I think so, but only if in combination with 1st order crossovers as used by Thiel and Vandersteen, revealing a coherent time-aligned step response.
218 responses Add your response
I recall seeing the terms "time coherent" and "phase coherent." Do these terms mean the same thing or are they different? Since this thread speaks to speaker alignment, perhaps someone could explain in layman's terms what causes speaker to operate out of phase. Does it have something to do with the use of caps and chokes in the x-over? Or perhaps the attribute of a dynamic speaker creating its own back EMF by reason of the voice coil moving in a magnetic field?? Incidentally, do all these electrical dynamics operating in tandem cause the electrical phase shifting that gives most amps a headache? Thanks |
Mofimadness answered correctly About time alignment.As far as phasing... it is simply having your individual drivers operating In unison at all frequencies... As you change crossover points and slopes, phasing changes between drivers.... Correct time and phase alignment is what seperates many ok speakers From world class speakers... These factors are why so many audiophiles like sigle drivers or point source.... I hope this helps, Tim |
Bifwynne, your entire post is a LOADED question! :-) we have discussed this way back in 2002. Here is the link to that thread (many very good & informative & technical posts by Roy Johnson of Green Mtn Audio speakers): http://forum.audiogon.com/cgi-bin/fr.pl?cspkr&1032037028 And here is a thread that Gmood1 started on Time & Phase as related to speaker operation: http://forum.audiogon.com/cgi-bin/fr.pl?cspkr&1060123575 And, finally, member Rbischoff (who started the 1st thread, above, that I provided the link) started another thread on First Order Cross-overs: Pros & Cons: http://forum.audiogon.com/cgi-bin/fr.pl?cspkr&1113252618 There is a great deal of really good info on this subject of phase & time coherency. Happy reading.... ;-) |
Bombaywalla, ... that;s for the links. I started reading into the threads. This is very complicated stuff. And I thought getting my arms around amp and speaker compatibility was tough. Speaker design is a beast. I will return to these threads over the weekend. In the end, as it seems to be the case with end end electronic gear, we're talking about design compromises. No perfect solutions ... just trade-offs that hopefully result in a product that sounds good. Thanks |
Bifwynne, welcome. I figured that these links would NOT be wasted on you. :-) indeed, speaker design is a complicated affair. Fully understanding what's going on takes skill, knowledge & experience as speaker design involves the mechanical aspect of actually building the speaker but also understanding what the electrical effect of the x-over is to the mechanical aspect. It's an inter-disciplinary expertise. In those links, there's a lot of chatter by members who don't know what they are talking about & by those who are airing their thoughts & experiences. You can skip over that which will shorten the reading time. Pay closer attention to the speaker manuf posts. Those are revealing in info. Well, now that you have the amp-speaker interface licked :-) it's time to go a bit downstream & understand what that electrical-to-sound transducer is all about... |
I think there is a lot of hype around this issue. A lot. If you have floor standing speakers and simply adjust the spikes so the front is slightly higher to accommodate your sitting position the you may have it. Is this just all a bunch of hype? Are there other more serious considerations? Cabinet construction, drive quality? Probably |
IMHO, the first link is perhaps the best thread ever to appear on audiogon, Roy Johnson of Green Mountain Audio was especially generous in sharing his thoughts. I've come to believe that some are more sensitive and/or prioritize these specific attributes more than others. Quite by accident, I've come to accept that these design principles are very important |
Bombaywalla ... I read as much of the discussions as I could before falling asleep at 1:30am in the morning. No surprise ... the physical science is way over my head. But I do have some take-a-ways that I would like to share. As a threshold matter, I ask whether we are "polishing a turd" as a practical matter. As a warm up, there was some discussion which addressed whether using mics to record a musical presentation accurately captured the complex sound wave information that emanated from the multiple performers and instruments. One poster said that all a mic could do was sample a point in 3-D space, thereby missing a considerable amount of sonic/acoustical information. Even using multiple mics, while an improvement, still left much sonic information "on the floor," figuratively speaking. Next ... the mastering process. So called sound engineers manipulate the sonic information that was imperfectly recorded at the live performance. The result is more sonic contortion and distortion. Let's skip the next steps relating to the reproduction and transmission of source material via the various media used today (e.g., LP, CD, SACD, internet downloads, etc) and the electronics used to decode the information back into analogue electrical signals that are fed into the speakers. Suffice to say that additional contortions and distortions are infused into the analogue signal before it even reaches the speakers. Ok, we're now at the speakers and a whole new set of issues present themselves. The bottom line challenge is that our speakers have to reproduce, more like reconstruct, the complex electrical analogue signals back into sound waves that are in phase through the whole harmonic spectrum. And here the engineering problems and challenges are almost insurmountable: designing a system using multiple drivers that are mounted on a baffle (sloped or not sloped) to reproduce a point complex source sonic wave front that is the same as the original signal, even as modified by the sound engineers at the studio. My take-a-way is that the speaker designer can solve one problem, but create 5 more. I don't want to embarrass myself by trying to faithfully restate what was said in the various posts. Suffice to say there are physical science challenges presented with every