Sanders Sound Systems - electrostatic

nsgarch

Well-Known Member
Apr 20, 2010
88
2
915
While the Innersound companies could not have survived the current economic recession, sales and income for Sanders Sound Systems have been growing during this time. This is proof that the Sanders' business model is successful. The future of Sanders Sound Systems is secure.
. . . . . . . despite the ignorant/arrogant utterances of a certain gasbag audio magazine publisher! :D As Kinky Friedman once so eloquently put it: "Let's let Saigon's be bygones . . . "

Best of luck, and welcome to the WBF.

Neil
 

Angela

WBF Technical Expert
May 24, 2010
141
0
0
Conifer, Colorado
. . . . . . . despite the ignorant/arrogant utterances of a certain gasbag audio magazine publisher! :D As Kinky Friedman once so eloquently put it: "Let's let Saigon's be bygones . . . "

Best of luck, and welcome to the WBF.

Neil
Hey Neil,

Thanks for the welcome and . . .

I would like you to know, that it just so happens, that I am an Honorary Texas Jewboy (#55)
(it's actually a great honor)

HTJ..jpg
Picture taken just now from my office wall!
(issued before Rog and I got married so it has my maiden name)

(. . . and I have read every book the man has ever written - they are laugh out loud funny! and have some original signed copies as well - small world ain't it?)

HA!
 

nsgarch

Well-Known Member
Apr 20, 2010
88
2
915
Angela, you are clearly a person with a LOT of backstory!!

-- N
 

Gregadd

WBF Founding Member
Apr 20, 2010
10,517
1,774
1,850
Metro DC
What are the prospects for a full range electrostatic.? ML has made the hybrid very popular. Everybody has capitulated except Quad. Soundlab IMO makes the best hybrid the Soundlab Dynastat II. I have not heard the 10c3. I just think a hybrid is a cop out. Although Roger makes a good argument, I'm not buying the whole limited dispersion argument. Maybe Roger could explain why he choose the Hybrid approach.
 

Angela

WBF Technical Expert
May 24, 2010
141
0
0
Conifer, Colorado
I sent Roger your questions and here is his response.

The Model 10c can be operated as a full range ESL. You can adjust its electronic crossover (or eliminate it entirely) to produce full-range operation anytime you wish.

So why do I prefer hybrid operation? Simply because I am not willing to compromise performance. You cannot beat the laws of physics. The fact is that the bass performance of a full range ESL is poor.


A speaker must do several things to perform well. In addition to obvious things like linear frequency response and fast transient response, a high performance speaker must also be able to play loudly enough to reproduce music and all its dynamics at realistically life-like levels. No full range ESL can do so.


High output is a tremendous challenge for an ESL. But technology has advanced to the point where it is now possible to make an ESL play at ear-bleeding levels from the midrange on up. But no full-range ESL can produce deep, powerful bass that can match what you hear from live sound.

Only magnetic speakers have that capability.

There is more to this issue of bass quality than just the amount of deep bass available. Not only is the quantity of electrostatic bass inadequate, but the QUALITY of electrostatic bass is dismal. Let me explain that statement.


One of the truly wonderful things about ESLs is their low Q behavior. By way of review, the engineering term "Q" refers to the "quality" of the sound with respect to control and damping.


In other words, a "low Q" driver exhibits very well-controlled behavior with fast transient response and without any overshoot and ringing. A "high Q"

driver is poorly controlled, has resonances, overshoots and rings after an electrical impulse, and has poor ("smeared") transient response.

It does not stop instantly as it should. Instead it vibrates for many cycles a relatively long time until it finally comes to a complete stop. It "rings" like a bell. This behavior adds extraneous frequencies to the original sound that corrupts it and makes it sound very unnatural.


The massless nature of an ESL means that it can accelerate instantly to follow the musical wave form. Even more importantly, the mass of the air around an ESL totally swamps the incredibly tiny mass of the ESL's diaphragm, so it cannot "ring" and resonate. It simply stops instantly, which gives it great clarity, detail, and transient response.


This is like trying to ring a bell underwater. It won't ring because the mass of the bell is swamped by the mass of the water around it. It simply is no longer free to vibrate and ring.


Therefore an ESL has very low Q behavior. This is one of the reasons it sounds so "tight", "crisp", "quick", and can extract the most subtle details from the sound and reproduce transients flawlessly. Low Q ESLs simply sound more "real" than high Q drivers.


However, this wonderful, low Q behavior of an ESL does not apply in the bass. An ESL is built like a drum, which is an extremely high Q device. An ESL has a stretched, "springy" diaphragm. The diaphragm is relatively large, so couples with a large amount of air in the room.


Air has mass. The relatively large mass of the air to which it is coupled will cause the ESL to resonate like a drum at some low frequency based on the mass and spring rate (tension) of the diaphragm. In ESLs of typical size, this resonance occurs between 50 Hz and 100 Hz. Tap on the edge of any ESL and you will hear this resonance.


This is the fundamental resonance of the ESL. It is the only resonance in an ESL, but it is huge (typically 16 dB) and like all resonances, it is extremely high Q.


Being high Q, frequencies around the fundamental resonance will be very poorly controlled and have a large amount of overshoot and ringing. The sound in this region will therefore sound very different from the low Q sound of the ESL everywhere else in the audio bandwidth. The result of this high Q behavior is that the quality of electrostatic bass is flaccid, poorly controlled, sloppy, and boomy.


Yet another problem with ESL bass is that it is very non-linear. The fundamental resonance causes a large peak in the bass followed by a sharp drop in output below it.


Most full range ESL manufacturers allow this peak to persist in a desperate effort to get their ESL to sound "bassy." But the result is very non-linear

bass response, with no truly deep bass output at all. It becomes a "one
note wonder."

In short, electrostatic bass has inadequate output, poor frequency response with no deep bass frequencies, and its high Q behavior results in poor bass quality. Therefore, I don't consider electrostatic bass to be satisfactory for a high performance speaker.


