Bruce Edgar Midrange Horn for 2445

[Dr.db]

Hi,


I´d like to build some wooden midrange horns for my 2445 2"-drivers.
I love the look of the Bruce Edgar horns, but I can`t find exact plans to rebuild some of them. Can anyone help me out ?

Besides that, I have a generell question on horns:

If the physical cutoff-frequency of a horn is 500hz, I would have to cross it over at 1000hz, right!?
For example this horn;
http://www.soniphase.com/speakersc.html
It has a 400hz cutoff, so crossover should be around 800hz.....!?

Another example, the JBL 2380;
http://www.jblpro.com/pages/pub/components/2380a.pdf
It claims to be useable from 500hz.... But the frequency-drawing shows, its very weak between 600-1000hz and actually starts rising efficiency sufficient at about 1100hz and up.
So without eq-compensation, even this horn shouldn`t be used below 1000hz!?


Thanks a lot and have a great week,
Olaf

[more10]

The reason you want to cut a horn one octave above cutoff is group delay and driver loading. This is a simulation of a 1100 Hz le Cleac'h horn:

https://dl.dropboxusercontent.com/u/106944188/LeCleach1100/Skärmklipp 2014-11-25 19.35.20.png
Black is resistive load, red is reactive. The reactive load causes phase shift. You want to cut at the first notch on the resistive curve. From this curve I would say 2,5 kHz.

This is group delay:
https://dl.dropboxusercontent.com/u/106944188/LeCleach1100/Skärmklipp 2014-11-25 19.37.08.png

Group delay is a measure on how much phase shift you have. 2,5 kHz looks ok here too.

With a real horn and driver, you can measure the impedance curve or use our ears.

Horn shapes have influence on the sound. Any angles in the horn causes reflections. Le Cleac'h horns don't have reflections. Tractrix horns on a baffle don't reflect either. On traditional horns you get less reflection problems the higher you cut.

[ivica]

Hi,


I´d like to build some wooden midrange horns for my 2445 2"-drivers.
I love the look of the Bruce Edgar horns, but I can`t find exact plans to rebuild some of them. Can anyone help me out ?

Besides that, I have a generell question on horns:

If the physical cutoff-frequency of a horn is 500hz, I would have to cross it over at 1000hz, right!?
For example this horn;
http://www.soniphase.com/speakersc.html
It has a 400hz cutoff, so crossover should be around 800hz.....!?

Another example, the JBL 2380;
http://www.jblpro.com/pages/pub/components/2380a.pdf
It claims to be useable from 500hz.... But the frequency-drawing shows, its very weak between 600-1000hz and actually starts rising efficiency sufficient at about 1100hz and up.
So without eq-compensation, even this horn shouldn`t be used below 1000hz!?


Thanks a lot and have a great week,
Olaf

Hi Dr.db,

I have understood that You want to make "tractrix" horn, so IF you have decide to use it from 500Hz, Rm-horn mouth radius has to be about 25cm, and the length would be about 50cm, ( fc=220Hz) but the "problem" of such horn type would be very narrow at higher frequency, so here about 23 deg (-6dB) on 10kHz. from my point of view, this too 'beamy' result for home-listening purpose. Almost the same results (but shorter horn) can be get from the JMLC horn type.

About the 'construction'

"......
HornCalc is based on this formula:

x = a * ln((a + sqrt(a^2 - r^2)) / r) - sqrt(a^2 - r^2)
where;
x is the distance from the mouth of the horn,
a is the radius at the mouth, and
r is the radius at distance x from the mouth.
....
Thus, P.G.A. Voight can be claimed to calculate: Fc=c/(4*r)...
c is the speed of sound, 340m/s, and "r" is the radius of the horn mouth for the fully expanded tractrix"

http://stereo-lab.de/Information-Spherical-Wave-Guide-Tractrix-Horn
http://www.sonicdesign.se/horncalc.html

http://darklanternforowen.wordpress.com/2014/04/13/tractrix-midrange-horn/
http://www.positive-feedback.com/Issue4/edgarinterview.htm

regards
ivica

[more10]

Aha, Bruce Edgar midrange horn is Tractrix! Then you can use Hornresp to simulate it and also generate horn profiles. You can quite easily make a petal Tractrix horn.

