Patch Ideas SPECIFIC MODULE PATCHES DISTRIBUTOR The outputs of the Distributor are also voltage sources when no input is present -- 0-to-+5V on Channels 1 to 3 and a constant +5V on Channel 4. FILTERS Voltage controlled resonance in Q150 • Split the filter signal output with a multiple or distributor • One signal to VCA 1, the other one to VCA 2 • Take the inverted output of VCA 2 and feed it back into filter signal input #2 • Resonance must be set at about a half Now you can control resonance through VCA 2. Low Pass Filter as a High Pass To use the Q150 as a high pass filter: • Split the signal from the oscillator(s) using a multiple or a distributor • One signal goes to mixer input 1, the other one to the Q150 filter • The output from the Q150 goes into the upper input of a Q125 signal processor • The signal processor gain is set around -100% to be inverted • The signal from the Q125 goes to mixer input 2 As a result, all the frequencies below the cutoff frequency will be removed from the audio spectrum. To be effective, fine tuning of the inversion gain is needed. If voltage controlling the filter, remember that this configuration will respond the other way around! You can get a highpass filter by subtracting a lowpass signal from the original signal. It turns out that this is one of the few situations where the phase reversal of the Q111 Pan/Fade module comes in handy. Here's how to patch it up: White Noise Source ----> Q111 Pan/Fade Input #1 | | Q111 Pan/Fade Output ---> Highpass Output! | |--> Q150 Lowpass Filter ---> Q111 Pan/Fade Input #2 Set the Q111 to "Fade" mode. Control all the way to the left will give you unaltered signal, all the way to the right will give you the lowpass filter sound. But in the middle, the lowpass is subtracted from the standard signal, and - tada - highpass! Filter Change: Lowpass into Highpass, Notchpass, and Bandpass Mix original out with filtered out and invert one to turn a lowpass into a highpass or vice versa. Half-volume-inversion will produce a notch filter. Also, if you have a filter with multiple pole outputs, combine two different pole outputs, invert one, and you get various bandpasses. Add OOMPH to the signal 'Oomph' from a high pass filter? Huh? Here's how - set up the filter with inverter & mixer as described above to make a 24dB HPF. Now control the filter with 1V/oct keyboard tracking to follow the osc pitch and tune the filter to the fundamental of the oscs. Crank up the resonance and the filter will now accentuate the fundamental for a huge beefy sine wave effect mixed in with the oscs. Use a second filter (before or after this setup) to do the regular filter sweepy stuff and just let this one track the oscs. State-Variable Filter A pulse wave run through the VCF can create drum sounds. Adjust the pulse width to a small spike (<1 on the Pulse Width), adjust the Cutoff Frequency for low or high sounds, and adjust the Resonance for the decay. Split the pulse first with a Multiple to send to an EG or S&H. When using the VCF in bandpass or notch mode, the Resonance acts as a bandwidth adjustment. Ringing the Filter for Drum Sounds If you send a small PULSE into the filter, you can get some cool drum sounds. Use the Q106 Oscillator PULSE output, adjust the width of the PULSE waveform, so it just the smallest spike, below the 1 on the Pulse Width Control, into the signal input on the Q107. If you use midi-cv use the gate output into the signal input on the Q107, reduce the lenght of the note in your midi sequencer to one tick. The filter's resonance acts as the drum decay. Use the filter's cutoff frequency to go from deep bass sounds to high pops/ticks. Use the velocity and pitch from the midi-cv interface for even more control. Send the PULSE to Q124 Multiples to trigger ADSR or S&H in time with the drum beat. GATES Splitting a Gate Patch the gate/trigger from your source (anything that's NOT the Q960 internal osc.), and patch it to a mult. Patch one leg of the mult. to a Q128 control-in. Patch another leg to the 'a', 'b', or 'common'. And finally, the output from your 'a', 'b', 'common' choice to whatever you wanna drive. Since it's usually two or more things... maybe patch to another mult., and from there, to whatever. Another way is to patch to the 'shift' of a Q962. Now, this'll make the 962 shift twice for every 'single' pulse. How the Q171/Q172 handles Gates/triggers... and after that- You all, by now, are aware that the Q171 /172 have channels -- 3 on the Q171 and #4 is on the Q172. The Q172 is NOT powered itself and MUST be connected to the Q171 to work.   Each channel has four jacks -- an 'input' and 'output' for the CV's (or Pitch, Notes, or whatever you want to call it) and an 'input' and 'output' for the gate(s).   Here's one deal. You'd almost think you need to patch both a CV 'pitch' and 'gate' input and, of course, the same on the outputs. Well, you don't HAVE to. The 171 (and 172) will create gate-outputs from the notes it processes from the input. But, and this is worthy of note, only when the incoming notes keep changing. For instance, an 8-step sequence, with all of the knobs (pitches/voltages) set to zero, or all at 1, or 4, or 7, or all C's, E-flats, or G's, or whatever where all adjacent notes are tuned to the same note, are the same to the 171's ears.  And in that case,  it will only output one gate. Now this is with no gate/trigger patched to the 171's 'gate-input' but the 171's gate-output is patched to the EG’s. And, of course, even though the 'pitch output' is patched, without a gate to activate 'an event', you'll hear nothing except the one per cycle of the Q960 as long as all notes are the same. If you plug a 'dummy' or unterminated patchcord into the gate-in, that will stop any output action from the gate-out. Even if you have as in this example the internal osc. of the Q960 patched to the gate-in jack, you still can't force it to fire a gate/trigger from the gate output. But this is not a 'problem'. Remember, we're "modulists" We don't 'have' to patch the sequencer’s or anything else’s gate/trigger thru the 171 to get continuous, steady trigger-firings for every step. Timing issues aside, if the gate-source is the steady flow of a sequencer, why does it need to be quantized? Normally, it doesn't. Just patch it to the EG’s and voila. Or, if you've done the WF-modification, and installed a switch on the Q963, you can patch from the Q963 (A or B) output, making sure to put that switch into the 'trigger' mode and not 'gate' mode, and voila some mo.   To recap- the 171 will produce it's own gate-signal(s), sync'd to occur when it senses a note (significant voltage) change... and when that happens... out goes a 'trigger' signal (refer to the spec-sheet on the specific duration). I do recall that a .2ms 'off-time' or 'space' is guaranteed (so you won't have one gate-voltage after another... one steady voltage that the EG's (and other modules) don't wanna see (for a gate-pulse). That space is needed. Robert Moog did it by making you use two different types of voltages... the 'S' and 'T' voltages, which had to be converted back and forth within the system (normally, in the 961 module). Since then, the industry standard has settled on... still, two voltages... one is the CV for pitch, modulation, etc., and the other is the gate, which is the on/off voltage for EG's & such (but there's also the trigger... needed only in certain modules)... so we call it trigger/gate or gate/trigger... the difference being : gate= one continuous 'on' voltage, which is on for the duration of the event (active stage... key depression, etc.)... and looks like to top part of a square wave. The 'trigger' is more like a sawtooth wave, quickly going to max. voltage, and quickly dropping off to zero, before the end of the event (active stage, keyboard 'tap' & release)... and modules needing a trigger (like the Q962 switch)(the Q128 switch uses gates), and they need to see that little space (the 'off' part)... , cause that's how it tells that the event is over, and the next one is a new event. This will be involving the Q960 sequencer (but you need to have done the variable pulse-width modification*)(* check the WIKI for how to's).... and the Q171... and a variable voltage source (Sig. Proc. offset, or Distributor, or Presets voltage-out, or even a variable voltage from an unused sequencer. Of course, if, when you did that variable-width mod., you opted to just install a pot/knob... you won't need a variable woltage. For a visual representation of what was happening, I patched-in a Q962>shift to the 171 gate-out... so it would alternate the stage lights with each event. (It' has nothing to do with how the patch performs). What this does, is let you hear and use both the rising, AND falling edges of the gate/trigger from the Q960 internal osc. (and I presume, any similar square/rectangle-pulse wave. So- what good is it? Well... for one thing... how may of you wanted/needed to double the gate/trigger-pulses, without doubling the actual frequency? Well... that's one thing you can use this for. I'll (and probably you guys, too) will find more uses for it.   Here's the patch part:   1. Patch-out from a Q960 row (CV)... (or if you must... 1, 2, or 3 rows > Q962 (a different Q962, that the one above, used to view the event)  962 output > 171 channel #1 'NOTE' input.   2. Patch-out from the Q171 ch. #1 'NOTE' ouput > Q106 osc. in (or Norm. module CV IN).  (If you turn the Gain of the VCA up (CW) (assuming you've got a standard patch set up), you should be hearing a pitch... turning the knob of the active stage of the sequencer, you should hear the Q171 quantizing it's a$$ off (okay... also assuming you've set the Q171 switches correctly. In this case... the important switch is the top one... let's you select ch. #1 (left), or none (center), or ch. #'s 1, 2, and 3 (right). Set it for 1, 2, and 3. And that should do it for the notes (as in 'tones'). Remember... this patch is about the gates. As far as the tones, and how to use this patch to affect the tones/notes... we can all come up with whatever we want. Meanwhile... back to the gates.   3. Patch the Q960 internal osc. output (or- a rectangle wave from a Q106 as a gate source)(and you can alter it's width manually, or via voltage-control, too)...  patch it to the 'PITCH" input of channel #2 (yes... PITCH of channel 2, not #1). Take the Gate output from channel 2... this is you main gate output for this patch... the 'magic' one. Where you 'stick it' is your own affair...  but- to first see what's happening, patch this to the 'shift' of that, otherwise unused Q962. (You can patch it to the 'shift' jack of another Q960 (and if desired, set it so that stage 3 resets... then, only stage 1 & 2 will work... giving you the same result as the Q962 option)(it's the 'trigger-outs' that you can put to use if you want). Have the internal osc. in the 'Lo' range, and set it somewhere between 11 o'clock and 3 o'clock (+/-). The range can later be turned up faster... but for experimenting initially, keep it close to this speed so you can see and hear the effects.   4. Now, you can patch from the 'trigger-out(s)' of the Q962 to whatever you find to be interesting, or what you want to try...  stage #1 will supply a trigger/gate on the leading-edge of the 960's internal rectagle wave... and stage 2 does the same, but on the trailing edge of the wave. If you pull the gate-out from the Q171 from the 'shift' of the 962, and patch it instead to the EG(s) (or Norm. module gate-in), you'll get double triggers for each stage/step... and these can be usable, to be sure. how you use them is up to you, and how much imagination you have.   This patch also demonstrates a few things... how the Q171 channels are completely discrete...  how the Q171 (and 172) generate gates/triggers from the changing notes of the inputted CV(s), and come out at the gate-jack... plus the original inputted is quantized, and outputted at the pitch-out jack. It alows the user to access both edges of the internal rectagle-wave for use... and, the user can access the leading edge, the falling edge, or both... which yields two discrete tiggers per stage... and, the user not only has access of the rectangle-width, but can voltage-control the width. Previously, to the best of my knowledge, the only ways of getting a 'doubling' effect for each beat (step/stage/event) was to either 1. Try patching the trigger/gate thru a rectifier... and, hopefully, if all goes well, and the frequency wan't too fast, it would be a definete 'maybe'. Or- 2. Do it "bass-akwards".... by using a main clock set at x2, or x4, x8, etc. the actual speed you wanted, then use a frequency divider (or a Q960, or Q962, or Q119 AS a frequency divider, by patching-out at the appropriate stage for your needs, while the main clock drives the sequencer. But, evidently, there are some people, who either don't want to use a sequencer for something like that... or, they don't have a sequencer to spare. Recording 'click-tracks' to drive the sequencer in the same way, for the same effect is a workable idea, albeit a variation of the above theme. Beyond that... I don't know of any other ways to get twice the number of trigger/gates from a single source. Do you? Would love to hear 'bout it/them. Extracting a Clock Pulse from a Stepped Output 1. Send the stepped output to a multiple. 2. Send one of the mult outs to a mixer, send the other through an inverter and then through a slew limiter (portamento) and then to the same mixer. At the mixer output, the inverter and slew will cause the two signals to cancel EXCEPT when a step change happens. The inverter & slew will cause a momentary pos or neg 'blip' as one signal suddenly changes and the other lags behind before it cancels with the first. 3. Send the mixer out through a rectifier to change all the neg 'blips' to pos 'blips'. Viola! A pos blip every time a step change happens. Use these blips as clock pulses. Self-Oscillating EG I'm always running out of VCOs, and I realized that I could possibly use an Envelope Generator as one. It took a little bit of figuring out, but I found if I plugged the output of the EG into the bottom portion of a signal processor (inverted, offset around 2) patched back into the gate input of the same EG, I found I could get some interesting self-oscillation (with some help from the signal processor which I am finding more and more useful every day... It used to be the most unused module in my setup.) Decay and sustain should stay at zero. (Of course if you want to do anything useful with this, I plugged the output of the EG into a multiples so I could route it back to the EG and other places as well. Unusual EG Uses Among other non-obvious uses for ADSR’s are: Signal input for sample & hold. (As a lengthy example, it may be interesting to invert the output signal before applying it to a control input, or to split the output into two outputs in a multiple and invert one of them, applying each to another control input, for instance two bandpass filters or one lowpass and one highpass, each panned hard in opposite directions. More interest might be obtained by mixing the above two filter outputs in a panner/fader and controlling the crossfade via another instance of the sample and hold output. But may be getting ahead of ourselves.) As control input for VCO (this technique is famously used by Keith Emerson in ELP's version of Aaron Copeland's "Hoedown", in which the sustain voltage is tuned to a certain note. As control input for VCO after treatment with sample & hold. As control input for two VCO s after treatment with sample and hold, the output of the sample and hold split, one half inverted, then sent to the two VCO s respectively. The outputs of two envelope generators can be mixed to obtain a more complex transient. The output of a single envelope generator can be used to control different elements of a single module, for instance the modulation index of a modulation input (via an intermediary VCA), pulse-width modulation, etc. Likewise several envelope generators gated by a single signal can control different elements of a single module, like frequency and resonance. INSTRUMENT INTERFACE In addition to bring in other instruments and signals for modification, it can be extensively used within the synthesizer itself to control amplification and to add break-out jacks for amplitude, gates, and triggers. To generate streams of random pulses, connect the slow random output of a Noise module to the Q118 signal input and adjust the threshold level until the preferred random rate is obtained. KEYBOARDS Tracking the Keyboard with Q107 Self-Oscillation Patch the audio output of your oscillator that you want to filter into the Q107 signal input, and the 1v/oct input is for a control signal like the CV pitch out of your keyboard or MIDI/CV converter. By patching the pitch CV to that 1v/octave, the filter's cutoff frequency will track evenly over each octave. The practice application of this is that you would have a different filter cutoff setting for each note you play, or if you're self-oscillating the filter until it just hums a tone (by cranking up the resonance and then setting the pitch of the tone using the cutoff frequency), you could then use that 1c/oct. input to play different notes, or fire off different pitches of self-oscillating noise. Partial Keyboard Filter Tracking with the Clipper Use the clipper to limit the filter's keyboard tracking. Suppose that you need the filter to fully track the pitch on the lower notes but you need for it to taper off part-way up the keyboard -- that is, full tracking opens the filter too much on the upper notes but it sounds good in the lower octaves. Just patch the pitch CV through the clipper on its way to the VCF's frequency CV input. With the clipping point to maximum, adjust the filter settings for the lower notes; then dial down the clipping point until the tracking sounds good for the higher notes. MIXER A Mixer can act as a preamp by running an output signal into multiple channels of the mixer, splitting it with a Multiple if necessary, and using the output. A 4-Channel Mixer can give nearly four times the output, and it can be attenuated by adjusting the channel outputs. NOISE The Q110 Noise Source is pretty straightforward, but it can also be used as a source of control voltages. Try running noise into the CV inputs of an oscillator's pitch or PWM to get a grainier or raspier sound. Another idea is to modulate a self oscillating filter's frequency with noise to create a new flavor of noise at the filter's output. This new flavor has a bit more sizzle to it and is especially useful in creating snares, hats, and other special effects. Crackly Noise Send noise to slew with separate up/down control, slew only up or down portions, not both. Experiment by breaking this rule. Some damn noise: Turn up amp as loud as it will go. Plug white noise output into amp. Do not move anything for 15 minutes. OSCILLATOR The Oscillator can be reset each time a key is pressed by routing a gate signal to the Hard Sync input. The range of an LFO Oscillator can be lowered by inputting a high voltage from a Signal Processor into the Linear Input. If the voltage is too high, however, the LFO will 'shut down' and cease to oscillate. The pulse output can double as a fixed +5V or -5V source. Set the Width knob full left (0) = -5 volts out. Full right (10) = +5 volts out. Using PWM as a transfer function This is actually a