<b>Has anyone seen any formulaes or proper theory covering the design of the socalled "spherical horns" used by Avantgarde and Doppenbergs "Oris" ??</b>
I bet they're a tractrix, because a trax flare launches a <i>spherical</i> wavefront. It was Paul Voigt's work in 1926 that lead to the "invention" of the trax flare, and Dr Bruce Edgar's work in the late 70's / early 80's that brought it back into the light (not to forget Dinsdale). The data from AG's website, including this picture
<img src=http://www.avantgarde-usa.com/pix/whitepapers/fig4a.jpg>
looks a LOT like a trax to me. I've built a few and studied the literature extensively, and have some trax designs being generated and simmed now, with mould to follow soon.
I've also read Bert Doppenberg refer to the Oris as a modified tractrix, and his mathematical description confirms it. How it's modified, if at all, I don't know.
<b>By first glance, it seems like a short exponetial section followed by a sperical prolongation to a 180 deg. opening of the flare......??</b>
It's actually closest to a hypex with an m of about 0.8 (0.83?) in the early part of the flare.
Cheers
I bet they're a tractrix, because a trax flare launches a <i>spherical</i> wavefront. It was Paul Voigt's work in 1926 that lead to the "invention" of the trax flare, and Dr Bruce Edgar's work in the late 70's / early 80's that brought it back into the light (not to forget Dinsdale). The data from AG's website, including this picture
<img src=http://www.avantgarde-usa.com/pix/whitepapers/fig4a.jpg>
looks a LOT like a trax to me. I've built a few and studied the literature extensively, and have some trax designs being generated and simmed now, with mould to follow soon.
I've also read Bert Doppenberg refer to the Oris as a modified tractrix, and his mathematical description confirms it. How it's modified, if at all, I don't know.
<b>By first glance, it seems like a short exponetial section followed by a sperical prolongation to a 180 deg. opening of the flare......??</b>
It's actually closest to a hypex with an m of about 0.8 (0.83?) in the early part of the flare.
Cheers
I believe spherical horns refer to those horns designed around the notion of propogating spherical wavefronts along the horn. The old notions of horn mathematics assumed plane wave propogation, which if you think about it seems odd, since a plane wave by definition cannot expand !
I believe the concept was first published in an AES paper by a group of Japanese engineers. I will look up the details.
Latterly, Earl Geddes has published extensively on the subject.
Andrew
I believe the concept was first published in an AES paper by a group of Japanese engineers. I will look up the details.
Latterly, Earl Geddes has published extensively on the subject.
Andrew
Found the AES paper amazingly qiuckly after pushing the "submit" button, amzing anyway for my filing (non)system!
The paper is "Improvements in Dome Loudspeaker Characteristics by Using a Spherical Wavefront Horn Baffle" by Junichi Hayakawa, Kenwood Corporation, JEAS Vol 36 No.7/8 1988.
Andrew
The paper is "Improvements in Dome Loudspeaker Characteristics by Using a Spherical Wavefront Horn Baffle" by Junichi Hayakawa, Kenwood Corporation, JEAS Vol 36 No.7/8 1988.
Andrew
- my horns are first part cone, then tractrix:
http://www.diyaudio.com/forums/showthread.php?postid=96329#post96329
inspired by bruce edgar saladbowls.
http://www.diyaudio.com/forums/showthread.php?postid=96329#post96329
inspired by bruce edgar saladbowls.
I think I once read on Avant Garde's homepage that they use the tractrix expansion.
Furthermore they don't have an electrical lowpass filter on the midrange horns. They exclusively rely on the acoustical lowpass formed by the compression chamber.
Regards
Charles
BTW: Till's horns look damnbloodycool !!!!
Furthermore they don't have an electrical lowpass filter on the midrange horns. They exclusively rely on the acoustical lowpass formed by the compression chamber.
Regards
Charles
BTW: Till's horns look damnbloodycool !!!!
Horn
Halo
I have designed my own tractrix horns made of kevlar/epoxy. I had Oris before with lowther PM4 but when I heared smaler tractrix horn with tad-4001, there was no doubt about what was the best sound, so I desided to design my own horn. The Oris with PM4 is also a verry good horn but in my ears I think that the TAD-4001 in correct horn sound more warm and detailed with even more dynamic punch. Because of this I made a plug out of aluminium after tractrix calculation made on CNC computer lathe bench. The horn has 300 hz flare and I will combine it with klipsch corner horn, also diy project.
Best regards
Torben
torben-m@online.no
Halo
I have designed my own tractrix horns made of kevlar/epoxy. I had Oris before with lowther PM4 but when I heared smaler tractrix horn with tad-4001, there was no doubt about what was the best sound, so I desided to design my own horn. The Oris with PM4 is also a verry good horn but in my ears I think that the TAD-4001 in correct horn sound more warm and detailed with even more dynamic punch. Because of this I made a plug out of aluminium after tractrix calculation made on CNC computer lathe bench. The horn has 300 hz flare and I will combine it with klipsch corner horn, also diy project.
