Phase Coherent

[dondd]

Hi, I'm new here, (and to the "high-end" audio scene as well) & I had a question regarding Klipsch speakers. From my research, looking at sites like Meadowlarkaudio.com, thielaudio.com, and a few online audio publications, it seems as though the use of a slanted baffle (to align the acoustic centers of the drivers), and a 1st order crossover are necessary to produce truely phase coherent sound.

I have been looking at the Klipsch KLF-30s or -20s and I was wondering 2 things:

1) Is the information that I have gathered so far accurate?

2) Does Klipsch employ these techniques in the construction of their speakers?

Thanks for any input you can give.

[pzannucci]

First off, acoustic alignment only occurs at one place in space. Move your head and alignment will change so hence part of your answer. Secondly, you don't need 6db per octave crossovers to stay phase coherent. The price you typically pay for 6db per octave crossovers is distortion from pushing the drivers the extremes due to the shallow slope. To keep this from occuring, you would probably want a 4-way+ speaker.

Of course, every company and designer has their own thoughts on this.

As far as the time alignment on Klipsch speakers, the horns make it very difficult but that doesn't mean that Klipsch speakers don't produce live sound.

You have to listen for yourself and decide what tradeoffs you can handle and what sound you like.

Peter Z.

[dougdrake]

I'm with Peter on his comment that you need to listen for yourself. I do not consider myself an audiophile in the sense that I would be able to use "wine terms" to describe sound ("round and perky bass" as an example). However, I know some people say that they can hear the difference between regular 12-gauge speaker wire and 12-gauge, electron-aligned, oxygen-free, organically smelted copper. (I could not.) So, I guess I'm just suggesting that you take all things audiophile in the context that some of it is pretty arcane and esoteric and may or may not make a difference to how things sound to YOU. Ear-dition everything, if you can. If you can hear a difference, and it's worth the $$ for it, then get it!!

[pzannucci]

Bill,

Why does Klipsch tell you the delay for each driver in their large horn systems if the end of the horns are time aligned?

Doesn't the speed of sound come in here?

Also, isn't a 12db per octave crossover really 180 degrees out of phase so that the drivers should be of reversed phase, and as PhilH put it to AL K., a direct radiating driver is 90 degrees out of phase from a horn loaded driver?

Isn't it the time/distance that the sound travels between each of the horns (and themselves) and the phase changes introduced in the networks really what might put them in back into alignment and not the mouths being on the same plane?

Please explain. I might have to re-think the speakers I am building.

Thanks,

Peter Z.

[pzannucci]

Oops, sorry. The only way to tell what you like is to listen to them.

I had just built a set of speakers that do a lot of things wrong (lots of compromises for a two-way) but they sound great. Technical merits of the pieces doesn't always amount to a wonderful finished end product. Listen...

Peter Z.

[discorules]

My understanding is delay is determined between the measurement instrument and the the actual drive mechanism in the horn. So unless there is some electrical delays, the drivers in a multi horn system is not time aligned.

I don't know the engineering behind much of the heritage line but the most crucial instrumnet are your ears. To mine, nothing else comes closer to real than the khorns. I've heard fine speakers from ML, Magnepan, Infinity, etc. with excellent performances. My ears tell me however that the Khorns are doing the right things when compared to live music.

[dondd]

Thanks for your replys, guys. I'm still slightly confused....

I think that alignment of the acoustic centers makes sense: sounds generated by a tweeter & woofer at the same time will only reach your ear at the same time if the acoustic centers of the tweeter & woofer are aligned. Does anyone know if the horns & woofers are acoustically aligned in the Klipsch Legend series?

Now, the crossover: If Peter was correct in his first post and you don't need a 1st order crossover in order to achieve a phase coherent speaker.....how is this done with higher order crossovers which (I have read) inherently introduce phase shifts?

Reason for all these questions: I have heard the difference between phase coherent & incoherent speakers, and the difference is huge.(phase coherent being better, in my opinion) Klipsch speakers have been given good reviews, and their brochure for the KLF-30s say they are phase coherent. However, the criteria that I have dug up about what makes a phase coherent speaker a phase coherent speaker (outlined in my first post) is not evident in the Klipsch literature. So I'm wondering how they are doing it.

I just want to know what I'm talking about before I go spending any money!!

[discorules]

Time delays can be introduced in a crossover network using a number of series inductors and shunt capacitors. I had the formulas for this stuff somewhere which give impedance, time delay and high frequency max.

[dondd]

quote:

---

Originally posted by mdeneen:

I'm trying to figure out what anybody (original poster?) means by "phase coherant" speaker?

I am not aware of any loudspeaker, regardless of the number of drivers or crossover elements, which maintains a "uniform" phase angle relative to the input signal - if that is what you mean by "coherant."