trick used in industrial motor controlling equipment. It's a sort of transfer function that works well for sending signals in electrically interfering environments. It's based upon PWM and here's the trick: Every waveform has a mean value, consider the simplest of them all the square wave which simply has it's mean value in the middle of it's peak to peak span. If the square wave spans from 0V to 5V it's mean value will be 2.5V since the square wave is symmetrical. Now a pulse has an equally obvious mean value. A pulse of 75%/25% value and spanning from 0V to 5V will have a mean value of 3.75V and so on. Going back to the motor controlling equipment a high frequency pulse wave is often used (like 10kHz and above) and then PWM'ed by a much slower (100Hz or so) sine wave to control an AC motor. The motor actually filters out the high frequency pulse wave with a low pass filter and what remains is the the mean value of the pulse, i e the sine wave. This is done because the pulse wave is much less sensitive to electrical interference than a pure sine would be and it is often critical that a motor runs clean and at a very precise speed. Now to use this trick in synthesis, what you do is crank a Q106 VCO up to a high frequency (you might need a voltage to offset it a bit), insert the audio you want to transfer into the PWM input of the Q106 VCO and listen to the pulse output. If the frequency is high enough (above your hearing level) you will not need to filter it through a low pass, you just patch it directly to your VCA. A good source of audio to transfer is a drum loop, your voice or a guitar or something like that. So where does the fun begin? You might notice that this transfer function actually works very well and that for high frequencies of the carrier VCO you don't really hear any discrepancies in the audio, well that is until you start disturbing the peace by modulating various parameters of the VCO. For example, try to offset the pulse width a bit and modulate the frequency heavily on the Q106 VCO. Oscillator Self-Modulation Send an oscillator into its own FM input. Modulating amount of FM can create vibrato. Linear input can be DC or AC with slightly different behaviors. Using Linear FM will provide good pitch tracking while Expo will have a bigger effect. Send oscillator’s pulse into itself with PWM to create vibrato + timbre changes. Harmonically-locked Oscillators Use a soft-sync-capable oscillator as a sync slave. Master and slave must be as in-tune as possible. Adjust sync and tuning of slave until it produces a clean waveform when playing. Send each into each others’ syncs for further precision. Pitch Snap Gate the pitch voltage going into the 1v/oct of an oscillator. This creates a nice snappy attack/release. Two pitches from one VCO Patch a square wave output from another vco into the exp input of the vco you're listening to. Do not change any knobs or experiment, it's good just the way it is. Completely silent sine wave: patch sine output of vco into input of vca. Do not patch anything into the cv in. Optional: do not patch the output of vca to anything, just to be safe. Creating a Saw Wave from a Sine Wave Tune two oscillators to the same frequency. Connect oscillator 1 to the hard sync of oscillator 2. Lower the frequency of oscillator 2 until you get something similar to a ramp wave. PAN/FADE Additional Function of the Pan/Fade Module The Q-111 can voltage-control panning and fading, but- the Q-111 can also be used to voltage-control outboard (or inboard) effects, such as echo (or slapback). If your main mixing console has effects sends and returns, you are in business. If your mixing console doesn't have effects sends and returns, you are still in business. If the only mixer you have is in inboard mixer (such as the Q-112 or Q-113), you are, yup- still in business. Basically what you do is: route the signal you want to your effects device, and patch it's output to the #1 input of the Q-111. Select "Pan" for the mode. Patch the #2 output to return to the signal mix. Tweak the control knob as needed. Some of the possible controllers could be: a keyboard; a sequencer; a sample/hold; an oscillator; and a foot-controller (pedal). The pedal allows you to articulate the effect's introduction into the mix as you play, or as other devices control the performance. (Note the "Look Mom... no hands", thing). Where both the Q-111 Pan/Fade module, and 'two Q-108 VCA's' can each perform voltage-controlled stereo panning, the Q-111 makes it a little easier to do, and takes up less space and fewer patch cords. By using a square wave as the control signal, a voltage controlled switch can be created. Use a sine wave and produce a fading effect between two signals. Same for a triangle wave (although the pan/fade will be with a different rate & overlap). Likewise, ramp/saw will pan/fade in one direction, and switch in the other. Remember that the Q111 outputs one of the signals as inverted. For audio, though, this doesn't make any real difference. PRESET "Switchpatch1" switches between three 'preset' independent patches or sounds. For the sake of time, the three patches are set-up as very simple "MiniMoog-like" sounds, that are easily identifiable as Sawtooth thru a lowpass filter, two square waves, and a sine wave with delayed vibrato. The actual 'master' patch grew to be quite complex, and would be very difficult to follow in either graphic or text notation. With that in mind, a description of what is happening should be easier to deal with. The Q-143 "Presets" module has a master A/B switch, with "OFF" being at the center of this 3-position toggle switch. So, it actually would be this: A/OFF/B. It has two "voltage-out" jacks, both of which output the knob A setting (-5 to +5volts), "OFF" (0 volts), or knob B setting (-5 to +5 volts). The voltage-out jacks can be patched to the "control" input of the Q-128 "Switch" module(s), which can control other elements of the "master" patch. The Q-143 also has two "A" inputs, two "B" inputs, and two "Common" outputs. These function not unlike the Q-128 "Switch", except that they are both controlled by the manual A/OFF/B switch at the top of the module. As a hypothetical example: if you had four Q-128 "Switches", and a Q-143 "Presets", you could conceivably switch 10 different things at once - probably more by adding on VCAs or Pan/Fades to work as switches. For this particular "Presets" patch, two sawtooth waves (in near unison), are patched to "A", and the 'common' is patched to a mixer, and then patched to a lowpass filter. A sine wave (2 octaves lower), bypasses the filter and is remixed with the filtered sawtooths before the VCA to provide the 'bottom-end' of the sound. The sine is initially tuned to unison, and then lowered by patching a Q-125 "Signal Processor" thru a Q-128 "Switch", and tuning the "offset" down by 2 octaves. The Q-128 "Switch" is opened by having the 'voltage-output' from "A" above 2 volts, which in turn lowers the sine. The two square waves come from the same oscillators as the sawtooths, but they are patched to "B". So what happens is: "A" gives 2 sawtooths and a 2-octave down sine wave. "OFF" gives only the sine wave (because it is not switched by the "Presets" module), and it is back up to unison because "Off" provides no voltage to open the 1st Q-128 for the negative voltage from the signal processor. "B" allows only the two square waves thru to the mixer. It also ouputs a voltage to a 2nd Q-128 "Switch", which turns off the sine wave. Other notable elements of this patch: · Sine & square waves are switched to an echo-send, and enabled by the Q-143 "Presets", and the amount of send is controlled by a foot pedal. The sawtooths are not sent to the echo-send. - The two sawtooths and two squares are patched thru a lowpass filter. The "Q" of the filter is determined by an 'offset' voltage from a Q-125 "Signal Processor", which is switched by a 3rd Q-128 "Switch". For the squares, the Q is set all the way open, and for the sawtooths, the Q is set about 2/3 closed, and is also modulated by an EG. · The sine wave alone, is modulated by a delayed vibrato patch, using an oscillator, a VCA, and an EG. The vibrato is only applied when there is no voltage coming from A & B of the "Presets" module (the off position) patched to a 4th Q-128 "Switch". In this demo, you will first hear the sine wave as the A/off/B switch is moved back and forth - 2 octaves down, then unison, then sine off (silence). After the second silence, you will hear the sawtooths and squares being mixed in at a low level. You will hear only the squares where there was only silence. And finally, the levels of the sawtooths and squares are brought up to full level, and then, a little bass part is played. "Switchpatch2" is meant as a more practical, and musical demo of the "Presets" and "Switch" combinations. Again, very basic Minimoog-like sounds are used for simplicity's sake, as well as a simple melody. This demo is all done in one take with the keyboard, meaning there is no multitracking, sequencing, or editing. This is to simulate a live performance. Some reverb is used throughout, along with sections where echo-send is switched with the patch. Starting out, there is a sine with delayed vibrato playing a simple melody. A simple bass part consisting of 2 sawtooths and a sine lowered by 2 octaves is played for 2 bars. The same 2-bar melody repeats nearly to the end. The presets are switched from "A", to "B", to "Off", and that pattern repeats. The second time around, the echo-send is enabled, providing echo for the squares and sine only, but not the sawtooths. QUANTIZER Even-Spaced Notes from LFO Triangle Wave into a Quantizer If quantizing a scale from an LFO Triangle wave, because