Horn Speakers - Is it me or.......

[Digital_Thor]

It is my experience, that as soon as a horn becomes larger than what you see in waveguides, like coax and typical waveguided tweeters. Then you have to think about the listening distance and what room you put the speaker in, the position and EQ - differently than with traditional speakers.

I always feel that bigger horns "pulls" me in to their near field, and that normal speakers quickly let you stay in the far field - in normal-sized rooms.
With a well-designed and controlled horn, you can hear the tweeter at a 100m - whereas with a normal direct firing tweeter... it's being absorbed by air and everything in the room... as soon as you move a few meters away. So I believe that the typical downwards tilt, you always see in normal speakers, require a whole different way of thinking, when you EQ the sound of a given speaker, and then measure in your listening position - just as Toole wrote - like a hundred times.

Alone, the difference between my KEF speakers and my friend's JM lab Alto Utopia - is huge. The way the JM-lab fires in all directions, "lights" up the room with a ton more reflections in the upper frequencies, than my KEF does. So my stereo perspective is way more sharp in the middle, but way less "sparkly" - simply because of the way of the speaker radiates in a given room.
So I think you need to EQ the horn to have an "artificial" downwards slope, to be able to listen to it that close, that you normally would a traditional speaker with direct firing drivers.
The "horn effect" that I had, with a set of smaller 8" horns on compression driver, was like you get "shot" in your face. Every note is like a "slam" in your ears, and you have an explosive dynamic range at around 3-4m. Some people like that almost violent dynamic range and very loud peaks, though.
A normal tweeter seems "soft" in comparison, and never feels loud, before it becomes annoying - IMO.

So to me, it's like a trade-off, where a waveguide gives me a bit of in-room directivity control. The direct firing tweeter is too much of a mess, in normal reflective living rooms, but the horn is too concentrated and "loud" for relaxed listening levels.

Big directive horns make a ton of sense when entertaining a thousand people in a huge venue, where you actually need 140dB, and we're in an enormous free open space - but I find it hard to enjoy, in a normal reflective living room.

 

[Bjorn]

Please understand that a waveguide is simply a short horn and has nothing to do with size in width and height. A waveguide can therefore also be large.

 

[Digital_Thor]

^
I agree. There were someone ones saying, that all horns are waveguides - but not all waveguides are horns.

 

[Teeter]

I LOVE horn speakers for DD MOVIES and dialog.

 

[egellings]

^
I agree. There were someone ones saying, that all horns are waveguides - but not all waveguides are horns.

Does a horn work acoustically in a manner similar to an electrical transformer (xfmr)? The horn takes a high pressure low excursion wavefront at its throat and transforms that into a low pressure high excursion wavefront at its mouth, if I have that right. An electrical transformer does a similar thing with voltage and current. The xfmr can take a high current low voltage input and change that into a higher voltage lower current one. Of course, the xfmr can go both ways with voltage and current. I suppose a horn could go both ways, too, as in the little horn held up to the ear, used as a hearing aid before electronic ones were invented.

 

[Keith_W]

With a well-designed and controlled horn, you can hear the tweeter at a 100m - whereas with a normal direct firing tweeter... it's being absorbed by air and everything in the room... as soon as you move a few meters away.

Is that a typo? Even if you meant 10m that is still larger than most listening rooms.

 

[Mr. Widget]

Please understand that a waveguide is simply a short horn and has nothing to do with size in width and height. A waveguide can therefore also be large.

Perhaps part of this is semantics, but while a horn will certainly act as a waveguide and will help define the directivity of a driver, a properly designed horn will act as an acoustic transformer and help match the air load impedance on the diaphragm offering substantially increased sensitivity. A typical waveguide will help control a driver's directivity, but it will not substantially affect the sensitivity of a driver.

This is very easy to see in comparing two speakers that use the same driver, where one is horn loaded and the other is not. The Klipsch LaScala which is horn loaded sees an increased sensitivity of about about 9dB over the Cornwall. Both speakers use the exact same 15" woofer. True, the response below 70 hz is rolled off in the LaScala due to the small size of the bass horn, but over its bandpass the 15" woofer is definitely horn loaded.

