ok this is my solution... though i'm not sure it'll work as i don't know if ground actually "pulls" into it, or is everything a result of positive voltage pushing everything in.... in terms of conventional current...
"R_z" limits the current so the "Z" zener diode can produce a voltage reference. the reference is then picked up by the "C1" and "D_av" avalanche diode (whose breakdown is equal to Z) compares it with the actual voltage from source in C2. when the voltage goes below a certain limit, as referenced by "Z", the avalanche diode breaks down and switches the thyristor "Q", which then shorts the power supply. causing a momentary voltage cutoff to the load. we assume the load, once disconnected will remain open, as if it was turned off by another thyristor as well.
ok. assuming this even works (i don't even know), being a complete newbie, in general case, my design is very limited... and, inelegant, even. we'd have to assume that the load will remain off, once turned off (as the case when it is turned on by a thyristor). can the supply handle momentary shorts?? we'd also have to find a way to turn off Q once it shorts. we'd have to assume that the voltage is falling to reach that certain voltage limit and not rising to it, etc, etc....
so i'd be happy if somebody can help me simplify this design.
EDIT:
another limit is that the zener diode has to very, very narrowly provide voltage reference only when the voltage supply is very close to reference.
View attachment 42519
