Rob Hordijk Triple-Input 24dB Filter

OSC HRM
The 2U wide Harmonic Oscillator module (OSC HRM) is used to create pitched waveforms with dynamically controlled timbres. Pitch control law is 1V/Oct and the module uses a platinum element for temperature stabilization of the pitch curve. Maximum frequency range is from 0.5 Hz to 16.000 Hz and the scale is perfectly tuned in the middle six octaves up to a pitch of 4.000 Hz. When playing pitches higher as 4000 Hz the 1V/Oct scale starts to break down, due to the necessary internal bandlimiting in the harmonic generators. The module uses a biquad sine/cosine oscillator at its core and through a process of recursion harmonic series of overtones are generated. There are two recursion paths, one that produces all harmonics and one that produces only the odd harmonics. By gradually opening the knobs that control the amount of recursion more and more harmonics are generated. When only the all harmonics path is used the waveform morphs smoothly from a sinewave to a waveform that closely resembles and sounds like a sawtooth or an inverted sawtooth. Opening only the odd harmonics knob will smoothly morph from a sinewave to a squarewave. When opening both knobs effects like pulse wave modulation are possible. Building up these harmonic series is under full voltage control and can be modulated from slow LFO speeds to fast audio rates to create FM timbres. When the waveforms are modulated there is a negligable amount of detune (less than 1 cent), though when modulating at audio rates an asymmetry in the modulating waveform can cause detune effects on deep modulations. There is an additional VCA incorporated in the module. The final output signal can be taken from a point just before the VCA and at the output of the VCA. This enables the module to be easily used in a situation where one wants to modulate another module by an audio rate signal and have the modulation depth under voltage control using e.g. a LFO waveform, an envelope voltage signal or a play controller that produces a control voltage, while still having the full output level signal available on the full output to serve different purposes. The waveforms have an exceptionally warm sound and when dynamically modulated have a deep spatial and organic character. With only one OSC HRM and one DUAL ENV module you can already have a voice with dynamic timbral and volume control that can do e.g. a pretty solid bass line

Dual Phaser
The Dual Phaser is 2U wide and has an internal CV voltage scale of 1V/Oct. Each phaser has a reasonably accurate one volt per octave direct control input that can track the keyboard voltage. Normalization is used, routing the V/Oct input signal of phaser1 into phaser2 when the phaser2 V/Oct input is left unplugged. Total control range is about 18 octaves. The Frq 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, so CV signals can pass the module equally well as audio signals. Only the internal resonance is AC coupled, so resonance drops off below roughly 10Hz. Additionally each phaser has a modulation input, also at 1V/Oct when the mode is set to sweep. When the mode is set to spread it behaves like the modulation sensitivity is halved, also when it is in half mode where only half of the poles in each phaser are modulated by this input. These inputs are 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. Audio routing is as follows: If a jack is connected to input1, and if input2 is unconnected, then the audio will route into both phasers. In this mode you can use the two phaser outputs as a stereo signal. Connecting a jack to input2 will override this internal input1->input2 connection and separate both phasers. If audio is routed into input1 and if input2 is left unconnected, and if a jack is connected into ónly output2, then the two phasers are automatically set to "inverse parallel" mode. Meaning that if both phasers are set to exactly the same knob settings the phaser outputs would be in exact reverse phase and thus result in almost silence. If audio is routed into input1, and if output1 is connected with a short cable to input2, and if output2 is taken as the overall output, the two phasers are in series and thus result in one 16-pole phaser. To summarize: you can use the phasers fully separated, parallel with two (stereo) outputs on one input signal, parallel with mono output but with one phaser in reversed phase before the mixing of the outputs of the phasers take place on output2, or in series. All this is accomplished by the internal switches in the connectors and only depends on which inputs and outputs have a plug."