 

[Bjorn]

Sure. The increased loading is because the horn is longer. Like previously mentioned; a waveguide is simply a short horn. Shorter horn means less loading.

The JBL 2384 is an example of a waveguide with decent dimension.

And one can design a waveguide with same dimension as the horn next to it.

 

[Digital_Thor]

Is that a typo? Even if you meant 10m that is still larger than most listening rooms.

No, I actually mean a 100m. Danley Sound demonstrate that on a parking lot, and shows that the tweeter hardly drops in level, even at great distance. Which will produce a way different result than the typical upper frequency drop off, like we see a normal speaker do, in normal rooms.
So just imagine this speaker at 3-4m in a normal living room. This would be almost linear at the listening position.

You can just make out the speaker in the distance:

 

[Digital_Thor]

The Odeon speaker, is the only horn speaker I actually liked.... it just seemed to work for me - the few times I got to hear it in Munich.

 

[garyrc]

Here are the results I get from my Klipschorns with Audyssey FLAT with tone controls FLAT in a treated room. The graph is cut off below 52Hz because I, and most people, would use a SUB below that. Actually, I can crossover my sub at either 60 Hz, or even 40 Hz, depending on the needs of the recording, the room PLUS the recording, and the listeners.

The sound is [or can be] ultra dynamic, exciting, and, to me, realistic.

I read the graph as about +/- 3 dB at and above the probable 60Hz crossover point. This is for flat-ish recordings, with music mostly from the Classical and Romantic periods, plus Jazz and Film scores.

 

[Digital_Thor]

But how do you measure this? Listening position?
There's a 5dB boost between 2 and 4 kHz, that looks like how the rest of the curve 'ripple'. So it seems like room-reflection-interference, which again indicates a measurement from the listening position - which should not be flat.
It's great that you are happy with your sound - that is the most important part
But to interpret your data, I need more.
Are your speakers measured anachoecly at some point... Also off-axis?

 

[Digital_Thor]

This one?
https://www.stereophile.com/content/klipsch-klipschorn-ak6-loudspeaker-measurements

 

[garyrc]

This one?
https://www.stereophile.com/content/klipsch-klipschorn-ak6-loudspeaker-measurements

No, that one is the AK6. Ours are AK4/5s, originally a much earlier model (from c.1980 CE, but given a stock upgrade in c. 2006 CE or so to AK5 ... after a few months, we, like some others over on the Klipsch Community Form, made a slight factory supervised change [one resistor, IIRC] to convert back to the AK4. A summary of the changes: then, and somewhat earlier, are:

1) New tweeter (the flat magnet K77F, redesigned very slightly when it was about 2 years old, i.e., c. 2003 CE), now front mounted).

2) New crossover network ("balancing network") bringing the crossovers to 450 Hz and 4,500 Hz, steep slopes including some EQ.

3) New midrange driver, the K55X.

4) Now has a non-metalic mid horn, with bracing ribs

5) Woofer and bass bin unchanged.

I agree that a whether an audiophile is pleased with their sound is the most important thing.

"There's a 5dB boost between 2 and 4 kHz, that looks like how the rest of the curve 'ripple'." Those are 1 dB divisions [10 dB between the bold type numerals] so I would tend to read the peak centered at 3 KHz as 3.5 dB or so.

I think the multicolored Stereophile graph of the new AK6 you linked, IMO, is of questionable validity, particularly in the bass. Stereophile tested the AK6 outside, raised on a furniture dolly, in the subjective reviewer's driveway. The AK6 was designed to be placed near a corner and flat on the floor. It still uses the corner and floor for boundary gain in the bass, or the beneficent floor and sidewall reflections PWK talked about which extend and partly load the bass horn. One doesn't want midrange/treble sidewall reflections right where a yardstick would hit the sidewall if pressed flat against the front of the mid-horn (top hat), so a little absorption on the wall there helps [see photo].

We have a wall to wall thick carpet and lots of diffusion due to many bookshelves and many of these diffusors my wife built to someone else's formula.

Objections to the requirements of the older Khorns [like mine] which had to be pressed fully into the corner [with neoprene or the like sealing them in] were that they could not be towed-in or towed-out to aim the tweeter and midrange right at one's ears, so the listener was pretty much "looking down the throats of the horns," as the company once said. I would guess that "near a corner" and "on the floor" means touching a wall with a part of the rear of the top hat, but with the Khorns towed, and on the floor. Also, they would not be so far into the room, and might look a little better in this position.

