[DawgByte]

Jim -

What is your philosophy regarding Time alignment and Phase Coherency? Given the following simplistic definitions:

1) Time-alignment means that sounds start at two different drivers at the same instant will reach your ears at the same time.

2.) Phase-alignment means that there are no phase errors between drivers: the same frequency waveform, for example, produced by two drivers (in their overlap region) will be entirely in phase with each other (and thus completely reinforcing).

Paul Hales, Jim Thiel, Sean Casey (Zu Audio) and Roy Johnson of Green Mountain Audio are well respected speaker designers and feel very strongly about this topic. Jim I was interested to get your take on this topic.

[DMurphy]

Jim -

What is your philosophy regarding Time alignment and Phase Coherency? Given the following simplistic definitions:

1) Time-alignment means that sounds start at two different drivers at the same instant will reach your ears at the same time.

2.) Phase-alignment means that there are no phase errors between drivers: the same frequency waveform, for example, produced by two drivers (in their overlap region) will be entirely in phase with each other (and thus completely reinforcing).

Paul Hales, Jim Thiel, Sean Casey (Zu Audio) and Roy Johnson of Green Mountain Audio are well respected speaker designers and feel very strongly about this topic. Jim I was interested to get your take on this topic.

Hi   I believe Jim is burning rubber en route to RMAF.  We don't want him texting while driving.  Are you asking what we think of each condition--1 and 2?   If a speaker satisfies condtion 1, it automatically satisfies condition 2.  But the reverse isn't true.    If two drivers start at the same time, and reach your ears at the same time, they will be phase coherent (condition 2).  But you can have phase alignment without condition 1.  Since I'm not driving, I'll give my opinion.  Jim Thiel (may he rest in peace), Vandersteen, and others believe condition 1 confers audible benefits in terms of imaging and transparency.  But it's a very difficult goal to achieve, generally requiring complex crossovers to ensure 1st order acoustic slopes over 2 or 3 octaves, sloped cabinets to align the acoustic centers of the drivers, a very robust tweeter (almost no ribbons need apply) that can stand up to the gentle slope at the bottom end, and very smooth drivers that don't run out of response too early to maintain the 1st order slopes (forgeddabout Seas magnesium woofs or mids).  And then the goal will be maintained only on axis.  Move up or to the side, and you lose condition 1.  Plus, the wide overlap between drivers will cause interferrence patterns that can be quite audible.  But is there magic to the sound when listening on axis?  I've never heard it, and I've participated in A-B demonstrations with active crossovers that switch instantly between condition 1 and the more common 4th order acoustic slopes, and I couldn't hear any advantage.  But that's not settled.  Suffice it to say that condition 1 is hard to achieve, and greatly restricts your choice of drivers.  And it's strictly a sweet spot sort of thing if it exists.  Condition 2 can be achieved by any properly designed Linkwitz-Riley 4th order crossover, which is what all of the Salk speakers use.  I go to great pains to achieve a high degree of phase alignment at the listening position, which I target at 3-4 meters.  That ensures that the drivers will sum flat, but requires a lot of futzing since Jim uses flat baffles, and the driver acoustic centers are not aligned to start with. But---that doesn't mean the drivers are playing exactly the same part of the program material at the same time.  The woofer and tweeter (in a 2-way) will be exactly one cycle apart.  The + and - parts of the wave will coincide, so there won't be any destructive interferrence, but they won't be at exactly the same point in the music.  Can the ear tell that?  I don't think so, but I'm not on a crusade.  We have to go with what we understand and believe, and I personally don't think Condition 1 is worth the cost and restrictions.  I do think that phase alignment is important, and I like to think that's why a lot of people find the midrange on Salk speakers very transparent.  But that's very controversial.  I'm just going with what I've found to work.  Other approaches may work as well.   

[Nuance]

Dang...that was a great response, Dennis.  I think I asked you something similar once, but that was a nice refresher.   

I'd have to concur that adhering to your strict standards in crossover design (especially the "number 2" you mentioned) is exactly why the midrange sounds so detailed and transparent.  Of course, there are many other variables that go into it, but I think it's safe to say that if someone else designed the crossovers for Salk speakers they would sound very different.  And in my opinion that would be a bad thing. 

[JLM]

We single driver speaker fans have solved that problem, and some others (like how two different drivers sound/radiate the same at crossover, and what about listening at different distances as Dennis alluded to), while gaining the very significant advantages of active amplification (one amp directly feeding one driver).

But I mention that to point out that some designers use extended midrange drivers to push those nasty crossover frequencies as far as possible from possible from frequencies where the ear is most sensitive to the phasing and other multi-driver induced "issues".  This gains most of the coherency of single driver designs while not limiting frequency response.  (Our own Brian Cheney is one of those designers who came from the multi-driver world.)

