Horn Theory Observation and K&A Opportunity

[boom3]

There was a message posted recently about (as I recall) the 'wavefront' model versus the 'transmission line' model of horn design, with some comments about reflectors at the corners of the folds. Got me thinking...

I've long been associated with engineers who work on high-velocity, high-volume air handling systems, and a thought has occured to me. First off, all us horn guys (myself included) are too close to the problem. We have heads full of Klipsch, JBL, Voigt, etc and we always refer to the problems in light of past commercial designs.

I think horn theory needs study by engineers (M.E.s) who are adept at air flow, not hi-fi (E.E.) types. Sound flow is air flow, after all, and high-velocity, high delivery airflow is studied by engineers using very sophisticated modeling and analysis tools.

I would hope Klipsch would lead the way in interfacing with this discipline of engineering, which deals with fans, blowers, propellors and air delivery ducting. It's a very significant business segment, one that dwarfs the horn loudspeaker industry.

[sunnysal]

certainly horn design could benefit from the application of engineering in air movement as you postulate BUT we are delaing with both issues, movement of air and wave propagation. both need to be taken into account. I suspect each becomes more and less important at different stages int he sound production. i.e. in the throat area air issues are more at the fore, I assume as the sound moves past the throat the wave issues take over until they dominate...regards, tony

[D-MAN]

The issues of air flow and sound wave propogation are quite different. Horns are not intended to move volumes of air from one point to another as would an air flow conduit of some sort.

Sound waves are both positive AND negative (compression and rarefaction) rather than a consistant and constant flow of a large volume of air in a single direction. Durations are therefore very short in time and change directions of air movement.

In the case of horns, the effect is NOT an continuous air flow coming out of the horn itself, but a vibrating sheet of air the size of the horn mouth. The limit of air movement (vibration) is constrained by the excursion of the driver cone (in the case of bass horns).

DM

[boom3]

----------------

On 4/28/2005 12:12:56 PM D-MAN wrote:

The issues of air flow and sound wave propogation are quite different. Horns are not intended to move volumes of air from one point to another as would an air flow conduit of some sort.

Sound waves are both positive AND negative (compression and rarefaction) rather than a consistant and constant flow of a large volume of air in a single direction. Durations are therefore very short in time and change directions of air movement.

In the case of horns, the effect is NOT an continuous air flow coming out of the horn itself, but a vibrating sheet of air the size of the horn mouth. The limit of air movement (vibration) is constrained by the excursion of the driver cone (in the case of bass horns).

DM

----------------

Horns are air conduits, so I disagree with your first point, and I will amplify below.

Yes, sound waves are positive and negative, as we all should know, and the advanced modeling and computing power that aerodynamicists (sp?) have access to should accomodate that.

I intuite that below some frequency-what I call a critical frequency (Cf) -any acoustic duct (horn, T/L what have you) acts more like a pipe constraining a fluid. Above that critical frequency, wavelength effects come into play, and we need to start pondering reflectors and numbers of folds. My view (I will not dignify it with the label of theory or even hypothesis)is that below the Cf we are dealing with a 'leaky' pressurized vessel and the mechanism for calculating flow effiency-a function of impedence matching between input (throat) and output (mouth) are well known in the fluid dynamics world. We've been trying since the Nine-Teens to approach this from an equivalent electric circuit model, and it's just not robust enough for the job.

Our horn design programs are really glorified spreadsheets cranking out results from the formulae of choice (Rayleigh, Marsh, Keele, etc).

Their software and techniques are simply more powerful than ours, so why not lay the questions before a fresh perspective with more modeling and simulation horsepower?

[D-MAN]

Inciting a riot:

Horns are NOT air conduits unless you use a fan in one instead of a driver.

DM

[dragonfyr]

"I intuit that below some frequency-what I call a critical frequency (Cf) -any acoustic duct (horn, T/L what have you) acts more like a pipe constraining a fluid. Above that critical frequency, wavelength effects come into play, and we need to start pondering reflectors and numbers of folds. My view (I will not dignify it with the label of theory or even hypothesis) is that below the Cf we are dealing with a 'leaky' pressurized vessel and the mechanism for calculating flow efficiency-a function of impedance matching between input (throat) and output (mouth) are well known in the fluid dynamics world. We've been trying since the Nine-Teens to approach this from an equivalent electric circuit model, and it's just not robust enough for the job."