electrical and mechanical component that makes up a complete speaker system. At best ... what reaches our ears is the product of price point driven compromises that are ultimately limited by the laws of physical science. My bottom line take-a-way is if "it" sounds good, "it" is good. Sure, we can and should audition speakers. Some will sound better than others. But to think any one speaker has perfectly solved all the engineering challenges and is able to faithfully reproduce a point source complex sound wave at the listening point is a pipe dream ... more like shear nonsense. Perhaps that may be the reason many audiophiles and reviewers say that while specs and stats are informative ... in the end, what counts the most is how a speaker **subjectively** sounds to THAT listener, plugged into THAT listener's rig, positioned in THAT listener's sound room, and so forth. And to me, that is what makes our hobby fun. Right now, I am listening to Eugene Ormandy, conducting the Philly Orchestra, performing a suite of various Saint-Saens selections. I am still enjoying the music even though it is being imperfectly reproduced by my imperfect rig. Cheers, Bruce |
Bombaywalla, I have a follow up question. How are small speaker manufacturers able to design speakers without the benefit of the R&D budget, engineers, and testing facilities that some of the larger manufacturers have at their disposal. For example, I recall reading that Focal, Harman and Paradigm have anechoic testing chambers, staffs of engineers and physical facilities that presumably enables these companies to make rational and informed choices when designing and building speakers. I also recall reading that these companies manufacture in-house their own drivers. That is a question ... not a statement. Bruce |
Hi guys, I don't want to pretend to be The Expert on this subject, but I've built hundreds of speakers and have worked and designed along side some of the best back in my Marcof/SpeakerCraft days... so, "but only if in combination with 1st order crossovers as used by Thiel and Vandersteen, revealing a coherent time-aligned step response" 1st order helps, but there is no speaker that is in complete phase at all frequencies... When you roll a speaker hard, it throws the driver out of phase just by the nature of steep slopes... Remember a speaker rolls in and out of phase up and down the frequency ladder, once you choose your initial crossover point, you need to remeasure the drivers phasing up and down the scale with your first crossover in place , then choose another point that gets absolute phase at the crossover point. You may see a speaker that has a 350 & 3750 crossover point and you know that the tweeter can easily be crossed at 2500 or so, as long as the mid is smooth out to a given frequency, the designer may choose to take the frequency where absolute phase comes back into place. Also Impedance compensation helps control phasing a also. For that reason, I prefer impedance compensation, normally a driver sounds better when it needs compensation and compensation is used. "I recall reading that Focal, Harman and Paradigm have anechoic testing chambers, staffs of engineers and physical facilities that presumably enables these companies to make rational and informed choices when designing and building speakers. I also recall reading that these companies manufacture in-house their own drivers". Anechoic chambers help tremendously in measuring driver frequency and the effects that driver modifications have on a drivers frequency response, but Impedance & phasing are measurements that can be done without having a critical listening environment, I'm not trying to say that a great listening environment is not important, only that impedance and phasing can be handled without an anechoic chamber. Again, time alignment & correct phasing are keys to an imaging champ. I've kept it as simple as possible, I hope this helps, Tim |
Thanks Tim. I'm listening to a HD CD of Dave Brubeck's, "Time Out." Perhaps I'm just used to listening to sonic swill, but the imaging is more than satisfactory and the music is quite enjoyable. My fronts are Paradigm S8s (v3), with beryllium tweeters. Btw, I think the original recording was made in 1959. Maybe that was the Golden Age of music recording, maybe not. But the HD CD redo sounds great. As an Fyi, I bought the CD from the Acoustic Sounds web site. AS is located in Kansas. Very good source of music in all formats. BIF |
Getting back to the original question concerning sloped baffles and time alignment, Mofimadness commented: "It is used to time align the drivers. Some manufacturers do this physically, (with the cabinet) and some do it electronically (in the crossover)." No knowing a whole lot about speaker design, my guess is that it is more difficult and expensive to design and produce a speaker with a sloped baffle. Wouldn't that imply that there is some perceived advantage to doing so? |
Sometimes sloped baffles are cosmetic only. My understanding for 1st order crossover speakers is the broad frequency overlap between drivers, requires a sloped baffle to keep the drivers on a single axis. When off axis the shared frequencies between drivers will cause cancellations/nulls depending on the distance. |
Psag, I don't think the last couple of comments posted before your comment (mine included) deviated from the OP's question about sloped baffles. I say that because sloped baffled may be used to address (in whole or part) issues relating to using multiple speakers and electronic crossovers. Bombaywalla and others have provided us with a cornucopia of information about time and phase coherance -- sloped baffles just being a subset of the larger issue. Quite honestly, I didn't realize that the time and phase alignment problem was so difficult to solve. Having said that, admittedly, I do not fully appreciate to what extent a speaker's sonic performance is compromised by time and/or phase errors. Perhaps, as Sounds_Real_Audio just posted, the issue may be more hype than real. I just don't know. Al and Ralph .... where are you??? :) |
