modules with different tubes?

[harry]

Hi Scott...

The detector on TomG's site is the board I layed out based on Bob Moog's

Etherwave Pitch to Voltage Converter (with permission). It works well on
any
regular waveform, but needs to have the fundamental cleaned up a lot for
guitar.
(this depends on your expectations... I'm really fussy!)

I looked into the idea of using the Hilbert Transform. You need a "dome
filter" also
described in electronotes, I believe.  I got the info from JH's pitch
shifter (frequencies only) and did a spice model of the filter. The
problem is... there is enough group delay
in the dome filter to wipe out the advantage of getting rid of the
harmonics...  you will
end up in the same place as the LPF anyway. That seemed like a great
idea... and probably would work for high pitches.

The Gold patent assigned to New England Digital gives a workable method
for
guitar pitch extraction. It will not work well on voice at all...
because it uses the facts that
guitars are almost always decaying exponentially... and that the
harmonics are...
er.... non-harmonic.  Vocal harmonics would be much closer. Guitar
harmonics are
 very sharp!

H^) harry

Scott Bernardi wrote:

> You run into the same problem trying to do pitch to voltage
> conversion. TomG has a pitch to voltage  converter on his site that I
> seem to remember uses a downslope detector to detect the true
> fundamental (his site seems to be down right now).
> There's also an Electronotes article EN#136 that discusses using a
> Hilbert Transform: take the signal and the 90 degree phase delay
> network, square both of them with multipliers, then sum the squares.
> It supposedly removes harmonics and strengthens the fundamental,
> improving the multiple zero crossing problem of complex signals.  Then
> he runs it into a downslope detector to extract the true fundamental.
> I plan on playing with it (sometime - you know how it goes) because I
> want to build a pitch to CV I can use with voice or guitar.
>
> harry wrote:
>
>> Aahhh....
>>
>> Nice try. Probably will not work.
>>
>> Problem... How do you decide the "waveform" crossing points.  I'm
>> thinking of
>> a Guitar wave... which is a bastard !  It may cross zero more times
>> than you
>> like.
>> Tracking the peaks will not work either.
>>
>> DDL  pitch shifters do exactly this. If they have very short
>> delays... the
>> glitches in
>> the waveform are severe. If you can wait a week... you could make a
>> really long
>> delay.
>>
>> You should tape record the whole thing... then edit out every other
>> cycle, then
>> play
>> at 1/2 speed...
>>
>> Or just hire a Psychic to set the unit properly for the "next" note
>> you are
>> about to
>> play....
>>
>> H^) harry
>>
>> Magnus Danielson wrote:
>>
>>
>> > From: harry <harrybissell at prodigy.net>
>> > Subject: Re: modules with different tubes?
>> > Date: Sat, 13 Jan 2001 11:44:39 -0500
>> >
>> >
>> >>  If ANYONE gets a really good octave divider for guitar, tube or
>> >>  otherwise,
>> >>  LET ME KNOW !!!
>> >>
>> >>  I'll build it even if it is TUBES....  ;^)
>> >>
>> >>  (of course I doubt anyone will ....)
>> >>
>> > Hmm... I just came up with an obscure idea using a pair of BBDs,
>> > but
>> > I am sure you're not interested ;)
>> >
>> > Theory: Do frequency tracking, fetch an overtone to sample in a
>> > full
>> > BBD, play it out at half clock-rate. Alternate BBD so that one is
>> > in
>> > "record" and the other is in "play" mode. Since they track
>> > frequency
>> > one snaps a waveform and plays that pitched down but on the same
>> > boundary. Simple? Heck no! Obscure? You bet!
>> >
>> > Reason for concepti!
>> >  on? Annoyment and ammusement of Harry! ;)
>> >
>> > There are loads of problems with this strategy thought. Can we
>> > learn
>> > anything?
>> >
>> > Cheers,
>> > Magnus
>> >
> --
> Scott Bernardi
> sbernardi at home.net
>
>