Pasi T Mustalahti wrote:

> PTM: This was not exactly what I was looking for, but I liked it.

  Well, I certainly didn't like the original.  I am still very
suspicious of the circuit published.  Transistors still have some gain
with emitter and collector reversed.  Unless the 2N3904 is somehow
immune to this (and I am SURE I have seen it deliberately used "upside-
down" somewhere to achieve a low saturation votage), when the PIC output
pin having been high and charged the capacitor, goes to ground, will it
not bias the transistor on so as to discharge the capacitor "up" to
ground as soon as the transistor's nominal collector falls 0.7V below
ground?

  Perhaps the LED will conduct first at 1.7V which is only 0.5V below
ground with a Vcc of 1.2V, but this is a bit close.  If the LED requires
2.0V, you are going to lose current through the transistor.

 It would be simpler to replace the transistor with a Schottky diode.

  Cheers,
        Paul B.