-------------------------------------------------------------

I am ultimately going to duck regarding this issue, but just a few points.

This seeming distinction between air and fluid dynamics

Air is a fluid. But being a gaseous 'fluid', it is distinct from a liquid fluid in that air is compressible, while liquids are not.

And while we posit seemingly disparate disciplines attacking the elephant, may I suggest that there indeed is a discipline that sees these hard distinctions simply as ones of convenience and doesn't suffer from the potential myopia of any particular area which Chooses to limit their frame of reference. It's called "physics", but many here don't like that word being used!

And while I have no dog in this hunt, and I don't posit electrical modeling as any 'special' panacea, I am curious to know why the electrical models are considered fundamentally limited, as the models really have very little to do with electronics in the broadest sense. And the individual components can be modified to allow for just about any behavior desired. They are simply mathematical algorithms, and its hard to see how mathematics is so limited! I might suggest that if we seek a limitation, that we look at those defining the models! As, after all, the question asked in large measure determines the outcome! Garbage in - garbage out.

So my curiosity is more with the perception of the 'systems' than with any particular system...

"Have fun" he was heard to say as he ran for the sidelines with a big grin!!

PS. But the horns employed here are not compressed gas horns nor about moving volumes of air! The are simply waveguides for the propogation of a waveform through (in this case) air, while employing an electro-mechanical/air boundary layer transformer interface.

[sunnysal]

I have to agree that audio horns are largely wave guides...the air movement is slight and only really would have any "movement" or pressure in the throat as I mentioned earlier, that is where the air issue might have some relevance...tony

[lynnm]

I am inclined to think that as pointed out above that the laws of physics ( both mechanical and electrical )have been and continue to be seriously considered by competent designers of horn loudspeakers.

What has been overlooked so far in this discussion is that in addition to PWK's Electrical Engineering background he was also an acknowledged expert in the field of ballistics having earned patents in that field.

I submit that anyone versed in the field of ballistics would have a keen awareness of the effects of air flow whether at the velocities generated by Howitzers or at the exit flare of a horn loaded loudspeaker.

I am not BTW arguing that Mechanical Engineers have nothing to contribute in this field and I would not be the least suprised to learn that some ME's are employed in the area of speaker design. I am aware that ME's are involved in structural design and acoustics and suspect that some of those skill sets are transferable across the usual boundaries of the various Engineering professions.

[D-MAN]

A horn is a acoustic pressure-reducing device. An acoustic transformer. The SAME elements of a horn can be applied to an air transporting conduit basically to match pressures, but acoustic horns have WAY more considerations to them than simply moving air from one point to another.

In an acoustic horn, high driving pressure at the throat is reduced to low pressure at the mouth as a sheet of vibrating air (think of a sheet of cellophane covering the horn mouth). The degree of vibration at the mouth (where it meets the atmosphere) is less than the vibration seen at the throat where it is at maximum pressure, of course.

So high pressure, low velocity particle VIBRATION is translated to low pressure, high velocity particle VIBRATION at the mouth, the electrical equivelent is a transformer. Did I get that right about the velocity part?

The point is: in the simplest terms, you SHOULD think of the horn mouth as a sheet of air vibrating in correspondence to the movements of the driving cone (bass horn) because the air in the horn channels COUPLES the horn mouth to the diaphragm (through the throat) and moves more-or-less as a WHOLE air column (in a manner of speaking) within its respective bandpass, but reduces the degree of movement in either direction as it is propogated along its length as determined by the cross-sectional area expansion, as the channels widen, pressure naturally drops accordingly, ESPECIALLY determined by the mouth area which has to match atmosheric pressure as best it can. Frequency cutoff notwithstanding, placement of the horn mouth can also have an effect on its required size, in that, for instance, corner placement allows for a smaller overall mouth area to be used than the same-sized horn placed next to a wall or out in center space, where larger and larger mouth areas are required, respectively.