Simply putting the voice coils in the same plane does guarantee a cohesive and coherent sound. Drivers that are placed close together are just as important for a cohesive sound. It is equally important to have a bass driver that is as fast as the midrange otherwise you end up with a bass that seems disconnected. I hear this in some very expensive speakers. |
Tim, although I don't understand the science, I recall reading that a sloped baffle is one component of achieving phase coherence. The proper cross-over order (1st, 2nd, and so forth), plus maintaining a specific listening position, plus, setting the cross-over frequencies at the right points, and so on and so forth also contribute to maintaining optimal phase coherence. In addition, assuming that the drivers are in proper phase alignment in the overlap region of the cross over points, presumably there should not be phase cancellation or augmentation which will mess up the frequency response. That is frequency response should remain flat. But what about maintaining the proper harmonic structure of complex musical passages? What other benefits are gained by locking one's head in a head vice in order to maintain the precise listening position, which presumably will enable the wave fronts from the various drivers to combine in an optimal phase coherent fashion at the listening position. So ... how much of this is hype? What other characteristics make for a good sounding speaker? I used to think I had a handle on this this issue ... but no more. BIF Al and Ralph ... are you catching any of this?? Throw me a line. I'm sinking in techno quick sand. |
Hi Bif, Well, you've got a bit of confusion going on. A sloped front is used to align the voice coils on the tweeter/woofer or tweeter/mid/woofer. The idea is to get all frequencies starting on the same plane... Using an easy example is a 2 way... A woofer may be 6 to 8 inches deep and it may be 5 inches or to where the voice coil meets the spider & magnet gap (that is your alignment point).... The tweeter however may be less than 2 inches deep. On the sloped front, you can move the tweeters voice coil backwards in alignment to the woofer's voice coil by moving it up the slope.... Guys, I am on solid ground here, you can ask, who ever you want, I may not be known like Al or Ralph and don't come close to their electronics knowledge, but Speakers, I've got a good handle on. There are tons of other issues that we haven't discussed. Moving the tweeter too far from the woofer (Depending on crossover frequency) can cause all kinds of problems with smearing, lobing and other dispersion issues. I was only trying to handle the original question of a sloped baffle and phasing. I've had great luck using 6/12 crossovers on 2 ways @ 2500hz and find most of the time (depending on driver) that the speaker is in phase at the crossover point with those slopes. I am currently using an MTM, crossed at 1700 htz with 12/18 slopes, it is phase coherent at that frequency and time aligned in the crossover and images as well as anything that I've sat in front of. Simple 6 db slopes cause less problems to deal with, I've used them with great success also, but its not the only way.... I hope this helps, Tim |
A good step response and square wave response can be a significant indication of wave form fidelity, things that any other piece in the system should readably be capable of. Much of the need for correct listening position for the appreciation of such design principles is in regard to distance from the speakers for proper driver integration, typically such designs have something of a wide horizontal dispersion, sometimes there is concern for the listeners vertical position. Many of us do serious listening while sitting somewhat centered between the speakers. Such designs at least offer a very good semblance of wave form fidelity at some position as opposed to never providing any semblance of wave form fidelity from any position. You don't have to sit with your head in a vice to get the virtues of such designs! |
06-21-14: BifwynneBifwynne, as Timlub wrote earlier, I don't think it's necessary to have an extensive R&D budget, scores of engineers, anechoic chambers & mutli-million $ machines to make a good sounding speaker *iffffff* the manuf really knows how to design a speaker. The companies that you stated need this to make-up for their lack of knowledge of speaker design. Focal has gotten better over the years. when I 1st heard their Electra series speakers, they sounded like sh**. Their ultra-expensive speakers have a big wow factor but nothing more. Those R&D budget, scores of engineers, anechoic chambers & mutli-million $ machines are good marketing hype items that sells. ;-) of course, my opinion based on listening to many of the cited manuf's speakers. If one really knows what one is doing then it is possible to select & buy speaker drivers made by OEM speaker driver manuf that are superlative in their specs & performance. The next thing to do is to design a speaker integrating those drivers into a cabinet such that sum of the two is greater than each part. Al and Ralph .... where are you??? :)You know Bifwynne, it is entirely possible that Al & Ralph cannot help us here. Raplh is an amplifier expert & the fine points of speaker design might not be in his 'quiver of arrows'. Same for Al. Unlike explaining electronics, one cannot explain way speaker design unless one has dealt with this complex task. Yeah, everything is a compromise - engineering is an applied science. By its definition it is supposed to make compromises to bring about a solution. The key is: which engineer makes the best compromise? I believe that the 3rd link I stated clarified your earlier question re. the difference between time coherent & phase coherent. The sloped baffle gives you a time alignment of drivers but does not tell anything about the time or phase coherence of the speaker. Another thing I learnt was that just because a speaker has a 1st-order x-over does not mean that the speaker is time or phase coherent. I found that a lot of speaker manuf hide behind their using a 1st order x-over. I found that many such speakers used a 1st order x-over but their drivers were some higher order (such a 2nd or 3rd) which means that the driver performance rolled off with a higher order well before the 1st-order x-over kicked in. So, in effect, such a speaker is a 2nd-order or 3rd-order speaker & not a 1st-order speaker. To be truly a 1st-order speaker the speaker has to be electrically & mechanically a 1st order speaker meaning to say that the driver's performance needs to extend beyond the x-over point so that the roll-off is being done only by the electrical x-over network. it's complex material that cannot be absorbed in 1 reading. I've those threads printed off & re-read them from time to time to refresh my understanding. Each time I extract new info from them. Like Unsound wrote earlier the 1st thread & I think the other 2 as well are some of the very best threads to have appeared on A'gon. |
