Acoustic Horn Design – The Easy Way (Ath4)

[augerpro]

Wouldn't log steps have more resolution at lower frequencies that are most meaningful, versus the top octave? Or did you change it because you modeled the top octave?

Not sure why, but in the solver file I have F1=1000 and F2=20000, but when I run it, it is still plotting out to 10000hz only.

 

[mabat]

Linear and log spacing between the (same number of) points -

 

[augerpro]

Increasing N steps would help find the best balance, at the cost of compute time?

 

[augerpro]

So where would I add the ellipse script for the mouth you posted above in the .cfg file? Would I have to delete another entry (I'm still working off demo1.cfg)? Here is what I have so far for a 180mm major axis ellipse with .618 aspect ratio:
GCurve.Type = 1
GCurve.Dist = 1.0
GCurve.SE.n = 2.0
GCurve.Width = 180
GCurve.AspectRatio = 0.618

What are Dist and SE.n? What if I just wanted a normal ellipse, not a superellipse?

 

[augerpro]

Ignore the above, I think I'm getting the hang of it. Will make some measurements of the SB26 dome, plug those values in, then post my .cfg in the morning and see what you guys think.

 

[mabat]

GCurve.SE.n = 2.0 means a normal ellipse, which is just a special case of superellipse - as you increase the 'n' it gets more "squarish": Superellipse - Wikipedia

(Funny thing is that n=0.5 gives you a parabola. These are all "conic sections" after all.)

 

[mabat]

OK, I've added a simple circular arc as another profile option. Hopefully I will be able to put a new release out soon.

 

[mabat]

And it's not bad

Waveguide 215 x 150 x 50 mm, 1" throat with flat wavefront:

 

[mabat]

This is just crazy - it seems that it produces quite nice spherical wavefronts.

 

[tmuikku]

Pretty cool! What if the mouth was circular too?

 

[mabat]

A circular one: 210 x 50 mm

 

[tmuikku]

wow, lathe time! hmm, a donut... getting hungry

 

[mabat]

⌀350 x 90 mm (infinite baffle):

In a finite baffle it should give a nice 90° coverage.

 

[mabat]

The profile:

 

[mabat]

Comparison with a smoothly terminated OS waveguide of the same size (⌀350 x 90 mm):

 

[mabat]

Radiation impedances of the two:

 

[augerpro]

Well that's encouraging! When do you think this version might be available? I've spent all morning measuring tweeter domes so I can start building models for sources.

 

[mabat]

I can give you a "beta" version 4.6.1 right now: ath-4.6.1-beta
At the moment, without a proper documentation, just use these lines:

Geometry.Definition = 2
Throat.Profile = 3 ; circular arc
CircArc.TermAngle = 1 ; angle at which the profile connects the baffle [deg]

Throat.Diameter = 36 ; [mm]
Length = 30 ; [mm]

GCurve.Dist = 1.0 ; guiding curve at the end of the profile!
GCurve.Type = 1 ; superellipse
GCurve.SE.n = 2.0
GCurve.Width = 200 ; [mm]
GCurve.AspectRatio = 0.7

PS. I also fixed a bug: GCurve.Width now sets the correct size for an ellipse (not half the size).

 

[augerpro]

mabat> it would be a great help if you could provide me one .cfg with the new features based on the dimensions below. From there I think I can tweak for any future tweeter/waveguide combo. Looking to model out to 20khz, with N frequency steps of 30 on linear.

Basic dimensions (sorry for odd sizes, converted from inches):
throat - 30.374
mouth major axis - 127
mouth minor axis (golden ellipse) - 78.486
vertical horn profile radius - 30.48
horizontal horn profile radius - 73.66
depth - 19.05
lip (see attachment) - 0.5

The dome attachment should be easy enough to read.

 

[mabat]

So the throat just touches the outer edge of the surround on the tweeter?