Once you got an initial signal, you can chain up any number of modules further up the signal path and/or go back again (feedback loops).
You could start with something that gives a single discrete pulse (like a click or burst) that you can use as an initial exciter into some feedback circuit, like a reverb, a delay, a phaser, a resonant filter or filter bank or combfilter or other module that has an internal feedback path. Or create an external feedback loop (like the one in my example). In that case almost any filter slope will eventually go towards self resonation if the signal has passed that filter often enough (in a feedback loop circuit).
You can also exploit (continuous) circuit noise to start things of. Airwindows has several noise or (intentially) noisy modules. Also, often digital filters have a very low level noise source built in, that will initiate self resonating when feedback/resonance is turned up far enough without the need of external input. Other noise sources could be amp(simulation)s, mixers and so on. In analog circuits the noise would be inherent to circuit and component āimperfectionsā.
Tuning can be achieved by precisely controlling delaytimes. The amount of āechoesā per second equals/determines the frequency (see Karplus Strong). Or you could use use small band or resonant filters and set/control the filter frequency.
I bet many tunable/controlable circuits can be created using these basic concepts.
They say that unintended use of stuff promotes creativity. If you want an example of what can be achieved when exploiting functionality/properties to the max, check out Jakup Ciupinski exploiting the Fundamental VCF.
Or this one, where various filter topologies are created, using other basic concepts/components/modules.
Anywayā¦
It pays of, delving deeper into basic core concepts of signal processing/synthesis. Understanding how/why stuff works further fills up your ātoolboxā and opens up all sorts of ways of exploiting these concepts for you own use cases/experiments.
