This module has two comparators, each of which takes an input voltage and turns on a gate whenever that voltage is above a threshold. It also puts out a trigger on one output when the voltage goes above threshold, and one on another output when the voltage falls below threshold. The threshold on one comparator can be from -8V to +8V, set using a front panel knob. On the other comparator, the threshold is provided by a CV with an attenuator. That CV is normalled to 8 V so with nothing plugged in, the attenuator knob will set a threshold in the 0–8 V range.
Yves Usson designed this module for ±15 V power. To get the same threshold ranges and output levels with ±12 V I changed some resistor values. I also changed some 12k resistors to 10k, not because they needed to change, but just because I have more 10k resistors than 12k. (They are in pairs whose exact value doesn’t matter, as long as the two are equal.) I considered changing the 10R series resistors on the power rails to 1N5817 Schottky diodes for power reversal protection, but decided to leave them alone since the thresholds and output voltages are sensitive to the power rail voltages. I kludged dual bypass caps from rails to ground in place of Usson’s single caps from one rail to the other, even though it probably doesn’t matter a bit, because I’m like that.
When I went to place the power header (there are footprints for a Synthesizers.com 6-pin MTA connector, a MOTM 4-pin MTA96 connector, or a 10-pin Eurorack connector) I was surprised to discover I couldn’t. Apologies for the poor focus in the photo but maybe you can see the Euro footprint is wrong. While the holes are spaced lengthwise at the requisite 0.1″, they are too far apart in the other direction.
Yes, I know, Yves Usson does not build Eurorack. I still find it remarkable this board went to production with this wrong footprint on it, and has yet to be corrected.
I considered a few options but ended up mounting the header and an upside down 1×5 pin header to a small piece of stripboard, and soldering that to one side of the faulty footprint.
Another problem I noticed was with the LEDs. The input signal on each channel goes to an inverting op amp comparator: If the signal is above the threshold the op amp output is close to -15 V (in Usson’s design, -12 V in my adaptation) and if it’s below threshold the op amp output is near +15 V. That output goes to op amps that produce the triggers and gate. It also goes via a 1.5k resistor to the cathode of an LED, whose anode is grounded.
If the op amp is putting out (near) -15 V then the LED is forward biased, so it conducts, and the LED lights up. If it is putting out +15 V then the LED is reverse biased… at 15 V. But typically the absolute maximum reverse voltage listed in an LED datasheet is around 5 V. Even in a ±12 V adaptation of this circuit the reverse voltage on the LED is way higher than it’s rated for.
I dealt with this by connecting a 1N4148 diode antiparallel to the LED — anode to cathode and cathode to anode:
Now when the op amp puts out +15 V, the 1N4148 is forward biased, it conducts, and the reverse bias on the LED is only about 700 mV.
And other than that, the module seems fine and works.
Panel design files, Gerbers, and documentation in GitHub repository at https://github.com/holmesrichards/frontpanels/tree/main/YuSynth_Comparators