In Reply to: horn colorations posted by angeloitacare on April 6, 2008 at 12:17:31:
I guess you know what I am going to say.
Conical horns have no sidewall deformities. The wavefront is not bounced around from one curving section to the next. The mouth reflection issue certainly is a problem.
Particular care must be made with the first 30% of the horn over the range of operation. The first 30% of "squeezing" contribute greatly to the honk sound. I have found that above 100Hz a conical horn is best, below this frequency, and exponential horn is best.
In some case, a modified exponential, (Tractrix) produces a little honk, but you get a good low cutoff.
Pinching a conical horn at the throat , increase the output past 8kHz , at the expense of some directivity.
A conical horn has poor low end loading. This is not too great an issue with compression drivers.
You may note that almost all the early Edison cylinder mechanical recordings were done with conical horns, used as a microphone. This is also true of Navy "talkback" speakers.
In my opinion the mouth issue has not been solved. It is a systemic problem that requires the entire horn to be reshaped. I am fooling with this now, but it is a long haul.
Foam in the mouth or throat has minimal effect. Acoustic lenses work the best at the present time.
Mouth shaping, by foam, ellipsoidal shape, or Karlson like slits also have trade-offs. This is also true for the TAD AFAST mechanical filter. This is a clever idea.
I have spent a lot of time trying to discover what shape is best. I built a horn a number of years ago with movable walls- so I could change the shape on the fly.
Here are some tests with a 700Hz conical horn with and without a mouth.
The mouth has a 5cm radius on this horn. I have also done this with a sharpish bend ( A la Keele)- this worked the best.
I have never seen curves of horn with different shaped mouths. Anyone who has done this, I certainly would like to see these tests. Here are a few …
A final note about mouths and conical horns. there is always a trade-off of horn length Vs adding a second flare at the mouth in terms of the volume the horn takes up in a PA system, or in a home.
I have always found the longer horn, is the better choice.
Look carefully at the mouth tests. Note while the conical no-mouth is worse to 3khz on axis- it is better off axis. your direction clues are oriented to the higher frequencies. You don't hear the boost at 1Khz too much. At 10 degrees it is gone. This is the endless trade-of horn design.
I choose my design based on these tests, and listening test, with ever driver on the planet, save for ALE.
This includes horn loaded Hiel. cones, dome, ribbons etc.
Confused? me too, there is no single answer.
Bill
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Some tests.. - RCA-fan 04/7/0811:52:54 04/7/08 (4)
- RE: Some tests.. - toxicport.e 15:14:23 04/7/08 (1)
- from memory...... - RCA-fan 17:27:22 04/7/08 (0)
- forgot this. - RCA-fan 11:55:22 04/7/08 (1)
- RE: forgot this. - angeloitacare 12:55:22 04/7/08 (0)