of the difference in intervals between notes within the scale, the notes don’t come out evenly spaced. Half-step intervals come out shorter than whole-step intervals. This technique causes all of the scale’s notes to come out evenly spaced regardless of the interval between notes. Hint: Imagine using a pulse to frequency modulate the triangle input to the quantizer. When do you want the pulse to be high and when do you want it to go low? LFO wave out --> Offset/Attenuator --> Quantizer CV in Quantizer CV out --> favorite VCO 1V/octave in Quantizer Trig out --> VC Switch reset in Master Clock --> VC Switch trig in CV Source 0V --> VC Switch input 0 CV Source 10V --> VC Switch input 1 VC Switch output --> LFO FM input Set the LFO Freq knob to the lowest position. You might need to calibrate the patch by wiggling the LFO FM input attenuator to get the right trade off between getting a quick scan to the next note and not going too fast to over shoot to the next note. Also it is crucial that the LFO can go really slow so that it is effectively paused while the FM input is 0V but still go fast enough when FM input is high to do a quick scan. A simpler solution, but one with a slightly less elegant result, would be to S&H the signal to be quantised. Then you just have to make sure that the slope of the signal is appropriate for the clock rate. It's actually fun to deliberately mismatch them for some nice skipping effects. QUANTIZER/AID MIDI Channel Selection The Q172's default MIDI-OUT channel is channel #1, but it can be changed...  to any channel -- #1 thru #16!   1. Power Off the system. 2. Patch the Q172 Channel 4 Gate Output back into the Gate Input. 3. Turn the Note Group rotary switch to select channel number 1-16. (The top-center position is Channel #1, the next one clockwise is #2, etc.) 4. Power On the system and wait 20 seconds. The new MIDI-Output channel is now set with the change saved in non-volatile storage, and it will remain in effect through power cycles. RECTIFIER Creating New Waveforms with the Rectifier Pass a triangle wave through the lower half of a signal processor and give it a five volt offset. Then, put it through the rectifier and listen to the result. It should still sound like a normal triangle wave. Now, slowly turn the signal processor offset from five to zero. As the offset is lowered the bottom points of the triangle wave will be rectified, adding some interesting harmonics. When the offset reaches zero you will have a simple triangle wave again but it will be one octave higher. You now have a new range of waveforms to explore. Experiment with various offsets. Checkout the new harmonics by sweeping your filter and listening to the band pass output. For dynamic voltage-controlled waveforms use a mixer to combine a triangle wave with an envelope and put the result through the rectifier. When you have finished experimenting with triangles, start over and try everything again with a sawtooth wave. A rectified sawtooth wave produces a triangle wave with the same frequency. You should also try sine waves. Pulse waves, however, do not do anything interesting when rectified. Clipper/Rectifier An Envelope Generator output run through the Clipper will clip the spike between the Attack and Decay values, creating a ‘punchy’ envelope reminiscent of vintage EG’s. A triangle output run through the Rectifier creates a triangle wave of twice the input frequency. A sawtooth or ramp wave run through the Rectifier creates a triangle wave of half the input frequency. A pulse wave run through the Rectifier creates up to a +5V (Lower) or –5V output (Upper). RING MODULATOR A Ring Modulator can be used as a VCA by patching the signal into one input and the gate into the other. The output amplitude is controlled by the gate input. Tune Q106 VCO #1 to a beat free A above middle C = 440Hz on the keyboard. Tune Q106 VCO #2 to a beat free D below middle C = 440 Hz on the keyboard. This will make the output of VCO #2 (about) a perfect fifth above VCO #1 if both oscillators are given the same control voltage input. Feed the output of VCO #1 to one Q116 Ring Modulator input and the output of VCO #2 to the other Q116 input. As long as both oscillators track the same input control voltages together, the output will have a fundamental one octave lower than VCO #1 with an overtone two octaves and (about) a third above the output fundamental. Use the sine or triangle wave output from both oscillators for the purest output tones. Using more complex waveforms (saw/ramp, square/pulse and/or variations thereof) can yield some exotic yet useful sounds for traditional music applications. +1 Octave Saw: Ring-modulate a triangle with its own square. Bowing the Ring Modulator There was a classic VCS3 trick called "bowing the ring modulator" which took advantage of the RM inputs being AC coupled. By feeding a tone in one input and the joystick in the other the RM would only output when the joystick was moved, analogous to bowing a violin string, with the amplitude proportional to the speed of movement. SAMPLE AND HOLD The Gate input also serves as an output when the internal timer is being used. This means you can trigger gate-driven devices from the output of the S&H when it fires. Use a ramp or sawtooth wave as your sample source, either alone or mixed with noise. Using a pulse wave as a source enables you to use the S&H as a controller, using the upper and lower values as controlling signals. The sample rate controls the flip rate of the controlling signal. Split the keyboard CV output, run one through an Oscillator, and the other through the S&H before going to another Oscillator. By adjusting the sample rate to your playing speed (for example, an arpeggio), the S&H will ‘sample and hold’ different notes that you are playing. Another use is sampling a voltage from a modulation source when a static rather than cyclical voltage is preferred. In the case of using a VCEG, for example, it is more controllable to have a parameter change once each time the EG is triggered (with the Q117 using the same trigger pulse) rather than having a sustain cv or attack cv changing during the course of the EG's cycle. In small amounts, with the input level close to zero, you can add touches of modulation to almost anything. One I like is triggering the Q117 with my MIDI-to-CV's gate with the input set just a sliver over zero and the output going into the oscillator's linear input so that every key press on my keyboard can be ever so slightly detuned for a bit of instability. Or you could open the input up a bit to completely change the pitch if you wanted. You could do the same thing with a sequence gate, too. Or whatever. Even though it's not a dedicated decimator, it does a pretty good job turning anything into garbles. In this case, you have to clock the Q117 with an oscillator at audio frequencies. The frequency of the oscillator will control the decimation amount. Use the Q960 at audio rates and then have an LFO (or programmed gate) trigger the sample and hold to sample the Q960's CV output. It can really randomize your notes and get a bit funky. Then have random gate sources triggering stages and you get a whole mess of awesome. If you had 3-4 sample and holds and a sequential switch you could rig one switch to route the gate from a keyboard to each sample and hold in sequence (shifting with each key press), then mult the CV from your keyboard into all of the sample and holds. Then you have rotating chord memory for 4 oscillators. Add another sequential switch to select the outputs of the sample and holds and you have a programmable sequencer (similar to an SH-101 but with 3-4 note memory and re-writeable while it is running). Important note about the Dotcom S&H: If you have the clock set to internal, you will get a trigger (or gate?) out of the trigger input. This can be useful or bad depending on how you are using it. It seems voltage outputs may vary as well so you might want to measure the output as to not fry something. SEQUENCER Q960 Patches and Patch Tips Random Function - Q117 S&H Output --> Q960 'Shift' Input. In this case, the Q117's internal osc. is the master. To use the 960's internal osc. as the master to drive the Q117 Sample-Hold: Turn the 960's int. osc. ON. Patch 960 Internal Osc. ouput > Q-117 gate input. Then Q-117 output > 960 shift input. Depending on whether you're using white or pink noise as the sample-source, or a mix of both, you can get more than just random effects. I've successfully managed to get patterns which favor alternating patterns, odd/even patterns, and even some backward patterns. Tweaking is everything. Also, patching this effect thru an 962 switch will allow you to have alternating patterns where the 960 runs normally, then changes to random, or one of the variations mentioned above. Multiple Sequencer in series - I have successfully patched two Q-119's and two Q-960's to run in series -- Q-119 #1 > Q-119 #2 > Q-960 #1 > Q-960 #2 -- then repeat. The patch would work even if you had 50 sequencers. Use the 'Done’ output on the Q-119's to both turn on the next sequencer, and to turn off the 119 that just finished running. On the 960's, you want to trigger the internal oscillator on, then when it has completed its' run, it should be triggered off. The sequencers could be kept in sync by using a common external oscillator to drive all, or use each sequencer's internal oscillator so that they can each be set at their own speeds. Another approach would be to let all of the sequencers run without stopping, and just switch which one is active in the patch. Terje Winther's 'Tangerine Dream' Sequencing in Series Method - Take two Q960s, and use the stop trigger (step 9) of one sequencer to start the