After the bridge rectifier, is there a filter capacitor to smooth the fullwave rectified AC into a DC voltage? Then what, a zener diode and an optocoupler, or something more precise like an analog comparator
What is the output stage of the voltage sensor? If it is a reasonably low impedance like an opamp or even an optocoupler, then one output can drive two circuits like the FET-A gate and a delay circuit. Why do you think you need a buffer section? From your descriptions, it looks like the voltage sensor output could drive FET-A directly, and it's negative edge could trigger a 555, so I without more information I don't see a need for a buffer, and without more information I can't design one anyway.
C1 is the power supply decoupling capacitor for the 555 IC. In my library the power pins are not shown on common parts.
The circuit in post #41 requires that the sensor circuit output be able to source and sink less than 1.5 mA. If it can sink 1.5 mA with an output voltage less than 1.0 V then it will work.
The circuit can be done with all discrete transistors. However, because of the long time delay it will take more than the minimum of 4. An extra stage is needed for hysteresis so the output waveforms to the FETs have rapid rise and fall times.
If the voltage sensor output is a low impedance that swings below 0.4 V in its low state, then Q1 is the buffer stage. If the low output is closer to 1 V, add a 1N914 or 1N4148 diode in series with the Q1 base.
Ohk and lastly, which is more feasible to implement - 1 resistor and 3 capacitor (connected to a rotary switch) or 1 capacitor and 3 resistors (connected to the rotart switch) considering lossesthat might occur in charging/discharging of capacitor?