gedlee said:
Dr. Geddes,
Speaking for the silent majority, and hopefully for those who may respectfully disagree, please lecture away. I tried to copy a portion of your design and apply an OB flair of my own and find myself unable to advance until I apply more science to my approach. In the meantime, I'm thoroughly enjoying my Summa wannabee OB's.
We are not involved with P-audio, except that our GM came from there. P-audio does not make the level of driver that I am interested in, ours are mostly B&C which I find to be a really good driver at a resonable price. Our products will not be the price leader, but I believe that they will be the value leader - best perfomance for the price - in the industry. As an engineer, I have come to respect value as it is the hardest thing to do.
Flight hoax.
I just read an interesting (non published) account of the true Wright brothers saga. They really accomplished nothing.
Both went to the library and found all the books on aeronautical and mechanical engineering written by the best educated and trained engineers. These books contained only the true and validated facts as the scientific community dictated.
There was no cutting and trying hundreds of times and guesswork used in building the airplane. They simply applied the accepted physics and off it went.
For validating my designs:
I expect 10 of my friends to give up working for a week so we can scientifically test and make sure the results are acceptable before I release any data on my designs.
Sadly, I have but two or three cheap pieces of gear to test my theories, but I do have my on board computer that fortunately runs in nanoseconds.
So, no more putting stuff together and listen. If it isn't already designed and tested by the more knowledgeable professionals it should no longer be tolerated.
Zene
I just read an interesting (non published) account of the true Wright brothers saga. They really accomplished nothing.
Both went to the library and found all the books on aeronautical and mechanical engineering written by the best educated and trained engineers. These books contained only the true and validated facts as the scientific community dictated.
There was no cutting and trying hundreds of times and guesswork used in building the airplane. They simply applied the accepted physics and off it went.
For validating my designs:
I expect 10 of my friends to give up working for a week so we can scientifically test and make sure the results are acceptable before I release any data on my designs.
Sadly, I have but two or three cheap pieces of gear to test my theories, but I do have my on board computer that fortunately runs in nanoseconds.
So, no more putting stuff together and listen. If it isn't already designed and tested by the more knowledgeable professionals it should no longer be tolerated.
Zene
My partner is a major designer and installer of audio systems in Asia so his focus is on pro. But I do see the home theater market as a major opportunity (but not in Asia) since our products offer high output (required for HT), reasonable prices (good when you are buying five speakers) and excellent sound quality (always a plus).
I am not sure how much "lecturing" that I will have the time for, but I'll try and come back with some comments. I am moving to Thailand at the end of the week for three months - family in tow. So I am quite busy this week.
I am not sure how much "lecturing" that I will have the time for, but I'll try and come back with some comments. I am moving to Thailand at the end of the week for three months - family in tow. So I am quite busy this week.
Not everyone can afford to do ellaborate subjective tests, and not every advance has to be so proven. But, one must always be careful not to claim unproven opinions as facts. WHEN scientific data is available use it, when you only have your own personal observations, use them too, but ALWAY be sceptical of their validity because the human psychy is not to be trusted (read Dr. Taleb if you want to see a proof of this).
My complaint is throwing out scientific facts when they exist because they do not agree with your personal opinion.
My complaint is throwing out scientific facts when they exist because they do not agree with your personal opinion.
Re: Flight hoax.
Zene,
This may very well be true for boxes, but when it comes to open alignments there aren't enough pros who qualify. Of course that's not stopping box pros from coming out with flawed OB designs. Other than SL and JK, the real OB expertise still lies primarily with a relatively small but growing group of DIYers.
Zene Gillette said:
Zene,
This may very well be true for boxes, but when it comes to open alignments there aren't enough pros who qualify. Of course that's not stopping box pros from coming out with flawed OB designs. Other than SL and JK, the real OB expertise still lies primarily with a relatively small but growing group of DIYers.
Re: Flight hoax.
So one can build crude working devices by cutting and trying? Who can argue with that?
I'll go one further: Sometimes you can go outside conventional design by applying educated guesswork and come up with something that has unique properties, and works better for whatever attribute is desired.
