak of the triangle waveform the S&H samples the LFO C modulation input signal and holds the sampled value during the slope that follows. This means that the S&H causes every upslope and every downslope of the triangle to have a different duration, defined by the momentary value the S&H happened to sample. However, the slopes will remain perfectly linear, only their steepness is affected. The effect is that there is a more or less random spread in time. On the pulse output this causes pulses of different duration, an effect also named clustering in time.

The output of the S&H is only available internally, but also normalizes to the LFO A modulation input. This can be overridden by inserting a jack with another signal into the LFO A modulation input. When no jack is connected to this LFO A input it will follow the S&H on the LFO C modulation input.

When no signal is inserted in the LFO C modulation input it is normalized to the output of LFO A. This means that when no jacks are applied to the LFO A and LFO C inputs, but their input level knobs and the LFO A fluctuation knob are opened, the result is constantly varying pitches and rates on both LFO A and LFO C, caused by the crossmodulation between LFO C and LFO A. On LFO A it will always sound like a stepped pattern unless you also open the fluctuation knob, as that will modulate the pitch slightly with the slopes of LFO C or with an external signal when the fluctuation input has a jack with a signal plugged in.

Note that for pitch or frequency modulation there is a behaviour that a fast pitch can easily be deeply modulated by a slow pitch or rate, but a slow pitch or rate is much harder to modulate deeply with a much faster pitch. A S&H changes this FM equation into a pure statistic function that becomes pitch independent, resulting in slow rates reacting much deeper to faster rate modulation signals as without the S&H. E.g. when a 1Hz LFO is modulated straightforward with a 1kHz audio signal there may only be a tiny bit of 1kHz zippery noise on the LFO signal, but the 1Hz will not seem to change much. But through the S&H the LFO will instead go all over the place, defined only by the average amplitude of the 1kHz signal.

There are maximum possible pitches for the three available LFO’s, they basically can not go faster as the maximum setting on their rate knobs. To use the S&H modulation on LFO C it is often a good idea to not set the LFO C rate knob to its maximum. There is not such a limit for the minimum rate times, though a very large negative modulation signal can cause a LFO to actually stop oscillating. If this happens when the LFO C rate knob is at its minumum, the modulation level knob is fully open, and a sampled negative signal is stopping the LFO, and thus also not sampling a new value that can start it again. In this case LFO C may appear to be frozen. In that case you can just open the rate knob a bit until it starts oscillating again. Or wait for a very long time for the S&H capacitor to eventually loose its charge.

A tip is to e.g. apply the LFO C Pulse output to the S&H input of one of the EnvGens while clocking that EnvGen with a stable clock. Then feed the output of that S&H to the Gate input of the second EnvGen. While the first EnvGen is triggered at a stable rate the second will trigger only now and then in an irregular rate but synced to the clock of the first EnvGen. If LFO C is deeply modulated the second EnvGen gates seem to appear in clusters in time.

And there are many other uses if you want to introduce a smaller or larger amount of variation in dynamics to an otherwise static pattern, or to spread events unevenly in time.

A video of the new feature is at the end of the page.

Here is the flowchart of the Triple LFO with LFO C selfmod.

Here is the flowchart of the newer Triple LFO with S&H.

Here is a video of the Triple LF-VCO. As I couldn't upload the one at Novars please follow this link : https://vimeo.com/239786705

Here is the explanation of the new S&H feature.

Thanks to Todd Barton for providing the picture of the module.

The Dual Fader

A Swiss Army knife for modular.

The Dual Fader is a voltage controlled mixer module with several options useful for different applications for dynamic signal. It can be used in various ways and at different points in your patches. It contains two RMS 'equal loudness' voltage controlled faders with both two inputs and two outputs. Depending on how jacks are connected the basic functions are crossfader, panner or alternating VCA.

The inputs are DC coupled which means it will process CV signals as well as audio signals. You could have one channel set up as a fader for CV signals and the other as a panner for audio signals.