The solution is to use magnetic drivers for the bass. Of course, bass from magnetic drivers is less-than-perfect too. While magnetic woofers can produce great power and depth that no ESL can hope to match, and they can be made to have excellent frequency response, they also have high Q behavior (although not nearly as high as the Q of the fundamental resonance of an ESL). So what can be done to fix the problems of magnetic drivers so that they can mate well with an ESL?


Much of the problem with magnetic woofers has to do with their enclosures.

The typical closed box (infinite baffle) and vented (bass reflex) enclosures greatly raise the Q of the driver and are largely responsible for the bloated, boomy, and muddy quality we usually hear from most magnetic woofers.

However, it is possible to eliminate these faults with the use of a transmission line loading and other techniques. This is another, long technical discussion that I do not have time to describe in this opus, but anybody who wishes can phone me (303 838 8130) for a detailed analysis and engineering solutions.


The bottom line here is that it is possible to achieve deep, powerful, low Q bass from a dynamic woofer. I do so by using transmission line loading, custom-built low mass and low Q drivers, magnetic damping systems, direct coupling to high damping factor amplifiers, and eliminating passive crossovers with their series inductors. The performance from such a magnetic bass system is far superior to electrostatic bass because it is loud, dynamic, has no resonance, is low Q, and has flat frequency response right down to 20 Hz.


The woofer systems in traditional hybrid speakers have failed to match the low Q behavior of an ESL. Furthermore, their passive crossovers have not had steep enough slopes or low enough crossover points to eliminate the woofer from affecting the critical midrange.


This midrange issue is very important. After all, we would all agree that no matter how good a woofer is, it simply cannot match the magnificent detail and delicacy of an ESL in the midrange. A superb hybrid must eliminate all woofer energy from the midrange. Traditional hybrids have failed to do so and therefore have earned a reputation as being unable to integrate the two different drivers.


But it does not have to be that way. I have resolved this integration problem by having developed a low Q, magnetic woofer system that has better sound quality and frequency response than any electrostatic woofer. I have eliminated the woofer from the midrange by using very low crossover points combined with extremely steep crossover slopes (48 dB/octave). I use electronic crossovers and bi-amp the system. The result is a no-compromise hybrid that easily out-performs any full range ESL by a huge margin.


This makes it possible for you to reproduce Row A concert hall levels and the sound of a grand piano or drum set at live volume in your listening room. All while maintaining the magnificent detail, transient response, and clarity of an ESL. Full dynamic range and spectacular performance is now possible.


So to answer your original question, I use a hybrid because it is the only way to achieve the high output, full dynamic range, flat frequency response, and low Q performance that is essential for a high performance speaker. A full range ESL simply cannot match the performance of my hybrid design.

Because I want the best performance possible, I do not manufacturer or sell full range ESLs anymore.

Using a hybrid is not a cop-out. Using a full range ESL is. It is much easier and cheaper to make a full range ESL than a quality hybrid. But I'm not willing to sacrifice the best performance for the sake of lower cost and ease of manufacturing that full range ESLs offer.


My Ultrastat panels can be ganged to make huge, full range ESLs. I have a few customers who have done so. But the laws of physics cannot be circumvented. The fact remains that ESL bass is of poor quality and simply cannot supply realistic sound reproduction.


I am not willing to deceive my customers to make sales. So I do sell inferior products like full range ESLs.


Turning to your question of dispersion, there is insufficient time to cover this topic in detail as this response is already far too long. If you have not already done so, I suggest you read my White Paper on the subject for a thorough explanation of my logic and reasoning behind my use of narrow dispersion panels. There I explain why I abandoned the wide-dispersion, curved ESL that I invented in favor of narrow dispersion, planar panels.

The link is:
http://sanderssoundsystems.com/technical-white-papers/dispersion-wp


In closing, let me say that once you hear my speakers, you will understand.

You are welcome to use my 30-day, in-home, free trial so that you can hear my speakers in your own home with your own associated electronics and familiar source material. You can then judge my engineering decisions for yourself and report your findings to your friends.

Great listening,

-Roger
 

MylesBAstor

Well-Known Member
Apr 20, 2010
11,236
81
1,725
New York City
Roger-

Thanks for taking the time to give us insight into your speaker designs and philosophy. I also think it's interesting to see all the people in the industry who contributed to the ML design--and certainly taking estat reliability a few notches up.

I know you can't comment on other manufacturers designs but Roger West of Sound Lab also talks about estats' problems in the lows, namely the "drum head" resonance you referred to and dipole cancellation effects (like what the Maggies run into). But had you considered an approach like Roger took to the resonance problem?
 

ggendel

New Member
May 26, 2010
17
1
1
One piece that shows the "drum head" resonances dramatically is ELP's Lucky Man. Near the end there is a tone that I've heard both Mark Wright and SoundLabs electrostatics go into fits over.

I use a British pressing as a mainstay for speaker comparisons. I've found the richness of the vocal harmonies come alive dramatically in electrostatics over any of the cone speakers I've tested.

BTW, I heard this piece on the Hill Plasmatronics and the vocals were too harsh. Now that was an interesting speaker.

-Gary
 

MylesBAstor

Well-Known Member
Apr 20, 2010
11,236
81
1,725
New York City
One piece that shows the "drum head" resonances dramatically is ELP's Lucky Man. Near the end there is a tone that I've heard both Mark Wright and SoundLabs electrostatics go into fits over.

I use a British pressing as a mainstay for speaker comparisons. I've found the richness of the vocal harmonies come alive dramatically in electrostatics over any of the cone speakers I've tested.

BTW, I heard this piece on the Hill Plasmatronics and the vocals were too harsh. Now that was an interesting speaker.

-Gary

Hill Plasmatronics? Now there was an interesting speaker. Hadn't thought about them in a while but remember an audiophile up in Westchester who had them and the old He tank too! He ended up with Apogees and Krell in the long run.

