"Perfect phase (coherence)" in a loudspeaker, just what is this?

spiritofmusic

Well-Known Member
Jun 13, 2013
14,601
5,411
1,278
E. England
I ask because the undoubted star of the AE Show this year was the Göbel Divin spkrs, and its main claim to fame is by far the greatest phase coherence over the uber competition
Apparently the most revered engineer of the AE world has lauded the Divin as exemplary here
Just what is this phase coherence?
 

RogerD

VIP/Donor
May 23, 2010
3,734
318
565
BiggestLittleCity
Did the speakers disappear? They should and also image like crazy, that's my experience.

I'm cheating.....

The importance of phase response in the audio chain has been brought to greater focus recently by equipment claims of phase coherency, (the output signal has the same phase relationships as the input signal). It is not particularly obvious that two different frequency components of a signal can go into a device at precisely the same time and emerge at different times, but it is extremely common. All audio components distort the phase of the signal to some degree-even air alters the time alignment of a signal, but the biggest offenders are loudspeakers and their crossover networks. Phase shifts in the audio signal destroy the wave shape of the important attack characteristics of many instruments and hamper our ability to perceive the localization of the image,smearing the apparent source. They can change the steady state waveforms of vocal sounds so that the singer seems to be ten feet wide.

Historically, the phase integrity of the audio signal has been considered much less important than amplitude and harmonic/intermodulation distortions, but as more of those problems are solved and the quality of reproduction improves, phase distortion stands out in greater relief. The question of the audibility of these distortions has become the object of heated discussions regarding the perceivability of absolute phase, frequency dependent phase shifts, and the rate of phase shift. Nonetheless we know that the ear is sensitive to phase and uses phase cues to help determine directionality. In the belief that proper attention to phase does produce better sound, I will discuss the design of crossover networks which ever minimum phase distortion.

https://www.passlabs.com/press/phase-coherent-crossover-networks
 

spiritofmusic

Well-Known Member
Jun 13, 2013
14,601
5,411
1,278
E. England
Thanks Roger
The report on Divin talked about a truly realistic rendering of the stage re a classical recording
It must be Divin exemplary phase allowing as optimal imaging as poss
Sounds like it may be a ground breaking product
An owner of Wilson Alexandria at the show said Divin was clearly superior
At c$300k, a veritable bargain in uber high end terms
 
Last edited:

DaveC

Industry Expert
Nov 16, 2014
3,899
2,141
495
I've calculated, measured, experimented with this. Time and phase are also related, they are nearly the same thing but it's possible to be in phase but not time-coherent. An example is a 4th order xo with 360 degree phase shift, you're still in phase but not time aligned anymore.

I feel this subject, like many in speaker design, is extremely murky as most won't share much and it ends up being a marketing term first and foremost. But it's not BS, it's very important in some circumstances and not others. Whether it matters depends on the speaker design and frequencies in question. For example think about a speaker with a supertweeter... wavelengths at high frequencies are short so moving one's head an inch would throw off any sort of phase/time coherence. And if you experiment with it you'll find it the position of the supertweeter can make a difference but it's not nearly as dramatic as, say, a normal tweeter crossed much lower that's not in phase, there you can easily have measurable interference. At the other end we often see sculpted front baffles arranging the drivers in an arc to keep them equidistant from your ears, but that doesn't work for all listening heights, only one. Tilting the speaker also destroys the spatial relationship. So for many speakers the correct time/phase may only exist at one location in space and if it's so important people would notice that, hey, my speakers only sound ok with my head clamped in this one position, but with rare exceptions of speakers with super-narrow dispersion patterns it's just not the case and if it is it's dispersion and not phase.

Also, this isn't anything new at all, we've been aware of the subject for decades. But there IS much debate as it's not always an absolute thing, in some cases and with some designs and some frequency ranges it matters a lot, others not so much. I've heard lots of different opinions on the subject from excellent speaker designers. It's an interesting subject but there's more important things and you still have to listen to the speaker. For a consumer and non-speaker engineer I feel it's use in marketing is often misleading and just causing confusion.
 