In a two inputs/one output configuration, the channels will act as individually voltage controlled crossfaders usingAX and/or BX outputs and the signal goes from one signal to another with the fade control. Fade parameter is voltage controlled.

In a one input/two outputs configuration, the channels will act as individually voltage controlled panners with "energy loss compensation" set at -3dB with the control centered (equal loudness curve).

It is good to note that when a signal is connected to only input A1 (or B1) and the output is taken only from AX/L out (or BX/L out) the input signal will go to both VCAs and the output will be taken from the mix of the VCAs.

Each channel has a 20dB boost available to, for example, adapt external line level signals to the signal levels inside the modular. This is also very useful to crack up the level of stompboxes to synth signal level. This way it is easy to use a fader to e.g. crossfade between the input and output of an external guitar stompbox effect or line level effects processor. With some clever routing using the matrix it is also possible to create external feedback paths for external effects and have the feedback under voltage control. You can watch examples in the videos at the bottom of the page. This 20dB boost is applied to the A2 & B2 inputs. A1 & B1 inputs are normalized to the A2 & B2 inputs respectively, so that when no jack is inserted in the inputs A2 & B2, the signals present at A1 & B1 are duplicated and can benefit of the 20dB boost. It now has an interesting feature, as when it is opened it will start to clip the input signal when a high level signals like from an Osc or the output of a filter is used. Now you can use the module as a distortion by letting the signal clip and basically set a mix between the clipped and the unclipped signal by the fader knob.

One application for this boost would be to feed a signal into e.g the A1 input and the input of a stompbox effect. Then route the output of the stompbox into the A2 input and crank up the gain from the stompbox output (that is often at line level) until it matches the A1 input level and now crossfade between the dry sound and the wet stompbox effect, e.g controlled by an envelope or a sequencer. Alternatively it is also possible to use a fader to set e.g. echo time on a stompbox, in this case one sends the output of a fader to the input of an echo-box, and reconnect the output of the echo-box to the A2 input, with the audio signal to be processed on the A1 input, while listening to the fader output. When the fader is turned left the dry input signal will be present on the output. By turning the fader to the right the output signal that is also routed into the echo-box will now route back a delayed signal into input A2 and the delayed signal will also be rerouted into the echo-box and thus start to create echo repeats. When turned more and more to the right (and the gain level set correctly) this patch will eventually create an infinite repeat. And this is now under full voltage control.

The AX TO B LEVEL knob receives a signal that is the crossfade mix of input A1 and A2. This signal is generated internally and is fed through the knob to the second fader and internally added to the B1 and B2 inputs.

The switch is in the line to the B2 input and it has three possible functions:

1) The AX MIX position.

The signal from fader A is passed through the AX LEVEL knob in an equal amount to both B inputs. In this mode you can for instance make a mix of two signals with fader A and then pan this mix to outputs BX/L and BR. So, you would e.g. connect two oscs to inputs A1 and A2, not use any of the A outputs, open the AX LEVEL knob and connect the BX/L and BR outputs to the left and right channels of a mixing desk or to two different audio processors. The FADE A knob sets the amount of mix of the two inputs and the FADE B knob pans the stereo position of the mix from left to right. The AX TO B LEVEL knob can be used as an overall volumeknob. Modulating the FADE B MOD at low rates yields tremolo / leslie effect-like results.

2) The AX>B position.

In this mode the internal mix signal from the A1 and A2 inputs is fed only to the B1 input. In this mode you can e.g. mix three signals to the BX/L output. Connect two signals to the A1 and A2 inputs and a third signal to the B2 input. The FADE A knob sets the mix between the A1 and A2 signals andthe FADE B knob sets the mix between the A1/A2 mix and the signal on the B2 input. This way you can mix three signals in any ratio with only two knobs and always have an output signal, as long as the AX TO B LEVEL knob is opened. You can even add a fourth signal by plugging it into the B1 input and set its input level with the B1 IN LEVEL knob.

