Re: [avr-chat] AVR on 1 AA bat with just 1 L and 1 C

[hanzl]

> As other people seem to be taking this seriously, I have to point out
> that you have missed the need to control the generated supply voltage.
> Without control, VCC will just increase until it blows up the AVR or
> some arbitrary limiting factor comes into play.  Perhaps you think
> that you have a voltage doubler here.  This is not so.  A boost
> converter requires feedback from the generated voltage to control the
> switching mark/space ratio in order to produce the desired output.

I thought either predetermined mark/space ratio could do (with known
power consumption) or the AVR could somehow use the A/D circuitry to
deduce the Vcc and act accordingly (thought I am not sure this could
be done without more external parts).

In any case, software error may be lethal and AVR is in danger of
suicide.

Or call it a feature ;-) - imagine that AVR knows some secret and it
may destroy itself as a last resort when under attack ;-)

> I think it is an interesting puzzle.  May I have your permission to
> use it in interview situations?

Sure, it is quite free ;-)))

(And somting like this may well be the best use for this 'design'.)

Regards

Vaclav