I'll be off the net for a couple of days, so I'll end my part of this thread here. Simple 6 db slopes do not tell the story at all, Most would be amazed that when you put a simple cap on a tweeter expecting a 6db slope, many times you might find 9 or 10 db slopes.... There is a difference between electrical vs acoustic slopes... On my MTM speakers that I referenced... I may be using 12/18 slopes, but the finished acoustic slopes are 24db per octave. Speaker design with new drivers from scratch requires measurements, you can get amazingly close with some of the better software however... Then lastly to throw another wrench in the fire... you can easily get time alignment between a tweeter & woofer without being phase aligned, but you can never achieve TRUE phase alignment between a tweeter & woofer without time alignment. I haven't read either of the threads referenced above, but hopefully a quality speaker manufacturer out there has explain it better. I was only trying to help, Tim |
Ok ... if Tim or Bombaywalla catch this, here the ultimate Q. How can one tell whether a speaker is time and phase coherent? Critical listening? Reviewer comments? Bench test? If critical listening is that important, the real challenge for us is, as many have written, that it is not easy to meaningfully audition speakers. So what's a person to do? I'll ask again, how important is time and phase coherence? FWIW, ... really more as an FYI, ... Paradigm's web site states that its "speakers have phase coherent crossovers designed so that the summed output of the drivers is completely and accurately rejoined." Is that hype? It is true at all frequencies? Dunno Thanks again. |
^Again good step and square wave response are indications of wave form fidelity (time and phase). Again, as to it's importance, well that's up to debate, with opposing opinions prevalent on both sides. I will say when I first began seriously auditioning speakers and without any real technical knowledge I was consistently drawn to those speakers that unbeknownst to me at the time shared those design properties, and those speakers designs continue to favorably impress. Others don't always seem to share the same sensitivity and/or priorities that speakers with these design priorities share. Only by listening to them yourself will you know for sure if it's important to you. Oh, btw, they usually share the similar quality of steady impedance loads, a subject of which I'm confident you will recall being discussed previously here on Audiogon. |
Hi Bif, I've found that time alignment is certainly worthwhile, but let me just say that all you may need is a way to visualize how and why it would be important. I'm not technically as well versed on the subject as some, but I do have some direct user experience with it. I'm in the process of triamping my rig. Each amp I have (and will be adding) will have a suite of (digital) tools which include a pair of xovers, EQ and time delay. Each pair of tweeters, mids and woofers will have their own amp...and each of those frequency bands can be delayed the appropriate amount - dialing in by ear alone, at the lp, is sufficient. I can tell you that in practice that process is far from complicated and can be done in a couple minutes by anyone. It just may seem impossible to grasp for you at the moment because (understandably) you don't have any way to vary the control over your time alignment one way or the other so there is no way for you to realize its effects at this point. But, try to visualize it in terms of what you may already have some ordinary experience with. Try this: turn on a TV in one room of your home and then turn on another TV in a different room, but turn that one up loud enough to hear it while you're in the other room listening to the 1rst TV. Make sure they are both on the same channel. You can't help but notice the substantial time delay - like a rather serious echo, but in a nutshell, that's really all there is to time alignment. You've likely heard this effect countless times before without thinking about it at places like car dealerships, fire halls, etc...places where a PA system is using more than one loudspeaker in different places outside (usually on the other side of the building), creating the very noticeable amount of delay. Now, the big difference (apart from the sheer amount of time in the delay) is the fact that, like with the TV example, all of the frequencies overlap...hence the echo effect, naturally. But, with HiFi speakers (except to a degree at the xover regions, which we'll ignore here) the frequency groups are largely separated in their delay. What this means is that (with any conventional, flat baffle arrangement, anyway) the highs reach your ears first, then the mids and then the bass. Basically it means that the highs are simply being projected out into the room and are imaged a little closer to you (in soundstage distance) than the mids, which are in turn projected a little farther into the room than the bass...it simply becomes apparent to you that the higher frequency band seems physically imaged a little closer to you. Time alignment resets this effect, moving the imaging of the higher band back into place with the frequency band below and gets all the instruments and voices that much more correctly positioned from top to bottom in the frequency range...a very nice and more realistic or natural presentation - definitely worthwhile. Once you've ever heard the effect for yourself it takes all of about 5 seconds to solve the mystery as to how and why it might be useful, but, with speakers if you visualize the difference in terms of the distance of the soundstage image from you with each