The paper circuit, shown above, is the only information that we have that pertains to the Swoop Spokes. It was a kind of mystery circuit that no one else (that I know of) tried to build or describe. I did it! Actually, I was instigated to build one when Dennis Verschoor send me a PC board that he had etched, based on this paper diagram. The first step was to draw a schematic diagram.
The schematic shows one of five identical sections on the board. I correctly guessed the function by studying this schematic. To the upper left an output emits from pin 7 of the TL082 op amp. Oddly the series resistor is 100k, rather than the 10k Peter Blasser convention. Notice that this op amp is buffering the voltage on the ‘hairy’ capacitor (more about those, later). At the top are the two sections of a CD4013 Dual D Flip Flop. Peter uses these as Set/Reset flip flops. On power up, both section are reset, but the same signal resetting these two goes to the next Swoop and sets them. So it would seem that the chips are being both set and reset on power up, but by two, probably slightly different pulses (generated by the action of the .01 uf capacitor being discharged. In typical P.B. manner, each section causes the next section to be set. Once started, each section cycles with a jumpy triangle wave. I was hoping these would run at LFO rates and that the module could be used as a CV generator. I was again correct. I chose five different values for the hairy capacitors: 0.1uf, 0.22uf, 0.33uf, 0.47uf, 1.0uf. The result is that the set of initial rates is in the range of a few Hz or so. (Smaller capacitors would probably turn this into an audio oscillator. But we already have Lil’ Sidrassi for that.)
Now look at the two inputs, the cross and double-cross. I theorized that these would impact the slew rates of the charge and discharge of the hairy cap. My theory was right, but there were surprises. I tried out the CV inputs by patching from other outputs into one or the other of the cross and double-cross inputs. When the cross input is patched, the output is altered, by injecting an odd pulse, or even in one combination the output becomes a staircase! Modifying one section changes all the other outputs, too. A surprise came when patching into the double cross. The output usually froze and held, or there was no impact at all that I could see. However, this was when patching from another output.
More about Cross and Double-Cross
I tried attaching the wiper of a bias pot (0 to 9v) to either of these inputs and found that within a certain voltage range, generally between 2 and 5 volts, there is a reduction in the rate of oscillation. Thus, using voltages in this range one can obtain control over the rates.
Interaction of sections with different hairy capacitors
I found, by making the largest capacitor to be 2.5uf, that while this longest cycle was active the others were able to have two or three shorter cycles. More experimentation is needed here.
My most recent idea for an enclosure is a plexiglass panel with 10 banana jacks, color coded for outputs and cross inputs, with a pot hooked to the double-cross inputs for initial rate control. I will add a 9 volt regulator, so I can run this off of a 12 or 15 volt wall wart. I also plan to add an LED to monitor each output, so you can see what it’s all doing. A ground jack will let me patch to other stuff.
For now, it’s looking like this.