Active Matrix
The 4U wide Active Matrix module is a fully buffered eight by eight matrix where any one of eight input signals can be added to any one of eight outputs. By using ¼-inch tip-ring-sleeve insert jacks for the matrix nodes (equal to stereo jacks) a whole range of applications become possible. First the column input signal is buffered and then routed to the tips of the nodes in that column. The ring signals of the nodes are basically summing inputs and summed to the final output signals at the ends of the rows. By connecting a stereo jack where the tip and the ring are connected directly together, a connection with unity gain is made from a column input to a row output. If the tip-ring connection in the jack goes through a resistor an additional attenuation can be accomplished. E.g. a 30k resistor will attenuate by 6dB and a 91k resistor by 12dB. When a stereo audio cable is soldered to a jack plug and on the other side of the cable a potentiometer is attached the potentiometer will act like a pot on the node, enabling to set the mix level by the pot. Basically each node is an insert, just like the inserts on a mixing desk. And can thus be used in the same way. So, using a jack with a pot means to ‘insert’ the pot into the signal path. This means that you can insert any other outside world device in the signal path by using an insert cable with a stereo jack on one side and two mono jacks on the other, provided signal levels match of course (e.g. 5V pp oscillator output signals will severely overload line level inputs on e.g. a digital effects rack or the guitar input of a stompbox). One could also connect a resistive sensor like a light dependent resistor (LDR) to a jack and make the node light sensitive. Each column also acts like a multiple. When a mono jack is connected into a node it will pick up the column input signal from the tip. But the ring input is now short circuited to the ground through the sleeve of the mono jack and will so disable any input from this particular node to the row output. This will not interfere with any other nodes in the same column or row, because of the full buffering of both the column inputs and row outputs. So, any node that is not used to route a signal to a row output can be used as a multiple output. Meaning that the matrix is also eight multiples with one buffered input and eight buffered outputs on each multiple. Matrices with bigger sizes can be built on demand. Input columns come in multiples of eight and any number of output rows are possible.

MiniBay
The 2U wide MiniBay is a smaller version of the Active Matrix. It has a four input by six output fully buffered matrix plus two passive multiples with five jacks each. The matrix section works exactly like the 8x8 Active Matrix.

Dual Envelope Generator
The 2U wide Dual Envelope Generator module (DUAL ENV) is a fully voltage controlled envelope generator specifically designed to be used with sequenced music. There are two different types of envelopes available, one is a four stage (attack, decay1, break level, decay2, release) envelope and the second is a one stage envelope (decay only). Both envelope generators share the same gate input, meaning that they can not be triggered separately. Triggering treshold is at roughly 100mV above ground and also accepts e.g. triangle waves. The ADBDR envelope is primarily intended to be used for volume envelopes. When the decay2 knob on the first envelope generator is fully open the decay2 acts like the sustain that you find on most of the traditional envelope generators. In this case the break control will act like the sustain level. There are CV inputs for the attack, decay1, decay2 and release rates. The CV for the attack is inversed, so increasing the CV level will shorten the attack time while increasing the decay times for the decay1, decay2 and release. This means that when e.g. the key velocity voltage is used a higher velocity will shorten the attack and increase the other decay times. Rate settings can be from really snappy to pretty slow. Care was taken that you still have good control over the rates when in the snappy range. The second envelope generator is intended as a modulation envelope generator to e.g. sweep a filter or control the harmonic waveshaping of an OSC HRM module. It has an extra output that is controlled by a bipolar mix knob that can invert the envelope shape and also gives some extra overall ‘sink’ or ‘lift’ when the output level is increased. Rate can be set from a glitch to about a minute. When modulating decay times with control voltages it is good advise to keep the voltage fixed while the envelope is developing. E.g. trying to modulate the decay time with an audio rate signal does in general not produce sensible results. A S&H is integrated into the module to sample the decaytime modulation input signal for the second envelope generator on every new gate trigger. This way the modulation amount will stay fixed until the module is triggered again by a new gate pulse. The sampled signal is also brought out on a connector, so it can be routed to a CV input on the first ADBDR envelope generator. Or be used in any other S&H application. The ADBDR envelope is designed in a way that is hás to finish its attack phase to reach its peak level before it can be retriggered. When used for sequencing this allows for complex envelope shapes that give interesting rhythmic effects, but when used for keyboard play it might feel a bit strange to play the module with long attack times and fast play. Note that there are no attenuator knobs to set the amount of modulation for the ADBDR envelope CV inputs, these inputs are at full sensitivity. They can be connected directly to e.g. the velocity CV or CC# CV outputs of a MIDItoCV converter, but when modulated from other sources one might need an extra CV mixer module to set the modulation levels properly

