Magick? Phase and pitch effect question...

[Vogon]

OK, something I've been aware of, but never stopped to scrutinize:
When we use a phaser or univibe with no dry signal, we get a pitch modulation effect.
If you arbirarily "stop" modulating the all-pass filters, and compare the pitch of the output to the incoming signal - the pitch would be the same.
A couple of set-cutoff allpass filters in series do not change pitch.
So the pitch modulation effect is only "happening" when the filters are moving...

Can someone explain this?

[earthtonesaudio]

It's a lot like playing a tape faster or slower.  But because they're phase shift stages the delay time is dependent on the signal frequency, so certain frequencies get sped up/slowed down more than other frequencies.

[frequencycentral]

Required reading: http://www.geofex.com/Article_Folders/phasers/phase.html

[Vogon]

It's a lot like playing a tape faster or slower.  But because they're phase shift stages the delay time is dependent on the signal frequency, so certain frequencies get sped up/slowed down more than other frequencies.

Right, the tape/flanger speeds or slows the frequencies equally across the board.
But with tape, if we stop modulating, and fix the speed at a faster rate, the audio stays sped up.
With an allpass, if we fix the cutoff at a higher point, are you saying that certain frequencies *stay* at higher shifted frequency depending on their relationship to the filter's cutoff?

Thanks FC, I've read that article a few times. I guess I can't figure out the some of the differences between shifted frequency and shifted phase.

[puretube]

It's a lot like playing a tape faster or slower.  But because they're phase shift stages the delay time is dependent on the signal frequency, so certain frequencies get sped up/slowed down more than other frequencies.

Right, the tape/flanger speeds or slows the frequencies equally across the board.
But with tape, if we stop modulating, and fix the speed at a faster rate, the audio stays sped up.
With an allpass, if we fix the cutoff at a higher point, are you saying that certain frequencies *stay* at higher shifted frequency depending on their relationship to the filter's cutoff?

Thanks FC, I've read that article a few times. I guess I can't figure out the some of the differences between shifted frequency and shifted phase.

The tape you`re talking about, is the one you (completely) recorded one hour of music on, yesterday...
today you`re going to play it back at (constant) double speed: one octave higher, in half the time...

The shifter (call it phaser or flanger) only plays back a short instance of what you just "recorded"
at a faster rate while you are speeding it up the very same/next moment...

only while the phase (or delay time) is constantly increasing, you`ll hear a higher pitch.

But: you can`t go keep on increasing all the way: so after a short while you`ll have to de-crease (lowering the pitch)...
in short: after going up, it has to go down: voila, Vibrato.

(soon someone else will algebraically derive the "rate of change" a.s.o. more precisely, I`m sure...
the technical descriptions have all been done in the 40`s & 50`s, during the electronic organ hightimes)

[Vogon]

The tape you`re talking about, is the one you (completely) recorded one hour of music on, yesterday...
today you`re going to play it back at (constant) double speed: one octave higher, in half the time...

only while the phase (or delay time) is costantly increasing, you`ll hear a higher pitch.

Aha! That's what I figured.

But: you can`t go keep on increasing all the way: so after a short while you`ll have to de-crease (lowering the pitch)...
in short: after going up, it has to go down: voila, Vibrato.

(soon someone else will algebraically derive the "rate of change" a.s.o. more precisely, I`m sure...
the technical descriptions have all been done in the 40`s & 50`s, during the electronic organ hightimes)

So the allpass is "transfering" the rise (or drop) in modulation *speed/rate* to the perceived pitch of certian freqs.
A sine at a given F will have it's waveform cycle "shifted" earlier or "later" in relation to itself as the allpass C sweeps up or down.
I'm beginning to realise what's going on, Thanks!
When the sweeping stops, so does the "stretching" effect on the sine, which remains at a constant phase shift - but now a constant perceived freq.
Sorry if I'm being slow! I have no training in these subjects, but I think this phenomenon is harder to grok than it first seems.
The usual "delay effects" principals to not bear much relevance at all.