But, if you want to understand why it works better and generalize those findings, then some real scientific rigor is required.
Sheldon
Zene Gillette said:
So one can build crude working devices by cutting and trying? Who can argue with that?
I'll go one further: Sometimes you can go outside conventional design by applying educated guesswork and come up with something that has unique properties, and works better for whatever attribute is desired.
But, if you want to understand why it works better and generalize those findings, then some real scientific rigor is required.
Sheldon
Martin, Earl, you may notice that this is a rather tightly moderated forum where personal attacks and consistent rudeness are swiftly dealt with.
If you feel that this thread is too woo-woo for the approach you want to take (and I am more than a little sympathetic to that view), then let's start a new thread devoted to the engineering approach to the Ultimate. I, for one, am extremely curious about OB, since I have loved the various dipole speakers I've had over the years, but I'm somewhat of a fogey in wanting to see real theory before tackling the intangible stuff.
The Ultimate OB is an interesting project, too interesting for you guys to bow out.
If you feel that this thread is too woo-woo for the approach you want to take (and I am more than a little sympathetic to that view), then let's start a new thread devoted to the engineering approach to the Ultimate. I, for one, am extremely curious about OB, since I have loved the various dipole speakers I've had over the years, but I'm somewhat of a fogey in wanting to see real theory before tackling the intangible stuff.
The Ultimate OB is an interesting project, too interesting for you guys to bow out.
MJK said:
If you perceived my comment as taking a "shot" at Earl - you are mistaken. (.if not, my bad - and in any event no offense taken.
)I think an AES forum would be an EXCELLENT idea, don't you think so? (..it certainly wasn't intended to "shoo-off" Earl to some other forum - he has already stated he has no intention of participating here, and in fact *I* think he would be a very valuable addition if he was actually contributing to this thread's purpose.)
As to the design and progress:
This IS Lynn's thread, he will be doing the engineering when he wants to and is physically capable. (He has said so numerous times.) Despite that a few others have already started on their own designs as an "off-shoot" (..which is particularly nice because they can add information on what did and did not seem to work - concurrently with this thread).
And sure enough Lynn has done OB modeling, as have I (when making an OB suggestion) - and no doubt several others have as well. I'm also fairly sure Lynn has also done some modeling of a 2.5.5.5.5.5.5 (etc.) crossover to see whats possible/practical. (..remember, just because it isn't explicitly posted doesn't mean it hasn't happened.)
The KEY point though is that the design isn't finalized. Sure, the shear length of the thread is daunting and often filled with off-topic comments (like ours here).
IMO the point of this thread is gathering ideas and to act as Lynn's personal "sounding board" to see what gets the "nod" for the final design. In other words he gets to conceptually "kick the tires" here while he goes through post-op waiting and physical therapy.
When he actual does finalize the design - we can only hope that he does so in another thread here.
As to others participating on the engineering-end ..open baffles (in a myriad of forms), and horns with CD's have been in discussion continually though somewhat sporadically - sadly, most who could interject with greater engineering depth on either topic have not. Perhaps its the "noise" level or the length of the thread?
Bottom-line: I'm fairly certain the Ariel wasn't conceived and built in a day, or even a few months. Now add-in a currently "bed-bound" engineer/designer and the amount of time it will take before prototyping will even begin. So if you are in a hurry to see something actually take form - you are in for a bit of a wait. In the mean time why not suggest something on the topic of open baffles that hasn't been covered and/or is more in-depth with an engineering perspective?
When the Cat's Away ...
The Mice Will Play - take off for a few hours to get a little fresh air, and things get really wild in here!
Later this summer, the magical MLSSA and supernatural SoundEasy will be churning out lots of pretty-looking graphs. Wizard's caps will not be required to interpret them. Whether this will overlap with the RMAF remains to be seen. At least I'll be out of the walker and wheelchair by then, at the rate I'm going.
Today I took my first step upwards into the house, a tiny but significant achievement. This replaces the sit-in-a-chair-and-swing-the-walker-around method of the last several months. Left-to-right symmetry on the walker is improving as well. I can also stand still, outside of the walker, and hold things in my hands, for the first time since January. Bit by bit, day by day, week by week.
