Electrostatic speakers?

[Gwreck]

According to Jens Blauert's 1969 research we perceive sound from our midline (nose to back of skull) orientation no matter where the sound is coming from a bit differently depending on the sound frequency we are exposed to. As graphed in the screen shot chart below around 300-500 Hz, and again 3,000 Hz we perceive that as in front of us. While sound peaking at 1,000 Hz and again likewise at 10,000 Hz we perceive that as in back of us. And lastly peaking sharply 1,000 Hz sound we perceive as being atop us.

This would explain your sense of being surrounded with sound from electrostatic speakers and when summed together envelop your head cresting the impression it came in your ears. My premise being the electrostatic's relative speed produces exceptionally clear audio frequencies which facilitates our evolutionary disposition for attention of content in sound clues. Other facets than this inter-relationship to bands of frequency co-evolved to perceive where the sound is coming from (locate) and distance (loud) in the direction of the sound source.

Note: I assume the below graph is of some average test subjects' data and individual human variation(s) surely exist. So I am not declaring it represents fixed ranges - allow me to present this as patterns to

According to Jens Blauert's 1969 research we perceive sound from our midline (nose to back of skull) orientation no matter where the sound is coming from a bit differently depending on the sound frequency we are exposed to. As graphed in the screen shot chart below around 300-500 Hz, and again 3,000 Hz we perceive that as in front of us. While sound peaking at 1,000 Hz and again likewise at 10,000 Hz we perceive that as in back of us. And lastly peaking sharply 1,000 Hz sound we perceive as being atop us.

This would explain your sense of being surrounded with sound from electrostatic speakers and when summed together envelop your head cresting the impression it came in your ears. My premise being the electrostatic's relative speed produces exceptionally clear audio frequencies which facilitates our evolutionary disposition for attention of content in sound clues. Other facets than this inter-relationship to bands of frequency co-evolved to perceive where the sound is coming from (locate) and distance (loud) in the direction of the sound source.

Note: I assume the below graph is of some average test subjects' data and individual human variation(s) surely exist. So I am not declaring it represents fixed ranges - allow me to present this as patterns to consider.

View attachment 461476

ESL’s are much narrower in dispersion so it functions almost like a natural crosstalk cancellation effect. I assume this is one of the reasons why they image like they do.

 

[gnarly]

This thread has leapt back to life and the myth masters seem to be having their way with it all over again.

What are the specific myths ?

 

[dlaloum]

ESL’s are much narrower in dispersion so it functions almost like a natural crosstalk cancellation effect. I assume this is one of the reasons why they image like they do.

With segmented setup and delay lines, you effectively have a multitude of speakers positioned and marginally delayed, so the end results behave like either a pulsing sphere segment or a pulsing cylinder segment - such setups do not suffer from the narrow dispersion effect.

It is interesting to note that Amar Bose in his early research also tended to point to a sphere as the ideal sound source (and his earliest prototypes reflected that - the later 901's were a derivation of that, as were the Bose Direct/reflecting family of speakers) - and no I am not talking about the Bose "lifestyle" systems, but their earlier HiFi ranges... which are oft forgotten (except for the 901).

 

[RayDunzl]

My impression of ESL's is that flat ones are very beamy; step off to the side of one just a bit and the highs disappear. Curved ones may prevent that. I've never heard a curved ESL, so I don't experientially know.

MartinLogan reQuest, 7 locations across the sofa, both speakers speaking.

The highs are still there, the traces are bumpy because of phase cancellations - different path lengths when off-center.

The big hole around 48Hz is a room problem when bass is mono. Not something you notice when listening.

 

[MattHooper]

i gave away a refurbished pair of esl63's about a month ago. they started buzzing, so they were free to the buyer.
always great sound. no bass. and they won't play loud.

I always found this view of the Quads a bit misleading.

Now, of course, it really comes down to somebody’s personal criteria.

But I found that the Quad 63s had quite wonderful bass, enough to make most music sound balanced, and they played about as loud as I would ever want to sit in front of.