RogerD

VIP/Donor
May 23, 2010
3,734
318
565
BiggestLittleCity
Thanks Roger
The report on Divin talked about a truly realistic rendering of the stage re a classical recording
It must be Divin exemplary phase allowing as optimal imaging as poss
Sounds like it may be a ground breaking product
An owner of Wilson Alexandria at the show said Divin was clearly superior
At c$300k, a veritable bargain I'm uber high end terms

Hi Marc,

Those Divins are huge http://www.monoandstereo.com/2017/07/ae-5th-super-high-end-show-2017.html#more

Well I speak from experience. The speakers that I purchased in the early 1980's are very coherent,disappear and do image like crazy. The crossovers were done using computer modeling and I thought they were exceptional and still do. The Divins are very big,but using only four drivers, I can't understand why this level of performance is not more common in this day and age. The good thing is the bar is pushed higher and may be the price may move lower....competition in high end is good,though 300K is a tough nut to crack...and even 50 off...not for me. YMMV
 

Folsom

VIP/Donor
Oct 25, 2015
6,024
1,490
520
Eastern WA
They probably are nice speakers. But are they so nice because of phase? NO. It's marketing. The phase of a single driver changes through the spectrum of frequencies it plays... if anything I believe most manufacturers have simply misunderstood other qualities for phase. Why? Because typical ways of trying to adjust phase mean taking acts like removing parts that cause other complications that are not commonly dealt with. It's more like they have better tuned the drivers than achieved anything special with phase. DIY pro speaker builders have been able to do phase and time alignment, so much so they don't necessarily ultimately value either as they continue projects with and without - that's people that actually know more than most smaller manufacturers of speakers.

If we were ever to see 100% correct phase speaker (including driver data) it would not and could not look like a normal dynamic driver speaker - ever.
 

Mark Seaton

WBF Technical Expert (Speaker & Acoustics)
May 21, 2010
381
141
390
47
Chicago, IL
www.seatonsound.net
I've calculated, measured, experimented with this. Time and phase are also related, they are nearly the same thing but it's possible to be in phase but not time-coherent. An example is a 4th order xo with 360 degree phase shift, you're still in phase but not time aligned anymore.

I feel this subject, like many in speaker design, is extremely murky as most won't share much and it ends up being a marketing term first and foremost. But it's not BS, it's very important in some circumstances and not others. Whether it matters depends on the speaker design and frequencies in question. For example think about a speaker with a supertweeter... wavelengths at high frequencies are short so moving one's head an inch would throw off any sort of phase/time coherence. And if you experiment with it you'll find it the position of the supertweeter can make a difference but it's not nearly as dramatic as, say, a normal tweeter crossed much lower that's not in phase, there you can easily have measurable interference. At the other end we often see sculpted front baffles arranging the drivers in an arc to keep them equidistant from your ears, but that doesn't work for all listening heights, only one. Tilting the speaker also destroys the spatial relationship. So for many speakers the correct time/phase may only exist at one location in space and if it's so important people would notice that, hey, my speakers only sound ok with my head clamped in this one position, but with rare exceptions of speakers with super-narrow dispersion patterns it's just not the case and if it is it's dispersion and not phase.

Also, this isn't anything new at all, we've been aware of the subject for decades. But there IS much debate as it's not always an absolute thing, in some cases and with some designs and some frequency ranges it matters a lot, others not so much. I've heard lots of different opinions on the subject from excellent speaker designers. It's an interesting subject but there's more important things and you still have to listen to the speaker. For a consumer and non-speaker engineer I feel it's use in marketing is often misleading and just causing confusion.

Well stated in pointing out that it's mostly a marketing term that each company tends to define in their own way. I would point out that more than a few cases are BS with little bearing on anything you could quantify with a microphone. While a microphone isn't going to give you a single measurement to define the sound of a speaker, it most certainly can quantify things like phase, group delay, and impulse response. In most cases, the more general the use of the term without qualifiers, the more likely it has little technical significance.

In the current mix of speaker manufacturers we have such a huge range of deep technical knowledge and test equipment on to the blissful ignorance of a skilled cabinet maker/artist who still does crossover design by plug-n-play until it sounds good. Remembering the Dunning–Kruger effect, it's important to not confuse the confidence in which something is claimed with it's actual validity.
 

853guy

Active Member
Aug 14, 2013
1,161
10
38
Hi Spirit,

Hope you don’t mind but decided to reply to you here since it seemed more on-topic. In the Audio Exotics Super Hi End Show thread you posted:

spiritofmusic said:
Stereophonic, all I know is that (the Divin is) claiming perfect phase coherence which someone has also said is only possible w DSP, and this has a purely analog crossover

My understanding is limited, so take this with a shaker of salt:

Group Delay is a measurement of time (not phase) in which frequency components of a square wave (can) become misaligned relative to one another. A filter (like a crossover) that passes a signal (square wave) in which misalignment is minimised can be considered constant, hence the term “constant group delay”, i.e. all frequencies arrive in time relative to one another, but delayed in time.

Phase Delay however, is a measurement of phase (duh) in which the frequency components of a square wave shift out of phase relative to their initial state. A filter that passes the above with minimal phase shift at all frequencies can be considered phase linear.