next sequencer. The real trick up the sleeve is to split the stop trigger from the first back into osc. on, so that the clock doesn´t stop. The oscillator will continue, and both sequencers will be perfectly in time, even if you shift tempo, but the stepping will stop/flip between the sequencers. Details Both sequencers step 9 set to stop Then patch cables: Sequencer 1 oscillator output -> Sequencer 2 shift Sequencer 1, step 9 out -> Sequencer 2 step 1 in Sequencer 1, step 9 out -> Sequencer 1 osc.on Sequencer 2, step 9 out -> Sequencer 1 step 1 in From there on you can add multiple sequencers, and multiple lines through Q962s and whatever else you want. Q963 Trigger bus Timing control - Patch Q963 bus 'A' to the lower part of a Signal Processor, and invert it. Patch that output to a mixer input. Patch bus 'B' directly to a mixer input. Patch the mixer output to the Oscillator Input of the Q960. (If you have made the WF-mod to the Q963, turn it off. Assign each stage to either 'A' bus, 'B' bus, or 'Off', according to the rhythm desired. Use the two mixer input pots to adjust the timing for all stages on each bus. Stages assigned to 'Off' will run at the frequency settings of the Q960 internal oscillator. Detailed tweaking of the two mixer input pots is crucial. This patch allows you to alter the 'timing' of the Q960, making instant, on-the-fly changes in patterns, a way better method than adjusting a row of knobs on the sequencer itself. Q960 PWM Patch - If you've added the PWM modification (jack vs. pot) to the Q960. A jack allows you to patch any source to control the pulse-width of the internal 960 oscillator, rather than a manual knob. The most obvious application is to patch the output of a row or rows from the sequencer so that each stage's pulse width can be set independently and in sync with the sequencer. Try this... patch an output from a Distributor to the oscillator input of the Q960. Look for a sweet spot near the high-end range. Too low, and the sequencer will stop... too high, and the sequencer will stop. But at the sweet-spot, I have found that I can get several effects from random, to reverse, to alternating stages. The odd thing is that this trick doesn't seem to work with any other voltage source, such as a Signal Processor, or a Preset module, or even a stage output from a second static sequencer. I'm not sure why... just chalk it up to FM* (friggin magic). Start/Stop of the 960 controlled by the keyboard. This one is great when you want the sequencer to run only when keys are played, but then to stop when they are not being played. Patch the KB gate signal to a mult. Patch the mult to the EG's for normal control there. Patch the mult to the input of the Pedal Interface. Patch the 0/5+ output to the Osc. On' jack of the 960. Patch the 5+/0 output to eith the 'Osc Off' jack, or the 9th stage (off). When a key is pressed, a voltage is sent from the 0/5+ jack of the Pedal Interface and triggers the 960 to turn 'ON'. When the key is release, that voltage is turned 'off', and the 5v/0 output voltage is turned on (in other words, this module will send out 5v when there is no voltage present at the input, or when a pedal/switch is patch to the input, so a voltage is sent to the 'off' jack, turning the 960 off. Delays can be set up by inserting EG's either in the 'on or 'off' part of the patch. There is also another output jack on the Pedal Interface which supplies -5/+5 volts, depending on the status of the input. This too can be used. Either the -5/+5 or the 0/5+ output can be patched to the trigger-input of one of the stages. For example, if you wanted the sequence to start on the same stage every time. Q960 as a Frequency Divider - patch external source to the 'shift' input. Use the stage gate/trigger output(s) to select the frequency ratio(s). The Q962 can do the same, but to a lesser degree. Start/Stop/Etc. Q960 with a Click Track The Patch Assume the click-track is 16th-notes (it can be anything, and any time signature, and any tempo). Take the click-track output from the recorder, and patch to the Instrument Interface. The I.I. doesn’t have to be used, but if you have one, it can make things easier in several ways. For instance, level control/threshold may be adjusted which can help if something other than an actual ‘click’ or ‘pop’ is used as the audio in the click-track. Also, usually as far as the sequencer goes, a gate or trigger will work for simple stage advancement. So one of the two can drive the sequencer (shift) and the EG(s) for the sequenced audio. So, determine if anything requires a gate, and if so, patch accordingly. If not, just make sure that the sequencer, any EG(s), and the Reset-EG are all being fed the clicks from the click-track. Patch either the GATE or TRIG output to an EG. (ADSR settings TBD*). Patch the EG output to the CONTROL-IN of a Q128 Switch. Find a spare voltage source (like +5v from a Presets, Signal Proc., Distributor, or even the trig-output of an unused Q962, etc.), and patch that to the ‘Common’ jack of the Q128 Switch. Patch jack ‘A’ from the Q128 Switch to the Trig-IN stage # on the Q960. Since there are no trig-inputs on the Q119, it’s kind of hard to use that one unless you arrange the tuning so that the reset-point will occur on the first stage, and about that- there ARE some variables that can only be established with each individual set-up and situation. For a common 16th-note, 8 or 16 stage pattern- it will often be stage #8 that will be the reset stage #. It’s often assumed that the reset-point will be stage #1, but consider the recording needs to be stopped/cued just BEFORE the first click in the measure, and likewise, the sequencer will usually need to be reset to the last stage # BEFORE the first note to be played. If it is set to stage #1, when the first click is received, the Q960 will instantly advance to the next stage while simultaneously the EG(s) are fired, which will not sync the sequencer properly unless one sets it up that way in advance. It’s considerably easier just to set up the sequence, tuning, etc. using the Internal Oscillator and then determine and assign which stage needs the reset point, and use the click-track to play it once you’re ready. Tweaking The only crucial tweaking is the Gain and Threshold of the Instrument Interface. Too low or too high, and the Q960 will not advance properly. Find the sweet spot. After that, tweak the ‘reset-EG’ ADSR settings. What you want is ZERO Attack, and the Decay, Sustain, and Release should be set so that while the click-track is playing, the EG is keeping the Q128 Switch on the ‘B’ side (which is zero volts). The Release setting will be a product of the frequency of the clicks vs. the amount of time between them. How It Works While the clicks are firing the reset-EG, the Release-Time is set longer than the time between clicks, so the Q128 Switch is kept in the ‘B’ position (zero volts output). When the transport is STOPPED, the clicks cease, the reset-EG stops firing, and as soon as the Q128 Control-In threshold is passed (1.5v), it switches to the ‘A’ position which has the +5v patched to it, thereby triggering the Q960 to go to the Trig-In stage # that it is patched to. And obviously, since the Q960 is being advanced via the ‘shift’ input and not the Internal Oscillator, nothing has to be turned off. The reset is triggered by the absence of the clicks on the click-track. The EG serves to hold a voltage on the Q128 Switch, until the EG is no longer receiving clicks. Now, if the tempo is very slow, it may turn out that you’d want to use the Gate Output from the I.I., and set the sustain on the reset-EG to max. There can be lots of variables to play with, meaning lots of options. It’s fairly quick to patch and set-up, and once it’s done, it works every single time. It works well for overdubbing in different sections of the recording and in multiple sections. I’ve used it where I only had clicks recorded in the sections I wanted and was able to run through the entire piece hands-free but chose to go back and add additional notes for chords. It saves a tremendous amount of time when having to do a lot of tweaking, tuning, and various other adjustments and you have to do a lot of practice run-throughs, especially if there are other musicians playing in the session. In some situations, I’ve used the Pedal Interface in a similar fashion. As long as it has a voltage applied to the input (or no voltage in the opposite situation), it is either outputting a voltage, or not outputting a voltage... depending on which output jack you choose to use. When the input condition changes, so does the output(s)... so, it can serve as a simple, yet dandy logic module. Random Notes on the Q960 using the S/H Do the 'random' function on a Q960 by patching a S/H > 960 shift-in. Tune the notes of one row so that it's like an arpeggio, where all of the notes will work with each other in any order. You can use another row to set the timing for a pattern, and in that case, you'd actually use the internal osc. of the 960, with the S/H's gate slaved to it, so that it follows the pattern. So- the S/H is not actually controlling the pitch-voltage, but rather, it's controlling which stage fires, and randomly, at that. Mixing a sawtooth or ramp in with the noise yields so degree of repetitious patterns, along with the randomness of the noise. Programming a Sequencer Rest 1 960, 1 962, 1 963, 1 Q128.... patch the normal 16-note melody with the 2 rows of the 960 > 962 > filter > VCA (+/-) ...... patch the 960 internal osc. > Q128 'B input'.... assign the one 'rest' note (or stage #) on the 963.... patch the output of the 963 (same bus) to the Q128 'A' input. Q128 'common' to EG input. EG out > VCA control-in. 960 stage 1 trigger out > mult. Split A > 962 shift.... Mult split B to Q128 control-in. Skip Stage on Q960 Sequencer There's a couple of ways to skip a stage you want to NOT play, using its trigger output to control whether the note is heard or not. 960 row a > 962 input 1 960 row b > 962 input 2 962 output > osc.