Hi,
The mystery is, it would be a simplest task to arrange a (nearly) similar environment like in a (typical) recording studio with some straigth forward effort if the goal would be set to hear the same as the recording engineer was hearing (this is often stated it seems) and get some of those similar 2-3 way box speaker and listen in the near field, BUT NO WAY, instead people invent the most strangest ideas on how the stereo should be reproduced at home ! 🙄 Poor recording engineers and studio designers ! All their efforts fall in vain ! Why is it people insist to build their own speakers that do not fit even closely to the studio 'standard' ? People do not appreciate good sound ? Isn't it the music should be enjoyed as served, and not modified at home for taste ? Is all the music only semi-finished when leaving the studio and it must be perfected at the listening end ? It's a mystery ! 🙄
- Elias
The mystery is, it would be a simplest task to arrange a (nearly) similar environment like in a (typical) recording studio with some straigth forward effort if the goal would be set to hear the same as the recording engineer was hearing (this is often stated it seems) and get some of those similar 2-3 way box speaker and listen in the near field, BUT NO WAY, instead people invent the most strangest ideas on how the stereo should be reproduced at home ! 🙄 Poor recording engineers and studio designers ! All their efforts fall in vain ! Why is it people insist to build their own speakers that do not fit even closely to the studio 'standard' ? People do not appreciate good sound ? Isn't it the music should be enjoyed as served, and not modified at home for taste ? Is all the music only semi-finished when leaving the studio and it must be perfected at the listening end ? It's a mystery ! 🙄
- Elias
Oliver Linearray,
Earlier there was discussion about "phase decorrelation" and bending wave speakers.
I found polar plots of, propably the most famous of this kind, Manger speaker.
3D3A Lab at Princeton University
It seems to provide two side lobes above 4kHz but the lobes are not very narrow. Also it seems to beam.
Now does this kind of bending wave speaker provide required amount of spatial phase decorrelation to avoid the combing effect in a conventional stereo triangle ?
I think not, but in order to provide the necessary decorrelation at the listening position in a room, the room reflections are mandatory !
And, does the recording engineer with his near field studio monitors have the level of decorrelation ? 😉
- Elias
Earlier there was discussion about "phase decorrelation" and bending wave speakers.
I found polar plots of, propably the most famous of this kind, Manger speaker.
3D3A Lab at Princeton University
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
It seems to provide two side lobes above 4kHz but the lobes are not very narrow. Also it seems to beam.
Now does this kind of bending wave speaker provide required amount of spatial phase decorrelation to avoid the combing effect in a conventional stereo triangle ?
I think not, but in order to provide the necessary decorrelation at the listening position in a room, the room reflections are mandatory !
And, does the recording engineer with his near field studio monitors have the level of decorrelation ? 😉
- Elias
Hi,
I think so too, and I think the explanation may not be even very complicated ! It is my understanding that the 'reproduction capability of the reality' is increased dramatically as long as the speakers are not localised as a sound sources by themshelves. And further so if this situation can be maintained in any circumstance in the listening session, including gross lateral movements and head turning. The increased 'reality' (even if fake) seems to be preferred over 'sound from a can'.
It just appears to be that wide dispersion speakers achieve this better than narrow beamers. The room reflections (again) ! 🙂 Additionally: Increased phase decorrelation. Decreased stereophonic comb filtering. Increased ASW. Increased spaciousness. All positive !
- Elias
I think so too, and I think the explanation may not be even very complicated ! It is my understanding that the 'reproduction capability of the reality' is increased dramatically as long as the speakers are not localised as a sound sources by themshelves. And further so if this situation can be maintained in any circumstance in the listening session, including gross lateral movements and head turning. The increased 'reality' (even if fake) seems to be preferred over 'sound from a can'.
It just appears to be that wide dispersion speakers achieve this better than narrow beamers. The room reflections (again) ! 🙂 Additionally: Increased phase decorrelation. Decreased stereophonic comb filtering. Increased ASW. Increased spaciousness. All positive !
- Elias
Or to put it in different words: A corrupted sense of spaciousness is better than none.
+1
I am wondering (since a while now)...if you take a mediocre narrow dispersion/beaming but CD speaker and add a side firing tweeter replicating the increased radiation around 3KHz from the pic I have posted...you would probably end up with a speaker that at least does not slap you into the face and that has some added "pleasantness".
Joachim Gerhard has added a decorrelated tweeter to his MPL dipoles (actually there are two meanwhile but I am referring to the one at the back). I am not sure now if it is band limited and what the FR is but I am convinced it can do something positive. He calls it the "G-Pole". Maybe you've read it already.
With the new 2nd one (HS-Devices AMT) he fixes the sound power of the beaming FRs somewhat / somehow.
Last edited:
A very interesting aside from the 'econut-pot' session.
Just before a "fact-finding visit" by one of our group (in his thirties), I arranged the nut-pots rather close together (less than 5 feet) and a pair of Infinity 2-way bookshelves outside of that. After listening to both, switching back and forth, we were playing the Chesky test tracks with the drummer behind the mic (a simple stereo X-Y pair so far as my info goes and our listening confirms) at various distances.
The Infinitys played the tracks well, but the rigidity of the stereo triangle/sweet spot was very obvious, and it took some "willing suspension of disbelief" to trust the images. But once the switch was made to the 'nut-pots', believe me if you can, the drummer moved progressively back (with some slight skew on account of the asymmetry of the room sidewalls), and the subjectively judged distance was a few feet less than the announced distance, in all probability as the wall behind the speakers is open latticework about 15 feet away (the room is on the first floor, and no other buildings in the vicinity), and so as good as free-field. And the 'nut-pots' disappeared TOTALLY, especially when the drummer was far behind the mic so that the reverberant field was progressively getting stronger than the direct sound. The icing on the cake was that all these happened with 'free seating/movements' on the listener's part.
While using full-range speakers, we remind ourselves again and again about the onset of beaming as we move up in frequency, especially with drivers having whizzer cones. IMO the only way you can lick the problem is to mount a 'nut' so that the beaming is taken care of fully, and dispersion is uniform, contributing to all-round satisfaction.
The above test tracks once again brought up the critical reminder about maintaining sober levels of playback that would approach the original acoustic situation-- if one is to reap the benefits of getting transported 'there' at least to a great extent.
Just before a "fact-finding visit" by one of our group (in his thirties), I arranged the nut-pots rather close together (less than 5 feet) and a pair of Infinity 2-way bookshelves outside of that. After listening to both, switching back and forth, we were playing the Chesky test tracks with the drummer behind the mic (a simple stereo X-Y pair so far as my info goes and our listening confirms) at various distances.
The Infinitys played the tracks well, but the rigidity of the stereo triangle/sweet spot was very obvious, and it took some "willing suspension of disbelief" to trust the images. But once the switch was made to the 'nut-pots', believe me if you can, the drummer moved progressively back (with some slight skew on account of the asymmetry of the room sidewalls), and the subjectively judged distance was a few feet less than the announced distance, in all probability as the wall behind the speakers is open latticework about 15 feet away (the room is on the first floor, and no other buildings in the vicinity), and so as good as free-field. And the 'nut-pots' disappeared TOTALLY, especially when the drummer was far behind the mic so that the reverberant field was progressively getting stronger than the direct sound. The icing on the cake was that all these happened with 'free seating/movements' on the listener's part.
While using full-range speakers, we remind ourselves again and again about the onset of beaming as we move up in frequency, especially with drivers having whizzer cones. IMO the only way you can lick the problem is to mount a 'nut' so that the beaming is taken care of fully, and dispersion is uniform, contributing to all-round satisfaction.
The above test tracks once again brought up the critical reminder about maintaining sober levels of playback that would approach the original acoustic situation-- if one is to reap the benefits of getting transported 'there' at least to a great extent.
Hi Elias,
Here is a paper of DML (NXT):
From http
See figure 3, there're 3 diagrams of polar patterns - 300, 1k, & 10kHz. Basically all are very fat bipole, or almost omni (null on the side is very narrow).
To my ears, it's pretty much like those diagrams. That thing is splashing everywhere, in all frequency it can play.
Here is a paper of DML (NXT):
From http
See figure 3, there're 3 diagrams of polar patterns - 300, 1k, & 10kHz. Basically all are very fat bipole, or almost omni (null on the side is very narrow).
To my ears, it's pretty much like those diagrams. That thing is splashing everywhere, in all frequency it can play.
There are different paradigms in choosing a bending wave speaker's
parameters.
One parameter is the coincidence frequency, where the propagation
speed of bending waves on the membrane is equal to propagation speed
of sound in air.
Above Fc bending waves are faster than sound in air and below they
propagate slower, this is because bending waves are dispersive:
Propagation speed rises with frequency. Fc as well as the dispersion
relation depend on the membranes design and materials used.
Manger e.g. has Fc above the audible range. That causes mainly the
pistonic portion of the modal membrane motion to be relevant for
sound radiation. In turn the speaker shows common effects we also know
from pistonic radiators like beaming towards higher frequencies.
But it also shows modal behaviour which is even visible in the impedance
curve from lowest resonance upwards to upper midrange ...
I looked up common translations of the word "famous"
into german language and due to some of them i have to
agree to your opinion.
Last edited:
Yes. Another famous designs of 'decorrelation tweeters' are Linkwitz Orions and John K NaO (where back firing tweeters are used).
I call it decorrelation rather than decorrelated, since the decorrelation is primarly achieved not at the tweeter itself but with the help of increased room reflections from that tweeter !
Often (allways !) the effect of back firing tweeter is explained by 'power response' issues, but I think the real benefit for stereophonic field at the listening position comes from the increased decorrelation ! 🙂
- Elias
Thanks ! I see NXT appears as an effective point source. It would mean the spatial phase is not random but follows closely the (ideal) point source field where phase isobars are coincided circles. No spatial phase decorrelation..
So no spatial phase decorelation from Manger neither..
NXT cannot do it.. Manger cannot do it.. Is spatial phase decorrelation only a hoax ?? 😀
My answer is.. Yes 😀
- Elias
In this case what you would mean and what is a physical reality differs
considerably.
A bending wave loudspeaker driven above the individual concidence frequency
neither is able to produce coincident circles as phase isobars, nor is able to have
polar radiation patterns without lobes.
This applies to all transducers of this kind above coincidence frequency
be those designs e.g. made by "Walsh/Ohm", "German Physiks" or even some
"DML" named design.
Here are some example 0 and 75 degree magnitude and phase curves
taken from a system i know myself quite well:
Model2 Messungen
Of course any results will look different, depending on how
much smoothing (over frequencies / over angles) you use.
And of course there are noteable differences in
- FR response smoothness
- "granularity" of lobes over angles and frequency
depending on the design under question.
________________________
Before any splattering may start now, i want to briefly underline
that i am not the one in duty to proof these concepts to those
who are not quite informed about bending wave loudspeaker designs.
CLS e.g. provided a link which may serve as a starting point and
there is a lot of information to be found on the web, being it from
that source or even from different (and earlier) sources.
Kind Regards
Last edited:
Do I understand that those are nearfield measurements "Nahfeldmessung" at the distance of 30cm "Mikrofonabstand", and your panel size is 48 x 118cm ?
At such a close proximity of course the amplitude and phase will show dependency on the angle.
How about at real listening distance 2-3 m ? This measurement reguires a huge space, outdoors maybe.
- Elias
Concerning the word "and":
I am sure you even spotted that bit above 3Khz ...
And i am sure you noted "some" similarity to the measurements
at 1.2m distance.
You can measure outdoors, with cold or warm microphone, in proximity or distant,
you even can turn yourself upside down during measurement ...
no bending wave membrane can radiate in a phase coherent manner above own
coincidence frequency.
You can influence "how" that decorrelation/lobing is shaped but there is no "if" due
to that taking place.
"Please refer to the literature".
Last edited:
Are you suggesting a bending wave speaker can provide spatial phase decorrelation at high freqs to an extent that in the listening position of a stereo triangle at a normal listening distance there will be no comb filtering field in nearly anechoic conditions without the support of room reflections, or cannot ?
That was the original issue, which I was looking at.
It would be great if it could, but I've never seen any measurements anywhere to confirm this.
- Elias
That was the original issue, which I was looking at.
It would be great if it could, but I've never seen any measurements anywhere to confirm this.
- Elias
I have some reasons, not to go into details due to that question.
But we should bare in mind that this question is useless if asked
in "yes or no" manner.
It is more that your question in fact rises a few more questions like e.g. :
"Do we want to broadband (spatial) decorrelation in stereo speakers ?"
I would say no.
"Would combing (under proposed unechoic conditions) be an issue, when
listening exactly in the median plane ?"
Obviously not.
Are unechoic conditions realistic for stereo listening at home ? ...
What kind of measure could be used for "spatial" decorrelation under
realistic home conditions ? ...
That could be a measure applicable to all kinds of speakers, regardless of
their directivity or phase coherence over frequencies.
If one answers these (and many more) questions (which have to be refined
and devised to greater detail), you then would be able to develop
a functional specification of a stereo speaker system, which hopefully
would yield better performance if designed and built according to that
specification.
That said specification will presumably fill a few more pages, than that of
most current systems, where some of the answers are void, simply because
there are questions which never have been asked, not in a detailed
and specific manner at least.
In the design of bending wave loudspeakers you have to be more kind of
"ignoramus", not to stumble into those question ineviteably, because
decorrelation and lobing is a thing you cannot avoid when having
coincidence frequency in the audible range. Which from traditional
"coherent speaker" point of view is "poo poo" anyhow ... see ?
To make it short: I am currently not interested in answering the question,
because that would force me to devise the question to a more detailed
and explicit manner, which implicitly would be a functional specification
for those already accustomed to think in bending wave speaker terms and
parameters.
But we should bare in mind that this question is useless if asked
in "yes or no" manner.
It is more that your question in fact rises a few more questions like e.g. :
"Do we want to broadband (spatial) decorrelation in stereo speakers ?"
I would say no.
"Would combing (under proposed unechoic conditions) be an issue, when
listening exactly in the median plane ?"
Obviously not.
Are unechoic conditions realistic for stereo listening at home ? ...
What kind of measure could be used for "spatial" decorrelation under
realistic home conditions ? ...
That could be a measure applicable to all kinds of speakers, regardless of
their directivity or phase coherence over frequencies.
If one answers these (and many more) questions (which have to be refined
and devised to greater detail), you then would be able to develop
a functional specification of a stereo speaker system, which hopefully
would yield better performance if designed and built according to that
specification.
That said specification will presumably fill a few more pages, than that of
most current systems, where some of the answers are void, simply because
there are questions which never have been asked, not in a detailed
and specific manner at least.
In the design of bending wave loudspeakers you have to be more kind of
"ignoramus", not to stumble into those question ineviteably, because
decorrelation and lobing is a thing you cannot avoid when having
coincidence frequency in the audible range. Which from traditional
"coherent speaker" point of view is "poo poo" anyhow ... see ?
To make it short: I am currently not interested in answering the question,
because that would force me to devise the question to a more detailed
and explicit manner, which implicitly would be a functional specification
for those already accustomed to think in bending wave speaker terms and
parameters.
As to current state of the art, i know of few bending wave loudspeaker
designs to be able to do what is quite easy with conventional pistonic
designs:
Perform down to 40Hz at least, having a performance that can be regarded
"non resonant like".
Most "bending wave speaker" designs which have LF capability use whole body
motion (pistonic motion) at LF, thus are more or less conventional drivers
de facto. In those cases you have a hard time in separating conventional
"fullrange drivers" from "bending wave speakers" because there are no
clear criteria for separation, it is more like a label.
Before being able to design close to even some specification, there are
technical barriers to overcome.
"Make a bending wave loudspeaker being at least on par with conventional
pistonic designs down to 40Hz and make it possibly go through the entrance door"
is a technical prerequisite, before talking about broadband specification
at all.
Muddeling around instead of learning how to make a certain concept a real
broadband capapable technology will keep that technology from crossing the
line to real "HiFi" for a long period of time.
The higher you go in frequency, the more fancy transducer concepts you find,
like plasma tweeters etc.
In some of those concepts, their being incapable of making a fullrange speaker
is obvious. In bending wave loudspeakers there are at least no such barriers
due to LF like e.g. efficiency.
The BWL concepts not relying on pistonic motion at LF are simply highpassed
in a multiway speaker at frequencies where performance gets too bad to expect
even the "common customer" to tolerate that kind of performance.
As long as that being state of the art in common designs, there is no path
leading to "HiFi", designing to some "audiophile" quality specs , or even
"get the most out of stereo format".
It is large and by more like a technology in a "pre- engineering" state
(still after being close to - or even more than - hundred years old in electroacoustics),
nowadays mainly used for sounding department stores and bathrooms.
That is what most systems are quite sufficient for.
designs to be able to do what is quite easy with conventional pistonic
designs:
Perform down to 40Hz at least, having a performance that can be regarded
"non resonant like".
Most "bending wave speaker" designs which have LF capability use whole body
motion (pistonic motion) at LF, thus are more or less conventional drivers
de facto. In those cases you have a hard time in separating conventional
"fullrange drivers" from "bending wave speakers" because there are no
clear criteria for separation, it is more like a label.
Before being able to design close to even some specification, there are
technical barriers to overcome.
"Make a bending wave loudspeaker being at least on par with conventional
pistonic designs down to 40Hz and make it possibly go through the entrance door"
is a technical prerequisite, before talking about broadband specification
at all.
Muddeling around instead of learning how to make a certain concept a real
broadband capapable technology will keep that technology from crossing the
line to real "HiFi" for a long period of time.
The higher you go in frequency, the more fancy transducer concepts you find,
like plasma tweeters etc.
In some of those concepts, their being incapable of making a fullrange speaker
is obvious. In bending wave loudspeakers there are at least no such barriers
due to LF like e.g. efficiency.
The BWL concepts not relying on pistonic motion at LF are simply highpassed
in a multiway speaker at frequencies where performance gets too bad to expect
even the "common customer" to tolerate that kind of performance.
As long as that being state of the art in common designs, there is no path
leading to "HiFi", designing to some "audiophile" quality specs , or even
"get the most out of stereo format".
It is large and by more like a technology in a "pre- engineering" state
(still after being close to - or even more than - hundred years old in electroacoustics),
nowadays mainly used for sounding department stores and bathrooms.
That is what most systems are quite sufficient for.
Last edited:
Just some strange line of thought ... to make clear the different
reasons for multiway
if some modified humans were some 70dB + more sensitive to sound,
there would be hardly any reason for them to invent midrangers and
woofers.
They would get along using (maybe slightly modified) 1 inch tweeters
rather well, equalized with rising 12db/octave slope below resonance,
that would make up a reasonable fullrange system for them.
Membrane size in pistonic drivers is from that point of view only due
to maximum SPL needed and due to efficiency.
Not so in common bending wave panels used in mid to high frequencies
as the performance is usually worsening with frequency getting lower,
even to our brothers and sisters we had to modify for the virtual
auditions.
reasons for multiway
if some modified humans were some 70dB + more sensitive to sound,
there would be hardly any reason for them to invent midrangers and
woofers.
They would get along using (maybe slightly modified) 1 inch tweeters
rather well, equalized with rising 12db/octave slope below resonance,
that would make up a reasonable fullrange system for them.
Membrane size in pistonic drivers is from that point of view only due
to maximum SPL needed and due to efficiency.
Not so in common bending wave panels used in mid to high frequencies
as the performance is usually worsening with frequency getting lower,
even to our brothers and sisters we had to modify for the virtual
auditions.