driver's operating range, then you just may at some point be able to recognize just how most speakers have somewhat less than ideal time alignment. And some will seem a little worse or better than others. If you use subs, however, it is always recommendable to take time alignment into consideration along with the other factors concerning placement, but I don't believe perfect time and phase alignment has ever been done in any speaker and I may never see it in my lifetime. But, triamping in this way for me is giving me the opportunity to pretty much 'fix' the time-alignment problem, which won't hurt me at all. Apart from building your own speakers though, it seems there are not too many designs that have actually tackled the problem, as near as I can tell. Until more makers become convinced there is a market for time-aligned cabinets (expensive because it's so labor intensive), then there may be only a handful of alternatives for us. Other problems for manufacturers include that in designing cabinets for true time alignment (meaning driver offset at something close to the correct distance) they start getting into a world of cabinet diffractions and reflections - not easy enough to successfully solve, particularly in any aesthetic sense. It's not horribly critical to get the distance precisely right, but the closer the better up to a point...like trying to position both your speakers the same - it might be quite good to get them within, say, a 1/4" of each other, but under an eighth of an inch??..at some point there ceases to be an improvement, but then again, with time alignment you would notice generally a better sense of scale and improvements in the macro areas of the soundstage...while the soundstage details would remain largely unchanged. Hope this helps. |
Bruce (Bifwynne), thanks for the summons. I haven't yet taken the time to read the links Bombaywalla was good enough to provide, but I am pretty much in agreement with all that has been said by the various posters above. Including the comment by Kiddman just above, which I believe to be a correct statement of the bottom line. And also including the comments by others to the effect that speaker technology is not one of my areas of expertise :-) And in that connection I'll say also that over the years I have found Tim's (Timlub's) posts concerning speakers to be highly knowledgeable, credible, and informative. Another member whose inputs I suspect would be particularly valuable on these issues would be Duke (Audiokinesis); perhaps he'll spot this thread. Regarding your basic questions about about how to gauge time and phase coherence, and about how important it may be among the innumerable tradeoffs that are involved in speaker design, I think that Unsound summed up the answers very well when he said: Good step and square wave response are indications of wave form fidelity (time and phase). Again, as to it's importance, well that's up to debate, with opposing opinions prevalent on both sides.Realize that a theoretically ideal square wave (which does not exist in the real world, of course), having infinitely fast transition times between its two states (i.e., risetimes and falltimes of zero), perfectly flat tops and bottoms, and no overshoot or ringing, consists of the summation of an infinite number of sine waves, one being at its "fundamental frequency" (the frequency with which its pulses repeat), plus others at every odd multiple of that frequency (i.e., the 3rd, 5th, 7th, etc. harmonics). The amplitude of each harmonic decreasing as its order (i.e., its frequency) increases. In order for a real-world approximation of that square wave to be reproduced accurately by a speaker, at a given listening position, all of those harmonics (which collectively are generated by a combination of all of the drivers in a multi-way speaker) have to arrive at that listening position properly timed and phased relative to one another. No speaker will do that perfectly, of course, but results will vary widely among different designs. In the measurements section of most of the speaker reviews which appear in Stereophile, John Atkinson presents and discusses the speaker's step response (which is pretty much as meaningful as the square wave response, a square wave consisting of a continuous series of positive-going and negative-going steps). For example, Figure 9 here is the step response for the (very expensive) Vandersteen 7, measured at a certain height and distance. The step response shown is probably about as good as it gets. (Keep in mind, btw, that the step response of any speaker will not **look** like a step because, among other reasons, for it to do so would require that the speaker's frequency response extend down to close to zero Hz). So perhaps some degree of insight into the importance of time and phase coherence could be gained by comparing that measurement for speakers whose sonics one is familiar with and that have been reviewed by Stereophile, and trying to judge the degree of correlation that exists between the quality of that measurement and one's subjective preferences among those speakers. Although to be meaningful, that would have to be done by a given listener for a considerable number of speakers. I've never done that, btw, and unfortunately I don't think that such measurements are available for most of the speakers I've owned or extensively heard over the years. Best regards, -- Al |
Bifwynne, if you don't mind a little more reading, I suggest Time and Phase Coherence by Roy Johnson from his website. This is very informative with illustrations, and you don't have to sort through confusing comments as in the threads from the links that were provided previously. Hope this helps. |
Al and others, take a look at John Atkinson's step measurements of the Revel Studio 2: http://www.stereophile.com/content/revel-ultima-studio2-loudspeaker-measurements JA notes that "the speaker is time-coherent rather than time-coincident: each drive-unit's step smoothly hands over to the next lower in frequency. This correlates with the superb frequency-domain integration between their outputs ...." Two observations on my part. First, the Vandy 7's step response plot shoots up like a rocket and then quickly rolls off and stays down. By contrast, the Studio 2 shoots up, rolls off and then rolls back up again. What does this mean? Why the differences? The other observation is that the Studio 2 has a ruler flat frequency response. I believe the Studio 2 uses high order cross-overs, like my Paradigm S8s. Interpretive comments are welcome. |
