Is Harmonic Tremolo = Phaser with 2 out of phase stages ?

[drolo]

I have been following midwayfair's and samhay's builds of an harmonic tremolo with much interrest.

At the same time, I'm working on an adaptation of the Lovetone Doppleganger. It occurred to me that the 2 doppelganger's phase stages, one lower and one higher frequency, oscillating out of phase from each other might be the same as what is happening in the harmonic tremolo?  (have not breadboarded the doppelganger yet .... leaving that one in my head until ... a bit later ..

So my question is, could the phaser be used to achieve a similar sound?

[gcme93]

My immediate thought from a theoretical perspective is that two waves out of phase will just cancel entirely. Two waves half out of phase will cancel 50% and so on... What you're after is a way to make the phase shift vary from 0deg out of phase to 180deg out of phase and back. Then the rate that this varies will be the rate of your tremolo.

Does this help at all or have I just stated what you already knew? 

[drolo]

My immediate thought from a theoretical perspective is that two waves out of phase will just cancel entirely. Two waves half out of phase will cancel 50% and so on... What you're after is a way to make the phase shift vary from 0deg out of phase to 180deg out of phase and back. Then the rate that this varies will be the rate of your tremolo.

Does this help at all or have I just stated what you already knew? 

*busy reading The technology of phase shifters and flangers at GEOFEX ... again*

[oskar]

From:
http://www.strymon.net/2012/04/12/amplifier-tremolo-technology-white-paper/
The Harmonic Trem is actually not a pure tremolo effect. It is really a dual-band filtering effect that alternately emphasizes low and high frequencies. The end-result is a soothing pulse that has shades of a mild phaser effect combined with tremolo due to the nature of the frequency bands that are alternated.

[gcme93]

I have to make it clear that what I wrote above is purely a theoretical model (one of those ones that requires perfect sine waves etc) and with the variations in each branch of the circuit I'm pretty sure the waves won't cancel completely even if we did use sine waves.

Having said that, (and not knowing a massive amount about phasers) I think a phaser varies the phase of the wave with time which is what we're after. It then combines it with a non phased wave and the combination of the two cancels at certain frequencies and not others, leaving the swooping sound as bass gets cut, then mids get cut then treble gets cut etc...

It's all getting a bit complicated because the phase cancelling is dependent on frequency. What we're after for a tremolo is for cancelling to be independent of frequency so that when the phase of the modified wave changes, every frequency is out of phase at the same time, providing a complete (or close to) silence. 

I'm hoping one of the big names will now step in, correct all my hand-waving theory and provide you with a nice clean solution!

Ultimately I think you need a frequency-independent phase-varying phase shifter. When I just googled that, it came up with lots of journals you have to pay to read, and quite unpractical looking solutions :/

[drolo]

*back from reading The technology of phase shifters and flangers at GEOFEX ... head hurting*

I was referring to the phase stages in the doppelganger, that has staggered caps, like a Univibe.

As I understood, this creates two different ranges of filtering, one concentrated in the lows and the other in the highs. The doppelganger an work as a traditional Univibe/phser with the LFO modulating all 4 stages the same way or it can modulate the first two with an LFO signal that is opposite to the one modulating the last two stages. That is the part that made me think it could create a similar sound to the harmonic tremolo. But it might just be "similar"

It's confusing to understand he difference between a phase shift as is created in a phaser and a phase shift that is created in a state variable filter for example, if you would hook up an lfo to it.

[samhay]

From:
http://www.strymon.net/2012/04/12/amplifier-tremolo-technology-white-paper/
The Harmonic Trem is actually not a pure tremolo effect. It is really a dual-band filtering effect that alternately emphasizes low and high frequencies. The end-result is a soothing pulse that has shades of a mild phaser effect combined with tremolo due to the nature of the frequency bands that are alternated.

That is essentially what I understand a harmonic tremolo to be - some filtering that splits the signal to a bass and treble component and then an LFO that pans between these two signals. Because the panning is usually not perfect, the cross-over frequency moves around during the sweep and thus you get some mild phasing.
At a guess, you might get something similarish with a dual-notch phaser if the notches were quite broad and at one end of the LFO sweep the treble notch shifted above your hearing range while and the other end of the sweep the bass notch shifted below your hearing range.

Edit - just re-read the first post. I had a dual notch (4 stage) phaser on the breadboard a while ago, which swept the notches in opposite directions. It didn't sound like a harmonic tremolo (actually, I didn't think it sounded great in general), but I expect the devil is very much in the details.

[drolo]

I guess it would be closer doing that with filter circuits with low and high pass instead of phasers  then ..

that means I will get to build a phaser AND a tremolo after all ;-)

[~arph]

A phase stage makes a notch in the frequency band (aka filters around one specific frequency)
The harmonic tremolo sweeps between a full high pass and a full low pass sound.
Totally different.

[artifus]

Totally different.

?

[~arph]

   

Yes, but is the crossover frequency moved around in the harmonic tremolo or is it panning between the two filters?
If it is, it is indeed just like a phaser, if not.. totally different  

[artifus]

sorry - ambiguous pre coffee posting, brain just waking up and struggling to make sense.

what happens to the crossover frequency as it pans between the two filters?

[~arph]

It stays in the same spot.

What is would look like in that graph above is the blue and red lines alternately move up and down. but the center dip remains where it is.

[artifus]

what is happening along the edges of the slopes as they move up and down?

*edit* am genuinely asking, not arguing - haven't really considered this before. i'm guessing the notch gets narrower and wider, varying q of phase shift? not as prominent an effect as the hi/lo amplitude modulation but there none the less, hence the sonic effect?

[drolo]

what is happening along the edges of the slopes as they move up and down?

*edit* am genuinely asking, not arguing - haven't really considered this before. i'm guessing the notch gets narrower and wider, varying q of phase shift? not as prominent as the hi/lo amplitude modulation but there none the less, hence the sonic effect?

i'm trying to animate the curves in my head ...
wouldn't that make the notch shift slightly left and right as one curve goes down and the other rises?

my feeling is that the tremolo has a bit of phasing going on, albeit not as pronounced as a real phaser.

[~arph]

i'm trying to animate the curves in my head ...
wouldn't that make the notch shift slightly left and right as one curve goes down and the other rises?

Actually you are right. It would

[samhay]

i'm trying to animate the curves in my head ...
wouldn't that make the notch shift slightly left and right as one curve goes down and the other rises?

Actually you are right. It would

And in addition, if you use a somewhat imperfect panning method it gets worse - I will post a sim of frequency response of the Anharmonic tremolo in that thread in a moment.