I've been reading Augusten Burroughs (Running with Scissors, Dry, Magical Thinking, and Possible Side Effects). I think it's had an effect on my writing, and I'm not sure if that's a good or bad thing. (Fair warning to the tender-minder reader - some chapters in the first two books are not an easy read.) He's a stunningly good writer, and works in a difficult form - the memoir.
The Mice Will Play - take off for a few hours to get a little fresh air, and things get really wild in here!
Later this summer, the magical MLSSA and supernatural SoundEasy will be churning out lots of pretty-looking graphs. Wizard's caps will not be required to interpret them. Whether this will overlap with the RMAF remains to be seen. At least I'll be out of the walker and wheelchair by then, at the rate I'm going.
Today I took my first step upwards into the house, a tiny but significant achievement. This replaces the sit-in-a-chair-and-swing-the-walker-around method of the last several months. Left-to-right symmetry on the walker is improving as well. I can also stand still, outside of the walker, and hold things in my hands, for the first time since January. Bit by bit, day by day, week by week.
I've been reading Augusten Burroughs (Running with Scissors, Dry, Magical Thinking, and Possible Side Effects). I think it's had an effect on my writing, and I'm not sure if that's a good or bad thing. (Fair warning to the tender-minder reader - some chapters in the first two books are not an easy read.) He's a stunningly good writer, and works in a difficult form - the memoir.
This Is Too Good To Let Slip By
Most fascinating discussion of "horn honk", by Jack Bouska. The simple rolled-up magazine test (mentioned in Part 3) makes the point, I guess maybe that's the reason I've never warmed up to conical horns - the resemblance to a simple megaphone is just too close.
EG's patented technique is of great interest, and I suspect there may be other ways to address the issue as well.
A quote from the linked post:
"In addition to the criteria of flat frequency response, I believe that the second biggest influence on the subjective quality of horns or waveguides is the unpleasant phenomenon known as “horn-honk”. This has been incorrectly attributed to a variety of causes, such as compression chamber and/or throat distortion, resonances in the horn cavity or walls, and even abrupt changes of DI between cone and horn transducers in a loudspeaker system. While the previously mentioned mechanisms can contribute to poor quality sound, the real culprit behind horn-honk is caused by poor mouth termination, and/or rapid flare rate changes within a horn or waveguide. In answer to Ian’s comment on subjectively poor performance of some CD horns, I note that many of the commercial devices on the market rely on diffraction slots and abrupt internal slope breaks (flare rate changes), while also displaying poor mouth to baffle impedance matching, all of which are responsible for imposing varying levels of horn-honk to the sonic character. Unfortunately, good CD does not guarantee good sonic performance."
<snip>
"Dr. Holland concludes that the poor sonic quality of many horns is attributable to internal reflections between the mouth and throat. The horn-honk caused by these internal reflections can be easily detected as a distinct tonal aberration, similar to talking through a cardboard mailing tube, or a small tunnel. (Send me private mail if you are interested in this article.)
Addendum: Please note that in Post #5 of my oblate spheroid thread, I incorrectly described this comb filtering effect as a resonance. I apologize if I mislead anyone by my use of that misnomer, as horn-honk is not caused by an internal resonance, but rather by one or more discrete reflections, between mouth and throat. In contrast, a typical fundamental resonance mode would exhibit a response peak much lower in frequency, which even for short horns would be below 500Hz. (eg: a 13” long horn would have a resonant frequency of 250 Hz, and a 7” horn resonance would be just over 500 Hz)
The broad-band comb filter effect, associated with horn-honk, is created by reflection, and summation of delayed signals, which are added back into the primary acoustic output of the horn (with alternating polarity). The delay & addition of any broad-band signal with itself creates a comb filtering effect, which on a log scale amplitude plot appears as a sinusoidal ripple in an otherwise smooth response. The period of this sinusoidal pattern is related to the time delay of the reflection according to the equations:
Time period = 1 / frequency
(This period is measured between peaks on a linear frequency scale spectral graph).