Without sounding like a shill (too late?), the Aries Cerat Symphonia’s outboard analog crossover is both time and phase linear relative to the mechanical alignment of its drivers. A speaker featuring a constant group delay filter (the Symphonia has three) also needs to be designed relative to the physical alignment of each of the drivers and vice versa with respect to both time and phase. In and of itself, physically aligning the drivers in space relative to one another achieves nothing other than a nice looking speaker. Fundamentally, physical alignment, time alignment and phase alignment are all interdependent variables, and must be considered together as a whole. That it also achieves a true 101dB sensitivity rating and has a maximum deviation of +/- 0.5dB over six and a bit octaves is kinda impressive (to me). At the very least, I don’t know of any other horn that achieves this, but am open to being wrong, which remains a constant possibility*.

It’s important to note that “linear” is a relative term apropos the real world (i.e. we’re talking about the “least non-linear”), but nevertheless it is indeed possible to build an analog filter that is both time and phase linear, like, for instance, in a Bessel filter (though I don’t know if that’s what AC use in the Symphonia - perhaps Flyer or Stavros can chime in).

“Perfect”, however, is an altogether different thing.

Best,

853guy

*If someone wants to school me on how woefully lacking my understanding may be, please go ahead and begin the lesson. I am but a humble student.
 
Last edited:

DaveC

Industry Expert
Nov 16, 2014
3,899
2,141
495
Hi Spirit,

Hope you don’t mind but decided to reply to you here since it seemed more on-topic. In the Audio Exotics Super Hi End Show thread you posted:



My understanding is limited, so take this with a shaker of salt:

Group Delay is a measurement of time (not phase) in which frequency components of a square wave (can) become misaligned relative to one another. A filter (like a crossover) that passes a signal (square wave) in which misalignment is minimised can be considered constant, hence the term “constant group delay”, i.e. all frequencies arrive in time relative to one another, but delayed in time.

Phase Delay however, is a measurement of phase (duh) in which the frequency components of a square wave shift out of phase relative to their initial state. A filter that passes the above with minimal phase shift at all frequencies can be considered phase linear.

Without sounding like a shill (too late?), the Aries Cerat Symphonia’s outboard analog crossover is both time and phase linear relative to the mechanical alignment of its drivers. A speaker featuring a constant group delay filter (the Symphonia has three) also needs to be designed relative to the physical alignment of each of the drivers and vice versa with respect to both time and phase. In and of itself, physically aligning the drivers in space relative to one another achieves nothing other than a nice looking speaker. Fundamentally, physical alignment, time alignment and phase alignment are all interdependent variables, and must be considered together as a whole. That it also achieves a true 101dB sensitivity rating and has a maximum deviation of +/- 0.5dB over six and a bit octaves is kinda impressive (to me). At the very least, I don’t know of any other horn that achieves this, but am open to being wrong, which remains a constant possibility*.

It’s important to note that “linear” is a relative term apropos the real world (i.e. we’re talking about the “least non-linear”), but nevertheless it is indeed possible to build an analog filter that is both time and phase linear, like, for instance, in a Bessel filter (though I don’t know if that’s what AC use in the Symphonia - perhaps Flyer or Stavros can chime in).

“Perfect”, however, is an altogether different thing.

Best,

853guy

*If someone wants to school me on how woefully lacking my understanding may be, please go ahead and begin the lesson. I am but a humble student.


Group delay can be expressed as a phase shift wrt frequency.

Honestly, I'd take any speaker manufacturer's claims with a grain of salt. The best thing to look at is a polar plot, it'll tell you near everything that matters that we can measure. Interestingly, very few provide this info and instead claim all sorts of stuff that may or may not make a difference. It's somewhat understandable as uninformed judgements are often made by folks who have no idea what they're talking about as a result of sharing measurements, so I can understand not sharing measurements, but all the other claims of perfect this or that, exotic construction materials, etc. should mostly be ignored imo.
 

853guy

Active Member
Aug 14, 2013
1,161
10
38
Group delay can be expressed as a phase shift wrt frequency.

Honestly, I'd take any speaker manufacturer's claims with a grain of salt. The best thing to look at is a polar plot, it'll tell you near everything that matters that we can measure. Interestingly, very few provide this info and instead claim all sorts of stuff that may or may not make a difference. It's somewhat understandable as uninformed judgements are often made by folks who have no idea what they're talking about as a result of sharing measurements, so I can understand not sharing measurements, but all the other claims of perfect this or that, exotic construction materials, etc. should mostly be ignored imo.

Hi Dave,

I agree, marketing BS is marketing BS, and should be called out for what it is.

