dumb question about horn orientation

[jdm56]

Why do some klipsch designs use a vertical horn orientation, while others use horizontal? It would seem logical to me that broader dispersion in the horizontal plane would always be more desirable than it would in the vertical. Klipsch' new THX designs are what makes me question this.

And, horror of horrors, could I be wrong all these years? The dispersion from a horizontally oriented horn is greatest in the horizontal plane, ISN'T IT?

[garymd]

Some earlier klipsch designs had the horns mounted vertically so the speakers could be laid on their sides. I say this because of pictures in some of the literature from the early 60s and the optional risers for the large side of the cornwalls. Also, some prefer the way vertically mounted horns image and use them standing upright. Many folks use the vertical cornwalls laid on their sides as center channels for HT. I don't actually have an opinion as to which sound better. I've only heard them once at IndyKlipschFans place used as his center channel and it sounded great.

[jdm56]

It just seems to my way of thinking that if one way is right, the other must be wrong. Maybe that is a simple way of looking at it, but I'd like to know why a designer chooses one way or another for a particular speaker.

Gary, I didn't even know there was such a thing as "vertical cornwalls" till I had hung out here awhile!

[JohnA]

To further add to your confusion, the K-77 series of tweeters were designed as diffraction tweeters. At lower frequencies, they have wider dispersion with the long axis vertical. So, the longer axis of the horn is not necessarily the axis of widest dispersion.

I am not familiar with the speakers you reference, but the engineer will choose the best combination of packaging and dispersion control for the particular application.

[Seadog]

As I recall, Trey's Cornwalls that were at the hotel in Indy were vertical models. We tried them upright and on their sides and they sounded about the same in that lousy, hot room (does anyone else that was there have an opinion?). I have a/b'd the vertical and regular Cornwalls at my house. When upright they sounded slightly different, but when we laid the verts on the side, I could not pick out which was which in a blind test. I'm using a vert CW for my HT center, turned on the side, and I am extremely satisfied. The dialog (or whatever else is on that channel) sounds very good from any position in the room.

[white_shadow]

I would guesstimate a few things.

The throat of the wave guide or horn plays more of a role than does horn facia.

The wave must propagate along the wave guide, meaning that the longer distance the wave must travel, the frequency must compensate by increasing, and vice versa. I think this propels the wave forward, rather than disperse up and down. This is probably wrong, so someone explain this to me.

I noticed that my square horns are also shaped in a vertical, rectangular manner. Meaning more tall than wide. The top and bottom of the throat is also angled rather than straight, like the sides.

[garymd]

Obviously, many use a vert cornwall for their HT center. Does anyone use a horizontal cornwall on its side as a center? I'd bet in a blind test it would be hard to tell the difference.

[Klewless]

Horns do favor one direction over the other based on it's physical structure. In olden days, some horns were called diffraction where the widest dispersion "circled" the narrow side. This means the horizontal dispersion occured when the horn was vertical.

However if you look at horn (they call them lens now) specs they have things like 40 degrees by 60 degrees, indicating the intended dispersion pattern. So one whould look at the spec to determine which provides the pattern you want. Round horns tend to be 45 degrees all around like a ball.

But to make matters worse, this is frequency dependent and not constant throughout the entire range. This little speech only touches the subject. There is much, much more.

I have listened to my Cornwalls both ways (mine can be layed on their sides). I cannot really tell much difference except the vertical position moves the highs a bit higher like the LaScala.

Since I always listen in my "sweet spot" dispersion is not a problem because I simply toe the speakers in a bit to get what I want. For those who must satisfy a larger group (wider seating), the dispersion could be more problematic and I would think the wider horizontal pattern would be preferable.

Now my brain hurts with all this fancy typing. Hope it helps rather than confuses!!!!!!!!

[jdm56]

Very interesting stuff. I never really considered horns to be so...complex. I used to chuckle to myself when reading klipsch ads that touted their horn "technology". "Yeah right", I thought. "It's just a horn. -not exactly rocket science". But I am starting to appreciate just how complex "horn technology" and theory is.

[WMcD]

Its a long story and I've written about it before.

The horn mouth aspect ratio does not work like there is a flashlight in the back. Pretty much the opposite.

You must also look at the exit angle. Basically, to maintain a narrow pattern you need a large dimension. And the higher the frequency, the bigger the mouth dimension you need.