---

I am not aware of any specific definition. However, you are correct in your assumption of the premise. A "Phase coherent" speaker is one which maintains a uniform phase angle with respect to the input signal.(as you stated) Trying to accurately reproduce the input signal in both the frequency and time domains.

This is the whole design philosophy behind such speaker companies as: Meadowlark (see: design philosophy at www.meadowlarkaudio.com),

Thiel (see: www.theilaudio.com), and Vandersteen (see: www.vandersteen.com/pages/Cpny_info.htm)

Here are 2 articles on the subject as well:

1)"http://www.stereophile.com/showarchives.cgi?221" <- without quotes

2)"http://www.soundstage.com/interviews/int07.htm" <-without quotes

I'm interested to see what everyone thinks.

mdeneen: I hope this helps to clear things up.....a little.

[Klewless]

dondd,

Some additional comments here:

1. Tilting the baffle to move the tweeter back makes it closer to the woofer/midrange voice coil is the first step in time alignment (and consequently maybe phase coherency). But the xover plays a part in that too, so the actual angle (tweeter setback) is probably determined experimentally.

2. The xover plays a big part in the phase coherency, which is different from time alignment. So as noted by others here one must carefully state just which parameter is being questioned.

3. It seems to me that if you have a xover then by default you don't have perfect phase coherency. It just comes with the territory.

4. If you have more than one driver, ie. woofer/tweeter, then you do not and cannot have true phase coherency if you include a xover. A speaker is also an inductor of sorts, so again there goes the coherency. The magneplanars are probably about as phase coherent as possible (except for the xover they present a resistive load). But alas, they have a xover too. There goes the baby with the water again.

5. The Khorn and LaScala and Bell Klipsch are no where near time aligned. You hear the tweeter then the midrange and then the woofer. Paul has demonstrated many times that on music you cannot detect this alignment until the drivers are way, way apart. Over 50 years of success have shown this to be true.

6. Time alignment was a big deal with the Altec "Voice of the Theater" speakers and was achived by moving the woofer back to line up with the midrange/tweeter driver.

7. Fast xovers are required by horn midranges and horn tweeters. Slow xovers simply do not protect their drivers sufficiently. Hence good xover design is a must for horn systems.

I am sure you have discovered by now that your question was not a simple one. Just goes to show how valuable this board is! I hope this adds to the discussion. There is a lot of history related to this subject.

I too must add that listening to the speaker is the only way to tell if you like it.

------------------

John P

St Paul, MN

[pzannucci]

dondd,

What many manufacturers refer to as phase coherent is that all the drivers are hooked up in phase with each other, and that is about where the similarities end. Some may speak for electrical phase and some may speak for time phase. These will be two different things and depending on the marketting hype, relate little to the overall sound of the speaker because there are too many other factors that enter into the equation.

If you look at many of the sights that sell drivers, you will find a phase chart on an individual driver because even one driver is not perfect. They all have inductance and capacitance to a degree which shifts phase around. The key is, how subtle or irratic are the phase changes and can the ear perceive them.

Bill,

I understand what you are saying about an acoustic transformer but I'm not quite sure it is 100% true in practice. Perhaps PhilH or BobG can update us. I've owned a couple pairs of K-horns but I don't think they were anywhere near time aligned. But again, as you said, they can be wonderful sounding speakers. That's why you have to listen to the speakers and not listen to the marketing hype. Remember, almost all high end manufacturers snub their noses at horns but the Advantgarde horns are considered reference level speakers.

Peter Z.

[jdm56]

dondd,

I don't know if I can add anything new here, but I just wanted to get my 2 cents worth in.

No, klipsch speakers for the most part are not phase coherant. Very few speakers are. Only Dunlavy and Thiel come to mind, though there probably are some more. But the audibility of time/phase alignment is very questionable. The phase aligned speakers you heard and liked so well may have sounded good due to other aspects of design. (frequency response, etc)

The fact of the matter is phase and time aligning of horn speakers is not very practical. Avant Garde does stagger their drivers to achieve a more phase coherant wave launch, but I don't know if their x/o's are phase coherant or not.

And as someone has already pointed out, even with the right crossovers and with physically staggered drivers, true phase coherance will only be achieved at one particular point in space; Phase coherance being the simultaneous arrival of all frequencys at the listeners ears.

In fact, the klipschorn, la scala and belle klipsch are probably three of the least phase and time coherant speakers on the planet, yet in my humble opinion, they are three of the greatest speakers on the planet. So much for phase and time alignment.

I would recommend not worrying too much about performance parameters of dubious audible consequence, and just listen for the speaker that sets your foot to tappin' and puts a big smile on your face -- the one that makes you feel good. Chances are it'll be a klipsch!

