>Hi,
>
> > I did not realise that something like this could be
> > so complicated and it would possibly be easier to
> > implement using analogue electronics but that
> > would not be as much fun to play with
>
>
>I'm not an expert in analog electronics, but how about
>this: Design a notch filter RLC or RC-Opamp circuit
>(RC-OpAmp is probably preferable -- no clunky
>hard-to-find inductors), and then depending on how
>many bands you want in your analyzer, use that many
>notch filters and select component values so that the
>upper cutoff frequency of a lower frequency filter is
>equal to the lower cutoff frequency of a higher
>frequency filter. Pass the input to all of the filters
>in parallel, and you'll get seven (for a 7 band
>analyzer) output signals that correspond to your
>bands. Pass each of these through an envelope
>detector, then measure the voltage of each signal.
>That voltage can then drive a bar graph LED (for
>example). There's a few circuits on the internet
>already designed that convert a voltage signal to a
>bar graph signal, just check some of the circuit
>archives kicking around and you'll probably find
>something.
>
>That's off the top of my head... I hope none of the
>analog pros on the list are shaking their head at me;
>Maybe one of them can comment on the feasibility of
>this approach. :)
>
>T.