I haven't heard the newest SLs but outside of the resonance, I found the early SLs horribly inefficient (one of the few worse was the Toltec from France; they were about 75 dB and barely played at a whisper at one CES) . A friend had a pair hooked up to a pair of Melos 400s amps that Mark had modded so as to put out around 700 wpc and even these clipped on the SLs :(
 

Gregadd

WBF Founding Member
Apr 20, 2010
10,517
1,774
1,850
Metro DC
You are welcome to use my 30-day, in-home, free trial so that you can hear my speakers in your own home with your own associated electronics and familiar source material. You can then judge my engineering decisions for yourself and report your findings to your friends.
If I come across a spare $13K, I'll take you up on that offer. More likely I'll catch it at one of the shows.
 
Last edited:

kach22i

WBF Founding Member
Apr 21, 2010
1,591
210
1,635
Ann Arbor, Michigan
www.kachadoorian.com
Hill Plasmatronics? Now there was an interesting speaker.
While looking up information on that speaker I found this interesting exchange:

http://www.audioasylum.com/cgi/vt.mpl?f=mug&m=149818
josh358
Industry Professional

Posts: 486
Joined: February 9, 2010
So you don't hear the mylar? Interesting, wonder why. I've heard it in every planar magnetic or electrostatic I've ever heard.
.......................................................................................................................

RE: Coherence and colouration, posted on March 6, 2010 at 18:38:59
DkB
Audiophile

Posts: 910
Joined: June 25, 2003
Certainly, I can hear "mylar clang" in many panel speakers, like Maggies and Martin Logans. In the Maggies, it is a plasticky undertone that underlies the music. In the Martin Logans, it is more like a tinny sheen.

I think you hear the tuning material because of the overall resonance of the mylar sheet, as determined by tension and mass (like the tuning of a drum skin.)

However, as mentioned in the above post, the Sound Labs have quite effectively dealt with this resonance by breaking up the panel's into a diversity of frequencies, spreading the resonances so there is no single overriding resonance frequency.

Certainly, there is no escaping the signature of any object producing sound, but for the SOund Labs, that trace has been suppressed to a highly miniscule level. More so than many transducers, panel or cabinet.

Has anyone heard "Mylar"?

The only time I have is in my DIY ripple speaker experiment, made a video/audio recording of it.
http://www.whatsbestforum.com/showthread.php?522-The-Point-Source-Principle-Ripple-One-Experiment
 

MylesBAstor

Well-Known Member
Apr 20, 2010
11,236
81
1,725
New York City
While looking up information on that speaker I found this interesting exchange:

http://www.audioasylum.com/cgi/vt.mpl?f=mug&m=149818


Has anyone heard "Mylar"?

The only time I have is in my DIY ripple speaker experiment, made a video/audio recording of it.
http://www.whatsbestforum.com/showthread.php?522-The-Point-Source-Principle-Ripple-One-Experiment

That person should at least give credit where credit is due for those remarks. Hardly an original thought and talked about by HP in TAS many, many times over the years. HP was in part trying to describe the "coloration" of both speakers and used this as an analogy. Unfortunately the problem is, there are so many variables in the design, how do they know it's the plastic? This is the same thing as for correlations or as my stat teacher said, just because you're born in a garage doesn't make you a car (yeah ok that was about the only thing I got out of the class :) ). Having played with Maggies, I'd attribute their "coloration" to their crossover components rather than the design eg. at that time they were using mylar caps, etc in the xover. In the ML, it's just as complex.

But how does this person know the reason isn't an impedance dip as with MBLs as Ralph Karsten eloquently pointed out on Audiogon not too long ago that drives amps crazy?
 

Angela

WBF Technical Expert
May 24, 2010
141
0
0
Conifer, Colorado
Roger-

Thanks for taking the time to give us insight into your speaker designs and philosophy. I also think it's interesting to see all the people in the industry who contributed to the ML design--and certainly taking estat reliability a few notches up.

I know you can't comment on other manufacturers designs but Roger West of Sound Lab also talks about estats' problems in the lows, namely the "drum head" resonance you referred to and dipole cancellation effects (like what the Maggies run into). But had you considered an approach like Roger took to the resonance problem?

reply from Roger

Hi Myles,

When you refer to Soundlab's approach to the resonance problem, I assume you are referring to Roger West's "distributed resonance" design. By way of review for readers who may not be familiar with this technique, this is where the speaker's diaphragm is broken up into sections of various sizes.
The idea is to produce several different, smaller resonant frequencies instead of one big one.

In theory, this technique should not only reduce the problem of extremely bad ESL bass frequency response, but it could be used to compensate for phase cancellation, which is a major reason that ESLs can't reproduce loud bass.

The different sized diaphragm sections are produced by using diaphragm to stator spacers that are at different distances from each other. Martin Logan also uses this construction method, although Soundlab claims to have a patent on it.

I have experimented with this technique extensively. Unfortunately, it has an unforeseen problem that prevents it from working as expected.

The distributed resonance idea works for isolated, widely-separated, electrostatic panels that can operate as individual drivers. Each will exhibit its own, unique resonance depending on its size and diaphragm tension.

But when you take those same drivers and bring them together to form a single speaker, something weird happens -- the panels now operate as one.
They no longer have individual resonances.

The reason for this is that the panels are immersed in air, which has significant mass. At audio frequencies, the air mass around an ESL takes on the consistency of a gel -- or at lower frequencies, a liquid. The air mass around the speaker then vibrates as a unit (as a bowl of Jell-O would vibrate as one mass), preventing individual resonances from forming.

In other words, distributed resonance does not work. You cannot break up the fundamental resonance of an ESL into multiple discrete ones.

However, the different spacer distances do have an effect on the speaker's fundamental resonance because the diaphragm no longer behaves as though it were at a constant tension. This reduces the magnitude of the fundamental resonance widens its bandwidth. So it does help to improve the massive irregularity of the frequency response, although at the expense of making more of the audio bandwidth adversely affected by the resonance.

Note carefully that even if the technique of distributed resonance worked, the bass performance would still exhibit high Q behavior with its awful effect on bass quality. After all, by definition, all resonances are high Q, and by having more of them over a wider bandwidth, you would simply be introducing the bad, high Q behavior of a resonance over more of the audio bandwidth. This certainly degrades the performance of the speaker.