The comb filtering effect can be band limited in the case where the internal reflections are not broad band. In Tractrix or Exponential horns, the higher frequencies beam down the center of the horn, and do not diffract or reflect from the mouth edge, and so the response ripples will be limited to the lower octaves of the horn bandwidth (which is in the human voice range, where we are quite sensitive). In the case of CD horns, if the mouth has some form of flair, or radius, the high frequencies will be presented with a better acoustic impedance match, thus limiting the reflections to the lower octaves, similar to Tractrix and exponential horns. The reflection induced ripples in the horn frequency response have the same root cause as the impedance bumps which might be observed in the first octave or two above cutoff, these are attributed to improper mouth dimension for a given flare rate."
Most fascinating discussion of "horn honk", by Jack Bouska. The simple rolled-up magazine test (mentioned in Part 3) makes the point, I guess maybe that's the reason I've never warmed up to conical horns - the resemblance to a simple megaphone is just too close.
EG's patented technique is of great interest, and I suspect there may be other ways to address the issue as well.
A quote from the linked post:
"In addition to the criteria of flat frequency response, I believe that the second biggest influence on the subjective quality of horns or waveguides is the unpleasant phenomenon known as “horn-honk”. This has been incorrectly attributed to a variety of causes, such as compression chamber and/or throat distortion, resonances in the horn cavity or walls, and even abrupt changes of DI between cone and horn transducers in a loudspeaker system. While the previously mentioned mechanisms can contribute to poor quality sound, the real culprit behind horn-honk is caused by poor mouth termination, and/or rapid flare rate changes within a horn or waveguide. In answer to Ian’s comment on subjectively poor performance of some CD horns, I note that many of the commercial devices on the market rely on diffraction slots and abrupt internal slope breaks (flare rate changes), while also displaying poor mouth to baffle impedance matching, all of which are responsible for imposing varying levels of horn-honk to the sonic character. Unfortunately, good CD does not guarantee good sonic performance."
<snip>
"Dr. Holland concludes that the poor sonic quality of many horns is attributable to internal reflections between the mouth and throat. The horn-honk caused by these internal reflections can be easily detected as a distinct tonal aberration, similar to talking through a cardboard mailing tube, or a small tunnel. (Send me private mail if you are interested in this article.)
Addendum: Please note that in Post #5 of my oblate spheroid thread, I incorrectly described this comb filtering effect as a resonance. I apologize if I mislead anyone by my use of that misnomer, as horn-honk is not caused by an internal resonance, but rather by one or more discrete reflections, between mouth and throat. In contrast, a typical fundamental resonance mode would exhibit a response peak much lower in frequency, which even for short horns would be below 500Hz. (eg: a 13” long horn would have a resonant frequency of 250 Hz, and a 7” horn resonance would be just over 500 Hz)
The broad-band comb filter effect, associated with horn-honk, is created by reflection, and summation of delayed signals, which are added back into the primary acoustic output of the horn (with alternating polarity). The delay & addition of any broad-band signal with itself creates a comb filtering effect, which on a log scale amplitude plot appears as a sinusoidal ripple in an otherwise smooth response. The period of this sinusoidal pattern is related to the time delay of the reflection according to the equations:
Time period = 1 / frequency
(This period is measured between peaks on a linear frequency scale spectral graph).
The comb filtering effect can be band limited in the case where the internal reflections are not broad band. In Tractrix or Exponential horns, the higher frequencies beam down the center of the horn, and do not diffract or reflect from the mouth edge, and so the response ripples will be limited to the lower octaves of the horn bandwidth (which is in the human voice range, where we are quite sensitive). In the case of CD horns, if the mouth has some form of flair, or radius, the high frequencies will be presented with a better acoustic impedance match, thus limiting the reflections to the lower octaves, similar to Tractrix and exponential horns. The reflection induced ripples in the horn frequency response have the same root cause as the impedance bumps which might be observed in the first octave or two above cutoff, these are attributed to improper mouth dimension for a given flare rate."
If I might suggest, the approach that I always take is to specify the objective, then see what of the alternatives best meet the objectives. It sounds to me like this thread has decided that "OB is the design, now whats the objective".