Just a supplemental fyi, here JA's bench test measurements for the Revel Salon 2: http://www.stereophile.com/content/revel-ultima-salon2-loudspeaker-measurements And one more, the might Magico Q7: http://www.stereophile.com/content/magico-q5-loudspeaker-measurements If nothing else, perhaps someone can explain how to properly interpret what these measurements mean. They do not look like the Vandy 7. |
Bifwynne, Ok ... if Tim or Bombaywalla catch this, here the ultimate Q. How can one tell whether a speaker is time and phase coherent?Critical listening is one of the methods - when a speaker is time-coherent then it will get the timing right for ALL genres of music. Recordings otherwise painful to listen to will be less painful (because the speaker is not distorting the music signal coming to it). In time-coherent speakers you forget the audiophile attributes (pin-point imaging, soundstage width/depth/height, etc, etc) & you focus on enjoying the music. The other method, as stated by Unsound & 2nded by Al, is the time-domain step response. Usually Stereophile (as indicated by Al already) & Soundatage reviews http://www.soundstage.com/index.php?option=com_content&view=article&id=16&Itemid=140 do a pretty good job of showing all the measurements they have done on various loudspeakers they have resp. reviewed. Thanks Tls49 - you have pointed Bifwynne to a really good website for reading on time & phase coherence. I always look to Roy Johnson when I have questions on this subject. To me he is skilled in this field & has dedicated his life to making time-coherent speakers since he began his biz in the 1980s. Green Mtn Audio speakers are the only speakers that I know of that are time-coherent across the entire audio band from my personal experience. I know that several people believe that Vandersteens & Thiel speakers are also time-coherent - I personally do not have any experience w/ either of those brands so I will not comment. First, the Vandy 7's step response plot shoots up like a rocket and then quickly rolls off and stays down. By contrast, the Studio 2 shoots up, rolls off and then rolls back up again. What does this mean? Why the differences?Why the differences? Because the Vandy 7 is time-coherent & the Revel2 is not! Bifwynne, if you click on the link provided by Tls49, you will see exactly why as you read the text there. Scroll down where Roy J talks about time-domain response & he shows how a step response looks when the speaker is time-coherent & when it is not. The Revel2 are clearly not time-coherent. I'll ask again, how important is time and phase coherence?To me it is everything! I used to own B&W speakers & I have heard speakers from many, many different brands now both auditioning at dealer locations & at multiple RMAF shows, at various friends' homes, etc. The sonics from a time-coherent speaker are in a totally different league. For me time-coherent speakers "get it" while all others simply do not. Once I heard Roy Johnson's Green Mtn Audio speakers & owned one for many years, I realized what the heck time-coherence was all about. Now, I cannot listen to any speaker if it's not time-coherent - very quickly I will realize the lack of time-coherence & I will immediately perceive that the music has no life & no PRaT & to me it will just good/superior sonics but not music. I want music from my speaker & not good/superior sound - I didn't pay top $ for good/superior sound; I paid top $ for music thru the speaker. These days I stick to time-coherent speakers - my selections are very few but it's worth every penny to me. This should be the case for everyone if you are interested in listening to music. Many of friends are into superior sound (& not music) so they are on a perpetual merry-go-round for speakers - speakers come & go thru their music rooms. They are always wow'd by a new speaker & later after a few months or a year, that speaker is sold & another one takes its place & the cycle repeats. In my mind I'm thinking that I've made my case for time-coherent speakers - you buy it once & then you sit back & listen to music because a time-coherent speaker will play every genre without tripping up because its physics is right. Of course, IMHO. YMMV. |
06-22-14: BombaywallaI agree. Bruce, taking the Magico Q5 as an example, whose measurements you also linked to, note that the frequencies of each of the four up and down oscillations that are shown in the step response figure become progressively lower (i.e., their periods/durations become progressively longer). That is because the first peak primarily represents sound arrival at the measurement mic from the highest frequency driver; the second peak primarily represents sound arrival at the measurement mic from the next highest frequency driver, and so on. So it can be inferred that the amount of delay between the start of the arrival of the sounds from each of those drivers and the occurrence of the step in the signal that is sent into the speaker are significantly different, and become progressively longer for progressively lower frequency drivers. While with the Vandersteen 7 they are not significantly different, resulting in the outputs of each of its drivers summing (at the position of the measurement microphone, at least) to a much closer approximation of an ideal step response. Best, -- Al |
Just a word of caution regarding some speaker measurements one might come across. In an effort to avoid room interactions skewing measurements, very often speaker measurements are taken closer to the speaker than where one would typically sit. Most speakers designed for time and phase integrity need to be taken further away to allow for driver integration (approximately 8'-12'), where listeners typically sit. Unless these further measurements are taken in an anechoic chamber, the room will now contribute more measurable distortions than if the same measurements were taken closer to the speaker. Most people don't listen to speakers at the distances many speaker measurements are taken. If speakers with time and phase integrity aspirations are measured at distance ranges other than where they were intended to be heard from; their square wave and step response, though probably still better than competing speakers without such design considerations, will have their ultimate measurable potential be compromised. |
