# automatic turn off mechanism

Discussion in 'The Projects Forum' started by electricmaniac, Dec 29, 2012.

1. ### electricmaniac Thread Starter New Member

Jul 25, 2011
13
0
Hi everybody.

I'm trying to simulate a capacitor charging circuit which automatically turns off the main power when the capacitor reaches a given voltage threshold. The system is based on a combination of an op amp, which compares two voltages (reference and voltage sampled from a voltage divider), and a relay that begins the charge (initial position closed) and should be opened when the cap reaches the desired threshold. Now, the circuit seems to work along these lines except for the following facts:

1) The cap doesn't retain the voltage reached: when the power turns off the voltage begins to decline and this apparently has nothing to do with the presence of a balancing resistor...

2) As a consequence the voltage sampled by the opamp begins to decline as well.

3) When the inverting voltage becomes lower than the non-inverting one there are periodic current spikes in the relay's coil (BTW the relay's current doesn't reach 0A when the inverting voltage is higher that the non-inverting one). The relay should stay at open position (i.e. main power off) as long as the cap retains enough voltage as to reach the given threshold...

I really can't figure out what's happening with this circuit...some hints?
Thank you.

2. ### wayneh Expert

Sep 9, 2010
12,103
3,038
A decent cap will hold voltage for days. You may want to test your's. C5 is it?

Any discharge path will greatly increase the rate of discharge. The main discharge path appears to be your voltage divider.

3. ### electricmaniac Thread Starter New Member

Jul 25, 2011
13
0
So the voltage divider slowly draws voltage from the main cap (C5) once the main power is switched off...The purpose is to charge the main cap to a reference voltage, switch off the main power, and make it retain this voltage as long as possible. Is there a method to make the main cap retain the voltage as long as possible in spite of the divider? And...how to avoid those current spikes on the relay's coil? I suspect that an SCR instead of a relay could do a better job but I really don't know how to rearrange the circuit accordingly...

Many thanks to all.