I have looked at OB several times with the idea to see if and how it meets the objectives. Every time I come up with one or more problems that seem to outweigh the advantages.
I really believe that to get to "the answer" we should first look at the "objectives". When those are clearly understood then it is much clearer how the design should follow.
A thread simply focused on "What should a loudspeaker do?" is long overdo. There is no simple answer here either since the room plays a very significant role and "there are many rooms". My expertise and focus is on "small rooms", they interest me the most as they are what we tend to live in on a daily basis.
Had I the time, I would start a thread on "The objectives of a loudspeaker in a small room" as this would parallel the central focus of my lifes work. I could start it, but I'm likely to disapear after a few days.
I have looked at OB several times with the idea to see if and how it meets the objectives. Every time I come up with one or more problems that seem to outweigh the advantages.
I really believe that to get to "the answer" we should first look at the "objectives". When those are clearly understood then it is much clearer how the design should follow.
A thread simply focused on "What should a loudspeaker do?" is long overdo. There is no simple answer here either since the room plays a very significant role and "there are many rooms". My expertise and focus is on "small rooms", they interest me the most as they are what we tend to live in on a daily basis.
Had I the time, I would start a thread on "The objectives of a loudspeaker in a small room" as this would parallel the central focus of my lifes work. I could start it, but I'm likely to disapear after a few days.
It is certainly easy to come to the conclusion that this is another of those thrown together projects, with one idea after another suggested and then thrown aside to be replaced by another, with no testing or direction. There are indeed such projects, and I for one have been guilty of participating in them on occasion. -sometimes that's fun in fact!
BUT unless you have followed from day one, you may be unaware that due to his severely broken leg, Lynn was barely capable of getting out of bed at the beginning and while he has made much progress is unable to make test baffles, set up test gear, even carry boxes of drivers into his basement where his equipment is.
So we are all having a good time suggesting possible areas that Lynn can investigate when he gets better. Things have stagnated a bit but his condition should be kept in mind when critisizing this. At the same time he has gotten lots (too much?) information, much of it probably well worth studying when he gets serious.
It is pointless to try to design the system for him -Lynn will do that when he is capable - after all , it's his speaker. He has pointed out that in the process he will use MLSSA, etc. , make lots of mistakes, learn a lot..... but that process has barely started, because it really isn't possible for him to do so.
While some estalished authorities complain of attacks rightfully, sometimes it seems that they are used to only agreement and interpret any questioning as an attack. Other members attack seemingly without reason just because they can. This is a problem here as in other forums, but this place isn't static. As SY has pointed out, we have started a strong effort to have the most civilized discussion of all the audio sites, and we are getting positive results. More useful discussions are taking place and more people who are participating have meaningful knowledge to offer.
In the future, we hope to be able to "publish" papers by people that have worthwhile things to contribute and wish to do so in a structured way. Changes are happening, but in their own time I'm afraid. A lot of this is going on in the background so isn't visible to most members.
Mr. Geddes certainly sums things up well in his most recent post.
"Not everyone can afford to do ellaborate subjective tests, and not every advance has to be so proven. But, one must always be careful not to claim unproven opinions as facts. WHEN scientific data is available use it, when you only have your own personal observations, use them too, but ALWAY be sceptical of their validity because the human psychy is not to be trusted"
Too bad he seems to have so many irons in the fire!
I think that another thread on Open Baffles is certainly not too many, and might prevent Lynn's thread from getting too wide ranging...
Variac,
DIY Audio Chief Moderator
BUT unless you have followed from day one, you may be unaware that due to his severely broken leg, Lynn was barely capable of getting out of bed at the beginning and while he has made much progress is unable to make test baffles, set up test gear, even carry boxes of drivers into his basement where his equipment is.
So we are all having a good time suggesting possible areas that Lynn can investigate when he gets better. Things have stagnated a bit but his condition should be kept in mind when critisizing this. At the same time he has gotten lots (too much?) information, much of it probably well worth studying when he gets serious.
It is pointless to try to design the system for him -Lynn will do that when he is capable - after all , it's his speaker. He has pointed out that in the process he will use MLSSA, etc. , make lots of mistakes, learn a lot..... but that process has barely started, because it really isn't possible for him to do so.