[DMurphy]

We single driver speaker fans have solved that problem, and some others (like how two different drivers sound/radiate the same at crossover, and what about listening at different distances as Dennis alluded to), while gaining the very significant advantages of active amplification (one amp directly feeding one driver).

But I mention that to point out that some designers use extended midrange drivers to push those nasty crossover frequencies as far as possible from possible from frequencies where the ear is most sensitive to the phasing and other multi-driver induced "issues".  This gains most of the coherency of single driver designs while not limiting frequency response.  (Our own Brian Cheney is one of those designers who came from the multi-driver world.)

Thanks for chiming in.  As I said, this is controversial.  I'm obviously not a one-driver fan, and I've tried just about all of them, including the super expensive and sophisticated German unit that shall remain namelss.  I think they have their uses in computer applications at short range, but I've never heard a single driver speaker that came close to a good multi-driver in terms of openess and smoothness, and I didn't hear any real advantage in transparency. 

As for using a mid over a very broad range, I've experimented and experimented with that.  I worked on various prototypes of the new HT4 for over a year using different mids and widely different crossover frequencies.  We ended up with what may be the world's most expensive midrange that is designed to be a full range unit.  In that respect, it is one of the best I've heard, with unbelievably broad dispersion.  But it ended up sounding best operating over a very narrow range (300 Hz - 1700 Hz), with the RAAL ribbon handling all of the treble region.  Now, if the FAL mid sounded as good as the RAAL up to, say, 6000 Hz, I probably would have crossed there.  Why cross in the upper mid, lower treble if you can cross where there are no fundamental tones in program music (the highest fundamental on a piano is below that point.)?  But the RAAL does sound better than the FAL and just about any other driver I know of above 1700 Hz, so we're back to trade-offs, personal opinions, and perhaps differences in individual hearing.  Which is why I enjoy speaker designing.   

[DawgByte]

Thank you Dennis for your candid response. This is indeed a highly controversial and contentious subject that has only recently (last 3 years) been brought to the forefront of audiophiledom by Srajan Ebaen's estatic 6 Moons review of Zu Audio's Druid speaker. Prior to that I recall this subject spawning over 100 posts on Audiogon.com. It was highly entertaining to say the least! I've spent a couple hours talking to Roy Johnson about this subject. There is absolutely no question where he stands. This subject was so important to Roy that he went back to engineering school to get a PHD. His goal was to master (as best as one can) the physics involved in getting condition 1  correct.

I also agree with your comments about single driver systems. It is physically impossible for a single driver to reproduce the entire audio spectrum like a multi-driver solution.

Dennis I recently saw a picture of the new HT4 - it didn't appear to have a sloped baffle, but the tweeter and midrange drivers seemed to be slightly set-back from the bass unit. Is that a correct observation?

I own time aligned speakers, and my ears hear the magic, however I'm not married to a single design philosophy. Based on the descriptions of what listeners are hearing from Salk speakers, I'm most curious!

[Carl V]

But the RAAL does sound better than the FAL and just about any other driver I know of above 1700 Hz, so we're back to trade-offs, personal opinions, and perhaps differences in individual hearing.  Which is why I enjoy speaker designing.

Well that's a nice segue....
when voicing a speaker, a 2 way perhaps (all things being equal-ha!)
what contributes most to the 'sound' the midwoofer or the tweeter?
Asked another way would a RAAL substitution in the HT1 or HT3
substantially change the sound.  Or would you expect a greater impact
by going to a different Midwoofer of equivalent pedigree/quality?

 

[fishinbob]

I love it when Dennis takes us to school.

[DawgByte]

I love it when Dennis takes us to school.

Bob - The class is still open... trust me!

[DMurphy]

But the RAAL does sound better than the FAL and just about any other driver I know of above 1700 Hz, so we're back to trade-offs, personal opinions, and perhaps differences in individual hearing.  Which is why I enjoy speaker designing.

Well that's a nice segue....
when voicing a speaker, a 2 way perhaps (all things being equal-ha!)
what contributes most to the 'sound' the midwoofer or the tweeter?
Asked another way would a RAAL substitution in the HT1 or HT3
substantially change the sound.  Or would you expect a greater impact
by going to a different Midwoofer of equivalent pedigree/quality?

That's a really hard question.  I think the RAAL would add a little air to the HT3, but only on some program  material, and it would allow a lower crossover point.  I'm not a ware of a better midwoofer than the W18, so just about any substitution would be at least a small minus, and possibly a big one, particularly if it weren't as smooth as the W18 between 800 Hz. and 2k.  But the bottom line is I don't think there's a single correct answer for your question. 

[Nuance]

I love it when Dennis takes us to school.

+1.  There is some great info here, whether you're a "noob" or "veteran."  I hope this discussion stays civil and continues to help me learn more. 