Now, this leaves out the various effects of expansion rates and types (parabolic, conical, hyperbolic, exponential, catenary, tractrix) on frequency response, overall distortion, and SPL consistancy, and otherwise ignores the requirements of horn design, but it should serve to illustrate the acoustic operation anyway.

DM

[boom3]

----------------

On 4/28/2005 2:49:23 PM D-MAN wrote:

A horn is a acoustic pressure-reducing device. An acoustic transformer.

(snip)

***

Yes, I hope we all know this. What I hoped for, am still hoping for, is that the K&A engineers, through the good offices of the Moderators, might notice this thread and contact some of these other disciplines. Maybe also those of us that are in engineering and/or science fields might foster this cross-pollination.

Part of creativity is taking seemingly unrelated facts, methods or arts and finding new, legitmate, testable correlations. Maybe a few the "mustard seeds" will find fertile ground.

[dragonfyr]

A horn is an electro-mechanical transformer. The medium the transformer is addressing is air.

It is not attempting to 'move air'. It is propogating sound waves within the medium.

And I would seriously question the qualifications of an engineer who did not have significant common coursework in the basic areas of (basic) physics, ciruits, mechanics, dynamics.

And if we are positing that a 'shop' such as Klipsch or any others would be myopic enough to simply talk with other EE's or to simply associate with only ME's tells me that we should all be looking elsewhere. Or better yet, perhaps they should be hiring Physicists who make no grand distinction between EE and ME, except to say that those disciplines simply stop at those respective chapters in an Intro physics course and focus simply upon the practical applications in those 'chapters', while physics goes beyond both! The physicists indeed finish the book and go on to other more advanced books!.

If only the world were so simple. Perhaps there is a use for physics after all, despite the cries of "burn the witch" that so often arise here when that word is mentioned!

[WMcD]

I can appreciate Boom3's comments. We'd all like to hope there is a deep mine of analysis and data in other fields which has been overlooked, and can be adopted with remarkable results.

However, "being too close to the problem" sounds like something from Forbidden Planet and psychology.

The cross pollination is also a bit funny. I had a filler job at a law library a few decades ago. An odd gentleman creaked in.

He assumed the role of a vast genius. Is this not the main campus? Where are the laboratories? Oh, this is a law school? Can you find me a student to lead me to the engineering school. What, law professors don't live with engineering professors? The essential cross pollination of any intellectual pursuit is obviously missing. I must consult with the Dean about this misfortune.

So he signed the log in with a big X. Excuse me, that is my name. Anyone will know me by that. He went away.

- - -

Seriously. The biggest advance to speaker design is that it is modeled with electrical circuits. So ME and EE's are doing the same thing and cross pollinizing (sp).

There may be something in the HVAC people's field of study that can help. Still, they are going in the opposite directions, as describe by others above. They are minimizing acoustic issues. Granted, this could yield some interesting data.

Gil

[markus111]

Please - don't bring in the physicists! I am an EE, and I work for physicists. The physicists I know would design the perfect speaker. It would be incredible, magnificent, sublime.... It would be the best speaker in the universe that could NEVER possibly be built by mankind.....

[dragonfyr]

The EE's could as well, if they had taken the rest of the courses!

Such a generalization! But ignorance is bliss! One understands the first principles of solid state physics and the other knows that there are solid state devices.

[markus111]

Yeah, we have some great banter. The engineers tell physicists jokes, and the physicists tell engineer jokes. And in both cases, that darn light bulb never gets changed...... I'm kidding about the physicists. We have a lot of great guys at work. Only one or two fall into the nutty professor category!

Seriously though, this is an interesting thread and I don't want to derail it. One possible reason for bringing in an "air" guy would be to study the sources and effects of turbulence in a horn. Reading Bill Fitzmaurice's articles in AudioExpress, he goes to extreme lengths to round internal corners and make the horn folds as smooth as possible to avoid turbulence. It may be interesting to do a very detailed model of the way air moves around horn foldings. ME's may already be doing this - I have no idea.

Mark

[dragonfyr]

Wave propogation is a well advanced field.

A simple way to model this without the MathCad or other advanced software models is to simply construct the barriers in a wave tank - a fancy name for a tray of water.

This makes the stimulus and subsequent interaction of the waves readily apparent...just a thought, as I suspect everyone here isn't an engineer or a physicist...