3) The RM position.

In this case the internal mix signal of the A1 and A2 input signals is fed to both the B1 and the B2 inputs through the AX TO B LEVEL knob, but the signal that goes to B2 is inverted before mixed with the B2 input signal. This means that when the FADE B knob is in its middle position the signal that goes to the B1 input is cancelled by the inverted signal that goes to the B2 input. If you start to modulate the FADE B knob with an audio signal on the MOD B IN input and opening the FADE B MOD knob, while the AX TO B LEVEL knob is open the output signal on BX/L will alternate between its normal signal and the inverted copy of the normal signal and ringmodulation will happen.

To do this just connect an osc to the A1 input (your carrier signal can also be a composite signal of A1 and A2), set the FADE A knob to its middle position, set the switch to RM mode, open the AX TO B LEVEL knob, plug a second oscillator into the MOD B IN connector, listen to the BX/L output, close the FADE B MOD knob and adjust the FADE B knob to the middle until you hear silence. Now slowly open the FADE B MOD knob and you will hear a ringmodulator effect. Tune the oscs so that they produce a nice sound and play both oscs with the same 1V/Oct signal from e.g. a keyboard. Changing the tuning ratio between the two oscs, setting their basic timbre with the ODD and EVEN HRM knobs and playing with the FADE B and FADE B MOD knobs will take you through a whole range of ringmodulator timbres. Due to the nonlinear crossfade curve there will be some distortion in the ringmodulation, and the modulation depth will define the amount of distortion. But as the curve now has sort of a compressing effect on the modulation waveform a triangle wave will be sort of compressed to a sinewave at a certain modulation depth and fine tonal control can be achieved by setting the modulation depth at certain levels for certain modulating waveforms. E.g. you can create the typical vintage 'diode-transformer' ring modulator sound from the fifties and sixties as used by e.g. Karl-Heinz Stockhausen or those SF-timbres from many sixties and seventies SF movies. The A2 IN GAIN control may be used to slightly overdrive carrier input into the modulator, for example leading to more 'vintage' sounding results.

The 'ducking' mode happens when two different signals are connected to e.g. the A1 and A2 inputs and the AX/L and AR outputs are both connected to two inputs of a mixing desk. In this mode the A1 signal goes straight to the AX/L output and the A2 signal to the AR output. When playing the FADE A knob one signal will increase while the other automatically decreases in volume. If the FADE A MOD input would now receive e.g. a side chain CV signal from a compressor that signal would automatically drop the volume on the A1 > AX/L line and open the A2 > AR line. This is often used when a narrative is added to a music track for a film or documentary, so that when the speech comes the music is automatically lowered in volume. It would however need an extra compressor with a side chain CV output or an envelope follower module. Of course you could also modulate this 'ducking' effect with an LFO or an envelope signal.

Here is the flowchart of the DualFader :

Check out these videos, you will find tons of good examples of patching this great module.

Thanks to Todd Barton for providing the picture of the module.

The TwinPeak Filter

The best filter ever?

In essence there are two independent 18 dB lowpass filters in parallel, only sharing the resonance knob and their outputs mixed in 'anti-phase', or in other words their output signals are subtracted from each other, instead of added. In practice this means that if both filters get the same input signal the LP band set to the lowest cutoff is subtracted from the LP band with the highest cutoff and this creates a bandpass response that only passes what is 'between the two cutoff frequency settings'. The filters can 'pass' each other, meaning it doesn't matter which filter is tuned higher or lower it is always the band between the two cutoff settings.

