Battery saver circuit (updates?)

[RP Henry]

About 2 years ago, there was discussion on this ng about an auto-off circuit
to disconnect a battery after a time. This led to the following ascii
schematic from Win Hill:

_|_
+9V -------+-----+--o o---+--- load
Vb | | | (<200k to gnd)
33k | Q2 |
| e pnp |
+-- b 2n3906 | Vb
| c 4403 | T = RC ln --- = k RC
3k3 | etc | Vth
C + | | |
,----||----------+--------' The k term vs. Vb and Vth,
| 100uF |
| .002CV d nmos Vb = 9V 6V
+----+-- g 2n7000 Vth ----- -----
R | _|_ s Q1 = 1.5V 1.792 1.386
680k /_\ D | = 2.0V 1.504 1.099
| | |
---+----+-----+----------------- gnd

Does anyone have any suggested improvements (because of new parts available,
more experience, change in the speed of light, etc)?

 

[Jonathan Kirwan]

Does anyone have any suggested improvements (because of new parts available,
more experience, change in the speed of light, etc)?

I missed it, before. But it actually looks decent. Very cheap,
small and common parts and non-critical layout. New parts would
just cost you more and be harder to find.

It doesn't retrigger, if you were looking for that. And you
can't just build this into a switch replacement, it's not an
in-line circuit.

I suppose that you might consider lowering the capacitor value
to reduce size or cost and increasing that 680k, but if you go
too far you may have to start worrying more about leakage.

Jon