Pseudo-random LFO

Sometimes there’s a need to get random voltages, you can do that with a S&H and a noise source (see this previous post), or you could do it with a pseudo-random voltage source. This is a pseudo-random control voltage source that uses four square wave generators, of one with controllable frequency, and mixes these four signals to form one moderately random pulse wave at the output. This illustration shows the mix of four square waves with different frequency.

Summing square waves. From Digital Generation of LFO’s for Modulating Effects. Copyright 2000 R.G. Keen

My design of a pseudo-random LFO is inspired by the design by R.G. Keen as well as a of Ken Stone’s design, and there are probably many other similar designs out there. It has four controls (Coarse rate and fine rate, Range of the pseudo-random CV (1-10V), and Slew (stepped CV or smooth CV), it also has two outputs (pseudo-random CV out and pseudo-random clock out (positive going clock pulses at +5V)).

pseudo-random-lfo

Pseudo-random LFO

The module uses +5V, +12V, GND, -12V, and needs a +5V option installed in the modular system, or have a +5V voltage regulator (e.g. 7805…). It would also be possible to use a HEF40106B Hex inverting Schmitt trigger instead of a 74HC14, and this would remove the need of +5V.

Mod wheels

I had two mod wheel from an old MIDI keyboard laying around, and thought it would be nice to have another control source in the modular, something that’s not that small that Eurorack tends to be… However, a mod wheel with only a volume control felt a bit boring, so I decided to add something else. One of the mod wheels had a spring making it return to the middle position and made it into Pitch bend wheel, with controls for the pitch bend range. I added three trim pots to adjust DC offset as well as range, it’s certainly not perfect but works good enough.

cv-wheel_schematic

Pitch bend wheel

It turned out a rather nice module, not very sexy or exciting, but useful. The second mod wheel did not have a spring, but stayed at the set level.  To make it into something more exciting I added an internal clock source. There’s also an input for an external clock source for syncing the module with other modules. The clock signal in turn, controls a stepped triangle or ramp wave which is sent via the input to the wheel and then (buffered) to the output. If any signal is connected to the input, this will override the internal stepped wave, and then the module might be used as an buffered level control.The stepped output might be used as an arpeggiator where the mod wheel sets the range of the arpeggio. An external CV quantizer i highly recommended to get correct tones…

cv-wheel2

CV wheel with built in stepped LFO

Both these modules turned out nicely. They are useful, and fairly easy builds, in the way I enjoy to make music with my modular setup.

 

Noise source and sample & hold

No modular is complete without a noise source. I use noise quite often when I make music, so I really wanted to add a nice noise generator in the modular. The noise generator is inspired by Eric G with some changes. My version outputs White noise, Pink noise, Red noise, and Blue noise. None of these outputs are really perfectly Pink or Red etc, but they are filtered noise which reduces the need of external filtering. The noise are based on a 1N759a zener diode.

noise_schematic

Noise generator

When the noise source were finished I had some space left on the faceplate, so I added a sample and hold unit to extract random voltages from the white noise. Many thanks to Magnus Danielson for ideas and discussions! The sample and hold is based on René Schmitz’s circuit, and the built in LFO is based on the ASM-1 LFO. The sample and hold has controls for LFO speed and CV range, and outputs for CV and Gate.

sampleandhold

LFO and S&H

This design works rather nicely. However, when making the pulse to the sample and hold (i.e. the FET transistor) short enough so the S&H doesn’t leak between samples the indication on the LED is very short leading to low light level and the output Gate is rather a short trigger. Therefore the pulse width of the LFO should be changed, by removing the 1N4148 diode and the 2k resistor. This would however decrease performance of the S&H why the circuit need an additional Gate to trigger circuit converting the square wave from the LFO to a short trigger pulse to the S&H, for example this one by Ken Stone.