[DawgByte]

I love it when Dennis takes us to school.

+1.  There is some great info here, whether you're a "noob" or "veteran."  I hope this discussion stays civil and continues to help me learn more.

Nuance - If you're interested in learning more about time alignment design, you can start with this thread and then continue to search the audiogon.com archives. If that doesn't satisfy your thirst for knowledge, or curiousity then I would recommend trying to find white papers written by Roy Johnson, or even contact him. He is very friendly and loves to talk about speakers and music, as I'm sure Jim Salk is as well.

http://forum.audiogon.com/cgi-bin/fr.pl?cspkr&1162943129&read&keyw&zztime+alignment

My only interest was to get Jim's opinion. However, since Dennis works with Jim his perspective is sufficient.

[Nuance]

Oh sorry, I didn't mean to imply that I didn't know anything about Time and Phase alignment.  I have owned Vandersteen speakers in the past, and I studied everything I could about Richard's (and Dunlavy's) engineering ideas.  I just meant I love learning from "Murphy University." 

[DawgByte]

Oh sorry, I didn't mean to imply that I didn't know anything about Time and Phase alignment.  I have owned Vandersteen speakers in the past, and I studied everything I could about Richard's (and Dunlavy's) engineering ideas.  I just meant I love learning from "Murphy University." 

Cool, thanks for the clarification. 

Nuance, I'm a big fan of the Vandersteen 5 series. How does the Salk HT3 compare to the 5's in your opinion? That will give me a good reference point to where my Hales fit into the picture.

[Rocket]

Hi Dennis,

What differences if any do you hear between the seas midrange that is used in salk sound speakers and accuton?  I have a pair of ht2's and the midrange is very good but I have also had positive experiences with other speakers that use accuton midrange drivers.  I have never compared the 2 together but I'm interested in your opinion on the matter.

Regards

Rod

[DMurphy]

The Accuton that competes most closely with the W18 measures almost identically to the W18, and when I compared that Accuton in the Ellis 1801 to a stock unit that used the W18, I couldn't hear any difference except maybe a very slight advantage for the W18 in the deep bass.  They make an excellent dome midrange, but it must be crossed at 800 Hz or higher.  They also have a 3.5" cone midrange that is quite good, but I found it tricky to voice.

[R Swerdlow]

?I'm a big fan of the Vandersteen 5 series. How does the Salk HT3 compare to the 5's in your opinion? That will give me a good reference point to where my Hales fit into the picture.

Zybar (who often posts around here) has owned both these speakers.  I hope he answers your question.

I haven't listened to the Vandersteen 5, but I am very familiar with the Vandersteen 3a signature and somewhat familiar with the 2ce signature.  I also own SongTowers.  I would rate the SongTowers as better than the 2ce but an even match with the 3a signature.  By even match I don't mean identical, but rather a toss-up.

[Nuance]

Oh sorry, I didn't mean to imply that I didn't know anything about Time and Phase alignment.  I have owned Vandersteen speakers in the past, and I studied everything I could about Richard's (and Dunlavy's) engineering ideas.  I just meant I love learning from "Murphy University." 

Cool, thanks for the clarification. 

Nuance, I'm a big fan of the Vandersteen 5 series. How does the Salk HT3 compare to the 5's in your opinion? That will give me a good reference point to where my Hales fit into the picture.

I have not had the pleasure with the HT3's, only the HT2 TL's, SongTower's and SongTower RT's.  I have, however, heard the Vandersteen 5A's, and as of right now they are my favorite speaker of all time. 

Zybar has had the HT3's and Vandersteen 5A's in house together, so hopefully he'll chime in.

?I'm a big fan of the Vandersteen 5 series. How does the Salk HT3 compare to the 5's in your opinion? That will give me a good reference point to where my Hales fit into the picture.

Zybar (who often posts around here) has owned both these speakers.  I hope he answers your question.

I haven't listened to the Vandersteen 5, but I am very familiar with the Vandersteen 3a signature and somewhat familiar with the 2ce signature.  I also own SongTowers.  I would rate the SongTowers as better than the 2ce but an even match with the 3a signature.  By even match I don't mean identical, but rather a toss-up.

I'd be inclined to agree. You've got some great ears there, young man. 

[Nuance]

Dennis, your response to Rocket's questions was a little difficult to find, so I fixed it up for you below. 

The Accuton that competes most closely with the W18 measures almost identically to the W18, and when I compared that Accuton in the Ellis 1801 to a stock unit that used the W18, I couldn't hear any difference except maybe a very slight advantage for the W18 in the deep bass.  They make an excellent dome midrange, but it must be crossed at 800 Hz or higher.  They also have a 3.5" cone midrange that is quite good, but I found it tricky to voice.

[DMurphy]

Thanks.  I just fixed the original.