Hello,

Doing a project and trying to figure out the best overall way to achieve this goal. I have a box with an LED and then the rest of the circuitry. Its power source if provided to it from another host device that determines when to send it power.

The intent is to delay everything but the LED from receiving power for the first second power is applied. The reason is that for low-power mode, the host device will pulse power to the box for less than one second, the result being an LED blinks to let the operator know the device still has power but is in low power mode, but doesn't waste power turning everything else on for a short time.

So I basically want to put a time delay between input power and everything else besides the LED. This will technically cause a one second delay on initial bootup of the box but that is fine. My most basic idea to do this would be a 555 timer. Part of me thinks there might be either an easier way or more reliable way? Looking for suggestions if anyone has any.

Input to the box will be 12V. Circuitry post-timer takes 75mA max.

Thanks!

 

Time Delay Module, Adj. 0-10Sec. Delay on Make, 12VDC

Item #: 33421 RL Sale: $2.95

http://www.mpja.com/Time-Delay-Module-Adj-0-10Sec-Delay-on-Make-12VDC/productinfo/33421+RL

Check their store. There are a few others.

 

A simple circuit would use a P-MOSFET, a resistor, and a capacitor, LTspice simulation below:

Edit: Note that M1 should be a standard MOSFET, not a logic-level type.

 

A consideration of just a PFET and RC gate charge network, is the
Trise of delayed pulse. There are min Trise specs on some parts/
logic, to eliminate gate oscillation as it stays in its linear region too
long and system noise causes it to give unpredictable results. An
easy way around this is single gate logic, a Schmidt inverter, added
to network or do the whole solution out of gates and add, if not
Schmidt gates, + feedback to get hysteresis, and fast edge times.

A crude sim of the problem, see attached.

Look at sink node for signal you will feed it to to see if there is any
issue with ramp/rise times at the input.


Regards, Dana.

 

Awesome, thanks everyone!