If you look at the 90 degree exit angle of the K-400 in the horizontal, the 17 inch size will maintain that down to about 400 Hz.

However the exit angle in the vertical is about 40 degrees and if you wanted to maintain that down to 400 Hz you'd need a very tall mouth. But we have only 5.6 inches. As a result, low freqencies are not controlled at low frequecies and the dispursion is very wide, almost 180 degrees at low frequencies. But, up at high frequencies the vertical dispursion is narrowing. The result is some on axis gain as frequency goes up.

The issue then becomes, where do you want the low frequency lack of control to go. In the K-Horn, it goes to the ceiling and floor. However, in the vertical oriented CW, it is kept off the floor and ceiling.

The so called tratrix horns are a bit better at control in both coordinates in that they have almost square mouths.

The new THX horns look to be designed even more in accord with what I'm saying about control. If you look very closely, the vertical exit angle is narrow, AND the height is greater. On the other hand, the horziontal angle is wider, and the mouth width is, correctly, narrower. I believe this was necessary to meet the narrow angles required by the THX spec.

All my information comes from a paper by Don Keele. He invented the constant directivity horn.

Gil

[kfalls]

Doesn't THX specify a limited horizontal dispersion to limit the overlap of the front channels to make dialogue more intelligible and to keep the sound more coheisive? Of course with a vertical horn the possibility of ceiling and floor reflections are a possibility.

[white_shadow]

I thought about its THX certification regarding horn orientation, but I dont think that is the issue or reason for the design. I think this because of the surround design. These are bipolar, I believe. That would cut short many arguments for vertical (or horizontal, with regards to imaging) horn dispersion. Unless certain speakers are designated specifically only as surrounds or mains.

The THX ultra2's are 7.2?

[jdm56]

Horn Theory 101: A tractrix or exponential horn with a rectangular "mouth" will always have wider dispersion at any given frequency in the plane of the longest dimension. Right? And this is regardless of flare?

[WMcD]

I looked at the Keele AES article again. He is speaking about his constant directivity design. That is hyperbolic at the throat, conical in the middle, and then a rapid flaring conical at the mouth.

None the less, you see the same effect in pure exponentials with mouths of just about any aspect ratio.

It makes sense if you're into RF antennae design. Low frequencies require big antennae. Our horn loaded woofers, squaukers, and tweeters follow that.

Also, if you want a narrow pattern of radiation off an antennae, the antenna must be larger than one with a broad pattern.

Keele's equation is that the mouth size for control is

X = 1,000,000/(angle * freq) The result there is in inches.

Looking at this, it means that for a 90 degree exit angle, you must have a size of about 1 wavelength. A 1 kHz signal has a wavelength of 13.5 inches.

If you look at the K-400, the 90 degree exit in the horizontal (big dimension) gives control to a bit under 1 kHz. Not quite down to 400 Hz though. You can crank the math.

The equation is described as being for Keele's conical type horns. It may be a bit hard to apply to an exponential. I.e., the last few inches of the exponential may be 90 degrees, which really isn't enough distance to count as such.

None the less books showing the radiation patterns of round exponentials show the same overall effect.

But going back to something like the Cornwall. When the long axis is up and down, the horn keeps most all frequencies off the floor and ceiling. Turn it so long axis is horizontal, and most all frequencies are kept off the walls.

Therefore we expect some different room reflections depending on orientation.

So, what is happening on the smaller axis? If you look at the horn the exit angle there may be 40 degrees or less. There is no control at low frequenies because the equation tells us that X must be very large, about 2 wavelengths and we have somewhere about 1/4 of that.

The result is that a low frequencies there is no control and the pattern is probably about 180 degrees.

However, looking at the equation, at some point in higher frequency, the X is satisfied, and there is a narrowing, which is to say control. This is the "controlled directivity" that PWK talks about in the interview by Edgar in SpeakerBuilder.

You should know that midrange drivers generally fall off in output at about 3000 Hz. However, this is just where the K-400 starts becoming directional (X is satified). Therefore the horn gives us some on-axis gain. This allows on axis flat output up to 6000 Hz. Further up, the gain of the horn is not enough to keep up with the fall in the driver output. But we turn on the tweeter.