Phaser Filter
The 2U wide Phaser Filter module combines 5 allpass poles with three lowpass poles in one module. The five allpass poles are configured as a phaser with positive feedback resonance control, creating two resonant peaks when opened. Then a crossfade knob fades between the input and the output of the phaser and this crossfade mix is the input signal into the lowpass filter. At the end of the allpass chain before the crossfader is a phaser monitor output. The lowpass section has a cutoff slope of -18dB/Oct and its own resonance control. Both the phaser section and the lowpass section have a 1V/Oct control law. The phaser has one modulation input and if it is not connected the audio input signal is used as the modulation signal. This allows for dynamic waveshaping of the input signal on the time axis without detuning to signal. The filter section has two modulation inputs, if the first is not connected the filter audio input signal is used to modulate the filter cutoff, allowing for even more dynamic waveshaping just like in the phaser section. If the second modulation input is not used it uses a signal from halfway the lowpass poles to self-modulate, thus producing all-harmonic distortion on the resonance peak when the resonance is set fairly high. When the 1V/Oct input jack for the filter is not used it inherits the signal from the phaser section 1V/Oct input jack. Both phaser and filter can sweep over a range of roughly 18 octaves and can be modulated up to really high audio rates. In this last case FM-type and ring-modulator-type effects occur, but with much more timbral control than traditional ring-modulators. E.g. when the outputs of two OSC HRM modules, set to sinewave output and tuned in some interval, are mixed and routed into the filter just slight amounts of the internal modulation on either the phaser or the filter will start to produce ‘undertones’ and ‘overtones’ that are sum and difference frequencies of the interval. This exemplifies the idea behind the Phaser Filter architecture, to not only take material away like a normal filter does but to also produce new material not present in the input signal and combine the both to create a vast range of possible timbres.

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'. E.g. in the Benjolin the last 3 stages of the shift register for a 3 bit code that is fed into a 3 bit DA converter. This DA eight 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'. Note that it is these properties of stepped, sloped and pulsed that are of interest in the waves. (The Dutch composer Jan Boerman formulated an idea in the 1960s about audio signals that are inbetween pitched and unpitched. Havoc waves are probably somewhere in that region, maybe a bit similar to granular synthesis stuff. I haven't really thought deeply about this myself, but Boerman has certainly always been an inspiration to me to try to go into that inbetween territory.)

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.

Dual Fader
The DualFader 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. There are some extra options, e.g. the output from the first fader is internally connected with an extra signal path to the second fader, depending on a switch this option adds the output of the left channel or the crossfade product to the next fader with its own level knob, or make the second crossfader act as a DC-coupled ringmodulator on the first fader crossfade product. Because the midpoints of the faders are at -3dB to get the equal loudness fader curve the ringmodulation has a compressive distortion that will increase when the modulating level is increased, which sounds like the vintage 'diode-transformer' ringmodulators that were used in the fifties and sixties.

Both second channels on both faders have an extra gain control that can boost up to +20dB, this is 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.

Basically the DualFader is an elaborate voltage controlled mixer module with several options useful for different applications for dynamic signal routing and sound synthesis.

Triple-Input 24DB Filter
As the input jacks are normalized, the filter can also be used to create a variable slope for one input signal. As when a signal is connected to only the LP input the normalization passes the signal on to the next HP and BP inputs. This way you can set the slope with the LP, HP and BP input level controls.

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 wíll be present on the BP input signal."

The lowpass and highpass curves are exact complements of each other, if both receive the same signal nothing seems to be filtered, even at high resonance settings. But when also the the bandpass level knob is opened the filter will act like an EQ, selective resonances can be applied to the EQ function, optionally with distortion in the resonance only when the mod2 level knob is opened.

The filter can also be used as a three input mixer, where the LP input gives the low end of a signal, the HP input the high end of another signal and the BP input a small band of a third signal. And all possible variations in between. 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.

Note that this module now has an extra exponential VCA on board and is no longer available without the VCA. This has increased the price from the old model without the VCA from 295 euro to 325 euro for the new model.

Faceplate = Dotcom but can be terminated to MOTM power and drilled accordingly. Cannot be resized, as is the case with all his modules except the Dual Phaser. The Dual Phaser can be made at the standard MOTM size.

Triple LF-VCO
It contains 3 CV controlable LFO's or as Rob calls them LF-VCO's. One LF-VCO has a triangle and an inverterd triangle output. Rate goes from several minutes to 100Hz.

An other LF-VCO has a sine output with "fluctuation" modulation. This is a combination of AM and FM witch softsyncs to the harmonics of the modulating signal.

The third LF-VCO has a triangle output and a pulse output. The triangle output can be modulated from ramp to triangle to saw, just like the MS20. This LF-VCO has a switch to change between hardsync or a "stop" function. The stop function stops the LFO on the current output level and from that it goes further on the rhythm of a modulating input signal or the triangle LF-VCO.

The modulation inputs of all VC-LFO's are normalized at the input connectors in such a way that everything can crossmodulate and sync.