Acoustic Horn Design – The Easy Way (Ath4)

[limacon]

Is there a scientific way to determine the optimum exit angle paramenter for a full range direct radiator?

 

[mabat]

Once the diaphragm breaks up, the real vibration pattern will be so complex that I doubt that the exit angle would have any meaning or that it could be optimized across a wide frequency range...

 

[limacon]

Fascinating insane
Yea, the full-range characted can be tempting.

Gloss black with a copper coloured driver mount / inlay , and copper back. Only possible with Ath4!

 

[TNT]

limacon - it would be very interesting to see measured polars on this - just maybe 15, 30 and 45...

//

 

[limacon]

limacon - it would be very interesting to see measured polars on this - just maybe 15, 30 and 45...

//

No problem. Would be my pleasure!

 

[gedlee]

Is there a scientific way to determine the optimum exit angle paramenter for a full range direct radiator?

AT LFs it should be straight, a tube, no angle at all. But then as the diaphragm begins to be non-rigid, there is no scientific optimal. It just becomes a mess.

 

[mabat]

Preparing a fillable throat part -
(Yea, I've learned to use the slicer modifiers... )

It could be also printed like this, i.e. virtually hollow -

 

[jzagaja]

Fine if your 3D print won't leak ;-)

 

[mabat]

Or this:

In any case there's no problem sealing the print first.

 

[mabat]

This was fast, it's only a light shell -

 

[sheeple]

Gloss black with a copper coloured driver mount / inlay , and copper back. Only possible with Ath4!

View attachment 1069326

I cannot stop thinking how nice a coaxial mid would be, with a paper cone profile according to an osse-formula and a compression driver. Insert in into a fitted waveguide, and there it is, the perfect three way.

 

[mabat]

I would love to see a measurement of the ONKYO GS-1. I came across these (pretty old) speakers recently and based solely on the appearance the horn looks surprisingly good... But I can't find any data.

 

[Ro808]

The horn is an 'optimized' radial design. Not much data available.

In room response:

 

[bmc0]

"Multipath ghost distortion" sounds a lot like HOMs by another name.

 

[Ro808]

It sounds more like a different terminology for "astigmatism"

 

[Ro808]

Judging by the profile, the horn will have a fairly steep rise in directivity with a maximum coverage angle of 90° horizontally at the low end to less than 50° above 10 kHz.

 

[Olombo]

Hi

slowly proceeding and did first few sims using ath.

anyway: need a clear design target! 2 way with widest possible coverage by compression driver/horn.
Dcx464 or HF1440? Still a bit worried about the hf1440 resonance below 1kHz which was schown by Mabats first test….

 

[limacon]

Equipment
TC9FD18 sealed.
Motu M4, MC2 T4-250, Umik-2.

Results
Very quick measurements. Will set up properly tomorrow.
~1m from cone. Horizontal - green on axis, increments of ~10deg, 1/6th smoothing.

 

[weltersys]

anyway: need a clear design target! 2 way with widest possible coverage by compression driver/horn.
Dcx464 or HF1440?

No question the B&C DCX464 could be crossed lower than the HF1440, simply due to it's greater Sd and more excursion potential.

B&C also makes the less expensive 2" exit DCX50 coax that can be used down to the 300-400Hz range.

 

[Olombo]

Of course not cheapest solution but I like the mylar ring version for midrange of the dcx more. Also 1.4 vs 2“ throat. So waiting for delivery and motivation then to process a bit quicker ;-)