Finally, distributed resonance would not solve the problem of poor bass frequency response. Instead of having one resonance, you would now have many. So the frequency response would be ragged over a wider bandwidth.

In fairness, the case could be made that the magnitude of the multiple resonances would be smaller than that of a single resonance. Some might find that preferable. But the fact remains that the frequency response that consists of a series of resonances would still be far from linear.

In summary, I find that distributed resonance does not work as claimed.
Furthermore, it degrades the otherwise lovely, low Q, linear frequency response exhibited by ESLs over the rest of their frequency range.

There are only two ways I have found to eliminate the high Q behavior in the bass of an ESL. The first is to avoid driving the ESL near its fundamental resonance. A hybrid does exactly that. But there is a second method that can be used in full range ESLs to make them have low Q behavior over their entire bandwidth.

I have been experimenting with motional feedback. By measuring the motion of the diaphragm and comparing it to the musical signal, the errors caused by overshoot, ringing, and frequency response errors can be identified. An error signal is thus produced that can be fed to the amplifier that is driving the ESL's diaphragm to actively stop the ringing and overshoot.

This works quite well. It not only stops the high Q behavior, but can be used to produce perfectly linear bass frequency response. It will also compensate for phase cancellation. Nice!

But motional feedback has serious problems with implementation. The major one being that a microphone cannot be used to produce the error signal as the delayed sounds from room acoustics get into and confuse the electronic system.

A laser can be used to measure the diaphragm motion. It is immune from room acoustics since it is not measuring air motion, but the motion of the diaphragm directly.

Such a system starts to get very complex, expensive, and impractical for home use. So we have also been experimenting with simpler systems such as using the speaker's own capacitance changes (the capacitance changes as the diaphragm to stator spacing changes) to measure the motion indirectly. This is simpler than using a laser, but preventing the drive signal from influencing the measurement is tricky. More work is in order here.

While motional feedback can solve the frequency response and high Q problems of an electrostatic woofer, it still doesn't address the main issue, which is poor output. The fact remains that a dipole radiator suffers from extreme phase cancellation and radiation resistance losses in the bass that prevent it from producing deep, loud, dynamic bass.

There is no apparent solution to this problem of feeble bass output. Until there is, the dream of a high performance, full-range, crossoverless ESL will remain an elusive goal.

Fortunately, hybrid systems have now reached the point where they perform as well as full range ESLs with regard to detail and clarity. Integration of the two drivers is flawless. And unlike a full range ESL, a hybrid can produce very high output so that the full dynamic range of live music is fully realized. So while I continue to experiment with electrostatic woofers, I don't really see any need for them.

Great listening,
-Roger
 

MylesBAstor

Well-Known Member
Apr 20, 2010
11,236
81
1,725
New York City
reply from Roger

Hi Myles,

When you refer to Soundlab's approach to the resonance problem, I assume you are referring to Roger West's "distributed resonance" design. By way of review for readers who may not be familiar with this technique, this is where the speaker's diaphragm is broken up into sections of various sizes.
The idea is to produce several different, smaller resonant frequencies instead of one big one.

In theory, this technique should not only reduce the problem of extremely bad ESL bass frequency response, but it could be used to compensate for phase cancellation, which is a major reason that ESLs can't reproduce loud bass.

The different sized diaphragm sections are produced by using diaphragm to stator spacers that are at different distances from each other. Martin Logan also uses this construction method, although Soundlab claims to have a patent on it.

I have experimented with this technique extensively. Unfortunately, it has an unforeseen problem that prevents it from working as expected.

The distributed resonance idea works for isolated, widely-separated, electrostatic panels that can operate as individual drivers. Each will exhibit its own, unique resonance depending on its size and diaphragm tension.

But when you take those same drivers and bring them together to form a single speaker, something weird happens -- the panels now operate as one.
They no longer have individual resonances.

The reason for this is that the panels are immersed in air, which has significant mass. At audio frequencies, the air mass around an ESL takes on the consistency of a gel -- or at lower frequencies, a liquid. The air mass around the speaker then vibrates as a unit (as a bowl of Jell-O would vibrate as one mass), preventing individual resonances from forming.

In other words, distributed resonance does not work. You cannot break up the fundamental resonance of an ESL into multiple discrete ones.

However, the different spacer distances do have an effect on the speaker's fundamental resonance because the diaphragm no longer behaves as though it were at a constant tension. This reduces the magnitude of the fundamental resonance widens its bandwidth. So it does help to improve the massive irregularity of the frequency response, although at the expense of making more of the audio bandwidth adversely affected by the resonance.

Note carefully that even if the technique of distributed resonance worked, the bass performance would still exhibit high Q behavior with its awful effect on bass quality. After all, by definition, all resonances are high Q, and by having more of them over a wider bandwidth, you would simply be introducing the bad, high Q behavior of a resonance over more of the audio bandwidth. This certainly degrades the performance of the speaker.

Finally, distributed resonance would not solve the problem of poor bass frequency response. Instead of having one resonance, you would now have many. So the frequency response would be ragged over a wider bandwidth.

In fairness, the case could be made that the magnitude of the multiple resonances would be smaller than that of a single resonance. Some might find that preferable. But the fact remains that the frequency response that consists of a series of resonances would still be far from linear.

In summary, I find that distributed resonance does not work as claimed.
Furthermore, it degrades the otherwise lovely, low Q, linear frequency response exhibited by ESLs over the rest of their frequency range.

There are only two ways I have found to eliminate the high Q behavior in the bass of an ESL. The first is to avoid driving the ESL near its fundamental resonance. A hybrid does exactly that. But there is a second method that can be used in full range ESLs to make them have low Q behavior over their entire bandwidth.

I have been experimenting with motional feedback. By measuring the motion of the diaphragm and comparing it to the musical signal, the errors caused by overshoot, ringing, and frequency response errors can be identified. An error signal is thus produced that can be fed to the amplifier that is driving the ESL's diaphragm to actively stop the ringing and overshoot.

This works quite well. It not only stops the high Q behavior, but can be used to produce perfectly linear bass frequency response. It will also compensate for phase cancellation. Nice!

