Securing the Load

[Raven Luni]

Greetings,

Just wanted to get some feedback on a wee circuit intended to cut power to a load when a fault is triggered (in the form of a logic 1).

Its based on a bistable mutivibrator. The capacitor sets the initial state by providing a brief path to ground (and discharges through the resistor when its off). It works fine in simulation but are there any potential issues with this? For example would the capacitor affect the pulse length needed to guarantee a fault is triggered?

 

[(*steve*)]

One question you need to (probably) answer is: "What happens if both test and reset are pressed?" Or more importantly, if the fault signal is asserted and you press reset.

If the power is cycled very quickly, will the circuit always start in the correct state?

 

[Harald Kapp]

What happens if both test and reset are pressed?"

The problem hre is not while both buttons are pressed, but in the sequence of release of these buttons. Depending on which button is released first, the latch will be in one of the two allowed states.

You could use a momentary switch with one center position (inactive) and two actoive positions. One of the active positions is test, the other is reset. This makes it impossible to have both active at the same time. If you can get hold of only a single pole switch, connect the center tap to GND, one actve tap to the base of the left transistor (test) and the other active tap to the base of the right transistor (reset).

If the power is cycled very quickly, will the circuit always start in the correct state?

To support that, add a diode from the 22nF capacitor to +12V. Anode to the capacitor, cathode to +12V.

 

[Raven Luni]

Cheers chaps. In this case it shouldnt be critical that the load is turned off immediately, just that it is within a few seconds and the user is alerted. If it was more critical maybe the reset could be changed to a short once only pulse or use a flip flop.