With a Radian diaphragm you can make the driver go lower. Just decide on how low you want cutoff, 200 Hz is reasonable if you can house such a big horn.

[Dr.db]

@ Mårten:

Thanks for the graphs, I think I got it :)
Phase-shift seems to be the biggest issue when crossing over to low...
A round tractrix horn like this below should have no reflections, but I think it is much more beamy than a bruce edgar tractrix-horn!? What do you think ?
http://gpoint-audio.com/wp-content/uploads/2012/03/T400-2.jpg


@ ivica: Thans for your calculations and links!
I don`t need to crossover as low as 500hz, I´m actually aiming 800-900hz as a crossover-point :)
I need a horn with flat-front and I´ve heard the JBL 2380 or 2370 are not very suitable for hifi-use, because they do not sound very musical :blink:
So I thought a tractrix-horn like the bruce edgar would be better sounding.
I have a 2405 to cope with the highs, so the midrange horn should behave gently to about 8khz. Is it allready too beamy at these frequencies ??

http://www.soniphase.com/speakersc.html
Any ideas, how such a horn would sound...!?
Any better than the JBL 2380 ?

I would try to decrease the measurements a little, to fit it into my enclosures.... My maximum width would be about 16". So cutoff should be little higher than 400hz, maybe 450hz I guess...

[ivica]

@ Mårten:

Thanks for the graphs, I think I got it :)
Phase-shift seems to be the biggest issue when crossing over to low...
A round tractrix horn like this below should have no reflections, but I think it is much more beamy than a bruce edgar tractrix-horn!? What do you think ?
http://gpoint-audio.com/wp-content/uploads/2012/03/T400-2.jpg


@ ivica: Thans for your calculations and links!
I don`t need to crossover as low as 500hz, I´m actually aiming 800-900hz as a crossover-point :)
I need a horn with flat-front and I´ve heard the JBL 2380 or 2370 are not very suitable for hifi-use, because they do not sound very musical :blink:
So I thought a tractrix-horn like the bruce edgar would be better sounding.
I have a 2405 to cope with the highs, so the midrange horn should behave gently to about 8khz. Is it allready too beamy at these frequencies ??

http://www.soniphase.com/speakersc.html
Any ideas, how such a horn would sound...!?
Any better than the JBL 2380 ?

I would try to decrease the measurements a little, to fit it into my enclosures.... My maximum width would be about 16". So cutoff should be little higher than 400hz, maybe 450hz I guess...



Hi,

If You decide to use 2-inc throat tractrix type horn over 500Hz you have to deal with the beaming problem

regards
ivica

[more10]

The Edgar midrange horn (http://volvotreter.de/downloads/Edgar-Midrange-Horn.pdf) is a biradial horn with conical vertical expansion and tractrix horizontal expansion. It cannot be simulated in Hornresp, this doesn't mean it is bad only it cannot be calculated.

The whole point of the tractrix expansion is to avoid sharp edges on the horn in order to avoid reflections. The conical upper and lower parts of the Edgar horn will cause reflections.

Any cylindrical horn will be beamy. This means that you will have a small sweetspot and a rising response directly in front of the horn. In practice you will have to EQ the horn or point the horns away.

[ivica]

The Edgar midrange horn (http://volvotreter.de/downloads/Edgar-Midrange-Horn.pdf) is a biradial horn with conical vertical expansion and tractrix horizontal expansion. It cannot be simulated in Hornresp, this doesn't mean it is bad only it cannot be calculated.

The whole point of the tractrix expansion is to avoid sharp edges on the horn in order to avoid reflections. The conical upper and lower parts of the Edgar horn will cause reflections.