Bombaywalla, I'm somewhat familiar with NRC test facility. But I am not sure which of the NRC tests speaks to time and phase coherence. Btw, I surmise that the reason many of the top speakers, like the Magico Q5, that have flat frequency response plots and probably use higher order cross overs is because the driver timing is tweaked at the cross over points to optimize wave cancellation and augmentation. But ... the drivers are not otherwise in phase outside of the cross over overlap region. So ... even if the speaker specs well on the bench, it may very well be distorting complex sound patterns as Roy Johnson explains in his articles. Bombaywalla, did you or Tim mention that one mark of a phase coherent speaker is one which has a flat impedance and phase plot. Take a look at the Magico S5's specs. Is there anything else apparent from the NRC tests that permit inferences about phase coherence? http://www.soundstage.com/index.php?option=com_content&view=article&id=1043:nrc-measurements-magico-s5-loudspeakers&catid=77:loudspeaker-measurements&Itemid=153 Al, here's another Roy Johnson article I clipped. Perhaps if you get a chance to read his articles, please share comments and thoughts: http://greenmountainaudio.com/speaker-time-phase-coherence/ Btw Al, you may find the square wave pics in the article to be of interest. It seems that using higher order cross overs creates an almost unsolvable electrical problem. BIF |
In speaker designs that I've only simulated for now, sloping the baffle 10 degrees allowed a slightly higher crossover with the same drivers, which is not so arbitrary. In a 3-way, it also allowed symmetric slope crossovers. In a 2-way, adjusted for BSC, it ended up asymmetric similar to the way I'ld do a flat baffle. Since a normal TM tends to lobe downwards, the sloped baffle tends to correct that and even allow a shorter height. The downside was that it was sensitive to listener height and distance. |
There are time and phase coherent speakers that are wonderful and there are those that are lackluster or worse. There are speakers that are not both time and phase coherent but are wonderful, and those that are lackluster or worse. Time and phase coherency are two out of many attributes that would be goals in designing a loudspeaker. But not having it does not mean the speaker is not great, and having it does not mean the speaker is great. You have to listen. Of my favorite speakers, one is and some others are not. It does not bother me that some are not. I cannot detect the effects of the lack of time and phase coherency of those that are not. None of this means that it is not a valid design goal, but it is only one of a very large number of valid design goals. |
Bombaywalla, did you or Tim mention that one mark of a phase coherent speaker is one which has a flat impedance and phase plot. Take a look at the Magico S5's specs. Is there anything else apparent from the NRC tests that permit inferences about phase coherence?Bifwynne, don't know whether I or Timlub wrote this - I don't recall writing this in this thread. Maybe in some other thread you & I contributed to? Anyway, looking at the Magico S5 measurements, even tho' the imp & ph plots are less bumpy than many others in the market, there is still quite a bit of phase shift. Plus, at 2KHz the speaker goes from positive phase angle (inductive) to negative phase angle (capacitative). This can be quite burdensome for the amp - the phase relationship between the current & voltage changes very suddenly & the amp has to react to that instantaneously. This is difficult to do. Looking at these impedance & phase plots I cannot confirm whether or not this speaker model is time-coherent. Probably not as I've not seen Magico design any speaker thus far that has been time-coherent. Maybe the S5 is a departure from that norm - don't know - but if I'd have to guess, I'd say 'no'. OTOH, for, say, Green Mtn Audio speaker I'm used to reading specs that look like this. For the C3 speaker model, for example: Response +/- 0.75dB from 40Hz to 21kHz, -3dB at 35Hz and My brother owns the Callisto bookshelf models & their specs are: Response Freefield at 2m: +/- 0.75dB from 55Hz In both the Green Mtn Audio speaker specs, now, we are talking flat impedance & phase response. :-) Nothing like that in the Magico S5. (Once again, I've experience with Green Mtn Audio speakers hence I bring them up time & again. If other members have similar specs for other time-coherent or phase-coherent speakers, please share those specs. Thanks.) |
Just took a look at this thread; the title didn't obviously say "read me" so it took me a while to check it out. Lowpass drivers lag behind highpass drivers in both time and phase, and a sloped baffle can introduce a limited amount of compensation for that, along with a tilting of the driver axes (which may be desirable in some cases). In and of itself, it's not enough compensation to make much difference, but could be part of a full-scale time-coherent effort such as Roy Johnson's magnificent speakers. "Phase coherence" and other phase-based terms have been used loosely and don't necessarily mean what we think they mean. For instance, the woofer can lag the tweeter by 360 degrees at the crossover point, yet the system could be called "phase coherent". Roy's designs are truly "time coherent", and that puts them in a different league; they're the real thing. His designs do everything we might have optimistically hoped for, such as waveform preservation, when we first came across the term "phase coherent". As for how small manufacturers can do decent design work without access to an anechoic chamber, one answer is time-gating. That's a measurement technique wherein the microphone is turned on briefly to catch the direct sound from the speaker, and then turned off before the first reflection arrives. Time-gating has limitations relative to what can be done with a true anechoic chamber, but it sure costs a lot less. Roy, I'd be very interested in reading any clarifications you care to post, and have zero problems with your making specific references to your products. The inherent rapid rolloff at the bottom end of the horns I use prevents me from ever doing what you do without totally changing directions (and going back to college), but that doesn't lessen my admiration for your work. Duke dealer/manufacturer |