(s)..... etc. Patch 962 trigger-out #1 to mult. Mult. split (a) > 960 stage trigger-input (which causes the two rows to alternate.... a, b, a, b). Mult. split (b) > 'control' Q128 switch. (Whenever 962 stage #1 is active, Q128 (A) will be active where you patch on the Q128 A or B) will determine the polarity of this part of the patch). Let's say stage #15* is the one you want to skip. Substitute whichever stage and adjust the patching accordingly. Now we have a way to separate all of the stages that WILL play from the stage that WILL NOT play. From this point, you have to decide how to patch the two. Not knowing your module complement, there could be several ways. You could patch 960 #15 trigger out to a Signal Processor, invert it, either add or subtract offset, and patch that to Q128 A or B again. Patch the Q128 common to VCA #2 control in, assuming that the EG that's driven by the 960's internal oscillator is patched to VCA control-in #1. With the offset adjusted, it should cancel out the pulse from the EG, but only when stage 15 is active. Several variables with this method… Variation: Try patching stage #15 trigger output to the 961: 961 > Signal Processor (invert) >VCA control input #2 Variation: Rather than using an inverted signal to silence the VCA, you could silence a VCF the same way. This works because the Signal Processor (inverted) and set at - X%, plus the offest adjusted, pushes or closes the VCA only when the selected stage is active, and that only happens every other pass of the Q962, because 962 stage 1 is controlling when the Q128 is switching. You could also patch the EG or the 960's oscillator output thru A and inverted #15 patched thru B, then to the VCA. Skip Stage on Q960 Sequencer If you have multiple stages active at once on the 960 (just as the 'master CV-out' on the Q119), you get the sum of all of the active stages, and this normally is not desirable, really, because unless you put oodles of time into it, the resulting pitch tends to be not what you want. Of course, that's my opinion. Surely someone out there will think it's the best thing since paid vacation. Reset- because of how the open-ended design is on the 960, often, you can find several ways to accomplish the same goal. Stage 9 is an 'off' stage – ‘off’ as in it puts out zero volts. There is a trigger in and out, and a skip/start mode switch. You can simply turn off the internal oscillator and back on again. But sometimes, it's not about using the 960 for its CV output; it's about using the trigger ins and outs to control any number of things. Currently, I have a patch running with two 960's. One is running normally, stages 1-8, but it's slaved (and sync'd) to another 960 which is using its internal oscillator as the master clock source. And on that 960, I have trigger inputs coming in from trigger busses which are fed from the other 960. Depending on which stage(s) are sent, and which stage mode switches are set into which mode, I get all sorts of different patterns --- some double-back, some reverse for X steps, all sorts of variations. The slaved 960 resets and resync's the master 960. (I know that sounds backwards, but that is what's happening.) So back to the subject- 'sync' and 'trigger' aren't really the same thing. I take 'sync' to imply locking with a running tempo. 'Triggering' causes something to happen, like you can trigger a stage-shift, an 'on' or 'off', or an EG to fire. Resetting can be done in different ways on the 960. It probably will depend on whether the 960 is continuously running, or starting and stopping, as to the patching scheme and what works best. Sometimes wrestling with the gate vs. trigger theme is an issue to be addressed, which is why I advocate doing the WF-modification to the 963's using a spdt switch so that you can select whether you get a gate, or a trigger. Oh, and triggering the stage 9 input does not reset it to stage 1. It triggers it to stage 9, and there it stays. Triggering to stage 1 will move it to stage 1. Sometimes, a patch may require triggering an 'on', and a trigger to the stage you want, or a couple of times, I found that triggering to the stage just before the stage I wanted gave me the result I wanted. But that was a while back and I don't recall the specifics. ‘Puzzle’ Problems and Solutions for Q960 Sequencing Overview: The following ‘puzzles’ were originally created by Morbius to inspire the author to not only figure out the means to perform the stated functions in them, but to improve the thought processes involved in determining how to make sequencers and other modules interact with one another in figuring out any problem posed in sequencing. There are probably many ways of creating solutions to these puzzles, depending on the modules available and one’s creativity. When solving these puzzles, one should strive for the elegance and simplicity that should be present in any patch. Puzzle 1: Make the sequencer play 1,2,3,4,5,6,7,8... and then 6,7,8... and have it repeat the process. (And the technique used should allow you to start over at some point other than 6, if so desired.) Solution: The switching offered from a Q962 is involved in doing this, using the Trigger inputs and outputs of both the Q960 and Q962. Here is the patch that enables playing the sequence as stated in the question... Q960 Stage 1 Trigger Out --> Q962 Shift Q962 Trigger Out 2 --> Q960 Stage 6 Trigger In This small patch causes the Q960 to run 1,2,3,4,5,6,7,8,6,7,8,1,2,3,4,5,6,7,8,6,7,8... repeatedly. One could use the shifting of the Q962 to make the repeated 6, 7, 8 go somewhere else if desired. Puzzle 2: Without using the Timing Row, use a row to SLOW each note of a sequence with a CLOCKWISE turn of the knob. Solution: This was one was where the Q125 Signal Processor comes in handy to control the Q960 internal oscillator speed. Choosing Row 2 to be the 'control row' with a characteristic sequence to control the timing of a sequence in Row 1, the bottom half of the Signal Processor is sufficient to handle the task... Q960 Row 2 Out --> Q125 In Q125 Out --> Q960 Control In Q125 set to -3V Inverted Using a negative voltage inverted causes the clockwise rotation of the knobs on Row 2 to slow the playing of the corresponding note in the Row 1 sequence. This uses Row 2 as the speed control so no function of the Timing Row would be indicated, since Row 3 was not used at all in the patch, nor was the Timing Row switch in the On position. Puzzle 3: Slow the notes of a row only at predetermined times, while other rows play normally. Solution: For this one, one can use mostly the same patch as the above, but, instead of Q125 Out --> Q960 In... Q125 Out --> Q962 Signal In 1 Q962 Signal Out --> Q960 Control In Q960 Stage 8 Trigger Out --> Q962 Shift This causes Row 1 to play with the speed controlled by Row 2 once, and then play normally the next time through. One could also patch in Row 3 with the Q962 for either the normal or controlled sequence. Puzzle 4: Start the sequencer when a key is released. Solution: There are probably many ways of doing this, but this solution approaches the problem with the Q125 Signal Processor (like the Q962, a very versatile module to have) and the Q109 Envelope Generator, as such. The keyboard indicated is the QKB-15S. Kbd Gate Out --> Q125 In Q125 Out --> Q109 EG Gate In Q109 EG Out --> Q960 Osc On Q109 EG settings: A=0 D=0 S=0 R=10 Q125 settings: +5 Inverted As stated in the problem, after pressing a key, the sequencer only starts when the key was released. Puzzle 5: Start the sequencer X seconds after a keypress. Solution: There are surely other and better ways of dealing with some of the problems I had with it, but this is one method. With the Q109 Envelope Generator handling the timing, the Q125 Signal Processor is the control voltage source, and the Q108 VCA does the controlling. The Sig Proc puts out sufficient voltage to start the sequencer (which proves that running the Sig Proc Out to the Osc On would start the Q960 at powerup). But even with the Q125 turned all the way up, and the Q108 Control #1 Level turned all the way up, it still won’t start the sequencer. So add the square wave output of a Q106 VCO to it, and it will have the 'oomph' to start the sequencer, using the following patch... Kbd Gate Out --> Q109 EG Gate In Q109 EG Out --> Q108 VCA Control In Q106 VCO Pulse Wave --> Q108 VCA Signal #1 In Q125 Out --> Q108 VCA Signal #2 In Q108 VCA +Out --> Q960 Osc On Q108 VCA settings: Exp with Control #1 Level=10 Q109 EG settings: A=time to start D=0 S=0 R=0 Q125 settings: +5 Normal The VCO was just outputs an audio-rate pulse around 15-20% duty cycle. One could probably use the top half of the Q125 rather than the VCO, but this works, though, enabling one to adjust the time the sequencer starts after a keypress by adjusting the Attack on the EG. Puzzle 6: Cause the above process in Puzzle 5 to happen after the second, third, or fourth keypress, Solution: This puzzle is intentionally left unsolved for the reader to work out. The main purpose of these ‘puzzles’, after all, is not to necessarily provide answers to how to perform various tasks, but to inspire the synthesist to think about the interfunctionality of sequencers and other modules. Q962 Patches The Q962 is bi-directional, meaning the audio signal (or control-voltage signal) can go in either direction. The Q962 can be used as a counter, without having to patch any audio signals thru