While some estalished authorities complain of attacks rightfully, sometimes it seems that they are used to only agreement and interpret any questioning as an attack. Other members attack seemingly without reason just because they can. This is a problem here as in other forums, but this place isn't static. As SY has pointed out, we have started a strong effort to have the most civilized discussion of all the audio sites, and we are getting positive results. More useful discussions are taking place and more people who are participating have meaningful knowledge to offer.
In the future, we hope to be able to "publish" papers by people that have worthwhile things to contribute and wish to do so in a structured way. Changes are happening, but in their own time I'm afraid. A lot of this is going on in the background so isn't visible to most members.
Mr. Geddes certainly sums things up well in his most recent post.
"Not everyone can afford to do ellaborate subjective tests, and not every advance has to be so proven. But, one must always be careful not to claim unproven opinions as facts. WHEN scientific data is available use it, when you only have your own personal observations, use them too, but ALWAY be sceptical of their validity because the human psychy is not to be trusted"
Too bad he seems to have so many irons in the fire!
I think that another thread on Open Baffles is certainly not too many, and might prevent Lynn's thread from getting too wide ranging...
Variac,
DIY Audio Chief Moderator
Re: This Is Too Good To Let Slip By
"the real culprit behind horn-honk is caused by poor mouth termination, and/or rapid flare rate changes within a horn or waveguide."
"Please note that in Post #5 of my oblate spheroid thread, I incorrectly described this comb filtering effect as a resonance. I apologize if I mislead anyone by my use of that misnomer, as horn-honk is not caused by an internal resonance, but rather by one or more discrete reflections, between mouth and throat. In contrast, a typical fundamental resonance mode would exhibit a response peak much lower in frequency, which even for short horns would be below 500Hz. (eg: a 13” long horn would have a resonant frequency of 250 Hz, and a 7” horn resonance would be just over 500 Hz)
The broad-band comb filter effect, associated with horn-honk, is created by reflection, and summation of delayed signals, which are added back into the primary acoustic output of the horn (with alternating polarity). The delay & addition of any broad-band signal with itself creates a comb filtering effect, which on a log scale amplitude plot appears as a sinusoidal ripple in an otherwise smooth response. "
My entire study of "waveguides" was aimed at one thing - to get rid of (God I hate this phase) "Horn Honk". Lets just say that "horns" have an unnnatural quality about them that it would be desirable to eliminate. In the late 80's I did several papers on horn modeling only to find that there was a big hole in the theory of "horns" that prevented me from really getting anywhere. Thats when I went the next step to "Waveguide Theory" which uses an entirely different set of equations than "Horn theory". This theory was in fact more complete and allowed for a better understanding of the details of wave propagation in these devices, but alas it was also far far more mathematically complex. SO complex, in fact, that the solutions were not really even know at the time of my first writing. This has indeed limited the understanding of this new approach as only a highly proficient mathematician can follow the development.
BUT, this new theory predicted some things that the old theory did not. That is the Higher order modes that can, and do, propagate in waveguides. These became of serious interest to me. I studied them for the next ten years - most of which was not published. Lidia nd I published our subjective analysis of these HOMs at the AES last year, but alas, as God would have it, we were BOTH called away to Lidia's mothers funeral and this paper was never presented. In it we showed how the ear is actually the nonlinear element and that the HOM, which are dispersive, are perceived as nonlinear distortion. Quite a revolutionary concept, it has received virtually no attention, and yet, to me, it is the answer to "Horn Honk".
Sorry for the complexity (I can't make brain sugery any easier), but after nearly 20 years of intensive investigation, I have come to this conclusion. And anyone who has heard my waveguides must agree they don't have the typical "Horn sound". They are as natural as any direct radiator. Mouth reflections are a factor (I would call this a "resonance") but HOM's are the real culprit to "Horn Honk". The current genre of CD (I say that with a smile) horns are based on diffraction which maximizes HOMs - net result, nearly CD with terrible sound. Virtually everyone today is steering away from diffraction techniques.
