USB only modification - measurements

FigA, TP1, left: framerate 5 f/s, right: framerate 60 f/s

FigB, at TP2, both: framerate 5 f/s, left: mic-slider low, right: mic-slider high

FigC, at TP2, both: framerate 60 f/s, left: mic-slider low, right: mic-slider high

FigD, at TP3, both: framerate 5 f/s, left: mic-slider low, right: mic-slider high

FigE, at TP3, both: framerate 60 f/s, left: mic-slider low, right: mic-slider high

FigF, at TP4, manually triggered to proof pulse-width is higher than input clock period

FigG, at TP4, both: framerate above 25 f/s, left: mic-slider low, right: mic-slider high

FigH, CH1=TP5, CH2=TP6, both: mic-slider high, left: framerate 5 f/s, right: framerate 60 f/s
charge transfer pulses are generated and differ slightly in lenght to the original pulses

The left picture of FigH proofs that this mod sometimes is working even with framerates below 25 f/s, but I wouldn't take it
for granted. During development I discovered that sometimes the left signal of FigD is enough to trigger the first multivibrator.
The additional 100k resistor (to guarantee operation at lower frame rates) decreases the input resistance of the first
multivibator, thus decreasing the signal amplitude.

FigI, CH1=TP5, CH2=TP6, both: mic-slider low, left: framerate 5 f/s, right: framerate 60 f/s
charge transfer pulses are not generated, output of 2nd multivibrator stays high