Of course, being an audiophile I had to push the performance levels and I ended up adding the gradient dipole subs (which didn’t go super low themselves). And yes, it made the quads sound even bigger. But before then I had found them to be quite satisfying as well.

Ultimately for me, it wasn’t a lack of base or a lack of loudness that led me to selling my quads. It was a lack of density and palpability to the sound, and I also preferred the tonality of some other conventional speakers.

 

[Gwreck]

With segmented setup and delay lines, you effectively have a multitude of speakers positioned and marginally delayed, so the end results behave like either a pulsing sphere segment or a pulsing cylinder segment - such setups do not suffer from the narrow dispersion effect.

It is interesting to note that Amar Bose in his early research also tended to point to a sphere as the ideal sound source (and his earliest prototypes reflected that - the later 901's were a derivation of that, as were the Bose Direct/reflecting family of speakers) - and no I am not talking about the Bose "lifestyle" systems, but their earlier HiFi ranges... which are oft forgotten (except for the 901).

Sounds like a phased array you are describing?

 

[Gwreck]

My impression of ESL's is that flat ones are very beamy; step off to the side of one just a bit and the highs disappear. Curved ones may prevent that. I've never heard a curved ESL, so I don't experientially know.

The narrow directivity “is a feature not a bug” … for myself anyway.

 

[MattHooper]

I have Sanders ESL’s and there are definitely reasons why one would choose narrow directivity speakers over wide dispersion speakers despite the common belief that board dispersion is the best. Roger Sanders who is not to be confused with Gayle Sanders of Martin Logan has multiple white papers and a pretty good white paper on dispersion on his website. He actually invented the curved esl panel that Martin Logan uses but came to the conclusion that it had too many drawbacks and he prefer flat panels.

Sanders Sound Systems - Technical White Papers

Electrostatic specialists for Electrostatic Loudspeakers (ESLs) along with our Magtech and ESL amplifiers.

www.sanderssoundsystems.com

Ultimately, I don’t get along well with really narrow directivity loudspeakers. Not a fan of the head in the vice feeling.

However, I’ve heard a a lot of great stuff about the Sanders electrostatics, and I have a little doubt that it would be a very impressive experience sitting in front of those speakers.
I’ve always wanted to hear them.

 

[Gwreck]

Ultimately, I don’t get along well with really narrow directivity loudspeakers. Not a fan of the head in the vice feeling.

However, I’ve heard a a lot of great stuff about the Sanders electrostatics, and I have a little doubt that it would be a very impressive experience sitting in front of those speakers.
I’ve always wanted to hear them.

I use them with BACCH which has head tracking and widens the sweet spot. If you lived in KC you’d be welcome to hear my pair. They are awful for hometheater though at least for more than 1 person. I think of them as Stuntman Mike(from Deathproof) speakers.

 

[dlaloum]

Sounds like a phased array you are describing?

I never associated the two, but there are substantial similarities!

 

[egellings]

The narrow directivity “is a feature not a bug” … for myself anyway.

When you're in he sweet spot, it's see-through sound, as if the speaker weren't there.

 

[Newman]

What are the specific myths ?

I'm not sure if you're just giving me more work to do, or have you actually been drinking the kool-aid coming out of these electrostatic speaker discussions? You haven’t spotted any issues?

The first myth, as usual, is the attributes coming out of the sighted listening reports by owners and auditioners. The unsurpassed midrange, the amazing imaging, etc. Nobody is mentioning that they are imagining these attributes, or that cognitive bias is responsible. And that's because they sincerely believe that the sound waves themselves are responsible. Myth.

Quad concentric rings. Then there is the myth that the Quad's concentric delay lines meaning that they don't have a beaming problem. Sounds good in theory but in practice the smallest one is four inches in diameter. Now, if someone released a conventional speaker with a four inch tweeter, it would be universally derided, and for good reason, because it would beam like a laser on steroids. But on the Quad it solves beaming? Myth. Sure it helps, but that’s because the untreated stat panel is so bad.

