Differences in phase altering pedals.

[H4T]

From reading the GeoFex article on the Tech Of Phasers and Flangers, I know the difference between the two most common phase-altering pedals, but what about delays and choruses? Are delays just phasers with longer frequencies, and choruses being just flangers with longer frequencies? I know I want a phaser on my pedalboard, but I'm not sure whether or not to go for a regular phaser, or a delay...or a chorus, lol.

[niftydog]

well, regardless of the topology of the circuit, if you want that phaser sound neither a chorus nor a delay will do it for you.

Delay and chorus is a very different beast to a phaser. Put it this way; a 1khz beep delayed by 1 second is the equivalent of shifting it's phase by 360,000 degrees!

[H4T]

Well forget about what I said about deciding on a pedal. I really want to know the information about the differences of the pedals on the mechanical level.

I know flangers are basically high-frequency phasers. And chorus pedals are slightly less frequency than flangers, but still much more than a normal phaser. So is a delay pedal basically a slow-frequency phaser?

[puretube]

R.G. @ GeoFex got this: http://www.geofex.com/Article_Folders/phasers/phase.html

H-C got this: http://www.harmony-central.com/Effects/effects-explained.html

[niftydog]

The real difference between a phaser and flanger is that a flanger always produces a large number of notches that are an even multiple of frequency apart.

The delay times in a chorus are larger than in a flanger

Having read those articles, here's my take on things:

Phasers and flangers are frequency dependent. Meaning they react differently to different frequencys.

Delays are frequency independent. They delay all signals by the same amount.

So, the "hardware" is different in that phasers and flangers have only short delays and there are frequency filtering elements controlling how the delayed signal is generated and mixed with the original. They are variable notch filters that just happen to be created using short time delays.

Delays are just that - delays. There's no filtering (excpet I suppose for anti-aliasing, but that's a whole other topic!) and the delay lengths are comparitively large.

So, when you're talking about "high-frequency" vs "low-frequency" I think you're getting a little confused with "short delays" vs "long delays - but I hope I'm following what you're trying to say.

Phasers and flangers use short delays and frequency selective mixing to generate a variable frequency notch filter. The difference between them is subtle, but relates to the amount of filtering and which harmonics are filtered and which aren't.

Chorus and delay use longer delays (and no filtering, necessarily) to generate a frequency independent time shift. The difference between them is twofold; Chorus effects have a modulating (varying) time shift that is short compared to the very long and static time shift of a delay pedal.

[Mark Hammer]

From reading the GeoFex article on the Tech Of Phasers and Flangers, I know the difference between the two most common phase-altering pedals, but what about delays and choruses? Are delays just phasers with longer frequencies, and choruses being just flangers with longer frequencies? I know I want a phaser on my pedalboard, but I'm not sure whether or not to go for a regular phaser, or a delay...or a chorus, lol.

I classify these devices based on the manner in which they shift the listener's attention, as well as their electronic parameters.

The first division is between delay line and all the other effects.  Generally, delays are fixed and not shifting.  Consequently your attention is drawn to the difference in arrival time between initial and later reflections/repeats.  

As regeneration is increased, the number of repeats until final decay is increased.  In many pedals, regeneration/feedback increases the degree of emphasis at certain frequencies.  For delay lines, when using delay times long enough to be perceived as echoes, such an emphasis is either so low in the frequency range or so well diffused throughout it, that you can't really hear it.  Consequently your attention is drawn almost exclusively to arrival-time differences - there you hear it again, and again.  As delay time is shortened, however, adding regeneration results in emphasis at higher and more audible frequencies.  When I say "higher", I mean in the range of room resonances.  So adding regeneration at 50msec delay time will produce a boxey irritating quality.  (if your intent is to use your delay line for rockabilly slapback, some highpass filter that removes low end from the regeneration path is probably a good idea).

I've never really thought about it before, but I guess one of the nice things about multi-head Echoplex-type tape delays is that you could get multiple repeats without being boxey sounding since regeneration was not needed to produce multiple repeats.

Chorus, phase-shifters and flangers all use modulated delay - i.e., vary the amount of delay.  These further subdivide into flangers/chorus on one side and phasers on the other.  As niftydog notes, there is a difference between applying the same amount of delay across the entire signal, and applying different amounts of delay depending on the actual frequency.

Phasers "shift" the phase of portions of the signal by a small amount.  If the amount of cumulative phase shift in the circuit is enough, then you end up with two "copies" of the same signal that are perfectly out of phase and cancel each other.  Because the LFO varies just where that point occurs in the spectrum, you end up with notches that move around.  

Note that phase shift is distributed such that there is always more phase shift higher up than lower down in the spectrum.  So, adding it all up across 4 stages, there may be (and I'm making these numbers up) 64 degrees of phase shift at 100hz, 128 degrees of phase shift at 200hz, and 180 degrees of phase shift at 400hz (it sort of plateaus over a certain point).  So, generally speaking the effect of phase shifters tends to be in the low mids to lower treble, since it is generally harder to nail notches way deep down in the traditional 4-stager.  Possible, but tricky.

