Constant Beam Width Transducers line arrays

Hi Patrick

The two are basically the same thing.

The more important concept to understand is that the vertical directivity of a line source is the Fourier transform of its elements. So any directivity in the range where the line array is longer than the wavelength can be approximated by the Fourier Transform of the elements. This will give the magnitude and phase that best fits the desired directivity. But of course the values will be different at every frequency so one "best fits" a magnitude only function to the values. This ends up being very close to a Bessel function.

If the array is arched then the transform becomes a Mellen (sp?) transform, but the exact same features as above apply.

With arbitrary magnitudes and phase to each element and any directivity could be maintained over a very wide bandwidth.

This is what is done in a torpedo head, only in two dimensions. A beam of about 10 degrees can be scanned continuously.
 
Have installed a couple of the JBL CBT70J with very good results. They are not meant to go into auditoriums as "stereo" pairs but rather as single point.



The bizarre thing about them is the amount of output from such a small unit and the very even field produced. The JBL simulation software is very good and they now produce more complex systems under their Pro division to cover auditoriums with multiple balconies.



A sub is required below 90 - 100 Hz and while the units do not resemble the pronounced arc that Keele used in his original designs, the shading and phase delays are simulated with LC networks.
 
The improvement of inverting one of the five drivers in the Bessel Array seems impossible to believe:

zyxwSOJ.png


An array of five Dayton ND64-16s wired as a Bessel array like this: https://www.angelfire.com/sd/paulkemble/bessel1.gif

RdcWM3f.png


The exact same array, but with the polarity of one driver flipped

And what I mean by "improvement" is this:

A Bessel array is basically a five element array where the outer elements are shaded. The "twist" is that one of the five drivers has inverted polarity. So the first picture is Bessel array wiring.

But it's difficult to believe that something so basic as flipping the polarity of one element makes such a tremendous difference. And I say this as someone who's built Bessel arrays!
 
That’s an impressive and encouraging result Patrick! I wonder what this looks like for a driver with a smoother native response like the FaitalPro FE25-16?.....that’s a compact but very capable system when paired with a subwoofer. I think about the implications for the wide even coverage in the average multi purpose media space