------------------

JDMcCall

[Ray Garrison]

There is a very specific definition of what "time coherent" or "phase aligned" means. To quote John Atkinson:

"Does a loudspeaker's time coherence matter? A "perfect" speaker, of course, would have both a perfect impulse response and a perfect frequency response (at one point in space). Another way of looking at a loudspeaker's time-domain performance is to examine its acoustic phase response, the phase angle between the pressure and velocity components of the sound plotted against frequency.

Again, this is an aspect of loudspeaker behavior that has proved controversial. One school of thought holds that it is very important to perceived quality; another, which includes almost all loudspeaker engineers, finds it unimportant. Floyd Toole, now with Harman International but then with Canada's National Research Council, in his summary of research at the NRC into loudspeaker performance that is described in two classic 1986 papers <32, 33>, concluded thusly: "The advocates of accurate waveform reproduction, implying both accurate amplitude and phase responses, are in a particularly awkward situation. In spite of the considerable engineering appeal of this concept, practical tests have yielded little evidence of listener sensitivity to this factor...the limited results lend support for the popular view that the effects of phase are clearly subordinate to amplitude response."

This is also my view. Of the 350 or so loudspeakers I have measured, there is no correlation between whether or not they are time-coherent and whether or not they are recommended by a Stereophile reviewer. However, I feel that if other factors have been optimizedon-axis response, off-axis dispersion, absence of resonance-related problems, and good linearitylike a little bit of chicken soup, time coherence (hence minimal acoustic phase error) cannot hurt. In my admittedly anecdotal experience, a speaker that is time-coherent (on the listening axis) does have a small edge when it comes to presenting a stereo soundstage, in terms of image focus and image depth. But time coherence does not compensate for coloration, poor presentation of instrumental timbres, a perverse frequency balance, or high levels of nonlinear distortion.

In 1990, Rodney Greenfield and Malcolm Omar Hawksford <34> used DSP-based digital filters to try to separate the audible effects of a loudspeaker's phase error from its amplitude response error. The point was made that a semi-reverberant environment will tend to mask phase effects. In addition, when typical recordings are played, which may have undergone many phase-altering stages during production, the audibility of phase differences becomes moot: "one is simply detecting a change in phase distortion and not a correction of it and as such preferences would most likely be personal." Nevertheless, the authors "very tentatively" concluded that equalizing a loudspeaker's excess phase error modified listeners' perception of the apparent soundstage.

It is important to note that there are phase responses and phase responses. Not only is phase error associated with a lack of time coherence, phase error is introduced by any departure from a flat amplitude response. The phase response of what is called a "minimum-phase" system is related to the amplitude response by a mathematical function called the Hilbert Transform. So, for any nonflat system like a loudspeaker, it is important to distinguish between the two sources of phase error. The MLSSA system allows its operator to subtract the calculated minimum-phase, amplitude error-related phase response from the measured acoustic phase response. The result is what is called the "excess" phase response mentioned above."

To see the whole article wence this came, go here:

http://www.stereophile.com/fullarchives.cgi?100

Ray

------------------

Music is art

Audio is engineering

[bill rawls]

Atta Boy Ray,

Nice reference. I wonder how many people are taking their tweeters out and placing them on top of the enclosure (elevated of course) and aligning the drivers with the squawker? I am thinking about it now and will probably have to try it just to hear the difference. If anyone has already done this, come clean and tell all. That tweeter sure is way up in there.

Screwdriver in hand,

Bill Rawls

[discorules]

This sure has been entertaining and informative. I too bought into time alignment in the late seventies and experimented with my tweeter and midrange by repositioning them. I did notice effects that were spatially related but I wasn't sure if the overall response was more accurate. As was stated previously, there are so many factors that can change phase and time alignment starting from miking, I'm not sure how beneficial an exercise it is to make phase alignment the key aspect in speaker design, esp. if this alignment can only be realized at one point in space in an anechoic environment. I don't listen to music in that environment.

[SamP]

To me phase coherence can be illustrated by listening to quick, short duration, impulsive types of sound like a drum stick being tapped on the rim. Any two drivers reproducing the same sound may interfere with each other in a destructive(180 out of phase) or constructive(in phase) manner. If you take any two identical wave forms, and superimpose them perfectly two things will occur...the output will increase, and the wave shape will remain narrow. Let the phase difference between them increase, and the "peak" will be reduced and the base will spread out. At 180 degrees they cancel completely. Transient signals by their nature are short duration, and lack of coherence will "smear" the audible output of the speaker system. At the 10 khz point where my K-77m's come in to bolster the falling high end of my Altec drivers, swapping the polarity of the tweeter OR moving the tweeter baffle forward/back about 0.68 inches made an easily detectable difference in the character of this type of sound. What you want is the peaks and troughs to be in phase as much as possible in the region where the two drivers are operating "together",i.e. at the crossover frequency, to prevent destructive interference from "smearing" the transients. Obviously any phase shifts caused by crossover components are going to effect the phase of the signals being passed, but by physically moving one driver in relation to the other and listening closely, a sharper "focus" to tics(drum sticks) and pings(chimes) will be heard.