>>>In both the Green Mtn Audio speaker specs, now, we are talking flat impedance & phase response. :-) Nothing like that in the Magico S5. Interesting. Any independent measurements you can point us at to verify your claims? The ones we see in SP are not that impressive. In fact, they are quit awful (and are not even time coherent as claimed). Even the so polite JA had to say that: ..the listening axes on which the speaker is (almost) time-coherent and offers the most neutral response don't coincide. Which is just one typical issue this concept have. Not to mention that distortion level with first order XO will be significantly high: occasionally thought I caught a hint of lower-midrange congestion. All and all, a concept that does not really work, due to all the other problems it creates. |
Hi guys, I'm back to the real world... first, Al, thank you for your gracious words... If Green Mountain audio meet specks, what Bombaywalla posted are truly amazing, it takes a lot of crossover work to bring phasing in line to plus or minus 3 degrees. I've never seen a graph of a speaker with specks that tight. 6db pads done correctly will keep you within 15 degrees or so. I'm not a pro and have nothing to lose by showing my stuff. I'm using HiVi D6.8 woofers (same as Totem Forest) and a low to mid Scan Speak Tweeter in an MTM. I've never used photo bucket, so hopefully this works, but I just uploaded my graphs & crossovers to photo bucket... Hopefully, you can view this, if not, I'll try to figure it out and post again. http://s28.photobucket.com/user/timlub01/media/XO_HiVi_D68_ScanSpeakD2608_MTMrev4_zps302ac070.gif.html |
it is true that the measurements by reviewers are few for Green Mtn Audio speakers. I don't know why? I believe that Green Mtn Audio speakers are different from conventional high-order x-over speakers that are prevalent in the market. If one is used to measuring the conventional high-order x-over speakers that are prevalent in the market that same technique is unlikely to work for Green Mtn Audio speakers because of the requirements of the distance & listening axis required to ensure that the drivers integrate. I believe that JA was not knowledgeable about this when he did the meausurements for the Diamante speakers. I also understand from the manuf that JA was advised by the manuf many times how to setup the speaker to measure it correctly but JA completely ignored the manuf's recommendations & did his own thing. The result is poor measurements & also measurements done incorrectly. Yes, the slopes of 1st order x-over are shallow compared to 2nd or 3rd or 4th order x-over so signal will bleed thru at high levels. While 4th order x-over speakers can be driven to 120dB SPL, 1st order x-over speakers can be driven to something like 105dB SPL. If this is a no-deal for you, so be it. But 105dB SPL for in-room/in-house listening is plenty loud by any stretch of the imagination. Yes, it's not outdoor rock concert SPL. Have you owned any Green Mtn Audio speakers by which you claim " All and all, a concept that does not really work, due to all the other problems it create."?? Or, are you one of those people who reads reviews & makes decisions based on the printed word? Or, are you a speaker manuf with an axe to grind? Because I speak from personal experience & from the experience of my friends & relatives who own & have owned Green Mtn Audio speakers. Everyone that I know of has had & is having an excellent experience with these speakers. Yeah they have limitations (which speaker does not?) but their limitations are of ommission rather than of commission. I do not intend to convert you to 1st-order speakers - never intended to for the get-go - please buy what you like & enjoy it. But I would not make such sweeping remarks without personal experiences. Anyway, this thread is about "sloped baffle speakers", "time & phase coherence". There are very few speakers that meet the time & phase coherence criteria & I believe that Green Mtn Audio is one such brand hence it gets brought up time & again in such discussions as an available commercial product. I'd prefer that this thread not take a downward spiral focusing on Green Mtn Audio speakers or Green Mtn Audio bashing. Let's keep the discussion on "sloped baffle speakers", "time & phase coherence". Thanks. |
Excellent comments gents. I'm hoping Al (Almarg) and Ralph (Atmasphere) get around to reading Roy's articles and sharing their comments. Although it has been said many times, designers of audio gear take into account many features and do their best to balance, maybe better said ... optimize, what comes out of the metaphorical oven. I'm sure phase coherence is just one of many important factors. That's why I'm hoping Al and Ralph weigh in here to help us better understand the relative significance of phase coherence as compared to other factors. Bombaywalla, I agree that one shouldn't base expensive purchase decisions solely on reviewer comments. But there's a problem; one that has been touched on by many A'goners including me over the years. There are very few hi-end B&M stores around anymore. And to the extent such stores are around, setting up meaningful and fair auditions is very difficult. Oh ... and as far as Green Mountain is concerned, no dealers near me. Not sure what else to say. Another potentially huge sleeper problem that is finally getting the press it deserves. |
Bandying about technical claims by manufacturers and trying to determine how it affects the final product, whether it be a car, motorcycle, amplifier, airplane, or missile is almost totally useless. Investing hours in driving to dealers, or flying, asking manufacturers to let you hear them, flying to audio shows, those are all much better ways to make an educated guess about how you will react to the sound in your home. |
