Voltage Controlled Current Source Problem, shutdown takes > 5 seconds!

[Shanjaq]

here is the schematic:

and here's the problem:
In the sim, dropping the input voltage to 0 results in an instant cutoff at the output P-FET.

On the breadboard, the output current from the P-FET sloooooooooowly drops to 0 over 5-10 seconds after setting the input/control voltage to 0...

I've build this circuit twice from scratch using different op-amps, same result. I'm using a TLC272 dual op-amp, 2n7000 N-FET and VP0104 P-FET
(supposedly a complement to the 2n7000?)

Why doesn't the second-stage amp drive the gate more forcefully into a cutoff state? I need it to instantly turn on and instantly turn off, or instantly shift to any level in between, as this is for an LED modulator!

Thanks for any help

 

[shar2013]

Are you feeding the opamp with same supply as the P-FET ? if so it needs to get to the rail to shutdown PFET and this might explain the delay.
Try feeding the opamp with higher voltage supply (say 12V) see if the situation improves.

Good luck !

 

[Shanjaq]

Are you feeding the opamp with same supply as the P-FET ? if so it needs to get to the rail to shutdown PFET and this might explain the delay.
Try feeding the opamp with higher voltage supply (say 12V) see if the situation improves.

Good luck !

Your message just jogged my memory. The TLC272 is TTL internally, which should have the same problem as TTL-based 555 timers driving FETs in that they can't quite reach the rail due to the inherent voltage drop of the BJT internals! This usually results in an undesirable bias on the gate drive signal, verified with some high-impedance probing. The solution in that case was to use a CMOS-based 555, like the ICM7555.

So, applying this logic to op-amps, I tried a CMOS dual op-amp (OPA2132) in the same circuit. Results: Response in either direction is snappy, as desired.

Thanks!