This is a simple sallen-key VCF using a miniature tube (1J17B) as the main amplifier. There are a few opamps in the signal path, but they are only used with a gain of 1 as inverting and non-inverting buffers. The circuit is based on Ken Stone’s tube VCA which uses the 1j24B. The cutoff frequency is controlled using a dual vactrol.
The core amplifier which is based around the tube is somewhat different than in Ken Stone’s VCA. I have removed the voltage control of the gain, and changed the tube’s biasing. The reason for this is that I noticed the tube going into cutoff mode with large signal inputs. That problem was corrected using an input resistor with greater impedance. The control grid resistor, screen grid, and plate resistors were adjusted so that the distortion is more or less symmetrical. A fixed gain of about 3 is used for the tube amplifier. Resonance is reduced rather than increased as the signal is attenuated in the feedback path. With this setup, the tube starts to clip on signals greater than about +-2V, but the distortion is what I was after. I think it sounds nice with signals at or above +-5V. The maximum signal output is around +-6V.
At high resonance levels, the filter will oscillate but requires some input to get it started. After oscillating starts, the signal can be removed. While oscillating, a large input is required to overtake the resonant frequency. Several different sounds can be achieved by adjusting the input level while at maximum resonance.
I have built a prototype circuit and tested with 3 different 1J17B tubes. I am planning a layout, but it is not a high priority (i have built to many filters for a while). below is the schematic and a recording made during testing. Its not an impressive sound but it should give a good idea of what this thing does.