Any cylindrical horn will be beamy. This means that you will have a small sweetspot and a rising response directly in front of the horn. In practice you will have to EQ the horn or point the horns away.

Hi more10,

You are right that Edgar horn has different flares: in vertical plane as conical horn, and in horizontal plane as tractrix. In the mentioned horn , as seems to me, either plane horn mouth would produce a kind of reflections in the lower frequency range ( say under 2kHz), but I would expect that the horizontal dispersion of the frequency over (say ) 4kHz would mainly affected by the horizontal plane horn flare, so the beaming would be present. I have such experience with such type of the horn with the Dynaudio D54 mid-range driver applied -look at the attached figure.
http://www.forker.de/downloads/300hz_d54_horn.pdf

Such driver is fare from being the same as JBL 2445 driver, but the beaming has been evident without the need for any kind of measurements, so from such experience I would expect that using 2445 driver with such as Edgar Tractrix (horizontal) horn would be beaming horn too.
May be looking at the JBL solution such as 2380, or PT-H95HF, or 18-sound XR2064 where a kind of narrowing sound passage in the horn throat would produce a kind of expanding dispersion , here , in the horizontal plane.

Regards
Ivica

[more10]

May be looking at the JBL solution such as 2380, or PT-H95HF, or 18-sound XR2064 where a kind of narrowing sound passage in the horn throat would produce a kind of expanding dispersion

These horns have diffraction slot for better horizontal dispersion. I believe the diffraction slot is bad for the sound.

I like the biradial horns like the Yuichi (https://www.google.se/search?q=Yuichi+horn&espv=2&biw=1520&bih=1007&source=lnms&tbm=isch&sa=X&ei=dQp_VK2uKqj7ywOToYHoBQ&ved=0CAYQ_AUoAQ&dpr=0.9), they don't beam much.

[more10]

You can find a lot of interesting horns at Auto-tech (http://www.horns-diy.pl/). They can be bought in USA from DIY Sound Group (http://www.diysoundgroup.com/horns/autotech-horns.html). They have a few Bi-radial horns, but no flat-front.

[Dr.db]

Thanks a lot four your help!!


Bruce Edgar horn will be beamy, because it is still tractrix.... :(

JBL 2380 won´t be beamy, but will have even more reflections caused by the diffraction-slot :(


Yuichi-horns won´t be that beamy, but they´re not flat front :banghead:



Are there any flat-front horns which are sweet sounding ?

[more10]

Olaf

Since you are building the horns yourself, I suggest that you build exponential or hypex horns from a hornresp simulation. Hornresp will help a lot generating the cut profiles. You can tell Hornresp to use conical vertical expansion.

The A 200Hz Exponential Midrange Horn for Altec 288B (http://www.geocities.ws/la1zka/hifi/Hornexperiments/Midhorn.html) has pictures of a larger horn. Interesting project.

If the horn beams you can eq it. If it sounds bad you have learnt something :-)

Mårten

[ivica]

Thanks a lot four your help!!


Bruce Edgar horn will be beamy, because it is still tractrix.... :(

JBL 2380 won´t be beamy, but will have even more reflections caused by the diffraction-slot :(


Yuichi-horns won´t be that beamy, but they´re not flat front :banghead:



Are there any flat-front horns which are sweet sounding ?

HI Dr.db,

That is why I would suggest You to pay your attention to: JBL PT-H95HF, or 18-sound XR2064 horns, as both are "flat" mouth, an either have a kind of "soft diffraction throat", and as seen from their horizontal plane dispersion I believe that they would be 'sonically' acceptable up to about 10kHz.
but not lower then 800Hz (I think).

http://www.jblpro.com/ProductAttachments/tn_v1n31.pdf

http://www.eighteensound.com/Portals/0/PDFs/XR2064.PDF


Regards
Ivica

[Dr.db]

@more10: Thanks for the idea :) But I wanna stuck to a completed design which is allready well tested.


