DCAudioDIY.com

For all you horny guys:

I am under the impression that the horn size (largest dimension at the horny end) sort of determines the low frequency the horn can willingly produce.
If I am right (I am know to be wrong some of the time), is there any advantage to using a horn that is considerably larger than the 3db down point of the xover being used for that horn?

The type of horn (flare) matters but you certainly want linearity in output response at freq above your xover point. Ever larger horn throats will start to cause issues with integration with other drivers in your system.

https://www.grc.com/acoustics/an-introd ... theory.pdf

I am sure some of the horn experts here will chime in shortly, I am no expert I just listen to them!

thanks
looks like a good read

Horn mouth size effects directivity in the same way as using a large diameter speaker. At a given frequency the larger the speaker (diaphragm) the narrower the beam pattern (dispersion).

You may wish to study Don Keele's constant directivity horns -

https://en.wikipedia.org/wiki/D._B._Keele_Jr.

_________________
Walt

Walt is correct about the crossover point. It is important for the drivers to have the same radiation pattern at the crossover point especially as the crossover slope gets steeper. If the radiation pattern is different you will get different room contributions and also off axis listening response problems as you go through the crossover point which will make the sound anything but seamless. When I did the crossovers for the eggs I took into account among other things the points where the drivers had the same radiation patterns at those frequencies. The results were good.

I think that the resistance to high order crossovers is mainly due to differences in the radiation pattern at the crossover frequencies. I have one system that has the same radiation pattern at the crossover points and it is successful with 48 dB/octave crossovers. Another speaker I have does not have the same radiation pattern between the mid and tweeter and the high slopes sounded bad. At 12 dB/octave there was enough overlap of the drivers that the room contribution problem went away.

Tom

Thanks guys
The reason for the question is that there was a vendor at CAF that used a 3ft(diameter) horn for a tweeter(xover point 1200 hz)
I was just curious

which one is that? Fern and Roby?

For a good approximation of the radiation pattern of a circular aperture see -

http://www.acs.psu.edu/drussell/Demos/B ... iston.html

k is the wavenumber 2*pi/lambda = 2*pi*f/c, where f is the frequency and c is the speed of sound. J1 is a bessel function of the first kind (appears in cylindrically symmetric systems). The are function packages for Bessel functions in most computer languages. I would use python if you wanted to evaluate the formula.

Tom,

You are correct about the matching of dispersion at xover. Major point!

This was a feature of the very first two ways from Western Electric around 1932-33. They put the 13.5" woofer in a short horn mounted in a flat baffle to induce controlled narrowing dispersion with frequency. More of a waveguide than an impedance transformer concept.

This design was the grandmother of the Voice of the Theater cabs, which were originally meant to be mounted in large flat baffles, but that notion got lost. The woofer horn in an 825/828 cabinet is not really there for impedance transformation...in fact the 120hz bump this small horn creates is a problem! but it does narrow the midrange dispersion of the woof to match that of a 500hz horn, maybe 60 deg horizontal.

Note that early RCA theater horns were stenciled "Directional Baffle"

As for the 1200hz round horn. Hard to say without knowing more. Perhaps a low slope xover was used and they needed loading down to a lower than expected frequency.

Also, the impedance of a horn usually flattens out and is optimal somewhat above its cutoff region, so this could be searching for the sweet spot.

Finally, given that it is a round horn, it may be a L'Cleach horn, quite popular in hifi circles. Jean Michel worked out the math for optimized terminations of the horn mouth to the free air. Generally, this is a critical point for horns. Poor termination leads to a plethora of mid-to-high order modes caused by reflections back from the mouth to the walls of the horn and the driver, visible in the horrible HF waterfall performance of the Altec 511 I posted earlier, for example. This is how the real (acoustic) ringing of horns is brought on. Major time smear and dispersion anomalies ensure.

The L'Cleach termination adds a lot of material to the edges of the mouth. Here's a pic of an Iwata profile horn with JMLC termination that we set up at ETF in Denmark last week. That I believe is a 500hz horn, but could be 300.

Here's one JMLC lecture: https://pearl-hifi.com/06_Lit_Archive/1 ... TF2010.pdf

I'll try to find a link for another good one. I have it on my computer as a pdf somewhere...

In any case, there may be very good reasons to use a 3' horn at 1200hz crossover but I think it could be done with less drama. Many pro 1200hz horns are like 8" wide!