Originally Posted by i_djoel2000
i checked the circuit, it is too big for portable use
There are at least two mentioned in that thread. The one I posted above the one Avro's link takes you to requires a dual op-amp, a handful of resistors, and a reference. Even with thru-hole, you could probably pack that into a corner somewhere. Do it with SMT, and it should be easy.
how do i check the opamp functionality?
The best way is to plug them into a separate, known-working circuit.
Lacking that, you can build this or something like it.
I doubt your low-battery problem killed the op-amps, by the way. I think it's more likely that your initial successes with the DC-DC conversion circuit were a matter of insufficient testing, and that you're now finding the downsides.
Try temporarily bypassing the conversion circuit and powering from a good linear power supply.
Originally Posted by Avro_Arrow
There is a simpler one in Tangents site that just turns off the LED
which should fit. I don't remember off hand which amp it is
You're thinking of the zener/CRD cut-off circuit in the META42.
You don't have to have a CRD. You can replace it with any constant current source. There are many in my op-amp biasing article, but easiest would be to use the one the PIMETA v2 already gives you, in its class A biasing circuit.
If you put a zener in series with LED2, you get the META42 circuit. The downside of that is that when the LED cuts off, it also cuts off class A biasing, though that's a small matter. You could even spin it as a positive thing, since it's a way to save on battery power.
If you don't like that, you could just add another transistor to the current mirror and run your zener-limited cut-off LED from that.
Edited by tangent - 4/24/11 at 3:11pm