The trick in the Twinpeak is how the input to both filters is mixed before they go into the two parallel filters. So, not only the outputs are mixed but there is mixing on the inputs going on as well. The A and B input connectors both have their separate input mixers. Basically the subtraction is not done at the outputs, those mix positive, but the input to one of the filters is inverted in phase. This has exactly the same effect as subtracting the outputs. For input A its signal always goes into filter one and the curve A knob sets how much of the input signal goes into filter two. So, if the curve knob is closed the IN A signal only goes into filter one and in the final output you only hear the effect of filter one on the IN A input signal. And thus it has a lowpass response as filter two does not get the input signal. That means that there is also not the subtraction at work that create the bandpass, as filter two gives nothing to subtract. Then, by opening the IN A curve knob, the signal level of IN A into filter two is increased and now there is something coming out of filter two and the final output curve changes towards a bandpass curve. For IN B it works the same, but this input always goes into filter two and through the curve B knob to filter one.

Basically it means the following options:

1) When both curve knobs are at 'lowpass' IN A goes into filter one only and IN B goes into filter two only. Now the Twinpeak works as two independent lowpass filters with their outputs mixed into one output connector. Because they are mixed in anti-phase the IN B signal part will be in antiphase on the output, while the IN A signal part is still in phase. Modulate each filter independently by the PEAK 1 MOD and the PEAK 2 MOD knobs only, as the CUTOFF MOD knob will modulate both filters in a linked way.

2) When curve A is at lowpass and curve B is at bandpass you hear the lowpass part of IN A, set only by the cutoff peak 1 knob. And from the signal on IN B you will hear the band between the cutoff peak 1 and the cutoff peak 2 knobs .

3) When curve A is at bandpass and curve B is at lowpass you hear the lowpass part of IN B, set only by the cutoff peak 1 knob. And from the signal on IN A you will hear the band between the cutoff peak 1 and the cutoff peak 2 knobs .

4) When both curve knobs are at bandpass both IN A and IN B will be bandfiltered with the same band between the cutoff peak 1 and cutoff peak 2 knobs.

5) And there are many settings in between, also when there are two totally different signals on IN A and IN B. It is best to judge the effects by ear.

If one only want to use one input on an external signal one can either close the level knob on the oscillator or one can plug the signal into IN A and plug a matrix jackplug into IN B  to completely break the normalization with the oscillator outputs.

The 1V-OCT input will affect both paralllel filters equally, like the CUTOFF MOD knob does.

The filter has a very good 'ping' characteristic at its maximum resonance. It lets itself also be modulated very well by audio rate signals. E.g. feeding a slow B PULSE from the TriLFO into e.g. IN A on the Twinpeak and feeding the A SINE from the TriLFO into the CUTOFF MOD on the Twinpeak will create a wide range of 'metallic' percussive sounds when the LF-VCO A RATE is tuned to audio rates.

The PEAK 1 MOD and PEAK 2 MOD have there zero level at twelve o'clock and turning them to the left will 'reverse' the direction of e.g. an envelope sweep.

The VCA can be totally independent if VCA IN is used for the audio input and the VCA CV with any unipolar, positive control voltage.

Here is the flowchart of the TwinPeak filter:

What if?... A TwinPeak filter bank (more than one module):

A single Twinpeak can be used to filter a single band of variable bandwidth and with 'corner-peaking'. The last means that the Twinpeak is not simply a bandpass filter with one resonance peak frequency in the middle of the band, like a 12dB state variable filter, but it means that the bandwidth can be set between two separately controllable cutoff frequencies and that when resonance is raised two resonant peaks appear, one at the low cutoff and one at the high cutoff of the band. And the cutoff slopes for both the low and the high cutoff slopes of the band are 18dB, which would only be 6dB on a state variable. So, in essence the Twinpeak is a 36dB bandpass filter.

A signal could first be panned in the DualFader module and then the AX/AL OUT and AR OUT pan outputs be fed to the Twinpeak A and B inputs and the CURVE IN A and B knobs set to lowpass response. This would allow the fader to fade from the A lowpass response, through a bandpass response, to the B lowpass response. Modulating the fader with a LFO or envelope can give quite lively filter results on otherwise static sequences and drones. Adding PEAK 1 MOD and PEAK 2 MOD and a little audio rate modulation on the CUTOFF MOD could be the icing on the cake.