In the case of the Symphonia however, having heard it twice - one of which was a private demo, I have to consider the possibility the crossover’s minimisation of time/phase anomalies and the physical alignment of each driver may have been a contributing factor to what I heard, despite the fact I can neither quantify their significance nor verify the claims objectively. So while it’s true all claims need to be taken with a grain of salt (mine included), at some stage the rhetoric needs to be considered in light of listening (despite the limits of human hearing and the subject’s perception) and an attempt made at correlation between what can be theorised and what can be heard (and vice versa), no matter how limited the value of the observation may be. Like I say above, so far, the Symphonia’s the only horn I know of that’s made claims for the use of three constant group delay filters in which the phase derivative is both constant and linear - so while my impressions limited to my perception are no more than that, they form at the very least a data point for consideration and potentially, a topic for further discussion.

However, that I can only offer a subjective take and offer my limited understanding of phase anomalies will not (and should not) be enough for anyone to unconditionally agree with the claims made for the Symphonia, but like you say, even were Aries Cerat to post measurements here, there would be those who would still doubt their veracity, and possibly more so in cases in which they made no effort to correlate those measurements with subjective listening. That is, I guess it would appear misinformed judgements are just as easy to make whether one wants to make those judgements solely via the experience of listening, or solely via the experience of analysing. Perhaps it’s true we hear what we want to hear and see what we want to see.

Nevertheless, discounting the value of any single variable and whether it may or may not make a difference to our perception of music when played through the device in question before hearing it is just as much a folly as it is to apply retroactive justification after hearing it. That I may be doing the latter shouldn’t prevent us from trying to understand how our hearing correlates with the theory, nor whether the human hearing mechanism’s limits can be defined without consideration of the neurobiological processes of the brain to those same stimuli. I don’t know the degree to which the claims made for the Symphonia bear out in real-life. All I know is I’ve never been as emotionally and intellectually invested when listening to music before nor since, despite having heard speakers whose claims of technological superiority (and price of admission) dwarf those of the Symphonia.

In any case, that I’ve reached my abilities to articulate what I experienced with what I “know” should be obvious. I leave the thread now to those whose grasp of such things exceeds mine.

Be well, Dave,

853guy
 
Last edited:

fish fingers

Well-Known Member
May 11, 2015
25
0
106
Alot of high end companies use 1st order crossovers to claim the above.. but that brings its own set of problems.. stressed drivers and distortion.
 

wisnon

Well-Known Member
Dec 12, 2011
3,525
635
1,200
Hi Spirit,

Hope you don’t mind but decided to reply to you here since it seemed more on-topic. In the Audio Exotics Super Hi End Show thread you posted:



My understanding is limited, so take this with a shaker of salt:

Group Delay is a measurement of time (not phase) in which frequency components of a square wave (can) become misaligned relative to one another. A filter (like a crossover) that passes a signal (square wave) in which misalignment is minimised can be considered constant, hence the term “constant group delay”, i.e. all frequencies arrive in time relative to one another, but delayed in time.

Phase Delay however, is a measurement of phase (duh) in which the frequency components of a square wave shift out of phase relative to their initial state. A filter that passes the above with minimal phase shift at all frequencies can be considered phase linear.

Without sounding like a shill (too late?), the Aries Cerat Symphonia’s outboard analog crossover is both time and phase linear relative to the mechanical alignment of its drivers. A speaker featuring a constant group delay filter (the Symphonia has three) also needs to be designed relative to the physical alignment of each of the drivers and vice versa with respect to both time and phase. In and of itself, physically aligning the drivers in space relative to one another achieves nothing other than a nice looking speaker. Fundamentally, physical alignment, time alignment and phase alignment are all interdependent variables, and must be considered together as a whole. That it also achieves a true 101dB sensitivity rating and has a maximum deviation of +/- 0.5dB over six and a bit octaves is kinda impressive (to me). At the very least, I don’t know of any other horn that achieves this, but am open to being wrong, which remains a constant possibility*.

It’s important to note that “linear” is a relative term apropos the real world (i.e. we’re talking about the “least non-linear”), but nevertheless it is indeed possible to build an analog filter that is both time and phase linear, like, for instance, in a Bessel filter (though I don’t know if that’s what AC use in the Symphonia - perhaps Flyer or Stavros can chime in).

“Perfect”, however, is an altogether different thing.

Best,

853guy

*If someone wants to school me on how woefully lacking my understanding may be, please go ahead and begin the lesson. I am but a humble student.
Doesn't Goldmund do all this with their Proteus/Leonardo DSP/DRC software and ultra wide bandwidth amplifiers (all built in into the active speakers)?
http://www.goldmund.com/en/technologies/the-leonardo-time-correction
http://www.goldmund.com/en/technologies/proteus-leonardo
http://attachments.goldmund.com.s3....mplifier_acoustical_grounding_white_paper.pdf
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing