Beyond the Ariel

Look up Wikipaedia - Compression Driver

You will see the pulsing sound in the phase plug and throat from the CD animation. It shows the diaphragm moving in and out. So with the complex musical signal it will be following that within the air within the small phase plug holes or slots, moving in and out to the diaphragm oscillations. This represents a turbulent flow of air to be repeatedly forced back and forth.

This is utterly wrong, incorrect, mistaken.
 
This is utterly wrong, incorrect, mistaken.
I looked at Wikipedia, and all I wanted was that little model. I did not read the text and there may be errors in that

I am sorry you feel that way. So far nobody including you has described the operation of CD in technical terms. Nobody.What does that tell you.

Be objective. What do you think goes on in a CD. Please describe step by step in a scientific and complete way exactly what happens in physical terms during the working cycle of a CD in its range of operation. Make point about how the phase plug works and what is happening with it.
I look forward to an objective detailed scientific appraisal.

Thanks for your interest
 
Last edited:
Boldname - I could describe how a CD works in technical terms in quite some detail, but alas I think that you would just ignore it as you have with everything that I have said. Your analysis of a CD is completely wrong, as I have said, and the facts do not support it at all. I described tests that were done to show that viscous losses in a CD were not a factor in the least and yet you continue to expound this falsehood. What would be the point of "describe step by step in a scientific and complete way exactly what happens in physical terms during the working cycle of a CD in its range of operation" since you will just ignore it.

I know, I know, you hear it therefore it exists.
 
Lynne

Clearly there is no correlation between cost and performance in the audio world. I sympathize with your views on how the very rich spend money with no consideration of value. It doesn't matter to them. To them it has to look expensive - and things like tube amps do, but lets not equate looks with sound quality. Those two things are likewise not well correlated.
 
Boldname - I could describe how a CD works in technical terms in quite some detail, but alas I think that you would just ignore it as you have with everything that I have said. Your analysis of a CD is completely wrong, as I have said, and the facts do not support it at all. I described tests that were done to show that viscous losses in a CD were not a factor in the least and yet you continue to expound this falsehood. What would be the point of "describe step by step in a scientific and complete way exactly what happens in physical terms during the working cycle of a CD in its range of operation" since you will just ignore it.

I know, I know, you hear it therefore it exists.

Earl, It is not really a case of dismissing all or much of what you have said. That was and still is your misconception. You may have a boiled it down to viscous issues not being significant enough to impact the performance. So perhaps that is the basic sticking point on phase plug debate. Then, as you say, you have quantified that and if you are singularly stating it is not significant from your studies, then I can trust that. Exaggerated broad disagreements from others is emotional and not constructive. You and I have been in this stuff too long to not have a clear analytical way of getting all the known variables, and then dealing with each one or dismissing it as insignificant. Hence your Summa and other developed speakers.

The CD from its operation has great benefits of control of directivity and power with even reduced levls of distortion from a direct driver. But they have their own limitations that may be more important to some rather than others. There are those who talk of coloration and HOMS but you have dealt quite well with these esopecially the latter but the horn sound ifor most is generic And still a noted characteristic. It however does not stop me from using either CD s or direct drivers, with wave guides and horns

Lets now let it rest.
 
Boldname - I am as familiar, or more, with "horn sound" and for the vast majority of devices, I too dislike it. The benefits of the CD/waveguide device have always been foremost in my mind, which is why I set out to keep the benefits while eliminating the "horn sound". I believe that I have done that, but very very few have actually heard this in reality. I find that a properly done waveguide makes a compression driver sound as if it were a very high performance direct radiator with no perceivable "horn sound". I believe that all of your issues stem from experience with what I will call inferior designs of the waveguide and that leads you to conclude that it is the fault of the compression driver.

Many years ago I was hired by B&C to study compression drivers with the intent to improve their audibility. This took several years and resulted in a couple of papers (which were never accepted well because of what they found.) In the end my conclusion was that the CDs themselves were darn near perfect and that not much could be done to improve them - which remains my position, and leads to my conclusion that CDs are basically a commodity.

Further study on my part has been conclusive, IMO, in that it is the waveguide that dominates the situation and nothing else. That few have heard a really good waveguide design is indeed disappointing because so many continue to discount CDs because of previously bad implementations of the horn.
 
Last edited:
Lynn,
I'm not suggesting, just as Earl is not suggesting that people don't purchase or lust after really expensive audio gear. There are those who will pay a thousand dollars a foot for a piece of wire, that doesn't mean that the product has any merit. Wilson sells speakers that cost mega bucks and basically makes boxes with other peoples commodity drivers, they do have a market they have cultivated over years, doesn't mean technically they are selling anything that I would even look at. Yes you can sell a two million dollar car but probably not many of them, but if you do and can you are doing well and just taking some rich persons money who is glad to part with it, they can't spend it fast enough in one lifetime, it is a game of showmanship.

I am not in any way trying to convince you or want to try and change your opinion on what sound you want. You do have a sound you are after, it is a very specific end result you are after, that is exactly what you should do. At the same time you have made the statement that your own conception is limited to a certain type of music, I don't personally believe that the ultimate speaker should be limited to any one kind of music, it should be able to play any kind of music, that should only be controlled by the original recording, not the playback system, that is my opinion on that.

Some here want to argue semantics. like the argument about what is constant directivity. The problem is there can be a technical argument and then the commonly agreed upon term which normally is used in a horn loudspeaker. I don't see an omnidirectional speaker like and MBL compared to one of Earl's speakers or a line array as being comparable types of speakers. I can't put them all in one category and call them all constant directivity designs, it just confuses the conversation to do that.

I get the power of marketing, without it Apple would not be the company that they are, they would have lost the market to all the commodity Android phones if it wasn't for the power of marketing on a cost performance basis.

I enjoy the conversation that goes on in this thread. I also am quit impressed with your understanding of electronics as it pertains to vacuum tube design. I haven't even thought about that since my electronics teacher in high school tried to teach to that when everything was already converting to SS. I happened to like some of my Mac tube amplifiers, I don't have a problem with feedback amplifiers. Actually many designers such as Bob Cordell and Doug Self would probably think not using feedback would be insane, that those who say it causes problems just don't understand the real physics behind the application of feedback. There is much more consensus for than against its use. You have an opposing view of this and I accept this and I understand your argument about harmonic structure. It is one way to look at the problem.

I still have my enclosures to make my high end horn loaded system, I didn't just throw them away, I even kept the tooling. I do have a plan for that, but it is not the time for me to do that right now. It is also at the same time a very different sound. Anyone who says you can't identify a system as horn loaded by the sound signature is either deaf or just not willing to admit it is distinctive. There are so many factors that affect sound, from design to materials used. These things do become preferences at some point, we each have our own opinions that are not determined by pure science, there is a life experience factor in all of this. Earl uses pure science at least at the level that he has been able to document and test. Does that mean I agree with everything he says, no. But I don't discount his science, I just think we have not been able to factually determine all the phenomena that we call sound, things are still confounding this whole issue we call reproduction.
 
There is only science or personal preference. The later can never lead to a consensus, nor can it ever be considered to be "Hi-Fi". Only the former has any ability to claim itself as the goal of "Hi-Fi" - the task that I set myself to some 50 years ago. I understand that there are those who only seek "personal preference" - I get that - but just like that is Lynne's goal, mine is and always has been "Hi-Fi". There needs to be the counter point here on these two separate goals, otherwise readers are lead to the believe that they are one and the same and they clearly are not. To me, accuracy dictates preference, not the other way around.
 
Earl,
If your design is as I understand it if you put me blindfolded in a room and I walked across the sound field and could here a sharp cutoff from your very controlled directivity horns I would know instantly that I was listening to a horn loaded design. I am not in any way saying that sitting in a centered position with your design that I wouldn't like it and find it very good, I'm just saying I would know I was listening to a horn loaded system. The sound field is generally distinctive. I don't think I would think it was all direct radiator devices. Perhaps that is not true, but that is what I expect. Perhaps in a very well treated room I would have a hard time identifying this, but that is so unusual as to be the outlier and not the general rule.

And I am not discounting your approach, I actually agree with it. What I am saying is there are still some things we have not been able to test for, something everyone is missing, we don't have a complete set of tools to predict everything, and don't ask me what that is, if I knew the answer I would put it out there. Nobody has been able to identify the tests that consolidate all the information to predict precisely the final results. It is more than all the distortion testing we now have in the test protocols, something is still missing to pull this all together so we can make a perfect reproduction chain, we can not reproduce a live sounding event, that is the bottom line.
 
Look up Wikipaedia - Compression Driver

You will see the pulsing sound in the phase plug and throat from the CD animation. It shows the diaphragm moving in and out. So with the complex musical signal it will be following that within the air within the small phase plug holes or slots, moving in and out to the diaphragm oscillations. This represents a turbulent flow of air to be repeatedly forced back and forth.

This does not happen with domes which can fairly linearly project straight out to our ears. So there must be prorate less losses. Jay and I as well as the fraternity have heard this perhaps for 40 years or more.

It does incur its own noise due to the motion back and forth with the diaphragm. With a 3" (75m) diaphragm with say 1mm stroke is 4.4mls i.e 4.4 cubic centimetres of air being thrust back and forth in the phase plug.

On high frequencies this will be naturally very fast incurring more frictional loss and sufferering higher degradation in the process. That also gets magnified by the efficiency of a CD. It is the main problem for CD operated at high levels. No free lunch

Ask Earl to explain it to you, and how CD's work. Take the phase plug out and it is a direct driver, and at least this phase plug issue vanish. How much does it matter. It matters to those of us that can hear it.

I hope this helps.


Here’s the problem and sorry for being blunt, but there’s no other way to say it; you demonstrate a complete lack of understanding of the physical phenomenon called wave propagation. I did try to explain it previously, but you’re still viewing the movement of waves as fluid flow which it most certainly is not. Wave phenomena are energy transport mechanisms, not matter transport mechanisms.

Waves are vibrational disturbances where each molecule oscillates around a ‘mean’ center (IOW its position in space is pretty much fixed) and through what are basically elastic collisions, kinetic energy is transferred through the medium. You must understand this fact if you want to understand the "describe step by step in a scientific and complete way exactly what happens in physical terms during the working cycle of a CD in its range of operation" You cannot describe anything step-by-step if you don’t possess a good understanding of the fundamental principles involved. This is why your insistence on turbulence simply doesn’t make sense - there can't be any.

Apparently, what I also said about the dimensions of the compression chamber and phase plug versus the wave lengths involved, must have gone over your head as you're still clinging to your micro-turbulence concept (in a round-about way) with the "incur its own noise" comment. Again, if this "noise" could possibly exist, it would be as I alluded to previously, way up at over 300 KHz. Will you now claim that this would make a difference?! And I'll say it again, seriously?!!!!

Lastly, please stop insisting on "high frequency degradation" and frictional losses in a mechanism that's basically ruled by elastic collisions and where shear stresses are not involved - it simply makes no sense.
 
Why people likes more the Compression drivers and the SET with no global feed back ? Just because the littliest cone surface of the CD are freeier of the bad come back of the "cone" effects on the amp ? Many testimonie for a more free sound and less thick mi-bass and medium (and more détails) with no global feedback amps (or at least with a very low amount of it) ?

In an other way many people showed there is no hope for good transcient in the High efficienty speakers below 30w/50W class A for frequencies below around 200 hz ! Is it hard to have a passive bi amp system with a good sounding with tubed or transistorised PP amps ? Circlotron design just for the bass drivers ? look at this paper from Rinaldo Bassi at la Revue de l'Audiophile ( bottom of this link); it explains because the many transcient peaks of symphonic music, you need Under 300 hz at least 50 W... measurements are made with a VOTT and a benchmark between a Radford of 30 w and an AudioResearch of 75W : very pleasant reading ! And all the measurement are made with a real music who goes very low : first 20secondes of Picture Exhibition by Nougorsky. Indeed lower than a Vott can play (60 hz ?).

The two amp are setuped at 85 DB average level (room is 30 m²) to allow a dynamic gap of a dynamic of 60 db (to allow to listen at least at 100 db the pianissimi Forte). to be short, they measured some peaks demand of 50 W on the Forte ! Conclusion : measure the amps on the bigges tdynamic peaks and if the PS is not collapsing before or fast enough to answer between two big bass peaks ! Above 300 hz (passive XO took as a ref for most of speakers) you need minima with this setup : 7 to 12 W... Only 7 w with the 30 W Radford amp because its supply is collapsing between two close peaks demand. the Audio Research has a strong and fast enough supply to supply the 5W than the 12W peak needs around this 300 hz frequency ! Above 5 Khz, Forte needs 3 W peaks on this reccording so an average 2W amp with a good and fast supply is enough above 5 W with this reccording and the limits given above (100 DB in a 30m² room) and an average 7 W SET amp well made above 300 hz (if it can provide 12w peaks on this reccording)... below no hope than an average 30W amp with a strong and fast supply. My low understanding of the paper.

It seems hard to make a low power system , even with very high efficienty drivers, without multi amp... amps should be choosed if I understand this reading in relation to the XO crossover if I understand it ! Chooe your poison first : the design of the speaker or the amp ?!
 
Last edited:
I don't personally believe that the ultimate speaker should be limited to any one kind of music, it should be able to play any kind of music,

Why not, if your musical preference is that limited :) After all, we can never had that "ultimate" speaker, can we?

I grew up with slow rock music from cassette tapes. Deep Purple, Scorpion, Nazareth, Black Sabbath, Twisted Sisters, etc.

When I started my audio system with tube amps [and FR speakers], I didn't recognize that my musical preference slowly moved towards country, ballad and vocal [from CD that I purchased]. But then I recognized that preference sway when I heard a rock music. I contemplated and realized that listening to rock music could give me more enjoyment than listening to even jazz. And that's the reason I left tube amps :)
 
Musical Noise,
Thanks for that post. I don't think some have any idea how small in comparison to the wavelengths we are talking about the area is between the diaphragm and the phase plug slot openings. It seems many confound the properties of the mass of the diaphragm with losses in a compression driver due to supposed viscous air losses. It is very much the same thing on the dome tweeter side where some seem to think that the dome tweeters work as perfectly rigid domes, that is far from reality.
 
Jay,
There is no perfect speaker. Baring that what I am trying to say is that if you have a very wide bandwidth speaker that also has very wide dynamic range and interfaces well with the room that should be able to play any kind of music that is well recorded. From there everything goes down hill, by that I mean that from the recording to the final design of the individual devices there are compromises, this is what we are fighting, all the additive little things that tell us we are listening to a recording and not a live event. I don't think you should have to have a speaker that is designed to make a rock sound, meaning that it shouldn't try and emulate a PA stack and what is going on in amplified music. It should just play back accurately what is on whatever recorded media. A speaker that was intended to make the kinds of distortion that you would hear from a guitar cabinet would just accentuate that type of distortion, in other words it would add more of that type of distortion. That type of speaker could never play back an orchestral sound or jazz sound correctly, it would be a distortion producer. The holy grail is a speaker that adds nothing and takes away nothing.
 
So it's certainly better to develop amps, powersupplies and speakers able to reproduce complex musics with a lot of high dynamic peaks close togethers with high DB presure multuple peaks in the low while having dynamics peaks above in the medium and micro détails with less dynamic: classical music do this very well. Also on the less complex jazz (if no big band of the beginning of the XX century) : you have a lot of fundamental, détails and peaks in the mid-bass régions. I surmise if the speaker/amp/Power supplies combo is not collapsing on symphonic orchestra it can plays Deep Purple or whatever !

I surmise Dr Gedle measure his speaker, drivers with powerfull rapid toneburst ?! And the others experienced listener here listen their amp/driver on complex powerfull music to merge to the same results too choose their stuffs ! In my mind I find more difficult to find those combo for music below 700 HZ 'said as an average XO often viewed here) also because the rooms ! What the hell fi you can listen to 5 000 hz but if in the same time nears transcient peaks are collapsing at low level and the harmonics are not played correctly ! Imagine a strong note of trumpett or piano at 90 hz for example which have reverberation ! What two do if the harmonics are collapsing or are hided by a thickness because the powersupply of the amp (or the driver damping material ?) can not folow ? 180 hz H2, 360 H3...720 H4... where the life of music is !
Too simple ? Ok a double 15" TAD Onken can play it more or less in wide band in a same enclosure... but with which amp ? Not saying below 200 hz our ears are flatting also a lot of things !
 
Last edited:
I measure a reduction in the reflections and the diffraction - or is this a trick question?

I showed how delayed aspects of the impulse response were reduced even after the frequency responses were made to be the same. These aspects were almost invisible in the frequency domain but became apparent in the time domain. I did a paper on this at ALMA many years ago. Maybe you should look that up.

An RC network would lower the highs and thus change the whole impulse response - not at all the same thing.

Don't you see that the foam reduces the highs, the diffraction and the reflections. Then EQing the highs brings them back, but the diffraction and reflections stay reduced. What part of this aren't you getting?

I understand now. I was trying to get at what you're actually measuring.
I do thank you for your response.
I do like the foam, but I like lots of things that are not currently measured.
 
Eldam,
Yes you have to get many things correct. That includes as you say an amplifier or set of amplifiers that have the slew rate and very low harmonic production products to do that. If you can't even get the music out of the electronics without changing the dynamic range or smearing the transient information what chance does the speaker have of reproducing the music correctly. Transient response and phase coherency in a speaker is so important also. There are many details to get right. In Lynn;s favor he is working on a coherent system with very specific requirements to meet his needs. As he says just a change in tube type will destroy his goals, this is what system design is all about, the total system and not the individual components in isolation.

Given that, most SE amplifiers I would consider as tone generators, they are not very accurate in passing a signal. Just because many like the sound of extra second harmonic distortion does not make that accurate. I am not including Lynn's design in this description, just the vast majority of commercial tube amplifiers sold as SE amps.
 
We get all like to get wound up around the need for accuracy and whether what we are doing is "real hi-fi". Then most (me included) will deadpan admit that the playback system can't sound right without room reflections (diffuse or specular) to make it sound more real - anechoic playback just doesn't work well for sounding real or making a decent stereo effect. So accuracy is essential in terms of frequency response and distortion, but just plain unnacceptable in that we need enhancement added by rooms, which is a much bigger change to the signal than a little nonlinear distortion? Seems very inconsistent to me.

My rationalization for this is that recordings do not actually capture an audio event - there are NO accurate recordings, every single one is a product of spicing added by recording or mastering engineers and producers. Even choosing what mics to use and where to position them is choosing a flavoring, and it's very rare for the seasoning to have gone no further. You don't ever have an accurately captured event to begin with, and there's much that doesn't fit in a stereo signal set.. The best you can do is use what you have to produce an acceptable illusion, in such a way that the tricks you use to produce it don't become too obvious, predictable sounding, or monotonous.