Best regards
Torben
torben-m@online.no
Kevlar/epoxy horn
That's quite interesting.
I have some experience with figerglass/epoxy boathulls, and I think this is also a possible route.
ABS is probably even more dead in terms of resonance, but I don't think a properly stiffened fiberglass or kevlar horn should produce any singing on its own....
What did you do to the surface finish ??
Did you get the TADs in Norway...??
That's quite interesting.
I have some experience with figerglass/epoxy boathulls, and I think this is also a possible route.
ABS is probably even more dead in terms of resonance, but I don't think a properly stiffened fiberglass or kevlar horn should produce any singing on its own....
What did you do to the surface finish ??
Did you get the TADs in Norway...??
This is an old discussion, but I thought I should continue it rather than start a new thread regarding the Spherical horn flare expansion. I have decided to try to build a pair of spherical mid/high horns for my new Baby Iron Lawbreakers. I used the Dave McBean Horn Response Analysis software to compare different flare types. One nice thing about it is that it includes modeling for both tractrix and spherical flare expansions so I can compare their frequency, phase and other responses.
In short, they looked similar enough to each other in their characteristics that I thought I could probably get similar sonic results from either one (at least I'm hoping so), but the spherical flare results in a shorter horn for a given mouth area which is preferable for my application. I suspect that both of these flares gain an advantage by approaching a 180 degree exit angle at the mouth, which should provide fewer problems with diffraction effects, I would imagine.
The McBean program even models the oblate spheroid flare that Earl Geddes favors. However, it shows a relatively ragged and nonflat frequency response. The group delay deviation near the horn cutoff is particularly low for the oblate spheroid flare expansion, however, so that may be contributing to its sonic benefit.
In short, they looked similar enough to each other in their characteristics that I thought I could probably get similar sonic results from either one (at least I'm hoping so), but the spherical flare results in a shorter horn for a given mouth area which is preferable for my application. I suspect that both of these flares gain an advantage by approaching a 180 degree exit angle at the mouth, which should provide fewer problems with diffraction effects, I would imagine.
The McBean program even models the oblate spheroid flare that Earl Geddes favors. However, it shows a relatively ragged and nonflat frequency response. The group delay deviation near the horn cutoff is particularly low for the oblate spheroid flare expansion, however, so that may be contributing to its sonic benefit.
I'm willing to accept Mr. Geddes' contention that the math used to generate the function is a bit abstruse, but the general unscaled equation for the oblate spheroidal taper is:
y = SQRT(a*2 + (Tan(theta)*2)(x*2))
Most of the taper is asymptotic as you move away from the origin to a conic profile with a y dimension of zero at the origin, but with a short radius added near the origin to match the throat diameter without slope discontinuities. As implemented in real world waveguides, a second larger radius is usually added to the mouth to minimize diffraction.
y = SQRT(a*2 + (Tan(theta)*2)(x*2))
Most of the taper is asymptotic as you move away from the origin to a conic profile with a y dimension of zero at the origin, but with a short radius added near the origin to match the throat diameter without slope discontinuities. As implemented in real world waveguides, a second larger radius is usually added to the mouth to minimize diffraction.
I also discovered the new OS waveguide functionality in Hornresp, but haven't had a chance to experiment with it yet. Looks like the new version came out 9/30, so it may be worth an update for those that don't have it.
The idea behind the Oblate Spheroid profile is that it is the 'smoothest' transition from a plane wave at the throat to a constant-directivity profile at the mouth. It also has the lowest 'higher order mode' generation of any profile due to this smoothness. The 'problem' is that no compression driver actually produces a true plane wave at the throat, although they are probably pretty close. Like any constant-directivity flare they require eq for flat response, whith the tractrix does not.
The idea behind the Oblate Spheroid profile is that it is the 'smoothest' transition from a plane wave at the throat to a constant-directivity profile at the mouth. It also has the lowest 'higher order mode' generation of any profile due to this smoothness. The 'problem' is that no compression driver actually produces a true plane wave at the throat, although they are probably pretty close. Like any constant-directivity flare they require eq for flat response, whith the tractrix does not.
I have made some tractrixhorns from epoxyresin one advantage is that you can determine its stiffness by adding flexible hardner.So within certain limits its is possible to make it reasonbly flexible. whats also works is to make the horn from epoxy half normal and half flexible hardner and when its is finished apply a layer off flexible hardner epoxy and glassbubbles. to make a thick coat, on the backside of the horn. this will dampen also quit nice.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.