need a little clarification of directionality

[thadman]

Im a little concerned about the directionality issues associated with ESLs. With line arrays (which are basically acting as a single acoustic transducer across their length) they are completely dead outside of their length (above and below). Would an ESL operating at 20khz (assuming the panel is 12" wide) throw a 20khz wave that is 12" wide or would it be less than the width of the speaker? Couldnt I combat the directionality of the ESL by making a very large width driver (say 48") so I could have a wide wavefront, and move around in that *large* sweetspot?

[Tommythecat]

It wont "throw a wave" as we imagine with woofers/pistons and low-frequency waves. You're going to get a beaming polar pattern, a bigger panel is only going to make it worse, lower.

[thadman]

?im just comparing the concept to a line array (as it should behave horizontally as it does vertically) but shouldnt it have energy directed at you across its entire width...and once you were outside the width of the speaker you would have almost no output.
___
| __
| __
| __>
|
| __->
| __
| __
|__

(I couldnt get this picture to come out right, imagine that as you graduate closer to the center the lines radiate farther outward)

The horizontal lines are the on-axis energy (sorry...I couldnt make a curved line just try to imagine it) and the vertical line is the ESL panel
_____________>
|
|
|
|
|
|
|
|____________>

Outside of the on-axis energy, there should be virtually no output. Which of these is the correct drawing?

Which one of these is correct? As I understand, a vertical line array behaves like this:

(no output above)
___________________> (sound coming at u)
|
|
|
| <------line of speakers in a vertical line
|
|
|
|___________________> (sound coming at u)

(no output below)

[ThomasW]

It would be nice were you to create a single thread about your potential ESL project. Then you could place all you questions there, as opposed to starting a new thread for every question you think up.

[thadman]

Sorry Edit the title if you wish, and this can become my dedicated ESL thread

[Dennis H]

You've been asking a lot of similar questions lately. Pictures will answer your questions better than words. You can model the polar response of lots of kinds of speakers, including ESLs, with the tools at the FRD consortium. The Radiation pattern tool is the best one for this job. You'll need Excel and you'll have to install the Analysis Toolpack and VBA (not installed by default with Excel) for the FRD tools to work.