This mechanics-forward project involves a simpler assemblage of electronics compared to my previous work. I’m doing a couple of electronics projects for other people at the moment, and it is becoming increasingly frustrating to try and create electronics with an acceptable degree of cleanliness to them. My purely electronic synthesisers are giving me a lot of grief with oscillator bleed; as far as I can tell, the oscillator design that I am currently employing puts a lot of strain on the power supply, creating a voltage ripple with each cycle. This slightly modulating supply voltage is then used to power the amplifers, and the changing power is manifesting as bleed. I am doing investigations into this; either some sort of pulldown resistor that ties the output of oscillators to ground (this is presuming that they only bleed when unpatched…. not 100% about that…) or a completely different oscillator design – transistorised oscillators are looking to be a good bet, a la Buchla.

The project after these ‘acoustic’ synthesisers that I am dreaming of is a synthesiser that can sound nice. My current finished synthesiser is a bit abrasive and obtuse, characteristics that are definitely befitting of a first synthesiser, but I have learnt a lot about designing electronics and have realised that there are many faults and mistakes present. Moving towards more triangle waves, perhaps an envelope or slope generator or two, but keeping the cross modulation (with room for attenuation of this cross modulation, a feature which Is sorely lacking in my current synth, which makes it very obtuse and not at all delicate) seems like a good starting point for the next design.

The electronics that will power the acoustic synths will be fairly simple, but as this will be my first venture into CMOS chips I will be learning a lot. CMOS chips have long been heralded as great for beginner DIY synths due to their relative simplicity. Stanley Lunetta, a pioneer of synths focused around these chips, is the namesake for the CMOS sub genre of DIY audio electronics, which has a particular cult following and a particular sound. Shift registers, clock dividers, lots of intermodulation, plenty of square waves: these all result in a very digital, bitty sound.

I will be using some CMOS counter IC, but I’m not fully decided on which one. Some can count upwards and downwards, which seems like an appealing feature. With independent modulation of clock pitch, count direction and pulse width (not sure how PWM would br possible, as I have read that the CMOS counters are quite particular about the clock signals that they receive, and I’m currently not aware of any methods of altering the pulse width of a signal after it has been generated). The 4017 is a pretty standard chip that can go up to 10 steps, but there are other, more curious IC’s at my disposal.

The image above is taken from the data sheet of the 4029 counter IC. This seems to be generating some seriously interesting rhythms, but I’m curious how much control there is over these rhythms. There definitely seems to be options out there.

These 4000 series CMOS counters will be used to either trigger solenoids or control the turning of a motor. A low pass filter can be applied to these square wave clock signals to make a smoother curve, which might be good for controlling motors.