Square wave perfect. Then we got the hoary old story of Peter Walker showing a perfect square wave coming out of a Quad, and how amazing this means the speaker is. This is a myth and a con job on two levels. First level, that ain't no perfect square wave. Any high resolution examination of the leading edge and the top corners would show grievous irregularities. Sure, plenty of speakers would be far worse, especially in its day, but if someone set out today to specifically design a conventional speaker to play a good looking square wave, the Quad square wave might not 'win'. Second level, there is no need for a speaker to reproduce a square wave well. It's a false goal. Basically, showing off the square wave for a speaker is a showroom stunt. A myth.

Dipole bass and sub. The idea that dipole speakers don’t integrate well with a conventional subwoofer and sound superior with a a dipole subwoofer eg a Ripole. The far more likely reason for one to prefer a subwoofer with diminished LF extension, like a Ripole, is because you are simply not doing a normal subwoofer correctly. You know what really causes a subjective impression of slow bass? A strong, deep-reaching subwoofer without any room correction. And the answer to that....well, I just wrote it in the description. But so many people just won't do correction and do it properly. So the next-least-worst option is to have less bass. It can't ring in the sub frequencies if it isn't there in the first place, right? And that's the reason why a dipole woofer will seem 'faster'. It has nothing to do with inductances and back EMF, as has been claimed in this thread. It would actually make sense to repurpose the drivers from a Ripole and put each driver into its own, normal, subwoofer box, position them around the room in accordance with multi-sub acoustical guidelines, and get interested in doing room correction properly to eliminate ringing and perceptions of slow bass. Better bass all round, and should be no integration problem as long as the panels are not doing 'bad bass' above the crossover frequency. You know, some of the ‘stunning midrange’ reports could be arising not only from sighted listening effects, but also from lack of seriously capable bass. When you under-deliver in one frequency range, the other ranges can sound extraordinary, but in reality it’s just imbalance.

Lighter than air. Don’t forget the general perception that stat panels have diaphragms that are so light that they have both incredible transient response due to the diaphragm not being ‘slowed down’ by its own inertia, and simultaneously supreme damping by the air being heavier than the diaphragm. Turns out that the diaphragm is several times heavier (not lighter) than the air, but even so, the impedance of the air does strongly dampen the diaphragm….at high frequencies. The net result is that the air load is present when it is not needed ie high frequencies whereby the diaphragm cannot respond instantly to transients in music because it is so over dampened, and not present when it is most needed, ie low frequencies where all the drum resonances of the diaphragm lie and also frame resonances. It is actually less responsive to transients than a good tweeter.

BTW I sometimes wonder about being asked to type unnecessary repetition. I get it that myth makers are tireless fans, so are bound to not listen to the realities when stated and therefore just repeat the myths next time the topic comes up, but why don't the myth readers learn when the mythbusting facts are being expounded and stop drinking the kool-aid?

Cheers

 

[traind]

I'm not sure if you're just giving me more work to do, or have you actually been drinking the kool-aid coming out of these electrostatic speaker discussions? You haven’t spotted any issues?

The first myth, as usual, is the attributes coming out of the sighted listening reports by owners and auditioners. The unsurpassed midrange, the amazing imaging, etc. Nobody is mentioning that they are imagining these attributes, or that cognitive bias is responsible. And that's because they sincerely believe that the sound waves themselves are responsible. Myth.

I am struggling to understand this statement. Of course sighted bias is real and people can imagine all sorts of things. However, are you saying that no one can listen to a set of speakers and determine they have a certain quality without a blind test?

Fwiw, my first serious listen to Martin Logan ESLs was with a used pair of 15As in a large dealer room. I wasn't there to look at them, I just dropped in to visit this store while travelling for business. They were the cheapest speakers in the room and several pairs of much more expensive, more traditional speakers were also set up in front of me. I did not know which speakers were playing when I sat down, I had to ask the owner because there were a bunch of amps on the floor in front of the seats that I didn't want to step over, lol. Although my time there was brief, during the demo I felt the imaging was the best I've heard to date, but I am willing to admit it might have been the room and at the least I believe he set up the MLs with care. But I do not believe there was much sighted bias at play because I did not know much about ESLs at the time of listening so did not know any of the myths. And they were the least expensive speakers in the room. Am I missing a source of bias with that assessment?

