Circuit to pump a poor battery empty

[Prototyp_V1.0]

Theory is this. I've got a poor batery, say 9V and its almost empty, but the voltage is still normal when unloaded. Say it's maximum current is 5 mA when shortcircuited.

And I got a beeping device* that runs on everything from 3 volts and above that. This (whatever it is) draws 50mA in order to work. The idea is that the 9V battery will provide the power. It doesn't need to run seameless, it may as well pulsate. The idea is that it's running as long as the battery can provide enough voltage.

So the solution I've thinking on is that the battery charges a quite large capasitor. And when the voltage over the capasitor reaches say 8 volts, then the capasitor discharges quickly over the beeper. And then the cyclus repeats.


* Beeping device or a lamp. It doesn't matter as long as it generate sound or light.

 

[Meridian]

Theory is this. I've got a poor batery, say 9V and its almost empty, but the voltage is still normal when unloaded. Say it's maximum current is 5 mA when shortcircuited.

And I got a beeping device* that runs on everything from 3 volts and above that. This (whatever it is) draws 50mA in order to work. The idea is that the 9V battery will provide the power. It doesn't need to run seameless, it may as well pulsate. The idea is that it's running as long as the battery can provide enough voltage.

So the solution I've thinking on is that the battery charges a quite large capasitor. And when the voltage over the capasitor reaches say 8 volts, then the capasitor discharges quickly over the beeper. And then the cyclus repeats.

* Beeping device or a lamp. It doesn't matter as long as it generate sound or light.

What type of 9V battery are you using? (e.a. heavy-duty, NiCad, NiMh, Lithium, etc).

 

[Prototyp_V1.0]

Doesn't realy matter wich type it is. I assume I have a battery with high internal resistance, but is capable to hold at least 5 volts under easy load (to power a beeper or whatever device).

But I though a while ago (after this thread was maded) that to make this thing work, I realy need another powersupply. Because it's likely that my circuit would be impossible to make whitout a schmitt-trigger or a mix of opamps and logic gates (hope you folks can disprove this). I have made a basic layout for this circuit by hand and it's likely to work even if it's not tested yet. But that circuit is useless for me because of the need for another stable voltage-supply (in addition to the battery which I want to drain empty in "pulses").

This is not a serious experiment that is very important to get working, but rather fun to see if it's posible to make up.
If this circuit's layout ever come a reality, I guess it takes at least two zeners, and a couple of lock-circuits.

The lock-circuit in this context is a basic one (see picture)