{"id":6129,"date":"2023-10-14T15:16:04","date_gmt":"2023-10-14T20:16:04","guid":{"rendered":"https:\/\/pugix.com\/synth\/?p=6129"},"modified":"2023-10-14T15:34:49","modified_gmt":"2023-10-14T20:34:49","slug":"fyrall-mutant-completed","status":"publish","type":"post","link":"https:\/\/pugix.com\/synth\/fyrall-mutant-completed\/","title":{"rendered":"Fyrall Mutant Completed"},"content":{"rendered":"\n
\"\"<\/figure>\n\n\n\n

I’ve understood it, modified it, built it into a plexiglass case. The hardest part was figuring out how it all interconnects internally. I won’t attempt to explain that here. Instead, I’ll give details of one third of each of the three main circuits and just hint at how they interconnect.<\/p>\n\n\n\n

The Oscillators<\/h3>\n\n\n\n

Fyrall has six oscillators, i.e. three pairs. Peter Blasser called one pair The King’s Face in his inimitable documentation. One of the pair “stabs up”, i.e. is a falling sawtooth, while the mate “stabs down”, i.e. a rising ramp. They share a common pot that acts as a modulation depth control, inversely between the two. When the modulation source is replaced by a constant voltage, the pot becomes a shared rate control, again inversely. Also shared is a single LED to give an idea what’s going on.<\/p>\n\n\n\n

The oscillators are connected in a ring, 1->2->3->4->5->6->1, each modulating the next. But the pairs are 1&4, 2&5, 3&6. These are color-coded blue, red, and yellow respectively (knobs and LEDs). My choice of capacitors results in 2&5 being the fastest, 1&4 being medium, and 3&6 the slowest. I use the same value capacitors in each pair. But I have range switches for each pair as well, with values chosen to keep these speed relationships across the ranges.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

The values are listed above. Switch in the center-off position is the fastest. The A side adds capacitance to the hairy cap on board to make it slower; B side even more slow. There are 27 combinations of different ranges. However, each range is pretty wide, because the modulations can make them go very, very slow. This results in a default behavior of being quite bursty. They’ll go subsonic for a while and then have a burst in the audio range.<\/p>\n\n\n\n

\"\"
A King’s Face Oscillator Pair<\/figcaption><\/figure>\n\n\n\n

The schematic I drew, above, shows the modifications I made:<\/p>\n\n\n\n

    \n
  • SPDT switches can break into the modulation points at ZE.<\/li>\n\n\n\n
  • Range switches next to the hairy capacitors<\/li>\n\n\n\n
  • Individual oscillator outputs with 10K series resistors to Green jacks<\/li>\n<\/ul>\n\n\n\n

    To add the modulation switches, I unsoldered the input end of the 100K resistors, which had been connected to the ZE nodes. You can switch between the ZE node (the output of the previous oscillator in the ring) and a fixed +9V or 0V through a 10K resistor. Brown jacks give access here, always available to apply external modulation. With the input selecting a fixed voltage, the oscillators run sans modulation. The pot then controls the two rates, inversely. Since each side always has the same capacitance, no matter the range, the rates become the same in the middle of the pot.<\/p>\n\n\n\n

    Peter is clever! The modulation out is hardwired to the next ZE point in the ring. It’s the product of the oscillator and the voltage incoming from the Orranger. This circuitry is on the right side of the above diagram.<\/p>\n\n\n\n

    Additionally, a count pulse is generated by each oscillator, which drives one of the three “Counten Temples,” one to count up, the other to count down.<\/p>\n\n\n\n

    Counten Temples<\/h2>\n\n\n\n

    Each of the 4-bit counters is driven by an oscillator pair.<\/p>\n\n\n\n

    \"\"
    One Counten Temple<\/figcaption><\/figure>\n\n\n\n

    Each Counten Temple yields eight<\/strong> outputs. Two step waves (Count and XOR) are sent to the U-strip for routing into the Orranger selector chips. I added the op amp voltage followers here for two reasons. First they are brought out to the panel on the Orange jacks through 10K resistors (not shown on the schematic), and second they go directly to the Orranger inputs. Those want to be low impedance to quick charge the Track & Hold caps.<\/p>\n\n\n\n