If one would have two or more Twinpeaks it would be possible to make a high quality 'filterbank' with 18dB cutoff slopes with several Twinpeaks, using each Twinpeak for one band. They could all receive the same input signal and the outputs could be mixed on the matrix or with a DualFader module. Or, as in the previous example, a DualFader module can be used to feed the total of four pan outputs into the A and B inputs of two Twinpeaks.

One of the most interesting effects of the Twinpeak is when the two resonant peaks are swept independently, as this creates vowel-like effects. The most obvious use for two Twinpeaks would be to sweep four resonant peaks independently and thus get even more pronounced vowel-like effects. For this use a DualFader to pan one audio signal to the total of four pan outputs and patch those to the A and B filter inputs on the two Twinpeaks. Then mix the Twinpeak outputs in the matrix using the two bipolar VCAs on the NodeProcs to get a complex modulateable filter with many modulation points, as the Dualfader can be modulated, the Twinpeaks can be modulated in several ways and the bipolarVCAs can be modulated. Or, instead of using the matrix/bipolarVCAs, connect the Twinpeak outputs straight to two different channels on a mixingdesk and set stereo panning on the desk. 

Quite some patching options here.

Thanks to Todd Barton for providing the picture of the module.

The 24dB Filter/VCA

This is a 4-pole (24dB) multi mode VCF with an unusual topology in that it doesn't have separate outputs for its different responses (LP, HP and BP), but instead it has three inputs and one mixed output. The three inputs are placed upfront of the filters and each input has a manual level control that determines the amount of signal that goes into each filter. 
The 24dB Filter has a 24dB lowpass slope, a 24dB highpass and the bandpass slope is 6dB highpass and 18dB lowpass.

As the input jacks are normalized, the filter can also be used to create a variable slope for one input signal. This way one can dial in specific responses by setting the levels of the respective inputs. As when a signal is connected to only the LP input the normalization passes the signal on to the next HP and BP inputs. But, one could also input a signal to the LP input, another signal to the HP input and a third signal to the BP input. So the filter can be used as a three input mixer with separate level controls. The output would then be a blend of the first signal treated by the LP filter, the second signal treated by the HP filter and the third signal treated by the BP filter. For example, setting the LP knob to 10 and the HP knob to 7 gives you a resonant notch filter with more emphasis on the lowpass section.

One can crossfade one signal to an other with LP input & HP input. When the LP and HP inputs are used for the same signal they can pass this signal unaltered at specific mixer settings, as the LP and HP slope are fully complementary, in this case it will even suppress the resonant peak, although that peak will be present on the BP input signal.

The LP, HP & BP inputs are DC-coupled so that also CV signals can be processed and it means it can also be used as a slew limiter when you feed it CV. With high resonance, it then imparts very interesting behavior on the resulting control voltage!

Both Cutoff frequency and resonance are under manual and CV control. There's a 1V/Oct CV input to the Cutoff frequency and two supplementary FM inputs with attenuators. The filter maintains a steady perceived volume level at all resonance settings. This filter will not go into self oscillation. 

Mod II is the level input knob of the second cutoff modulation input. If there is no jack in the second modulation input opening this knob will introduce a smooth, subtle and very natural distortion effect on the resonant peak that is similar to tube distortion (all harmonic distortion). When a jack is connected it sets the amount of cutoff modulation. Often the first cutoff modulation input and level knob are used to control the tracking of the filter to a keyboard, and then the second modulation input can set the level of e.g. a filter sweep.

The filter reacts exceptionally well on audio rate modulations signals, which can add a further range of new and interesting timbres to the already quite versatile module.

Next to the main filter output there's also a VCA output for amplitude modulation of the filter (unattenuated CV input to the VCA and no control over initial gain of the VCA). The VCA has an exponential curve.