But motional feedback has serious problems with implementation. The major one being that a microphone cannot be used to produce the error signal as the delayed sounds from room acoustics get into and confuse the electronic system.

A laser can be used to measure the diaphragm motion. It is immune from room acoustics since it is not measuring air motion, but the motion of the diaphragm directly.

Such a system starts to get very complex, expensive, and impractical for home use. So we have also been experimenting with simpler systems such as using the speaker's own capacitance changes (the capacitance changes as the diaphragm to stator spacing changes) to measure the motion indirectly. This is simpler than using a laser, but preventing the drive signal from influencing the measurement is tricky. More work is in order here.

While motional feedback can solve the frequency response and high Q problems of an electrostatic woofer, it still doesn't address the main issue, which is poor output. The fact remains that a dipole radiator suffers from extreme phase cancellation and radiation resistance losses in the bass that prevent it from producing deep, loud, dynamic bass.

There is no apparent solution to this problem of feeble bass output. Until there is, the dream of a high performance, full-range, crossoverless ESL will remain an elusive goal.

Fortunately, hybrid systems have now reached the point where they perform as well as full range ESLs with regard to detail and clarity. Integration of the two drivers is flawless. And unlike a full range ESL, a hybrid can produce very high output so that the full dynamic range of live music is fully realized. So while I continue to experiment with electrostatic woofers, I don't really see any need for them.

Great listening,
-Roger

Roger: Thank you for a VERY informative and detailed response. I learned an awful lot from your post!!!

I think one issue is that many feel there is always a tradeoff between the "speed" of a dipolar bass system such as the Maggies and a dynamic speaker system. To my ears, the Maggies have an incredible ability to define say notes on a standup bass as opposed to the dynamics and amount of air moved by a cone driver. That said, what do you think the lower limit is and what factors might influence how low you could extend the frequency response on an electrostatic panel before crossing over to a dynamic driver or say your transmission line approach.
 

Angela

WBF Technical Expert
May 24, 2010
141
0
0
Conifer, Colorado
Roger: Thank you for a VERY informative and detailed response. I learned an awful lot from your post!!!

I think one issue is that many feel there is always a tradeoff between the "speed" of a dipolar bass system such as the Maggies and a dynamic speaker system. To my ears, the Maggies have an incredible ability to define say notes on a standup bass as opposed to the dynamics and amount of air moved by a cone driver. That said, what do you think the lower limit is and what factors might influence how low you could extend the frequency response on an electrostatic panel before crossing over to a dynamic driver or say your transmission line approach.

More from Roger
-------------------------------
Hi Myles,

Thank you for your kind compliments on the educational value of my responses. I sincerely try to help audiophiles find the true reasons for what they are hearing. There is far too much hype and voodoo science in this industry, which makes it very confusing for many music lovers. Hopefully I can help with this issue.

Turning to your questions, you bring up several issues that need to be clarified. First, there is no question in my mind that you hear a very different bass quality from Maggies than you do from conventional closed box (infinite baffle) or vented (bass reflex) woofer systems. However, there are several possible causes that could explain the differences you hear. So you need to be very careful drawing conclusions about cause/effect relationships.

There are many differences between Maggies and box speakers besides the fact that one is a dipole and the other is a monopole. While the dipole/monopole issue COULD explain the difference you hear, there are several other differences that could also explain the phenomenon. So you should not assume that the Maggie woofer sounds better simply because it is a dipole.

Specifically, there is a big difference in Q between conventional bass systems and the Maggies. Also there is certainly a big difference in frequency response.

In my tests, I have found that the Q of the typical, conventional woofer system (closed box/vented systems) is much higher than that of a Maggie. This causes a lot of overshoot and ringing in conventional woofer systems that adds a lot of energy to the reproduced sound that is not present in the bass instrument itself. Of course, this dramatically colors the sound.

In other words, conventional woofer systems corrupt the sound significantly. Their overshoot and ringing adds energy that masks the subtle harmonics that give bass instruments their detail and clarity. By comparison, a Maggie's woofer has far lower Q and is therefore "cleaner", which makes it possible for you to hear the detail in the bass instruments much better.

Yet another factor is frequency response. Being a dipole, the bass from a Maggie lacks real depth, just like electrostatic bass. However, unlike an ESL, the Maggie is magnetically driven, so it has a lot of excursion and can produce fairly high output from the mid-bass on up (assuming many hundreds of watts of amplifier power are available). Still, the lowest frequencies will be weak compared to a conventional woofer system.

When you reduce the deep bass, the mid-bass and higher frequencies are more apparent since they are not masked as much by the deeper frequencies. So the mid and upper bass from the Maggies may sound more clear simply because these higher frequencies are not masked by deep bass.

You can easily prove this to yourself by using a DSP (Digital Signal Processor) or graphic equalizer to roll off the bass below 50 Hz on a conventional woofer system. The overall bass sound quality will become "cleaner," although the overall "punch" and depth will be somewhat reduced.

I worked on this problem of high Q behavior in conventional woofer systems for 17 years as it was the main problem in getting a magnetic woofer to integrate well with an ESL. The problem mostly has to do with enclosures. Here is the problem:

Consider an infinite baffle (closed box) enclosure. Its sole purpose is to stop phase cancellation, which it does perfectly by preventing the front wave of the woofer from being cancelled by the rear wave, which is 180 degrees out of phase.

But this closed box introduces several problems. First, think of what happens when the woofer is driven into the enclosure -- it compresses the air inside the enclosure.

Now consider what happens when the woofer needs to come back to its neutral point and stop. The compressed air inside the box acts like a spring that pushes the woofer towards the neutral point. It doesn't stop pushing until the woofer actually reaches the neutral point. This means that the action of the compressed air spring strongly pushes the woofer BEYOND the neutral point.

Of course, the relatively high mass of the driver means that it has a lot of inertia. Since there is very little to counter the kinetic energy of the woofer and bring it to a quick stop, the woofer coasts far past the neutral point until its suspension (and electrical damping by the amplifier) brings it to a stop. The combination of the compressed air spring and inertia results in severe overshoot.