    The Count stair step output is monitored by an LED.<\/p>\n\n\n\n

    The two Most Significant Bits (MSBs) of the counters are sent to the panel on Red jacks, as well as controlling the selection made by one of the Orrangers.<\/p>\n\n\n\n

    Finally, the four inverted bits are passed on to the next Counten Temple for preloading.<\/p>\n\n\n\n

    Each Counten Temple sports three control inputs, all binary. Purple jacks can be used to not<\/strong> invert the inverted MSBs. The Yellow jack causes a preload event when pulled low.<\/p>\n\n\n\n

    Orrangers<\/h3>\n\n\n\n

    The Orrangers consist of double 4-way selector switches. Each takes in a signal from the U-strip and routes it over to one of the oscillators (see above). The selection made by an Orranger is controlled by the two MSBs of a related Counten Temple.<\/p>\n\n\n\n

    \"\"
    One of three Orranger switches<\/figcaption><\/figure>\n\n\n\n

    I added 100nf capacitors to the switch output buffers to hold the last voltage value when the voltage on a Blue jack is high. Track\/Hold.<\/p>\n\n\n\n

    Here’s where I tapped the two Counten Temple MSBs out to the Red jacks.<\/p>\n\n\n\n

    Each Orranger Select output is hard-wired to one of the oscillators (see above).<\/p>\n\n\n\n

    A single LED monitors both outputs. Peter must have run out of room on the PCB to use the same LED driver circuit that’s on the other signals. So these LEDs seem to stay partially on all the time.<\/p>\n\n\n\n

    Studio Outs<\/h3>\n\n\n\n

    The three Grey jacks correspond to the three top 3.5mm jacks for studio output. These are three identical FET buffer circuits that AC-couple the outputs and drive them into a a bit of warm distortion. Signals can be stacked onto a gray jack, resulting in an averaging mix.<\/p>\n\n\n\n

    Spesal Cuck Input<\/h3>\n\n\n\n

    The fourth 3.5mm jack is the signal input for modulating the MIDO voltage. The modulation depth is controlled by the Black knob. A signal patched here affects every circuit! I’ve found it fun to patch out from Fyrall through an effect and then back here.<\/p>\n\n\n\n

    Range Capacitors<\/h3>\n\n\n\n

    Who knows what values to choose for Hairy caps in Ciat-Lonbarde designs? It’s experimental. The caps soldered into the board can’t easily be changed. But I wanted the lower range caps to be easy to experiment with. So I came up with a socket approach with headers.<\/p>\n\n\n\n

    \"\"
    Twelve range capacitors<\/figcaption><\/figure>\n\n\n\n

    The six pairs of range caps are soldered to pin headers and plugged into female headers soldered to the range switches.<\/p>\n\n\n\n

    \"\"
    Low range caps, 1uf electrolytic and 6.8uf film<\/figcaption><\/figure>\n\n\n\n
    \"\"
    Four pairs of mid-range capacitors<\/figcaption><\/figure>\n\n\n\n

    The mid-range capacitors can even be swapped into the opposite pair, if I want to.<\/p>\n\n\n\n

    \"\"
    Range caps showing through transparent bottom<\/figcaption><\/figure>\n\n\n\n
    \"\"<\/figure>\n\n\n\n

    Completed Fyrall with plexiglass bottom installed<\/p>\n\n\n\n

    Demo Recording<\/h2>\n\n\n\n
    \"\"<\/figure>\n\n\n\n

    This is a first patch. The audio recording was made at the same time as the video. I’ve shortened the audio a bit. It’s intermittent bursts at first. Then I started flipping switches, making some patches, and trying feedback through the Spesal Cuck input. At about two minutes in I added effects. The first effect is a phaser, the second is Data Bender, and the third is Mimeophon.<\/p>\n\n\n\n