active bleeder circuit for capacitor discharge

Question

I have a circuit with a large capacitor, a solenoid and an N-channel FET fed by a 60VDC supply. When I trigger the FET the capacitor discharge activates the solenoid. This all works as expected (there are additional components -- resistors and diodes and such).

Now I need to add a safety circuit such that if the power is turned off and the supply line quickly falls towards ground potential the capacitor will quickly (e.g. <2sec) discharge to prevent accidental firing of the solenoid. Due to various constraints a bleeder resistor will not work.

I have done a pretty thorough online search and haven't come across a simple circuit that does this. It seems that a P-channel "safety" FET connected across the capacitor in series with a bleeder resistor, with the gate connected to the power supply would do the trick -- when the supply is high the FET is off, and vice versa.

Is this a reasonable approach? Some of my concerns: What happens when the supply is turning on? Will the safety FET also be on causing a large load on the supply? What about limiting the source-gate voltage to something reasonable (or is there a FET that can do the job and handle 60V)? I'm trying to get away with as few components as possible due to severe space constraints.

Here is what I had in mind (solenoid and trigger FET not shown). A further constraint of the problem is that each of the 64 solenoid circuit "units" (the dashed box) is independent and, because of cabling/connectors and such, additional wires to it are not allowed. So, there can be no separate "discharge" signal to each unit -- the only notice a given unit gets is that power has dropped.

^{simulate this circuit – Schematic created using CircuitLab}

Answer 1

Tried the circuit, with 300K resistor from 60V to gate, and 100K from gate to source, using an IRF9540 P-channel MOSFET. It worked but not as well as I'd liked. I changed the 60V to gate resistor to 100K, and replaced the gate-source resistor with back-to-back 5.1V zeners. This worked much better because it tried to maintain a high gate-source voltage even when the supply voltage bled down.

BTW: Disconnecting the supply output from the solenoid unit worked well, with fast discharge of the cap, but disconnecting the input side did not. I lowered the supply bleeder resistor on the solenoid unit from 10K to 1K, and that worked better for supply input disconnect, but also boosted the supply dummy load to a few watts for each solenoid unit. Something in between works fine for my needs.