Last edited:
May be worth kicking this thread back to life again, who knows ...? 🙂
The most "radical" technique, which I've used with great success for many, many years is to eliminate as much low level distortion as possible, what some people call "lowering the noise floor". This doesn't rely at all on worrying about speaker technology, positioning, etc, etc, but rather getting rid of all the silly weaknesses in the way the speaker has been assembled, ensuring that the carcase is extremely tightly coupled to the room structure, and then working back through all the electronics to the source component, and sorting out all the issues, big and small, degrading the sound.
If you clean up enough of the setup this way "outstanding stereo imaging" will automatically pop its head up ... well, it does for me! 😀
Frank
The most "radical" technique, which I've used with great success for many, many years is to eliminate as much low level distortion as possible, what some people call "lowering the noise floor". This doesn't rely at all on worrying about speaker technology, positioning, etc, etc, but rather getting rid of all the silly weaknesses in the way the speaker has been assembled, ensuring that the carcase is extremely tightly coupled to the room structure, and then working back through all the electronics to the source component, and sorting out all the issues, big and small, degrading the sound.
If you clean up enough of the setup this way "outstanding stereo imaging" will automatically pop its head up ... well, it does for me! 😀
Frank
Well, my last two surround sound projects never got finished, and instead evolved into what I am doing now. In the past I thought I wanted the ultimate in surround sound. I studied it to death and designed a great system, and built much of it. Now I'm building a stereo acoustic hologram (as I call it) single box speaker system that goes under my 42 inch HD TV. It will be supplimented by side channel speakers up at the ceiling aimed directly across at one another. This single box is 3 way, wideband (30HZ - 20kHZ), 4 pole active x at 150 HZ (and active woofer EQ), 1 pole passive x at 7kHZ, but the big feature is that it uses the Polk technique of creating the stereo hologram.
For decades I've been excited about the Carver electronic hologram, but your head has to be in exactly the right place for it to be an improvement. Plus the center image has been described as "phasey". With this "soundbar" acoustic hologram, the listener position isn't nearly as critical, and there is a separate L+R center speaker included to give solidity to the center image.
Previous center speakers sounded too confined or small in comparison to when program content would turn on the rest of the speakers. It bothered me. After two rebuilds of that I decided that the center speaker should do it all, in full stereo.
There's a quick and easy way to drive the speakers passively for the hologram function, but it can be done better with active electronics ahead of the poweramps. I'm doing the processing electronics as thoroughly as I know how.
I'll be having 10 circuit boards made very soon for this project. The box has two 6.5 inch Peerless woofers side firing near the back (to take advantage of the wall for acoustic amplification), five 3 inch Peerless/Vifa drivers across the front (same ones Roger Russel of McIntosh uses in his $15000 vertical line arrays), and three 3/4 inch Dayton dome tweeters also on the front (which Zaph claims are ruler flat from 4kHZ to beyond 20kHZ). I'll try to upload a picture looking in the bottom of the finished cabinet here. Well, apparently I can't upload an image unless it's already on the web. Oh well.
For decades I've been excited about the Carver electronic hologram, but your head has to be in exactly the right place for it to be an improvement. Plus the center image has been described as "phasey". With this "soundbar" acoustic hologram, the listener position isn't nearly as critical, and there is a separate L+R center speaker included to give solidity to the center image.
Previous center speakers sounded too confined or small in comparison to when program content would turn on the rest of the speakers. It bothered me. After two rebuilds of that I decided that the center speaker should do it all, in full stereo.
There's a quick and easy way to drive the speakers passively for the hologram function, but it can be done better with active electronics ahead of the poweramps. I'm doing the processing electronics as thoroughly as I know how.
I'll be having 10 circuit boards made very soon for this project. The box has two 6.5 inch Peerless woofers side firing near the back (to take advantage of the wall for acoustic amplification), five 3 inch Peerless/Vifa drivers across the front (same ones Roger Russel of McIntosh uses in his $15000 vertical line arrays), and three 3/4 inch Dayton dome tweeters also on the front (which Zaph claims are ruler flat from 4kHZ to beyond 20kHZ). I'll try to upload a picture looking in the bottom of the finished cabinet here. Well, apparently I can't upload an image unless it's already on the web. Oh well.
- Status
- Not open for further replies.