@ivica: Thanks, those are some nice looking alternative flat-front-horns! But I am afraid, that a JBL 2397 or Yuichi Horn will still sound a lot better, right :confused:

I have the feeling that all these flat-front horns are just some kind of compromise. They might sound allright, but they´re never really good sounding!?
So perhaps I have to rethink my design to be able to integrate a 2397 or Yuichi...

[Hoerninger]

You should remember the KLANGFILM Kugelwellenhorn, which is well documented in this forum.
Avantgarde Acoustics uses this type of horn:
http://www.avantgarde-acoustic.de/de/hornlautsprecher/service/downloads.html?file=tl_files/avantgarde/downloads/produkte/allgemein/kompendium/kompendium_der_horntechnik.pdf

A Kugelwellenhorn integrates well with the baffle as a tractrix does.
But it is shorter, so the dispersion is wider.
____________
Peter (HH)

[more10]

Dr.db, are you going to build or buy?

2397 are nice because they are easy to build. But they have separate apparent vertex horizontally and vertically. Vertically the apex is at the mouth. Horizontally at the throat. This means it is hard to pinpoint the sound source.

If buying, the polish biradials seem ok to me. The are almost flat front, just protuding a little bit.

Emilar have made similar horns, a friend of mine has a pair which sounds very nice:
63777

If carpenting maybe Astos horn (http://www.diyaudio.com/forums/multi-way/227939-astos-biradial-lecleach-wooden-horns.html) will give you some inspiration:

http://i857.photobucket.com/albums/ab138/selmashusse/8F16B991-E8C0-4253-B58C-EB4078CA3995-4951-00000798CEBBDC65.jpg

[Lee in Montreal]

Plenty of cheap 800Hz horns out there. Emilar EH500 horns can be had for dirt cheap, as well as very high prices, depending where you find them. JBL's 2350 also has a personality. Very dynamic and definitely not subtle. Love them. ;-)

http://img.canuckaudiomart.com/uploads/37/190821_large_040b55ef0f6d967714963ea94a514e28.jpg

JBL 2350 is theorically a 500Hz horn.

http://img.canuckaudiomart.com/uploads/large/203206-jbl_2350_horns.jpg

[Horn Fanatic]

Dr.db, are you going to build or buy?

2397 are nice because they are easy to build. But they have separate apparent vertex horizontally and vertically. Vertically the apex is at the mouth. Horizontally at the throat. This means it is hard to pinpoint the sound source.

If buying, the polish biradials seem ok to me. The are almost flat front, just protuding a little bit.

Emilar have made similar horns, a friend of mine has a pair which sounds very nice:
63777

If carpenting maybe Astos horn (http://www.diyaudio.com/forums/multi-way/227939-astos-biradial-lecleach-wooden-horns.html) will give you some inspiration:

http://i857.photobucket.com/albums/ab138/selmashusse/8F16B991-E8C0-4253-B58C-EB4078CA3995-4951-00000798CEBBDC65.jpg

If you are able to locate the original 1949 article and drawing for the Bob Smith horn you will notice he calls it a DHS, Distributed Source Horn. The 2397 however, is not a Smith horn and bears little resemblance, as there is nothing exponential about it. I have built many 2397 horns, but my favorite remains the Smith Horn.

As for the "Edgar" horn. Although round Tractrix horn are mathematically correct, IMHO they sound awful. I have built several rectangle Tractrix horns using a 250Hz fc, and they sound much more pleasing.

[Dr.db]

@ Hoerninger:

The Klangfilm Kugelwellenhorn will be less beamy, because it is shorter ? Is it a big difference ? Does it sound as "good" as a 2397 for example ?
If I look at the Kugelwellenhorns from Jabo and the Tractrix-horns from stereolab, there proportions (diameter/lenght) are within 10% equal....