As the woofer reaches the limits of its suspension's excursion, it reverses direction and sails past the neutral point once again, but in the opposite direction. There is some friction from the air and damping from the amplifier that uses up some of this energy, so the distance that the woofer's cone overshoots the neutral point the second time is less than the first time. As a result, the woofer gradually comes to a stop after several cycles of overshoot (known as "ringing").

This high Q behavior introduces a great deal of extra energy to the original sound that should not be there. It can certainly be considered distortion, and subjectively, it adds a "heaviness" and "muddy" quality that degrades the quality of the sound.

But that's not the only problem. Consider the energy that the woofer generates in your room. I think you would agree that a conventional woofer can produce really powerful bass in a room that is hundreds of times larger than the woofer's enclosure.

A woofer generates the same amount of energy on both sides of its cone. The high energy that is enough to rattle the windows and floor of your listening room is magnified hundreds or thousands of times as you cram all the energy into a relatively tiny enclosure. So you can start to appreciate that there is a lot of extremely powerful sound energy inside that tiny woofer cabinet!

What happens to all this energy? Well, since there is really nothing to absorb it, the energy must escape from the enclosure somehow. Some of it is released by bending the flat sides that are present on most woofer enclosures. This is the infamous "box talk" with which you are familiar. It is a major contributor to the relatively high distortion and coloration in conventional woofer systems.

But most of the energy is going to come through the weakest point in the enclosure. That point is the woofer's cone, which is very thin and weak compared to the thick wall of the enclosure. The result is that there is a very large amount of delayed energy coming from the woofer's cone after it has bounced around inside the enclosure for a short time. This "box energy" is loaded with resonances, distortion, and phase error.

A vented enclosure is even worse. It is designed to resonate as certain frequencies. This causes even worse overshoot and ringing. It frequency response is far more non-linear than a closed box. And the "garbage" is different at different frequencies.

All this "garbage" is added to what should otherwise be a very pure reproduction of the original bass instrument. And when you add the garbage, plus all the overshoot and ringing, it is a wonder that conventional woofers are even listenable. Actually, the only reason we tolerate conventional woofers is because our ears are quite insensitive to flaws in bass frequencies. Nevertheless, any thoughtful audiophile recognizes that most bass speaker systems are seriously compromised. They don't sound very realistic at all.

In summary, conventional closed box and vented enclosures are truly awful designs. The only reason they are in common use is because they are cheap and easy to build and relatively small woofer systems can be made with them. But I do not consider them satisfactory for high performance speaker systems as they simply fail to produce "high fidelity" sound.

A transmission line eliminates all these problems. By way of review, a TL (Transmission Line) is essentially a long (typically around 8 feet), tapered pipe that is filled with fibrous material. The woofer is mounted in one end and the other end is open. Now let's look at how this affects the problems of overshoot, ringing, and "garbage" (enclosure energy/distortion/resonance/box-talk).

First, the TL is long enough that phase cancellation cannot occur at audio frequencies because the rear radiation has to travel so far to get around to the front that the phase is shifted well beyond 180 degrees. In fact, we generally design TLs so that the phase is shifted 360 degrees at some deep bass frequency so that the rear radiation from the woofer actually helps boost the front radiation. This really helps support the deep bass frequencies, which are always falling in all woofer designs due to radiation resistance losses. This is one of the reasons that TL's are known for having really powerful deep bass performance and flat frequency response to lower frequencies than conventional enclosure types.


As an aside, since no woofer can produce flat deep bass response due to radiation resistance losses, all woofers need to be equalized below 50 Hz or so. One of the really nice things about digital crossovers is that they offer this feature so you can get truly flat frequency response right down to 20 Hz (assuming adequate woofer excursion capability and many hundreds of watts of amplifier power).

Secondly, consider the compressed air spring problem. Because the TL is open at the far end, the woofer cannot compress the air inside the TL. The air simply escapes through the port rather than being compressed.

Even more importantly is the effect the fibrous material (damping material) has on the air as it passes through the TL. The air runs into the millions of fibers in the TL, which cause a lot of friction.

Therefore, the air cannot pass freely and quickly through a TL. It is slowed down by the friction from the damping material and a lot of energy is needed to push it through the line.

Now what happens when the woofer needs to come back to the neutral point? There is no compressed air spring to push it. Furthermore, the woofer's inertia is used up as its kinetic energy is converted to heat by dragging the air back through the friction of the damping material inside the TL. It is like having a little "shock absorber" attached to the woofer.

As a result, the woofer simply stops at the neutral point. There is no overshoot. Without any overshoot, there can be no ringing.

Now let's look at the rear energy coming off the woofer's cone. For the mid-bass frequencies on up, the air motion is relatively small and virtually all of this motion is reduced to heat by friction with the damping material. There is no possibility for free motion that would cause resonances inside the enclosure that could escape and color the sound. Also, the TL is tapered so will generate an infinite number of infinitely small resonances rather than just a couple of huge ones as is the case in typical, rectangular enclosures. So there is no box talk and no "garbage" coming through the woofer cone.

The deep bass frequencies contain a great deal of power and very long wave lengths. For them, the TL is not long enough to completely convert them to heat. Therefore, a lot of deep bass energy is released from the port at the end of the TL. But remember, it has had its phase shifted so that it comes out in-phase with the front of the woofer, so it does useful work supporting the deep bass frequencies.

Note carefully that this port energy is not produced by any resonance phenomenon. It is clean. It is the original musical signal that has been phase shifted. So it does not produce distortion that degrades the sound.

When a low mass, low Q woofer is used in a TL, and is driven by a high damping-factor amplifier, the problems of overshoot, ringing, and distortion from a conventional cone woofer are eliminated. As a result, a good TL sounds just as clean, pure, and detailed as a dipole. But it has the advantage over a dipole of having really deep bass and being able to play at earth-shaking output levels.

There is one more detail to bring up in this quest for a low Q woofer, and that is the effect that passive crossovers have on amplifier damping. Understand that the Theil-Small parameters consist of Qms, Qes, and Qts. Qms refers to the mass of the driver. It is considered to be the mechanical Q of the speaker and is determined by the mass of the moving parts of the speaker and the behavior of the suspension system.