You can use this filter to get some really interesting slowly evolving waveforms. Example: patch a fat PW modulated square wave into the LP input, and a triangle wave through a slowly sweeping wavefolder into the HP input. Now sweep the filter cutoff with a mixture of very slow sine LFO plus an audio rate sine that is just slightly detuned from the two audio input waveforms... Bizarre spectral 'crossfading' from one timbre to the next, shifting from the PWM osc to the wavefolded triangle, and producing complex combinations at the points in between...

Here is the flowchart of the 24dB Filter/VCA :

Here are some videos of the module. The second one at Novars workshop greatly explains how the filter is designed.

Thanks again to Todd for providing the picture of the module.

Basic Electricity #15 in Berlin

On April 10th 2015 Basic Electricity #15 took place in Berlin as a special event dedicated to Rob Hordijk instruments and musicians playing them.

Rob himself gave a hands-on demonstration of his modular synth with a special emphasis on the Dual Env (Rev 3) and the S&H function. 

You can watch and listen to his explanations in the videos.

Thanks to Richard and Navs for setting up this event.

The Blippoo Box

Ah! The fabled Blippoo Box. There are many adjectives that describes the sonic character of the "bent by design" box : organic, unique, chaotic, unpredictable, etc...
But you really have to play it once to have a small idea of its potential.

Rob also offers a banana version of the Blippoo Box for those who want to interact with their Buchla/Serge/Banana systems.

There is a great page from Richard Scott's website about the Blippoo Box, including an album and videos with and without the Benjolin.

Hans Tammen has an instructive page about the Blippoo Box too with videos of different applications.

But the most complete and deeply instructive article about the Blippoo Box is in the Leonardo Music Journal issue #19 of 2009. Be sure to read and own it. Wonderful explanation about the concept and the design of the box.

Here is the flowchart of the Blippoo Box.

Check the Blippoo Box and Blippoo hashtags on Instagram for short videos of the box.

Also Youtube has some great gastric sounds. Check for more.

EEEM 2012 : the first series of videos from Rob.

This is an "old" playlist of videos from Mallorca. 

The case is different, lot of updates in the modules since, new modules, etc...

But still, this is a nice ressource and there are already some nice patches in it. Enjoy!

Thanks to Pedro Trotz who filmed and shared theses videos.

Modular Meets Leeds synth tutorial

In this video, Rob is showing a great use of the Shaper module. One can create a totally complex sequence using a waveshaper.

And it is also nice to have more modules explanation.

Live performances of Rob Hordijk

To my knowledge there are only 2 videos of Rob performing live. I really like the one in Rotterdam, great set.

So enjoy !!!

Rob Hordijk at the Noodlebar December 22th of 2016.

Rob Hordijk at Modular Meets Leeds 2017.

Rob Hordijk Masterclass November 21th 2015

A masterclass with only Hordijk users took place at Rob's workshop in The Hague in November 2015.

Anyone who had specific questions could ask and Rob would gave the answer.

The masterclass was more oriented in noise and complex random, self-generating patches.

Thanks to Chris who recorded, edited  and shared the videos.

The Novars Workshop 3-4 June 2017

At Novars Research Center in Manchester, UK was held a long workshop explaining the Hordijk Modular in June 2017. 

Philosophy of concept, functions, design were discussed and each module was explored in depth. There was a lot of technical explanations as well as demos. This is a great and wonderful ressource for any synth enthousiasts, sort of a video manual.

Novars Hordijk Workshop

You will find some of these videos in the modules description.

Thank you very much to Mark Pilkington for making the videos and to the Novars folks for sharing this treasure.

A pic of the Manchester Hordijk Modular.

The Hordijk Minutes

Todd Barton, the famous synth wizard, offers a series of small tutorials/demos of instruments in his Instagram account. 

He just wants to share what he is learning about the Hordijk with others. And we are glad he does.