@ more10:

The 2397 won`t produce a precise soundstage, did I get you right? And this is because the top and bottom "plate" is not conical, but parallel to each other ?
I´ve got a pair 2345, they do sound great. But I´d like some wooden horns.
Your wooden horns look very well made, respect! I think I´d prefer a flater horn like the 2397 though...


@ Lee:

The 2350 must be a great horn. But it is a little too big for me. 80cm width is heavy ;)
This horn has no seperation-fins like 2397; does it mean this horn would be more beamy ?


@ Horn Fanatic:

The only difference between smith horn and 2397 I can tell is, that the seperation-fins in the 2397 are a lot shorter. You might help me out explaining the exact differences :)
Maybe you have a original drawing!?

Regarding bruce-edgar horns: Did I get you right, that round tractrix is crap, but the rectangular like the http://www.soniphase.com/speakersc.html sounds good ?
Are they as nice sounding as a smith-horn ?
Whats the differences you have noticed ?

[more10]

The 2397 won`t produce a precise soundstage, did I get you right?

Correct. I have read it somewhere but cannot find it online, maybe The JBL Story (http://www.amazon.com/The-JBL-Story-Sixty-Innovation/dp/1423412818)?

I say, build a large rectangular Tractrix!

[Ruediger]

The apparent apex serves a useful purpose when stacking horns. If the horizontal and vertical apices are different, the horn can only be stacked in one dimension. Imagine a PA system where you combine several horns to achieve a certain coverage angle.

It is not the same as the "acustical center".

I do not see how one can determine the acoustical quality of a horn by considering its apparent apices. The sound from all acoustical apices will be in phase with the sound coming from the driver, it does not experience any variation by passing such an apex.

Two different apices will result in what one calls astigmatism in optics, but I doubt that one can hear that. I may be wrong.

I have put a paper covering this subject in the "Technical References" thread in the "General Audio Discussion Forum":

Altec Lansing Technical Letter 262: Coverage of Multiple Mantaray Horns

Ruediger

[sebackman]

Hi,
Try the JBL 2332 if you can compensate in the active or passive filters. Sounds good if the throw is less than 8m. And better if the alu is dampened. I use a bunch of them with 2451Be, 2450SL and also a pair with standard 2451Ti. They are drilled dual pattern so they fit all drivers. Flat front an 12" wide. Thay are CD's so they need EQ.

I use a BSS BLU800 with digital input cards as active XO and also EQ.

Otherwise I think the newer JBL waveguide horns that Ivica is mentioning would be a good choice. There are a few different models and they are all flat front. I'm a bit unsure how they behave below 1k though...

regards
//RoB

[Maron Horonzakz]

The Smith Horn And the JBL 2397 are distributed source horns in the horizontal plain,, Difraction in vertical plain

[Horn Fanatic]

@ Hoerninger:

The Klangfilm Kugelwellenhorn will be less beamy, because it is shorter ? Is it a big difference ? Does it sound as "good" as a 2397 for example ?
If I look at the Kugelwellenhorns from Jabo and the Tractrix-horns from stereolab, there proportions (diameter/lenght) are within 10% equal....


@ more10:

The 2397 won`t produce a precise soundstage, did I get you right? And this is because the top and bottom "plate" is not conical, but parallel to each other ?
I´ve got a pair 2345, they do sound great. But I´d like some wooden horns.
Your wooden horns look very well made, respect! I think I´d prefer a flater horn like the 2397 though...


@ Lee:

The 2350 must be a great horn. But it is a little too big for me. 80cm width is heavy ;)
This horn has no seperation-fins like 2397; does it mean this horn would be more beamy ?


@ Horn Fanatic:

The only difference between smith horn and 2397 I can tell is, that the seperation-fins in the 2397 are a lot shorter. You might help me out explaining the exact differences :)
Maybe you have a original drawing!?

Regarding bruce-edgar horns: Did I get you right, that round tractrix is crap, but the rectangular like the http://www.soniphase.com/speakersc.html sounds good ?
Are they as nice sounding as a smith-horn ?
Whats the differences you have noticed ?