Generally, for a given suspension, the higher the mass, the more the driver will tend to overshoot and ring. For a given mass, the softer the suspension, the less the driver will overshoot and ring. In other words, a low Qms woofer will have low mass and a very soft suspension. These factors will also reduce its free air resonance point.

Qes refers to the electrical Q. Understand that when you move the voice coil through its permanent magnetic field, an electrical current will be generated in the voice coil. If you place a load across the voice coil, this current will do work. Work requires energy. The energy comes from the motion of the woofer cone.

If you short out the voice coil, the most energy will be required from the woofer cone. If you leave the voice coil as an open circuit, no energy will be required from the woofer cone.

You can easily feel this. Just disconnect a speaker from its amplifier, then push the woofer cone back and forth with your fingers as you alternately open and short across the speaker binding posts. You will clearly feel a resistance to the woofer's motion when you short the voice coil.

This is electrical damping. It becomes greater if the permanent magnet is more powerful.

Qes is highly desirable. It is at its best when the woofer is connected directly to a very low impedance amplifier. The "damping factor" is the ratio between the impedance of the voice coil and the amplifier's output impedance.

For example, if you have an 8 ohm voice coil in your woofer and you connect it to the 8 ohm taps of a tube amplifier, the damping factor will be 1 (one). If that same 8 ohm voice coil is attached to a powerful solid state amplifier whose output impedance is 0.02 ohms (a typical value), then the damping factor would be 400. The lower Q of the solid state amp system causes the bass to sound "tighter" than from most tube amps.

I have simplified the damping factor issue somewhat because I did not mention global feedback. Suffice it to say that global feedback incorporates the speaker load into the feedback loop and significantly increases the damping factor of an amplifier. So a moderate amount of global feedback is very helpful at controlling the woofer's motion.

It is very important to note that the addition of any series resistance between the woofer and the amplifier will tremendously degrade the damping factor. This is where passive crossovers are so evil. They always have one or more inductors in series with the woofer.

An inductor consists of a long, thin piece of wire wound into a coil. It has significant resistance. So it will always degrade the damping factor of the system.

This is one of the major reasons that passive crossovers should not even be considered for high performance speakers. Electronic crossovers are far superior because their amplifiers are directly connected to their drivers without any intervening inductors or resistors that ruin the damping factor.

As an aside, I always get a bit of a chuckle when I hear how adamant some audiophiles are about using huge, low-resistance speaker cables when used with passive crossovers. They think they are helping the situation by using a very low resistance cable, but they simply don't realize that they are connecting it to tens of feet of tiny magnet wire that forms an inductor. This eliminates any beneficial effect the large speaker cable might have offered. They might as well have used a tiny speaker cable since the wire in the inductor is far longer and has more effect than their huge cable.

The Qts is the combination of the mechanical and electrical Q of the driver. As you can see, it will be inaccurate if an enclosure is designed using it without consideration of the effects of a passive crossover.

Finally let me turn to your question about the lower limit of dipole operation and what crossover point to use. There is no simple or direct answer to this. There are many factors that must be considered and none are cast in stone. But I will list a few that will drive your decision.

First, as described in a previous communication, the ESL must not be operated at or near its fundamental resonance or it will exhibit high Q behavior. Exactly how close to the resonance you can go depends on the type of crossover filter used and how steep its slopes are. In my speakers, I find that I can get to within about one octave by using 48 dB/octave slopes in the crossover.

If you use gentler slopes, you must stay further away from the resonance to avoid exciting it. For example, if you use 24 dB/octave slopes, you will need to cross over about 2 octaves above the resonance, etc.

Diaphragm tension and suspension spacing (the ratio of the diaphragm to stator spacer distances compared to the diaphragm to stator spacing) will have a very powerful effect on the fundamental resonance. Lower diaphragm tensions or larger suspension spacing will reduce the frequency of the fundamental resonance (as will altitude, temperature, and humidity).

But lowering the resonance point will also reduce the stability of the diaphragm, reduce the polarizing voltage possible, and reduce the sensitivity of the speaker. So a series of compromises is necessary to reach an optimum among these factors.

The minimum dimension (usually the width) of the ESL will determine the phase cancellation frequencies and therefore how much midrange equalization is required to produce flat frequency response at the crossover point. Each dB of equalization has the effect of reducing the overall output of the speaker by the same amount. So as you reduce the width of the ESL, you quickly find that you run out of available excursion or amplifier power or both because of the added equalization required.

Not all ESLs use midrange equalization. But they must somehow compensate for phase cancellation or they will sound very bright, thin, and anemic. Those that don't use EQ usually increase the amount of radiating area with decreasing frequency. This is a big compromise that is an entirely different subject that I don't have time to discuss in this article. But the problem is the same regardless of whether EQ or variable radiating area is used.

So you will need to trade off a low crossover point for adequate output. This is a critical design compromise where the width of the ESL, the output desired, and the crossover point all have to be juggled to reach a reasonable compromise.

Diaphragm excursion is a critical factor. At some high output level, depending on frequency, the diaphragm will slap the stator, and this produces an absolute limit on output.

The diaphragm to stator spacing determines the maximum excursion possible, which has a profound effect on the maximum potential output of the speaker and its sensitivity. A larger diaphragm to stator spacing can potentially produce higher outputs at lower frequencies. But it reduces sensitivity and requires much higher drive voltages, which are very difficult to produce.

Obviously, if you can get more excursion with high output, you can use a lower crossover point for a given output level. But there are severe limitations on how much excursion you can get from a practical perspective. So you will have to trade off total output against the crossover point.

One nice thing about digital crossovers is that you can store many different crossover arrangements. This makes it possible to use a relatively high crossover point when you want or need very high output levels, and then quickly switch to a much lower crossover point (or even full-range, crossoverless operation) for quiet listening.

As you can see, there are a whole host of factors to consider when deciding what crossover point to use. Therefore, I cannot give you any hard and fast rules without knowing all the various parts of the speaker design that affect the crossover point.