Yes, the measurement I included in my recent post was taken from the main listening position, centered, but not perfectly, and 13 feet away [4 meters?], in a 4,000+cu. ft. room.

It may have many ripples because the psych smoothing [I'm told] applies more weighting to peaks to simulate how we actually hear (dips being less noticeable -- is this true?), and uses 1/6 octave above 1KHz, but 1/3 octave below 100Hz, and variable smoothing [1/3 to 1/6] between 100Hz and 1KHz.

While most of my measurements may not be perfectly centered, I did no deliberately off-axis ones except the following truly ancient one. The people at Audyssey had answered a question on how to simulate an REW type measurement, with Audyssey's super secret "fuzzy logic" kind of device on. They said that all they could think of would be to take a RMS average from all 8 mic positions we had used when running Audyssey, but no guarantee that the results would be similar.

So here is the result FWIW -- ignore the "X" it was my crazy attempt to draw in what the terrible "X" curve would look like, should I want to try it. Now that I've read Amirm's paper on it, I know I don't want! I don't remember what the blue line was.

Here is another old one, smothered in EQ, then psyched, made from an average of the Khorns that I measured for extremely bass shy recordings:

 

[Newman]

There were someone ones saying, that all horns are waveguides - but not all waveguides are horns.

It would have better served you to not believe them. It's back to front.

 

[Andysu]

JBL 2380A screen channels with LFE.1 the stage bass cabs are muted , the 2380A can play motion picture soundtrack loudness with clarity here on the , lucasfilm ltd THX sound system

i not been in my THX cinema for 5 days been in bed , its hard tell even on laptop speakers if the sound sounds like JBL 2380A , unless i'm actually in the THX cinema then it sounds like professional JBL THX that is felt at 40kw !

Bear cat approves of JBL motion picture loudspeakers

 

[Keith_W]

It would have better served you to not believe them. It's back to front.

How is "all horns are waveguides but not all waveguides are horns" incorrect?

As I understand it:
- "waveguide" - shapes the radiation pattern of the speaker
- "horn" - improves impedance mismatch between cone and air and increases efficiency

The difference between the two is the depth of the throat, so it is a continuous spectrum when "horn" becomes waveguides".

For it to be back to front, i.e. "all waveguides are horns", this is untrue. Waveguides have a shallow throat, they can not improve efficiency.

Perhaps I misunderstood you.

 

[Duke]

This is the way Earl Geddes put it: "All waveguides are horns, but not all horns are waveguides."

In other words, all such devices which prioritize an approximately constant-directivity radiation pattern are both waveguides and horns; but not all such devices that improve the impedance mis-match also result in approximately constant-directivty radiation patterns, which is what characterizes a waveguide.

Even a shallow waveguide improves the impedance mis-match at the lower end of its frequency range, and this shows up as increased efficiency at the lower end of the frequency response curve.

That being said I understand the idea that it's a continuum, but according to Earl (who I believe was the first to use the term "waveguide" to describe a shallow smooth-sided constant directivity horn) the word "horn" applies across the continuum, while the term "waveguide" only applies to one end of the continuum. And yes there is a fuzzy transition zone on the continuum.

 

[Newman]

How is "all horns are waveguides but not all waveguides are horns" incorrect?

"All waveguides are horns, but not all horns are waveguides." - Earl Geddes, PhD, author "Acoustic Waveguide Theory".

Sometimes, in matters of classification and definition, the truth just is what it is by definition, and arguing any other way, even though logically, is just a byline.

The basic idea is that a horn is literally anything acoustic and shaped like an animal horn, ... starting from cutting the horn off of an animal and blowing through it, through to musical instruments, then moving through to putting them in front of a vibrating playback diaphragm, and the acoustic waveguide is a special subclass of the latter that has specific objectives for the beam width of the wavefront.

Cheers

 

[Keith_W]

This is the way Earl Geddes put it: "All waveguides are horns, but not all horns are waveguides."

Ah, I see. Thank you and @Newman for the correction.