There are some other interesting observations on the K-400. Looking at it in plan view (from the top) it looks an awful lot like the Keele design or the tratrix designed by Voight! The curve is exponential but we're in the spot where it comes close. PWK says he fished the design out of the wastepaper bin and it was by Jensen. So, all great minds have been fishing in the same waters.

Best,

Gil

[dougdrake]

Dude, where's my car?

[jdm56]

I'm feeling faint - could someone please bring the smelling salts?

uh, thanks Gil. I do think I have a better understanding - just enough more to know how little I really know!

[colterphoto1]

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

On 7/30/2004 6:01:14 PM William F. Gil McDermott wrote:

""But going back to something like the Cornwall. When the long axis is up and down, the horn keeps most all frequencies off the floor and ceiling. Turn it so long axis is horizontal, and most all frequencies are kept off the walls.

Therefore we expect some different room reflections depending on orientation.

So, what is happening on the smaller axis? If you look at the horn the exit angle there may be 40 degrees or less. There is no control at low frequenies because the equation tells us that X must be very large, about 2 wavelengths and we have somewhere about 1/4 of that.""

Help me out on this Gil, I'm really confused now--

Here is a quote directly from the Architects' and Engineers' Specificaitons for the Cornwall:

"The high-frequency system shall comprise straight-axis horns; horns employing bends will not be acceptable due to inherent errors. The horns shall provide a coverage of 80* horizontal, 30* vertical, with less than 10 dB loss at the stated extreme angles."

I've got lots of old klipsch data sheets and ad literature and almost none show any dispersion data. Am I understanding you correctly when you say that with a standard Cornwall cabinet setting upright (vertically) with a set of horizontal horns, that the dispersion pattern is vertical (up and down) "When the long axis is up and down, the horn keeps most all frequencies OFF the floor and ceiling."? I thought the old idea for stereo speakers was a wide (horizontal) dispersion pattern so the sweet spot for listening in normal rooms was wide.

Please elaborate!?!??!

I do understand the narrowing of horizontal dispersion for HT and THX systems, I guess the idea is to keep the directionality high so the listener can discern the source of the sound. Keeping early reflections from the floor and walls helps immeasurably in this regard as the 'Bose' and bipolar/dipolar schools blur the sound source by purposely adding reflected sounds. Nowadays this is obviously a no-no.

Thanks, I do appreciate your engineering approach.

Michael

[Piranha]

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

On 7/25/2004 9:55:07 PM garymd wrote:

Some earlier klipsch designs had the horns mounted vertically so the speakers could be laid on their sides. I say this because of pictures in some of the literature from the early 60s and the optional risers for the large side of the cornwalls. Also, some prefer the way vertically mounted horns image and use them standing upright. Many folks use the vertical cornwalls laid on their sides as center channels for HT. I don't actually have an opinion as to which sound better. I've only heard them once at IndyKlipschFans place used as his center channel and it sounded great.----------------

Wait a minute. Dave is right. You heard Trey's at Comfort Inn in Indy. So that's twice.

[IndyKlipschFan]

One lesson... simple to remember in placement.. If you can see down the drivers throat to the back where the driver usually has some kind of screen attatched to the horn.... (the back) Then you have proper placement. Any speaker in this position will sound fantastic.

In the case of vertical horns left vertical, if you toed them in to where you are sitting they are just wonderful. This is especially true if sitting low or high in a couch/chair because your covered. They have less of a left to right if you walked in front of them sweet spot. When put on their sides... (these older Cornwall II's with vertical horns now horizontal) They have more of a sweet spot, because they are now horizontally placed, horn, driver, lenz wize.

I have had a lot of people to my house that have Bells, and La Scalas as centers. They are both incredible speakers, but no bottom end compared to a Cornwall. Especially if you are using K horns for left n right. YOU need that bottom end and not just in the sub. Some people have bought an Academy too.. Nice speaker but with "Heritage" especially bigger Heritage speaker boxes I STILL feel this is THE speaker to have. Same drivers as a Klipschorn, la scala, belle, and a perhaps faster (perceived) and lower than ANY speaker Klipsch has except for the K horn Heritage wise.

With some prodding others on here have found this out as well. There is just too much dialoge in the center channel to have an inferior speaker here. Classic Disney, current Pixar/ Disney to action films with great sound tracks you will hear it.

And thanks for the Kudos from those from the Indy Klipsch event too.