Greetings Dr.db -

Instead if describing the differences, I'll link the article, including the article for the DHS tweeter. As you can see, out side of the approximate width dimension, the Bob Smith horn is much different than the 2397, also that it was designed for a small format driver. The drawing indicates six separate exponential cells, vs. the short tapered dividers on the 2397. Calculations will reveal that the only part of the 2397 that resembles the Smith horn is approximately nine inches from the throat, the rest was added by Bart Locanthi. You'll also note, that the divider arrangement on the TAD horn follows that of the 2397, only much shorter. My guess is that the first section of the TAD horn approximates the DHS tweeter with a radial exponential manifold section in front of it.

The Smith DHS is just that, six separate sound sources distributed across an arc. If you pass from one side to the other with information put through it, you can hear each individual cell which function as a point source. The main reason I like the Smith horn, is that with his simple calculations a larger version can be designed to accommodate a 1.4" or 2 " driver.

As for the "Edgar" horn? To be blunt. I have heard the round Edgar horn in a few different listening environments, and I stick to my opinion. Round Tractrix horns sound awful, especially with a two inch driver. You may as well be listening to a Hartsfield horn without the lens. The Tractrix horns that I build were for John Tucker of Exemplar Audio back in 2003 and 2004, of which were his design in collaboration with Jeff Marquardt. That horn was used in conjunction with his Tractrix bass horns. The high frequency driver was the GPA 909-8A, and the woofer was the GPA 515-8G. That combination along with the crossover John and Jeff designed made for an awesome sounding system. I haven't heard the Soniphase horn. That chap lives about 20 minutes from me, although we have never met.

Here ya go. Although I used better material, I chose to remain faithful to Bob Smith's plans. I built the horns out of 6mm Baltic birch, and poplar. For the throat transition I was given two throat transitions for the 604 Mantaray horn by Bill Hanuschack at GPA. I cut the back piece down to about an inch square. The horn really needs a proper adapter in my opinion, perhaps a 3/4" thick circular to square transition. I wouldn't think of asking Bill to provide any more back plates, as he would be left with unusable horns.

It really boils down to individual taste. Some people like diffraction horns, and some don't.



http://www.audioheritage.org/vbulletin/showthread.php?27413-Smith-Horn-Article (http://www.audioheritage.org/vbulletin/showthread.php?27413-Smith-Horn-Article)

http://www.audioheritage.org/vbulletin/showthread.php?9871-Smith-Selsted-Tweeters

The Exemplar system I built: http://www.jenalabs.com/images/vsac2008-12-r.png

Good luck,

H.F.

[ivica]

Greetings Dr.db -

Instead if describing the differences, I'll link the article, including the article for the DHS tweeter. As you can see, out side of the approximate width dimension, the Bob Smith horn is much different than the 2397, also that it was designed for a small format driver. The drawing indicates six separate exponential cells, vs. the short tapered dividers on the 2397. Calculations will reveal that the only part of the 2397 that resembles the Smith horn is approximately nine inches from the throat, the rest was added by Bart Locanthi. You'll also note, that the divider arrangement on the TAD horn follows that of the 2397, only much shorter. My guess is that the first section of the TAD horn approximates the DHS tweeter with a radial exponential manifold section in front of it.

The Smith DHS is just that, six separate sound sources distributed across an arc. If you pass from one side to the other with information put through it, you can hear each individual cell which function as a point source. The main reason I like the Smith horn, is that with his simple calculations a larger version can be designed to accommodate a 1.4" or 2 " driver.