When asked about why he created it Todd says :

"I like short forms: in poetry-haiku, in literature-short stories, etc. so I thought it would be great challenge to offer some short tutorials. Oh! Ans I was new to Instagram and there is a & minute video limit for content :-)"

He had a lot of positive response from both Hordijk practitionners and other analog synthesists that could translate what he was doing to their formats.

But the most amazing and encouragement came from Rob :

"Thank you very much for the ‘one minutes’ [tutorials], they get a really good reception with users. I must admit they inspire me too, especially as I have come to the habit of immediately setting up complex patches and then getting lost in how to play them. These simple examples let me go back to the basics, with often better results. –Rob Hordijk"

There will be more Minutes to come in the future but for now I invite you to check the ones that are online in Todd's Youtube playlist

Some extra videos are in the playlist. Be sure to watch them too.

The OscSync

The OscSync has at its core a sawtooth oscillator with a downgoing slope. On the flank of the sawtooth it syncs a second built in osc that has a waveform that can be morphed between a triangle and a sawtooth.

A hardsync sound basically has two sonic components, one is a formant effect that when swept gives a flanger-like sound, the other is a pulse that is produced by the transient flank introduced on the syncing moment. This transient has a level that depends on the frequency of the synced osc and when swept it goes up and down in level in basically an unwanted way. As this new transient has a lot of sonic energy in the high end of the spectrum it doesn’t sound good on e.g. sine and triangle waves that by nature do not have one or more transient flanks in their waveform. And so it is of interest to suppress only this sync transient effect. For this a sync transient suppressor is built into the osc and this makes the sync sweep sound as smooth as silk, no matter the type of waveform.

Both the original sawtooth and the ‘transient suppressed’ synced waveform are brought out on separate output connectors.

There is a subsquare divider built in that goes into a built in VCA. This signal can be used to modulate the sync sweep. The effect is that on every alternate wavecycle the formant structure caused by the sync is alternated by two settings that depend on the level of modulation. This gives a ‘talkative’ effect on the sound. This subsquare is also brought out on a connector to modulate e.g. a filter that has the basic sawtooth wave as its input.

The sub out can also be used in a mix with either the sawtooth or the synced sound to add a suboctave bottom to the sound.

There also is a built in S&H that triggers on the flank of the main sawtooth osc. It has its own input and output, but by default the input is normalized to the internal transient suppressor and samples the level of the sync transient. The output is normalized to the modulation inputs of the main pitch, the subsquare modulation level and the sync sweep CV inputs. The sync transient depends on the pitch of the synced oscillator in an ‘irregular way’ it can be fed back to the sync sweep input to cause a ‘chaotic’ modulation on the sync sweep. This effect is not unlike the effect of a rungler circuit. When also fed back to the main pitch modulation input it will change the length of the wave cycle with an amount that will be different for each cycle. As the oscillator can go into the LFO range this can also create quite nice effects.

The normalizations can of course be overridden by plugging in a cable at the CV input connectors or close the modulation level knobs.

The S&H can also be used with an external audio signal to ‘bit crunch’ that signal.

The sync transient suppression is what makes this osc unique. The S&H and subsquare make it great for a vast range of different sonic effects.

As this oscillator can do a range from quite traditional synth and sync sounds, but with a very smooth sound, up to rungler-like chaotic sounds the Twinpeak is the prime suspect to do filtering on the several output signals from the osc. The two Twinpeak ‘peaks’ can e.g. be modulated separately by the subsquare output, the S&H output or an audio range signal from e.g. the OscHRM or an LFO and/or envelope.

Here is the flowchart of the OscSync : 

Chris David aka Osc1899 filmed these videos of Rob explaining the module :

Also this video of the Novars workshop is quite nice :

I also recorded a small series of demos when I received the OscSync Rob Hordijk OscSync demos

And there are also more videos and audio available. Check Todd Barton's channels and be amazed ;-)

Thanks again to Todd for providing the picture of the module.