Square vs round horn mouth dispersion/arrayabillity?

[kessito]

Hi all,

The following has been bothering me for a while and though I have searched pretty extensively, I haven't been able to find a satisfying answer yet;

I am wondering what the difference in dispersion at the mid/high frequencies are between a round horn or a square/rectangular horn. (In both cases starting round at the throat for mating with the compression driver)

- Is it so that the round horn will produce a round dispersion pattern and the square horn will produce a (more) square dispersion pattern?

- If so, does this mean that a square horn will automatically be more suitable for arraying multiple speakers because the patterns have a better match? (meaning that with 2 round horns there will be gaps because of 2 circles which don't connect well?)

- Or is it so that the wavefront will always be round and it will never acoustically "see" the corners of the square, suggesting that the square shape will be sub-optimal and only causes disturbing reflections/fluctuations in the wavefront?

I have been playing around with axidriver (BEM-modelling) quite a lot and begin to understand the finesses/tradeoffs of the horn design process. But since axidriver is axisymetric, it doesn't allow for square horns.
For this difference to simulate I would need to use ABEC (full 3D BEM), but this has a rather steep learning curve, and it will cost me some more time to use it.

Any references/opinions/ suggestions/ measurements will be greatly appreciated.

Many thanks,
Kees

[whgeiger]

Originally Posted by kessito

Hi all,

The following has been bothering me for a while and though I have searched pretty extensively, I haven't been able to find a satisfying answer yet;

I am wondering what the difference in dispersion at the mid/high frequencies are between a round horn or a square/rectangular horn. (In both cases starting round at the throat for mating with the compression driver)

- Is it so that the round horn will produce a round dispersion pattern and the square horn will produce a (more) square dispersion pattern?

- If so, does this mean that a square horn will automatically be more suitable for arraying multiple speakers because the patterns have a better match? (meaning that with 2 round horns there will be gaps because of 2 circles which don't connect well?)

- Or is it so that the wavefront will always be round and it will never acoustically "see" the corners of the square, suggesting that the square shape will be sub-optimal and only causes disturbing reflections/fluctuations in the wavefront?

I have been playing around with axidriver (BEM-modelling) quite a lot and begin to understand the finesses/tradeoffs of the horn design process. But since axidriver is axisymetric, it doesn't allow for square horns.
For this difference to simulate I would need to use ABEC (full 3D BEM), but this has a rather steep learning curve, and it will cost me some more time to use it.

Any references/opinions/ suggestions/ measurements will be greatly appreciated.

Many thanks,
Kees

Hi Kees,

In the far field, wave fronts emanating from a horn loaded driver will be spherical with energy concentrated in a three dimensional polar pattern that reflects the aspect ratio of a horn section near its mouth. Note that this section may not be flat as in the case of a sectorial horn. Typically these horns have dispersion patterns that are wider in the horizontal plane and narrower in the vertical plane, and are deployed in short-throw applications.

Horn farms are typically arrayed vertically. Some vertical lobbing in the crossover frequency overlap region will be unavoidable.

For further study, you will need to do journal searches at

AES
AES E-Library

ASA
Browse - Journal of the Acoustical Society of America

and other websites.

For a start, the following article may be helpful.

Track down and contact authors directly. In many case they will send you a copy of their article and additional information as well. Just tell them what you are trying to do. Be brief, so you at least get a response.

Regards,

WHG

Modeling of horn-type loudspeakers for outdoor sound reinforcement systems
A. Schuhmacher* and K. B. Rasmussen

Department of Acoustic Technology, Building 352, Technical University of Denmark, DK-2800 Lyngby, Denmark

Received 5 November 1997; revised 13 March 1998; accepted 13 March 1998. Available online 13 October 1998.

Abstract
A waveguide oriented numerical simulation model for horn loudspeakers has been developed. The model is able to predict the output from horns with square openings under general conditions. The primary resonance of the driver is included in the simulations and comparison with measurements on two horn-type loudspeakers is made. The agreement between measured and calculated results is very good provided that a sufficient number of modes are included in the simulation. Simulation models of this kind represent one of the first steps towards a CAD tool for outdoor sound reinforcement systems.

[kessito]

thanks WHG, i will check it out!

Kees