As for the "Edgar" horn? To be blunt. I have heard the round Edgar horn in a few different listening environments, and I stick to my opinion. Round Tractrix horns sound awful, especially with a two inch driver. You may as well be listening to a Hartsfield horn without the lens. The Tractrix horns that I build were for John Tucker of Exemplar Audio back in 2003 and 2004, of which were his design in collaboration with Jeff Marquardt. That horn was used in conjunction with his Tractrix bass horns. The high frequency driver was the GPA 909-8A, and the woofer was the GPA 515-8G. That combination along with the crossover John and Jeff designed made for an awesome sounding system. I haven't heard the Soniphase horn. That chap lives about 20 minutes from me, although we have never met.

Here ya go. Although I used better material, I chose to remain faithful to Bob Smith's plans. I built the horns out of 6mm Baltic birch, and poplar. For the throat transition I was given two throat transitions for the 604 Mantaray horn by Bill Hanuschack at GPA. I cut the back piece down to about an inch square. The horn really needs a proper adapter in my opinion, perhaps a 3/4" thick circular to square transition. I wouldn't think of asking Bill to provide any more back plates, as he would be left with unusable horns.

It really boils down to individual taste. Some people like diffraction horns, and some don't.



http://www.audioheritage.org/vbulletin/showthread.php?27413-Smith-Horn-Article (http://www.audioheritage.org/vbulletin/showthread.php?27413-Smith-Horn-Article)

http://www.audioheritage.org/vbulletin/showthread.php?9871-Smith-Selsted-Tweeters

The Exemplar system I built: http://www.jenalabs.com/images/vsac2008-12-r.png

Good luck,

H.F.

HI Horn Fanatic,

Many thanks for the Bob Smith DSH horn "reminder". Owing to its evident large diffraction in vertical plane, I think that Yuichi A290 design would be better solution for the home listening environment, but sure from the point of the DIY realization Smith DSH horn would be much easier to be done.

Regards
ivica

[Dr.db]

Thanks a lot for all your detailed answers :)


To summarize;

Round tractrix-horn sounds awful. A rectangual tractrix as the soniphase might be good sounding, but still beamy.
But none of you has listened to these soniphase and they don`t seem to be very popular...

Yuichi probably better than 2397 or smithhorn because of less diffraction in vertical plane.
How will I notice the differences ? To be honest, I have no clue how "much diffraction in vertical plane" will sound...

Smithhorn sounds better than 2397... But how about the Westlake Audio version of this kind of horn ??

[Ruediger]

<snip>
...
</snip>
I have no clue how "much diffraction in vertical plane" will sound...
<snip>
...
</snip>


Read an article about "Huygen's Principle" and about "Diffraction at a slot". The latter simply widens the beam. It does not "sound" and it does not add any harmonics, so it does not add distortion.

At the 2380 series of horns the area of the diffraction slot is too small, so part of the sound gets reflected back into the horn, this is what causes the resonances.

At the diffraction slot these horns radiate into quite large a solid angle:

2382 - 120 deg -> 1/3 space
2380 - 90 deg -> 1/4 space
2385 - 60 deg -> 1/6 space
2386 - 40 deg -> 1/9 space

If the slot was longer and thus the slot area was larger, there would be less reflection and still the same diffraction.

Ruediger

[deanznz]

There is a quick and dirty overview of different horn designs here:
http://www.excelsior-audio.com/Publications/QTWaveguide/QTWaveguide_WhitePaper.pdf

more in depth articles
Horn Theory: An Introduction part 1 and 2
http://kolbrek.hoyttalerdesign.no/index.php/publications
figure 1 in part 1 is missing but it is the "Original article draft" next link below the above two.

I have done a bucket load of reading lately and the best mid range horn available to the average person (what ever that means) is the K 402 klipsch horn (modified Tractrix). The people who own them have never heard better, same with the people who go to listen. You can hardly find them at all second hand (desperate for cash or downsizing only generally) One catch is they only sell them new with a compression driver, last I heard they were about $1,000 USD each. Another catch (for some) is they are not small, and they aren't sexy either. They are used as part of a system for klipsch pro behind the cinema screen sound systems. Not many people know of them actually. It's the horn in the middle-ish (what's the middle of 4) of the stack over here:
http://www.klipsch.com/behind-the-screen

I have not seen someone attempt to make a diy version, but I know its make up of a mix of 3 horn designs, Tractrix (mouth) and 2 others ..... ha ha, oh and conical is not one of them.