it. The Q962 can be used as a voltage-controlled preset patch controller... esp. in conjunction with the Q128 Switch(es). The Q962 controlling Q128 switches via Distributor modules. If you have installed toggle-switches in your Distributor(s), you can realize much more usefulness. If you patch the trig-outs from a Q962 to the input of the Distributors, then patch Distributor outputs to the control-input of the Q128 Switch(es), you can patch specifically tuned oscillators thru the Q128 switches and back to your oscillator mixer. As an example, using four oscillators (two tuned to unison and slightly detuned; one to unison -1 octave; and one tuned to a 5th)... patch the -1 octave and the 5th thru the Q128's before the mixer. This will allow you to have manual control over when these two are heard, or to be switched on/off depending on which stage the Q962 is on. This could also be done with a Q960 instead of a Q962, or even a Q119. You just need to have a voltage output which can sync to the sequencer(s). If you turn off the input toggle switch installed on the Distributor(s), you then have manual control of whatever oscillators are patched thru the Q128's. When you turn on the input jack(s), you have both automatic and manual control. Obviously, this controller patch will work with much more than just audio oscillators. It is very useful when you might want a major chord to change to a minor chord... stuff like that. With a Q-119 The Q-962 can be used with the Q-119 Sequential Controller, although, not quite as creatively as the 960. With one Q-119 in 3 X 8 mode, use one row as a trigger (gate) with only one of the pots turned up, normally, #1. This will send a trigger every time the 119 resets to stage X (#1). The other two rows can be set to alternate by patching them into the Q-962's signal inputs, and using the signal output from the Q-962. You could also use the "Done" output, which is a gate. With two Q-119's, you sync them together, and use one to provide the trigger(s), and the other for the CV. As a simple 'preset' device One or more 962's could be used to set up preset patches for live or studio performance. Each 962 could switch up to three CV's to one output (at two jacks). Trigger outputs are generated each time a stage is switched, and these triggers can be used to control other things, such as: sequencer(s) on-off, shift (advance); envelope generators; Q-128 switches; open VCA's or Pan-Fade modules; and even sync oscillators. Each 962 can also accept up to three triggers (gates) at the trigger-in jacks, as well as the 'shift' jack. The manual momentary push-buttons are ideal for 'one-time' or one-shot operations, such as starting or stopping something... and as long as a stage is active, it is outputting and holding a voltage... until another stage is activated. Some other ideas as a 'preset' device The Q-962 is NOT just for use with the Q-960, or any sequencer. But besides using it to switch the output of rows, you can use it to trigger the 960 to go to a certain stage, at a certain time, or even turn the 960's oscillator on or off at a certain time. Use it to turn a Q-119 sequencer on/off, or to switch between Q-119's... or to switch between two sequencers where one is tuned, and one is tuned but set for random mode. A Q-962 can be used manually, as well as with VC. This is worth mentioning again... remember it! You could set up a 'pitch-bend' using one, or up to three envelopes, each preset with a different response time, and the amount of 'sustain' of each would determine how much the bends are- a half step, two steps, 5 steps, an octave... your choice. Controlling the EG(s) with signal processors will give you much more control over the characteristics of the bend(s). The 962 would give you the ability to choose which of the three bends you wanted to use at any given time or place. The Q-962 can also do things like allow you to switch between oscillators to make and play different chords, and change them manually on-the-fly, or by VC. Another preset idea is to let the 962 switch in or out reverb. This is a simple patch: osc., mixer out, or VCA > mult. Split 1 > (without reverb) to 962 sig. input 1. Split 2 > reverb input. Reverb output > 962 signal input 2. 962 sig. output back to the next patchpoint (after VCA, before VCA... wherever you would have patched the reverb return before you added the 962. You can now switch the reverb in or out with the touch of a button, or with VC. The same can be done with just about anything else... a filter, or up to three filters, all set differently... an oscillator, or bank of oscillators... switch between sequencers... between a sequencer and keyboard... switch between a mono output, to a stereo output... yes, you can switch audio as well. This is a 'must have' module, much like a sequencer for automated, self-running patches. I would bet that you could have four, and still have need for more. Like the 'multiples' module, I think it would be hard to have too many Q-962's. I am at six right now, and have already thought... "yeah... I need another"... but, I was able to use a Q128 switch with the 962, and that satisfied the patch I was working on. Don't forget about, or exclude the Q-128 switch for use with the 960 and 962. They work great together! As an Octave Divider Roger verified /2 or /3 up to several khz. As a mini sequencer with a Q125 The Q962 is basically a 2 or 3-stage sequencer but provides switching, not voltages. You can use a Q125 to provide those voltages and wha-la, mini sequencer. Trigger (gate) output to filters, oscillators, etc... Also, the Q962 provides a trigger (gate) output for each stage. This level is about 5 volts and can be patched to most filters, oscillators, etc to change parameters. Since most of these modules have attenuators (sometimes inverting), just dial in the effect. \As a 1-in/3-out or a 3-in/1-out switch Of course the obvious switching of any signal into/out of the signal chain. Unlike the Moog 962, the Q962's switch will operate both ways. 1-in, 3-out --or-- 3-in, 1-out. Additional ideas Trigger the Q962 with a Q106 oscillator by patching into the shift jack. You can also trigger the Q962 with the Q119 sequencer. Oh, and you can shift with a gate signal from your keyboard. Q963 Patches Q963 Trigger bus Timing control - Patch Q963 bus 'A' to the lower part of a Signal Processor, and invert it. Patch that output to a mixer input. Patch bus 'B' directly to a mixer input. Patch the mixer output to the Osc.' input of the Q960. (If you have made the WF-mod to the Q963, turn it off. Assign each stage to either 'A' bus, 'B' bus, or 'Off', according to the rhythm desired. Use the two mixer input pots to adjust the timing for all stages on each bus. Stages assigned to 'Off' will run at the frequency settings of the Q960 internal osc. Detailed tweaking of the two mixer input pots in crucial. This patch allows you to alter the 'timing' of the Q960, making instant, on-the-fly changes in patterns... a way better method than adjusting a row of knobs on the sequencer itself. SIGNAL PROCESSOR Using the Q125 Signal Processor to attenuate the velocity output from the Q104 MIDI interface This C/V velocity is sent to a Q108 Amplifier control in with an envelope from Q109 as a second control in. The Q108 Amp signal in from a lowpass filter with velocity applied to the cutoff and can then be limited when the filter opens. This provides a means of eliminating harsh gain runaway with brightening of the filter. Basically set the Q125 gain to -100% and the offset to -3 for a limiting effect on the velocity C/V. The Signal Processor outputs –5V to +5V depending on the Offset value if no input is present. Dynamic Modulation can be achieved by running the output of the Envelope Generator through the Signal Processor before going to the Exponential Signal input of the modulating LFO. SLEW LIMITER A gate signal can be run through the Slew Limiter before going to the Envelope Generator for legato effects. The Envelope Generator will not begin a new cycle if a key is not pressed until after the EG finishes its last cycle. Modulation delay can be accomplished by running the Envelope Generator output through the Slew Limiter before going to the Oscillator acting as the modulating LFO. SWITCH Simple Logic using the Q128 Switch I see that some have requested a module with some simple logic functions. I'd like one too but in the meanwhile a switch can provide the most simple ones. AND An AND gate is the easiest one, simply patch one of the logic signals you want to AND into the A input of the switch and the other one to the control input of the switch. Now both signals must be on for the Common output of the switch to be on and the input signals are effectively ANDed?. OR The OR gate is a bit harder but still fairly simple. Split one of the logic signals you want to OR with a multiple and patch one part of it to the A input of the switch and the other part of it to the control input. Now patch the other logic signal to the B input of the switch and at the Common output will be your ORed? signal. VCA Bipolar VCA A bipolar VCA would would invert or 'negatively amplify' the signal when the CV signal goes negative. Split ("mult") your signal in 2 (good place for those Radio Shack gadgets I like). Patch the signal into Pan/Fade input #2, and also patch it through the Signal Proc and then into Pan/Fade input #1. Use the Signal Proc (the lower section is fine) to invert the signal. Switch the Q111 into "Fade" mode, which means "cross-fade between the inputs". Basically, we are fading from (a) 100% inverted signal + 0% positive with no CV input to (b) 0% inverted + 100% positive with maximum CV input. In the middle of the range we've got 50% inverted signal