Well thats enough "lecturing" for tonight.
"the real culprit behind horn-honk is caused by poor mouth termination, and/or rapid flare rate changes within a horn or waveguide."
"Please note that in Post #5 of my oblate spheroid thread, I incorrectly described this comb filtering effect as a resonance. I apologize if I mislead anyone by my use of that misnomer, as horn-honk is not caused by an internal resonance, but rather by one or more discrete reflections, between mouth and throat. In contrast, a typical fundamental resonance mode would exhibit a response peak much lower in frequency, which even for short horns would be below 500Hz. (eg: a 13” long horn would have a resonant frequency of 250 Hz, and a 7” horn resonance would be just over 500 Hz)
The broad-band comb filter effect, associated with horn-honk, is created by reflection, and summation of delayed signals, which are added back into the primary acoustic output of the horn (with alternating polarity). The delay & addition of any broad-band signal with itself creates a comb filtering effect, which on a log scale amplitude plot appears as a sinusoidal ripple in an otherwise smooth response. "
My entire study of "waveguides" was aimed at one thing - to get rid of (God I hate this phase) "Horn Honk". Lets just say that "horns" have an unnnatural quality about them that it would be desirable to eliminate. In the late 80's I did several papers on horn modeling only to find that there was a big hole in the theory of "horns" that prevented me from really getting anywhere. Thats when I went the next step to "Waveguide Theory" which uses an entirely different set of equations than "Horn theory". This theory was in fact more complete and allowed for a better understanding of the details of wave propagation in these devices, but alas it was also far far more mathematically complex. SO complex, in fact, that the solutions were not really even know at the time of my first writing. This has indeed limited the understanding of this new approach as only a highly proficient mathematician can follow the development.
BUT, this new theory predicted some things that the old theory did not. That is the Higher order modes that can, and do, propagate in waveguides. These became of serious interest to me. I studied them for the next ten years - most of which was not published. Lidia nd I published our subjective analysis of these HOMs at the AES last year, but alas, as God would have it, we were BOTH called away to Lidia's mothers funeral and this paper was never presented. In it we showed how the ear is actually the nonlinear element and that the HOM, which are dispersive, are perceived as nonlinear distortion. Quite a revolutionary concept, it has received virtually no attention, and yet, to me, it is the answer to "Horn Honk".
Sorry for the complexity (I can't make brain sugery any easier), but after nearly 20 years of intensive investigation, I have come to this conclusion. And anyone who has heard my waveguides must agree they don't have the typical "Horn sound". They are as natural as any direct radiator. Mouth reflections are a factor (I would call this a "resonance") but HOM's are the real culprit to "Horn Honk". The current genre of CD (I say that with a smile) horns are based on diffraction which maximizes HOMs - net result, nearly CD with terrible sound. Virtually everyone today is steering away from diffraction techniques.
Well thats enough "lecturing" for tonight.
Lynn,
From the various prototypes that I went through with my current waveguides, I can't agree that horn honk has anything to do with the mouth termination, but instead is caused by reflections between the surfaces of the horn itself. That puts me firmly in Earls camp in that regard. I tried quite large roundovers and even tried significant foam padding near the terminus wrapping around to the back (similar to what Peavy does with its quadratic WG's). My current WG's also had that same unnatural tonality until I lined the interior surfaces with 1/2" foam, which cured the problem entirely even though I left the terminus as bare wood. It still have unresolved issues with the terminus, but they seem to affect imaging, not tonality, just as the edge termination does with any speaker.
From the various prototypes that I went through with my current waveguides, I can't agree that horn honk has anything to do with the mouth termination, but instead is caused by reflections between the surfaces of the horn itself. That puts me firmly in Earls camp in that regard. I tried quite large roundovers and even tried significant foam padding near the terminus wrapping around to the back (similar to what Peavy does with its quadratic WG's). My current WG's also had that same unnatural tonality until I lined the interior surfaces with 1/2" foam, which cured the problem entirely even though I left the terminus as bare wood. It still have unresolved issues with the terminus, but they seem to affect imaging, not tonality, just as the edge termination does with any speaker.