Olaf I have never seen someone reply with an overview of other peoples suggestions before, I think that is very polite and quite a smart idea.

regards,

Dean.

[more10]

I have read it somewhere but cannot find it online

I found it! Smith Selsted article (http://www.audioheritage.org/vbulletin/showthread.php?9871-Smith-Selsted-Tweeters&p=101069&viewfull=1#post101069) posted by Steve Schell.


This type of horn has an interesting performance characteristic. When listening to it one can not tell precisely from where the sound is coming. This is due to the fact that the focus of the sound source is not the same in the vertical and horizontal planes. The source of sound in the horizontal plane appears to be at the throat of the horn, while in the vertical plane it appears to be at the mouth. This effect is startling in a similarly designed horn for the 500 to 5000 cps range.

[Dr.db]

Thanks Morten for searching the article.


I´ve listened to a JBL 2397 the last days and I really liked it. I´m going to build these in spring.

But actually I´m willing to rebuild the westlake audio-version, also in walnut :)

[ivica]

Thanks Morten for searching the article.
I´ve listened to a JBL 2397 the last days and I really liked it.
I´m going to build these in spring.
But actually I´m willing to rebuild the westlake audio-version,
also in walnut :)

HI Dr.db,

Some words that has been shown at the end of the Smith's paper are interesting form me, and I think that would be the differences between Yuichi A290 and JBL 2397

reagrds
ivica

[Dr.db]

Thanks again everyone for your big help, I really appreciate that!! :)


I believe a 2397 is not the ultimate horn, but still "better" than many other designs (like the 2380, bruce edgar, round tractrix, etc....).
A Yuichi would be better in the vertical-behaviour, but I think this wouldn´t be a major difference.


Regards,
Olaf

[more10]

I´ve listened to a JBL 2397 the last days and I really liked it. I´m going to build these in spring.

Good decision! You can build these with hand tools.

But if you have access to a CNC machine you can build much more difficult stuff. Matsj built these a few years back: Jbl Diy K2/Array (http://www.audioheritage.org/vbulletin/showthread.php?17779-Jbl-Diy-K2-Array) .

http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=27037&stc=1&d=1185998577

They are now for sale: Jbl k2-9500 (http://www.blocket.se/kalmar/Jbl_k2_9500_57919267.htm?ca=4&w=3&last=1)

63896

[Dr.db]

I´m a carpenter and I know a few guys with cnc´s.... Building difficult horns wouldn´t be the problem.

But I also like the flat appearance of the 2397 rather than the bulky look of a yuichi horn. They appear a bit threatening in a living room in my opinion...

[ivica]

I´m a carpenter and I know a few guys with cnc´s.... Building difficult horns wouldn´t be the problem.

But I also like the flat appearance of the 2397 rather than the bulky look of a yuichi horn. They appear a bit threatening in a living room in my opinion...
Hi Dr.db,

Unfortunately, for acoustic horn, visual outlook and acoustical behaviors are not usually "going in the same direction".
If You want to suppress the reflections from the mouth, mouth surface curvature has to be "Yuichi horn" look.
If you want to use the horn to lower frequency, the horn size would rise.
If you want higher frequency wide coverage then a kind of horn diffraction (in the throat, or in the mouth region) has to be applied, or multicell construction has to be applied,
on my opinion, the diffraction in the throat is better concerning the "imaging" of the sound.

regards
ivica

[more10]

or multicell construction has to be applied,

The construction of a multicell horn (http://www.diyaudio.com/forums/multi-way/136964-construction-multicell-horn.html). This guy is not even a carpenter :-).

https://dl.dropboxusercontent.com/u/106944188/Horn-43.jpg