and 50% positive, which should cancel out. Use the Initial Control on the Q111 to adjust the midpoint to "zero out" the inverted and positive signals. You may also need to use the Q125 (the upper section) to put some offset and/or control the range of your control voltage. SPECIFIC EFFECTS Buchla Vactrol Imitation Osc 2 sine > Osc 1 softsync input set at 3-8 (play with this) Osc 2 was fed some notes from a Monopoly arpeggiator but you can use a sequencer or play real time Noise > S&H > Osc 1 1/V oct input. I triggered the S&H with the Monopoly gate out so each note is different. Fast EG with no or low sustain to State Variable filter cutoff A moderate amount of S&H output also sent to filter cutoff Osc 1 and Osc 2 > mixer > SVF audio input (use mostly Osc 1) SVF output to VCA Fast envelope with low or no sustain to VCA control input VCA > audio out The trick (beside using the softsync in its middle range) is to find the exact envelope settings. It needs to be fast but vactrols have "ring" and also "thwip", meaning they don't come on or shut off instantly, even hit with a square wave, so the envelope attacks have to be very slightly higher than zero - just one or two tick marks. You can use a small amount of sustain and short decay and release (but longer than the attacks). A vactrol is a trade name for a device that has both an LED and a photocell in one pkg. They are light coupled but not electrically coupled. The photocell has a slight lag time coming on and an even longer lag time to shut off (known as 'ring". These lag characteristics are known to be musical so they are used in things like Buchla Low Pass gates, but also in classic compressor designs - wherever you see "opto" in the name or description. A Simple Delay Patch Requirements: • 2 Q106 VCO • 2 Q108 VCA • 1 Q109 Envelope • 1 Q124 Multiples A delay is a repetition of a sound that goes more and more silent as it progresses. This is an approximation of that effect that is interesting to try to get the hang of the delay effect as well as a pretty useful source for further treatment by other effects. One Q106 VCO is your sound source, this can be replaced with any patch you've made and perhaps should consist of more than one Q106 VCO and a Q107 filter but to keep it simple we just use one Q106 VCO. The other Q106 VCO is your repetition driver so set it to a low frequency and watch the LED blink in time with your delay. One Q108 VCA is together with the second Q106 VCO the delay and the other Q108 VCA is the main VCA that, when controlled by the Q109 envelope, forms the decay of the repeating sound. • Patch your sound Q106 VCO to the signal input of the first Q108 VCA • Patch the output of your first Q108 VCA the signal input of your second Q108 VCA and take your audio out of that one • Patch the down sawtooth wave of your control Q106 VCO to the control input of the first VCA • Open up Q108 VCA 2's gain and listen to the sound repeat itself over and over, it should be a short decaying sound • Close the gain of Q108 VCA 2 and patch your Q109 Envelope to the control input of Q108 VCA 2 • Split the gate from a MIDI->CV module or your keyboard with a Q124 multiples • Patch from the Q124 multiples to the gate input of your Q109 Envelope • Patch from the Q124 multiples to the hard sync input of your control Q106 VCO • Set your Q109 envelope to 0 Attack, 5 Decay, 0 Sustain and 5 Release Phase Shifting If you take the voltage controlling your oscillator frequency (and this idea only works if you're controlling it externally, so if you're not, you'll have to re-arrange your patch to do so, but that's not hard), pass it through a multiple, and then toss it to two different oscillators, then take the pulse output of the second oscillator, invert it with a signal processor, and patch that to the hard sync of the first oscillator, won't the second oscillator's pulse width knob effectively act as a phase knob for the first oscillator? (As usual, I'm not sitting in front of my system right now to try it, but it seems that way to me -- turning the knob will effectively delay how long into the wave cycle the wave will trip high, but the cycle will still be precisely the same length as the wave on your primarily oscillator, so you'll wind up with a wave of the same frequency being triggered with a controllable delay.) Sync + EG Patch the sawtooth output of VCO1 into the hard sync input of VCO2. If VCO1 is set to a higher frequency than VCO2, VCO2's pitch will be controlled by VCO1. But, the timbre ("tone") of VCO2 varies as its pitch is changed. Next, patch an EG into the FM input of VCO2 (the one with a level control, ideally, because we will need to adjust the amount of FM). I'd say that the typical patch uses an envelope that ramps from zero to maximum during the duration of the note. So, dial S to 10, set D and R to 0, and adjust the A time appropriately. The result is a kind of "ripping" sound. You've heard it before, certainly! Adjust the pitch of VCO2, the amount of FM (pitchbend) caused by the EG, and the levels of the VCOs until it sounds good. SPECIFIC SOUNDS Cymbals Try ring modulated LFO square waves at various frequencies Kick Drum Ping a filter. Then use two VCO’s, two VCA’s, and two AD-style EG’s. EG1 opens VCA 1 passing VCO 1. EG2 opens VCA2, passing VCO2 to VCO1’s FM Input. Make EG2 really short, simulating the click at the beginning of the hit. This could potentially work with any percussion sound that has two parts with different envelopes. Canon Patch Patch the pitch out to oscillator 1, and to the input of S/H. Patch the gate to the EG, and to the signal processor. Invert the gate signal and offset with 5. Output to the S/H gate. Patch output of S/H to oscillator 2. The two oscillators go to an amplifier controlled by the EG. Now you play a two voice canon; both the note you play and the previous played note will sound. But you have to play with a detached touch, no legato. Chaos Patch A good starting point is to set the VCOs to Low (LFO) range, and put the FM input levels up to around 8 or 9. Watch your monitoring levels! The output frequencies will cover pretty much the whole audio range. Small adjustments can make big changes to the output. >The basic patch: > VCO1 sine out --> VCO2 1V/Oct. input > VCO2 sine out --> VCO3 1V/Oct. input > VCO3 sine out --> VCO1 1V/Oct. input and, using a multiple... > VCO3 sine out --> your usual audio monitoring input To attenuate the audio output, I use a VCA, setting the Gain as needed. Experiment with feeding VCO1 or 2 to the VCA's control input. It's okay but not amazing. More interesting is modulating the VCA with a frequency-shifted version of VCO3. Not many folks have frequency shifters, though. Fat Moog Bass Sound Use a single sine wave either 1 or 2 octaves lower than your sawtooth(s), but make sure the sine is in perfect tune with #1 sawtooth if you use two. Using two, slightly detuned, will give you the fat Moogy sound, and the sine wave will give you the low-end bass. Or instead of using 2 sawtooths, use 1 sawtooth and 1 square/pulse. A good way to "fatten" the oscillator waveform is to overdrive it for extra partials. Use the mixer and feedback the signal to itself. Split the output from the mixer and return it to a channel of the mixer for a fattening effect. It is similar to using the external input on a Minimoog for feedback overdrive. Sawtooth Animator Patch Another trick you can do with the clipper is it can be used as a sawtooth animator to add an effect similar to pulse width modulation, but applied to a sawtooth instead of a pulse wave. 1. Connect a sawtooth to a mixer and connect the mixer out to a clipper. 2. Connect an LFO sine or triangle wave to a second mixer input, but with the level a little lower than the level of the audio sawtooth. This will 'bias' the sawtooth positive and negative, which will cause the clipper to flatten just the top of the saw , then just the bottom, etc. 3. Connect the clipper out to a multiple. Send one multiple out to the final destination (filter input, VCA, etc.) and send another multiple out to a 3rd mixer input to form a feedback loop. 4. By adjusting the mixer levels of the LFO and clipper feedback input, you can dial in the saw animation effect. Now, instead of the simple sawtooth waveform, try it with more complex waveforms like the output of a resonant filter. Dual Saw Mix Saw and phase-aligned Square, invert one, modulate pulse with a triangle or sine, each having its own effect. Triangle causes trill at high speed while sine causes vibrato. for the Dual Saw to work, you need to mix until you hear a +1 octave saw (pulse must be centered and square). This should not produce "trill or vibrato" as described above unless, instead of mixing until hear 1+ oct saw, you mix until you only hear the pulse sound disappear. I'm explaining it 100%. Basically you will be able to achieve a true (but limited) phase-modulation effect by modulation the PW. Now modulate that pulse width at high frequency using a second oscillator with its pulse output. Experiment with pulse width of the second oscillator to achieve some gnarly effects akin to Bi-N-Tic. Unstable Saw Linear-FM a saw with itself. CONNECTIVITY Dot Com Attenuation to Outside Systems The signals in a Synthesizers.com modular are very 'hot' meaning they are often 20Vpp or more. These high levels help with noise reduction. Since amps and mixers are designed to deal with much smaller signals, you'll need to attenuate the signals from the synth to prevent clipping. This is very easy since most synth patches end with a VCA (Q108 Amplifier). Simply use the Control #1 attenuator to reduce the output level to whatever you need. Another way is to use a Q125 Signal Processor to attenuate the output signal. 18