[SamP]

Hate to state the obvious, but at 350 hz the mid output level will be 12dB less than at 700 hz. The woofer output will be hotter. Unless the drivers are outputing equal levels at the same freq, total cancelation CANNOT occur. Same at 1500 hz. The woofer should be downn 12 dB, aand the mid hotter. PHYSICAL separation on the baffle is a reality we multiple driver system owners have to live with. My concern was the sound wave as it travels forward in space from the driver to the listener. Specifically at 10khz, where my tweet comes on line to bolster the falling high end response of my mid. That mid is playing along at a 10 khz. sine wave for example. If the drivers are not brought into acoustic phase, the two signals will interfere, either constuctively or destructively. Look at two "in phase" sine waves on paper. "ADD" them. The peak gets larger, but the base stays narrow. Now shift the phase on one wave, and "add" them. The peak is lower and the base is broader. That's what happens to transient signals, they get "smeared" by poor ACOUSTIC PHASE conditions. Also note the "new" waveform riding on the original wave. Can you spell heterodyne? I know that sound bounces all around the room, there are baffle interactions, etc. It doesn't matter that the tweeter is not physically aligned with the mid driver, just their acoustic phase must be in sync to prevent the interference. This is the same kind of "lobing" in freq resonse caused by drivers being mounted far apart on baffles. Yes it is room position sensitive. My point was to consider the WORST CASE lobe situation, which is happily the only one easily corrected.

[pzannucci]

Bill,

We keep each other straight and learn from each other. You do have to think about it though....

I'm with SamP. You had better have acoustic alignment through the crossover range (as much as possible with mechanical and electronic effects) or your sound will not be as articulate and crisp. Once you are out of the crossover band though, all bets are probably off. The only problem is that -- Can you hear the difference with a couple of microseconds between drivers and is the wavelength small or large enough at those frequencies to notice smearing or loss of focus due to the two arrivals mixing?

One can just strive for perfection but when it's good enough, put a bow on it.

Peter Z.

[oosting]

Smearing?

If you take two 1000 Hz sine waves of the same magnitude (peak voltage or SPL), one at 0 degrees and the second 120 degrees out of phase and combine the two, you end up with a 1000 Hz sine wave of the same magnitude as the originals but 60 degrees out of phase. The base doesn't broaden, it stays exactly the same. Otherwise, you would have a frequency shift along with amplitude and phase shifts. That is impossible.

When you have multiple sources of the same frequency, you always have constructive and destructive interference, even in an anechoic chamber. This is one of the penalties of stereo. If you have two perfect point sources, you still get interference with stereo (and other multi-channel sound reproduction).

The simple fact is that other factors far outweigh phase problems in the overall listening enjoyment or we would all stick to headphones where there is NO interference.

Ed

[WMcD]

There has never been a question about it. The drivers in the K-Horn are located very much distant from each other. As a result, the sound from the mid arrives later than that from the tweeter. The sound from the woofer arrives later than the mid.

The Heyser / Audio review shows this without question.

Within the limits of the K-Horn design, humans just can't detect the differences in arrival times. Or at least the studies show.

You should consider the structure of the inner ear. The little hairs respond to different frequencies. Somewhat like the display on a real time analyzer. So the brain would have to interpret the timing of the firing of the say, treble hairs, and compare them to the time of the firing of the midrange hairs, and the bass hairs.

Evidently, our brains don't do that very well unless there are large delays.

Phase is a term used very loosely. Sometime, but not always, it is used to measure a time difference.

PWK is very careful about this when he writes. E.g. you can wire one of two drivers in reverse polarity, and all reproduced sound is 180 degrees "out of phase" but not delayed. On the other hand, a delay can put sound at a single frequency 180 degrees out of phase because of the delay. However, at twice the frequency, the delay causes a 360 degree shift. But the resulting combination is "in phase" at that frequency despite the delay.

It takes a lot of diagrams to show this.

There are examples where "phase" is actually used to indicate a dirivitive, or slope. Like in calculus.

The current and voltage in a pure inductor or a pure capacitor are said to be 90 degrees out of phase. This is really to say that when one is a sine wave, the other is a cosine wave, one being the "slope" or dirivitive of the other. The slope of a sine wave is the cosine wave.

Some smart cookie realized that the sine and cosine waves are 90 degrees apart. Therefore they talk about current and voltage being 90 degrees out of phase. However that does not mean there is a delay in time of 1/4 wavelength.

Gil