 

[Arindal]

The first myth, as usual, is the attributes coming out of the sighted listening reports by owners and auditioners. The unsurpassed midrange, the amazing imaging, etc. Nobody is mentioning that they are imagining these attributes, or that cognitive bias is responsible. And that's because they sincerely believe that the sound waves themselves are responsible. Myth.

I have been testing (rather competent) dipole electrostats under true blind test conditions, comparing them with conventional ones.

Would say, a part of the myths have a base, others are just a myth. Would not say they sounded anyhow natural or realistic over overly precise in terms of localization/imaging. Rather the opposite, phantom sources appear unnaturally broadened and pulled vertically, depth-of-field is exaggerated. On the other hand, the imaging is pretty stable and offers reduced proximity, which some people like. I prescribe this to excellent attenuation of side-wall reflections, absence of band-limited localization instability (´jumping´ phantom sources) and good tonal balance of the rear wall reflections being responsible for quite some degree of diffuse field in the room while on the other hand the sheer size of the panel, the rear SPL maximum become audible.

Excellent midrange clarity I could also confirm, but would assume this is rather an absence of ´muddy´ lower midrange reverb thanks to balanced directivity than a unique quality of the speaker. Other concepts like cardioids, electrodynamic dipoles or line sources can achieve something similar.

The idea that dipole speakers don’t integrate well with a conventional subwoofer and sound superior with a a dipole subwoofer eg a Ripole.

That is not a myth. At least if you are dealing with a fair amount of room modes, increasing RT60 in the lower mids, and hybrid stats like Martin Logan with a higher x-over req. The versions with true dipole woofer (like the one @traind has mentioned) are superior which is particularly measurable between 100Hz and the x-over freq between dipole sub and stat panel.

 

[Selkirks]

Reminds me of when I had Elac FS 607 X-Jet speakers in room beside my Martin Logan ESL 15A's. I spent the day enjoying blissful and immersive electrostatic speaker performance from the Martin Logans only to realize the Elacs were hooked up the whole time.

 

[Arindal]

ESL’s are much narrower in dispersion so it functions almost like a natural crosstalk cancellation effect. I assume this is one of the reasons why they image like they do.

Sounds like a rather unlikely explanation. The response is not changing that dramatically over the course of a fraction of a degree. And if it would, imaging would be dramatically unstable, with phantom sources collapsing the moment you move your head a few centimeters.

As mentioned, attenuated side-wall, ceiling and floor reflections along with frequency-independent radiation pattern and strong rear-wall reflections, might rather explain the specific yet pretty stable imaging.

According to Jens Blauert's 1969 research we perceive sound from our midline (nose to back of skull) orientation no matter where the sound is coming from a bit differently depending on the sound frequency we are exposed to.

This is correct, but is basically a very simplified model of HRTF-induced shifts in tonality for sound sources coming in from directions with no interaural difference.

This would explain your sense of being surrounded with sound from electrostatic speakers and when summed together envelop your head cresting the impression it came in your ears.

Not sure if we can conclude such from Blauert´s theories. To feel ´surrounded´ by sound, our brain would need to get signals with no or very little interaural difference and a tonal reference, which is both not the case with the direct sound from two-channel speaker reproduction. It might apply to the reverb in the listening room (and there are strong indications that the tonality of the reverb is influencing our perception of where it comes from) because our brain gets it after the direct sound as a tonal reference.

What big planar speakers, particularly dipoles, do achieve, is a relatively good tonal balance of reflections in the room. So the reverb field created in the room is rather unlikely to be perceived as tonally shifted or coming in from the wrong direction. One important take from Blauert´s theories is that reverb which is upper-midrange-heavy (in many cases due to overrepresented indirect sound in the 500-2,000Hz band) is rather perceived as coming from behind or from the sides, hence detached from the frontal phantom sources which sound like ín a vacuum´ with many speakers offering increasing directivity.

