Acoustic Horn Design – The Easy Way (Ath4)

Jump to Latest
#10,781
Only a bit to do with ATH: I am more familiar with abec and wanted to simulate a 2 way now with a circsim generated ATH profile. The project runs fine in Abec but importing the file in Akabak leads to issue "size zero in component....". Strange: even if I copy the nodes by hand and set up a new project the issue remains. Something wrong with circsym mode perhaps? Usually use 3D Mode....
 
#10,785
Hi Marcel

More and more I learn about influence of the single paramters. Unfortunately I habe to try around with round horn shapes to keep sim time low for faster learning. The dcx464 is now here waiting 🙂
The rectangular horns need several minutes to solve so I am not so fast in optimizing. Whatever I try starting from your post 10653 I cannot get same quality as you in

https://www.diyaudio.com/community/...-design-the-easy-way-ath4.338806/post-7132867

Can you give me an idea of what you changed (especially: do you keep or alter first segment (round->rectangular) parameters?). Maybe also what leads you to which change? Learning from your design thougts may help a lot to ger better own optimization strategy
 
#10,786
mabat said:
Yes, it should be possible now (haven't checked for a while):
https://www.diyaudio.com/community/...-design-the-easy-way-ath4.338806/post-7046022
Click to expand...

So far so good, i think. Inifite baffle. Making progress...

Code: 
ABEC.SimType = 1
ABEC.f1 = 200               ; [Hz]
ABEC.f2 = 16000             ; [Hz]
ABEC.NumFrequencies = 19
ABEC.MeshFrequency = 1000   ; [Hz]

ABEC.Polars:SPL_H = {
  MapAngleRange = 0,90,19
  NormAngle = 10    ; [deg]
  Distance = 3      ; [m]
  Offset = 102       ; [mm]
}

Device width x height = 231.03 x 242.02 mm (9.096 x 9.528")
Mouth aspect ratio =    0.955 (1.048)
Device length =         100.00 mm (3.937")

1667424197512.png
1667424230425.png
 
Last edited:
  • Like
Reactions: mrk7
#10,787
Olombo said:
Whatever I try starting from your post 10653 I cannot get same quality as you in

https://www.diyaudio.com/community/...-design-the-easy-way-ath4.338806/post-7132867

Can you give me an idea of what you changed (especially: do you keep or alter first segment (round->rectangular) parameters?). Maybe also what leads you to which change? Learning from your design thougts may help a lot to ger better own optimization strategy
Click to expand...
I just try things. If it was (comparatively) good then it's a pity, as I'm afraid I don't have that set anymore, nor I remember what I did. As it's not only bigger but also shorter, hard to say what it was.
 
Last edited:
#10,788
It probably won't get much better than this -

Code: 
HornGeometry = 2
Length = 232
Throat.Diameter = 36
Throat.Angle = 0

Horn.Adapter = {
  L = 120
  k = 4
  Width = 120
  Height = 50
  SC = 0
  Segments = 10
  ZMap = 0.5,0.4,0.5,0.5
}

Horn.Part:1 = {
  L = 1
  Segments = 16
  H = {
    r0 = 50
    a0 = 30
    k = 4
    s = 0.96
    a = 34
    n = 4
    q = 0.995
  }
  V = {
    r0 = 30
    a0 = 10
    k = 4
    a = 24
    s = 0.55
    n = 4.0
    q = 0.995
  }
  ZMap = 0.5,0.3,0.5,0.9
}
 
  • Like
Reactions: limacon
#10,789
It probably is not necessary to have it any better....

1667508709612.png

Loading at low frequencies look sufficient. I did not yet plot die DI but anticipated from the polars it might be fine. Although having basically what I want now comparing old/new scripts and try to understand whats happening 🙂
Still some small wishes that at least will be tried to be implemented. Awesome work Marcel....!
 
#10,792
limacon said:
This is for a an array of 30 miniature speakers (each rated 102dB@10cm, flat 1000-20k)

Aperture will be 13.6mm wide x 300mm tall.

Vertical C-to-C between speakers = 9.1 mm

View attachment 1105889View attachment 1105928
Click to expand...
is this an attempt at very narrow vertical directivity by constructing the wavefront that would be in a conical horn (of great depth) using this quantity of small drivers closer to the mouth?
 
#10,797
Could I curve the pcb and throat face a little? 2mm pcbs are bendy. So the physical drivers are on a large perpendicualr radius, not a flat wave front. I think I could sim that with Vituix so the top and bottom drivers are set back some mms. Or could ATH4 allow a curved driver surface at the throat?

Another option could be some small guide surfaces or lens for the first few mm in the throat? I've seen that done on a few CD horns for PA.
 
Last edited:
#10,800
I think its realy cool, the right time delays on element quads and you could make it have constant directivity behaviour in the vertical but a very tight pattern (might not be required for desktop usage though as you could just be aligned with the array). Whats the small driver your intending to use? (If your willing to share this info).

I initialy thought you where making a desktop device like the Danley J series where layered combiner elements are used to make a curved wavefront to shorten the depth of the horn.l

1667606414821.png
 
You must log in or register to reply here.
Top Bottom