I hope this information has been helpful.

Great listening,
-Roger
 

Attachments

  • 695..gif
    695..gif
    985 bytes · Views: 760

Gregadd

WBF Founding Member
Apr 20, 2010
10,517
1,774
1,850
Metro DC
Well while we have your attention... I suppose I prefer electrostatic bass because of its transient response. That is to say I tend to prefer less ringing and slightly decreased output to increased output and more ringing. Many have tried a "servo" to get the woofer to start and stop properly. What role does a servo play?
 

Angela

WBF Technical Expert
May 24, 2010
141
0
0
Conifer, Colorado
Well while we have your attention... I suppose I prefer electrostatic bass because of its transient response. That is to say I tend to prefer less ringing and slightly decreased output to increased output and more ringing. Many have tried a "servo" to get the woofer to start and stop properly. What role does a servo play?

(Reply from Roger):


A servo is another name for a motional feedback system. The general idea of such systems is to compare the motion of the speaker's cone to the musical input signal, identify the difference between the two, and use the resulting error signal to drive the amplifier to force the cone to move exactly like the musical signal.

These can be made to work in several ways. For magnetic woofers, the most common technique is to add a couple of turns of wire around the voice coil former that is electrically distinct from the voice coil itself.

As the voice coil moves back and forth in its permanent magnet field, this detector coil generates a small current. This current will accurately describe the motion of the voice coil. It therefore can be used to compare to the input signal to see how closely the woofer follows the input signal.

The current from the detector coil is fed into one of the inputs of a differential opamp and the musical signal is fed into the other. The opamp then generates an output signal that reveals any difference between the two.
This is called an error signal.

The error signal then has its phase inverted and is fed into the woofer amplifier's input along with the musical signal. The amplifier will then apply an opposing drive force to the woofer to make it to move only as allowed by the musical signal.

Motional feedback systems can offer a significant improvement in bass response. But they are not quite perfect and have several problems that limit their use.

The main problem is that woofer cones are not rigid. Therefore, the woofer cone does not follow the motion of the voice coil precisely. So even if the voice coil is made to move exactly as the input signal, the cone will not follow it perfectly (although it will be much better than without motional feedback).

The biggest problem in this regard is the phenomenon of cone "break up."
Large woofer cones only operate as pistons at low frequencies. At some higher frequency (usually a little above 100 Hz), the mass of the cone will produce so much inertia that the entire cone will not be able to follow the voice coil. At this point, the cone material will flex as the outer edges of the cone tend to remain stationary while the inner section of the cone move rapidly with the voice coil. The cone is then considered to be operating in "breakup mode."

Of course, the outer edge of the cone doesn't actually come to a complete stop. It tries to follow the inner section of the cone and by failing to do so, it produces distortion. Also, the frequency response is altered as the radiating area of the woofer changes. The flexing cone also produces waves in its surface that cause harmonic distortion.

Woofer driver manufacturers expend a tremendous amount of engineering effort trying to get their cones to break up smoothly, with minimum distortion, and with reasonably smooth frequency response. This is why you will note that woofer cones are not straight-sided and made of stiff material as you would expect if the cone acted like a rigid piston.

Instead, most woofer cones have shallow, curved surfaces that are deliberately made to flex in a controlled and desirable way. They usually are made of soft material like baxtrene, polypropylene, or paper since these materials tend to have excellent damping characteristics and therefore minimize distortion and ringing. They produce "soft" breakup characteristics.

A lot of work has been done in recent years to produce cones of stiffer material like aluminum, titanium, and carbon fiber. Sometimes these are made into very stiff shapes like straight-sided cones with steep sides.

Such woofer cones operate to higher frequencies before experiencing breakup.
But they have rather abrupt break-up characteristics and therefore can sound quite harsh. So they are best used only in the deeper bass regions where they operate as rigid pistons. These ultra stiff cones follow their voice coil motion more accurately and should work better with motional feedback systems than conventional, soft, woofer cones.

But the fact remains that motional feedback only works well for woofers used below their break up frequency. So it can't effectively control the flaws in a conventional woofer in the upper bass and midrange.

Of course, the biggest problem with servo systems is that they are complicated. They must have box of a dedicated electronics and ideally, they should have their own, dedicated woofer amplifier.

Most audiophiles are very resistant to using complex systems. Only a few will even use electronic crossover systems because of the added complexity and the need for multiple amplifiers. So they continue to suffer the poor performance of passive, high-level crossovers. Getting them to use motional feedback systems is a real challenge.

However, in certain applications, the benefits of motional feedback systems can be both successful and audiophiles will accept them. A good example is in large subwoofers, which have their own, internal amplifiers and crossovers, so adding motional feedback is easily accomplished. But for the typical, full-range speaker, including adding motional feedback is not practical from a marketing perspective, so is not used.

I find it very odd that while most audiophiles will spend tens of thousands of dollars in their quest for musical nirvana, but they fail to take advantage of well-established techniques for improving sound. The best example of this is the presence of passive crossovers in high-end speakers -- even in speaker systems that claim to be "reference" or State-of-the-Art. No speaker can be considered high performance if it uses passive crossovers. Motional feedback systems also offer improved performance, but generally are not used.

Anther excellent example are room correction systems. Even if you have a perfect speaker system, it will be degraded by room acoustics. Room correction systems work very well to deal with this problem, but most audiophiles fail to take advantage of them.

-Roger
 

Gregadd

WBF Founding Member
Apr 20, 2010
10,517
1,774
1,850
Metro DC
Thank you for that informative explanation. I guess the next question was set up by you Roger. What is the difference between active and passive crossovers? Did I hear you say you prefer a brick wall crossover between the woofer and stat panel? At your leisure of course.
 

MylesBAstor

Well-Known Member
Apr 20, 2010
11,236
81
1,725
New York City
Thank you for that informative explanation. I guess the next question was set up by you Roger. What is the difference between active and passive crossovers? Did I hear you say you prefer a brick wall crossover between the woofer and stat panel? At your leisure of course.

Thanks Roger for giving us the designers eye perspective :) Your posts have been highly educational!
 

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