 

[Gwreck]

Sounds like a rather unlikely explanation. The response is not changing that dramatically over the course of a fraction of a degree. And if it would, imaging would be dramatically unstable, with phantom sources collapsing the moment you move your head a few centimeters.

As mentioned, attenuated side-wall, ceiling and floor reflections along with frequency-independent radiation pattern and strong rear-wall reflections, might rather explain the specific yet pretty stable imaging.



This is correct, but is basically a very simplified model of HRTF-induced shifts in tonality for sound sources coming in from directions with no interaural difference.



Not sure if we can conclude such from Blauert´s theories. To feel ´surrounded´ by sound, our brain would need to get signals with no or very little interaural difference and a tonal reference, which is both not the case with the direct sound from two-channel speaker reproduction. It might apply to the reverb in the listening room (and there are strong indications that the tonality of the reverb is influencing our perception of where it comes from) because our brain gets it after the direct sound as a tonal reference.

What big planar speakers, particularly dipoles, do achieve, is a relatively good tonal balance of reflections in the room. So the reverb field created in the room is rather unlikely to be perceived as tonally shifted or coming in from the wrong direction. One important take from Blauert´s theories is that reverb which is upper-midrange-heavy (in many cases due to overrepresented indirect sound in the 500-2,000Hz band) is rather perceived as coming from behind or from the sides, hence detached from the frontal phantom sources which sound like ín a vacuum´ with many speakers offering increasing directivity.

https://3d3a.princeton.edu/sites/g/files/toruqf931/files/2022-12/sanders_sound_systems_model_11_h_contour_plot.png

Here is the directivity plot of a pair of Sanders ESL’s similar to what I use. They are so directional that one has to be fairly deliberate in placement or you will lose all high frequency sounds. I use these with BACCH as this narrow directivity enhances the XTC effect significantly. The narrrow directivity improves the impulse response as compared to wide dispersion speakers so they are more like a pair of headphones. Also im not sure about the frequency dependent HRTF effects for my system as BACCH corrects for frequency response at the ears. Even with the XTC filter off and the frequencies corrected the imagining is still the same with the ESL’s. With corrected frequency respondents imaging actually improve a bit.

 

[sq225917]

Mrdogs 989s beam like a bugger, I should know I'm the one usually sat outside the image on his sofa.

His subs are fully dsp PEQ, with bags of clean class d power up them.

I've heard a lot of nice home hifi, some very spendy. Only his makes me think I might be missing out.

 

[RayDunzl]

As mentioned, attenuated side-wall, ceiling and floor reflections along with frequency-independent radiation pattern and strong rear-wall reflections, might rather explain the specific yet pretty stable imaging.

Unsmoothed room-corrected frequency response

MartinLogan reQuest vs JBL LSR 308

JBLs located next to the reQuests. The JBLs are my daily drivers.

The only explanation I have is sharp cancellations due to the wide dispersion JBLs working the walls and the floor and the ceiling and working the walls some more.

 

[Arindal]

They are so directional that one has to be fairly deliberate in placement or you will lose all high frequency sounds.

That is common with all flat true-planar transducers as the sheer width causes the radiation angle to narrow down towards higher frequencies. If the panel is flat but not too broad (50cm max), this effect of ´you are suddenly leaving the zone of any available treble´ is restricted to the highest octave, live 8K+. It is very audible in terms of tonality (getting dull), but it does not necessarily mean a significant impact on localization and HRTF-induced diffraction. The latter is mainly happening in the 500 - 5,000Hz bands and in the latter band the listening zone is still okay, in the region of +-6deg, so interaural crosstalk should not be affected.

The narrrow directivity improves the impulse response as compared to wide dispersion speakers so they are more like a pair of headphones.

You are referring to the same phenomenon of significantly attenuated early reflections? That is certainly significant, I find the term ´impulse response´ a bit confusing, would rather call this reflectogram.

sharp cancellations due to the wide dispersion JBLs working the walls and the floor and the ceiling and working the walls some more.

Sounds like we mean the same phenomenon.