You can find a lot of posts here, under the cocoquantus and the quantisise tags, where I have explained the architecture of the five oscillator cluster in the Cocoquantus. The cluster is sometimes called the Quantussy. It consists of five voltage-controlled oscillators, each internally clocking a pair of sample and holds. Each VCO puts out a triangle wave, represented by a star shape on the graphics. One of the sample and holds samples the triangle wave of one of the other four VCOs and its output is called the Castle output. The frequency modulation input to each VCO comes from an analog switch, which is used to select the Castle output of one of the other four VCOs. The same switch also selects the same ‘partner’ VCO’s triangle wave as the input to the Castle sample and hold.
Here’s the schematic published by Peter Blasser.
Please refer to the last page of the schematic in another tab while reading this explanation. Resistors are in ohms (e.g. 104 is 10 followed by 4 zeros, or 100,000 ohms) and capacitors are in picofarad (e.g. 104 is 100,000 pf, or 0.1 microfarad).
The schematic is divided down the middle into two parts. On the left is INNEROUTER SPIRALS, FM OSCILLATOR, and on the right, PROCESSOR OF DAGGERPULSE. The two parts are connected! Peter’s schematic has shortcuts. Find the little dagger. On the left side, the dagger connects to the square wave output of the VCO. I recognize a standard VCO schematic on the top left. Note the three-position toggle switch that selects a capacitor for the range. The VCO core is made from an Operational Transconductance Amplifier (OTA) and a dual op amp. (The op amp part number is not given, but it’s probably a TL082.) Half of the op amp buffers the integrating capacitor to make the triangle output. The other half of the op amp is a comparator that outputs a square wave. This goes into the OTA to reverse the current and start the triangle slewing the other direction. But the square also clocks the sample and holds through a pulse counting network. The dagger connects them. Note that the 4052 chip is drawn on both sides. It’s the same chip!
Processor of Daggerpulse
It’s a counter made from a pair of Flip-Flops (FF) and a network of diodes. The part number of the flip-flop is not given. Note the three LEDs: red, yellow, and blue, the visual indicators located in the center of the Cocoquantus. The Processor of Daggerpulse’s purpose is to drive the CD4052 chip, which serves both as a dual switch and a dual sample and hold. The CD4052 is a dual four-to-one selector. The inputs are indicated the two spirals of four wires each. One set comes from the triangle outputs of the other oscillators; the other set come from the castle outputs. The two are selected in tandem. That is, the triangle and castle outputs of the same VCO are selected together. The CD4052 has a two-bit selector input (A and B). These select which of the other four is chosen. But the CD4052 also has an Inhibit input. When Inhibit is enabled, then no other oscillator is selected. But, since there is a 1 microfarad capacitor on the output, the last voltage will be held, hence creating a sample and hold. Technically, it’s a follow and hold circuit that gets pulsed to make it sample and hold.
There’s also a small circle (which indicates a brass panel screw, or a green banana jack on the new model) that is the input to a trigger generator that clocks the counter. If you touch this, the LEDs seem to jump randomly. If you clock that input, by patching to it, you can then see the binary counting. I admit that I don’t fully understand the switching matrix of diodes. It is a bit maddening that the schematic shows four wires glommed into three inputs (A, B, C) on the 4052. I’m certain that C is the Inhibit input. I can’t tell which of the four diodes are connected to C, but my guess is it’s the two in the middle. In any case, what it does is to select one of the other four VCOs and also whether it is being tracked and held or sampled and held.
The lower left side of the schematic shows the CV section. It consists of a VCA, which controls the depth of FM, and a current source for controlling the frequency of the VCO. There are three inputs to control FM, two panel controls and one CV. The first panel control is an initial frequency setting. It affects all five oscillators. The CV comes through the selector switch from one of the other Castle outputs. This passes through a VCA to control the depth of FM. The Chaos panel knob controls the FM depth simultaneously of all five VCOs. There is no way to patch a CV into this VCA. (I added this feature to the Quantisise.)