Of course, when you think of it, bumping over a 2khz wave by 180 degrees (the point where copy A and copy B cancel) isn't really much of a delay.  One complete cycle at 2khz takes 500usec and half of that is 250usec...not the sort of delay time you can actually "hear" as a delay.  This it-doesn't-really-sound-like-two-copies aspect is part of what contributes to the different sound quality of phasers.

A second contributing factors is the number of notches produced.  Normally, one notch is produced for every two stages of phase-shift. So, an 8-stager can produce 4 notches and no more, while a traditional 4-stager produces only 2.  No matter how high or low it sweeps, you don't hear more "disturbances" in the spectrum than those 2 or 4.  We'll come back to this in a moment.

Flangers and choruses are essentially the same basic beast, with a few minor differences.  One difference is that the sweep speed of the one (chorus) generally needs to be faster to notice the effect (more in a moment).  A second is that the one (flanger) benefits from use of regeneration of the delay signal while the other doesn't.  A third difference, however, is that one occupies a slightly different range of delay times over its sweep, and the perceptual consequences of this are huge.

Because flangers and chorusses apply a constant time delay over the entire signal (rather than focussing "wave-delay" incertain portions of the signal as a phase-shifter does), they will always generate many more notches than a phaser (unless it is a huge and complex phaser, the likes of which rarely see commercial production).  Where the number of notches produced by a phaser remain constant over sweep, the number of notches produced by full-bandwidth time delay increases, and start lower downin the spectrum, as the delay gets longer.  The resulting phenomenon is that the number of notches you hear increases as the pedal moves "downward" in its sweep cycle.  Typically, if the minimum attainable delay time is short enough, many of those notches will be inaudible because they are too high up.  One of the consequences of this for flangers is that as the sweep starts to move downward, the number of audible notches increases dramatically, both as more notches are produced and as more become audible.  The signal starts to get "infected" with notches, and then "cured" as the sweep starts to move upward again (towards a shorter delay).

With chorusses, the minimum delay time (and we'll say it is in the 5msec ballpark, though it varies from model to model) is generally long enough that there IS no point where the notches are inaudible, and the contrast between least number and greatest number of ntoches is small.  Moreover, where the change in time delay for a flanger, especially when slow, produces little discernible change in pitch in the delayed signal, the generally faster sweep (LFO rate) of a chorus, and longer delay time, produces very noticeable pitch change in the delayed signal.  Indeed, it is the pitch change one is aiming for in chorus pedals, since that is what helps to create the illusion of two "copies" of the instrument/voice trying hard to track each other but not quite succeeding.  One can often mimic some aspects of a chorus using a flanger by setting the initial delay time to near max and setting the LFO faster.

One of the big differences between chorus and flange in terms of the consequences for the listener, however, is the audibility of the time delay.  In the case of flanging, the delay time is very short, and the sweep long and slow, so the listener's attention tends to be directed towards the notches.  In the case of chorus, the sweep rate is faster, and the pitch change more audible.  The delay time is also audible enough that attention is turned away from the notches and towards the time-staggering and pitch-staggering of the two copies.

So, in a sense, even though flangers and chorusses are very close in terms of technology, the issue of the half-dozen milliseconds difference in the delay range they cover redirects human attention so that different sorts of effects are more noticeable and noticed by the listener.

[H4T]

Wow, thanks for that great post, very thorough, I learned a lot from that.

Seems as though deciding one which of these pedals is "right" for me will be rather tough. I understand the differences in these pedals now, but I think I'm going to need to go to a music store and play a delay and play a phaser and see which I like better, lol.

One quick question, I read this throughout the Geofex article, just want to clarify. When we're talking about notches, are notches simply areas in the signal where an inverted, shifted signal cancels out with the regular signal, causing no signal for a brief moment?

[puretube]

just & only for that 1 frequency (at which the notch(-es) occur(-)s at that moment)

[Mark Hammer]

Which you like best?  WHICH you like best?!  WHICH YOU LIKE BEST?

You need them BOTH, my friend! :lol:

[H4T]

Heh. Well I'll try em both and see how it works for me lol.

The delays and phasers still seem very similar to me. Here's a crazy question, in the Pink Floyd song "Money," is the cool warbble effect a delay pedal or a phaser? They both seem so similar lol.

[psw]

Money = Tremolo I believe...different again!!

Here's a question though...

I thought chorus' and flangers were time based effects and phasers were entirely different beasts...could be wrong...

I'd like to know how the phase notches are created in a phaser...I understand that there are cancellations of frequencies but how are they produced...(is it a time delay?)...is it some kind of filtering effect..

Also, how are these notches created electronically...is there a simple example of a method by which to change a signal's phase manually and/or at different frequencies...and how is this achieved.

interesting topic...nice replys too :wink: ...p

[puretube]

WHY not click the 2 linx in the 4th post of this thread?

or take this 1: http://diystompboxes.com/sboxforum/viewtopic.php?t=28936&highlight=phasers

(yes, needs another click in the 1st post, too...)