Acoustic Horn Design – The Easy Way (Ath4)

[mabat]

Yesterday I could spend all day working on speakers

...and today working on the horn molds. I want to try concrete with these.

The printed parts are basically hollow shells to be filled - probably gypsum to keep the costs down.
(The blue one is not yet complete, the mouth layer is still being printed.)

These are 640 and 540 mm wide, 1.5" and 1.4" throats. The drivers at the moment are the ND4015Ti2 and HF1440.

 

[mabat]

Not sure if it's clear from the pictures but this is the idea:

No additional mounting plates for the drivers should be necessary, only bolts.

 

[jzagaja]

Gypsum or concrete will make the job of course but heavy. From other side 3D printed structure allows very light and stiff structure. PLA+GF is about 35 euro/kg (Azoty). A way to go is PP+GF.

 

[mabat]

I actually think the molds could be just PLA, if printed with a denser infill. I wanted the print job quick (these are already rather large pieces) so these are basically hollow with only a very sparse gyroid infill to keep the shape during pouring. These were very easy to print, without any warping etc. I'm quite satisfied with the results so far.

- This was about the biggest I could print on my Ender 6 (could be a bit larger but not much) :

 

[mabat]

A "bottom" mouth piece: http://at-horns.eu/img/timelapse_BASE_2x-C.mp4

- The whole model (each part shown only once):

 

[Tom Kamphuys]

What are the dimensions?!

 

[mabat]

The blue one is 64 cm, the same width as the OB frame shown on the previous page. The red one is 10 cm smaller.

 

[gedlee]

Not sure if it's clear from the pictures but this is the idea:

View attachment 1179108 View attachment 1179109

No additional mounting plates for the drivers should be necessary, only bolts.

As one who has study resonant structure quite a bit, I'd find a way to brace the grey part into the waveguide. Cantilevering a heavy driver on a small ring is not such a great mounting.

 

[Dkalsi]

The above shape:

OSSE = {   ; #33/T34A-ROSSE
  r0 = 19
  a0 = 41.06
  a = 71.25
  k = 1.3
  L = 19
  s = 1.12
  n = 2.467
  q = 0.999
}

Hi @mabat I'm trying to follow along with how each factor impacts the shape of the waveguide.

In an attempt to do, I'm using the inputs above and modeling them using the OSSE calculator you linked on your website.

I've attempted to populate what I can, but I know I'm obviously on the wrong path because there is no place for me to specify the "a=71.25" input and the "t=5" default input below doesn't tie to anything above.

Can you please help?

 

[fluid]

Use this one instead

https://www.desmos.com/calculator/3iswclzjzh

t is the same as a0, the opening angle of the throat which is or can be different from a which is the coverage angle (alpha symbol in desmos graph you linked).

 

[mabat]

As one who has study resonant structure quite a bit, I'd find a way to brace the grey part into the waveguide. Cantilevering a heavy driver on a small ring is not such a great mounting.

Earl, the grey structure is a part of the mold to make the bolt pockets as shown below. I really can't see how to make this better.

 

[beeharry]

Thank you for creating an easy program to design waveguide. I am studying with ATH-Dom-Template file. When I used akabak, I had a problem of substituting LE values, so I would like to try again using abec3. The questions are as follows

1. I couldn't find in the manual what "Mesh.ZMapPoints = 0.3,0.2,0.7,0.9" means.

2. I understand that aath can be constructed by adjusting parameters and constants of OS-SE or other mathematically defined curves. But this needs to be adjusted manually. Is there any way to optimize this automatically? I can use MATLAB student license.

 

[mabat]

Use this one instead

Yeah, and of course feel free to change the slider limits. They might be a bit too restrictive as they are.

 

[mabat]

1. I couldn't find in the manual what "Mesh.ZMapPoints = 0.3,0.2,0.7,0.9" means.

It defines the rate of change of the z-coordinate of the slices (how densely is the shape sampled with axial distance). It was explained here (I have to do the search myself as I don't remember): https://www.diyaudio.com/community/...-design-the-easy-way-ath4.338806/post-6102810

2. I understand that aath can be constructed by adjusting parameters and constants of OS-SE or other mathematically defined curves. But this needs to be adjusted manually. Is there any way to optimize this automatically? I can use MATLAB student license.

No, you would have to devise your own.

 

[mabat]

...and today working on the horn molds. I want to try concrete with these.

I won't cast the concrete but will work with a clay consistency instead, perhaps I wasn't clear on that:

It's possible to make almost vertical walls that way even with a simple one-part mold.

 

[beeharry]

Because I needed that...
It was the first step to control Ath from Matlab in order to replace the Matlab MMM_toolbox solver by ABEC:
Matlab for Genetic algorithm, Ath to generate the profile and ABEC project, ABEC Demo to solve and finally import the results back into Matlab.
No more IB limitations.
Now it's all done, oddly it was not as painful as I had imagined. @Matbat and the R&D team are legends.
I still need to make sure everything is fine and see if I can leverage the Ath report function as a bonus instead of rewriting the whole thing but it's looking good ATM.
CNY is coming, I will look into that in depth then.

(Much) further down the line, I am looking into optimizing the box + waveguide as a whole, as l prefer the looks (I know, superficial...) and the better usage of the physical foot print (larger internal volume) of a "shoebox".
I suspect that it won't be lost on the midwoofer upper range to have an optimized box either.

Any tips on how to optimize the sim parameters to have shortest calculation time?
That would be greatly appreciated!

@maiky76 I've found someone who's already tried what I'm looking for! It's hard to get much information from what you've written. Can you share more about your project?

 

[mabat]

... there's a catch with the sources compound of more than one driving groups - what is actually plotted is the impedance of one of the surround rings

This is a workaround - you would have to reorganize the ABEC scripts this way. I'll make this an option.

solving.txt

Elements "SD1D1001"
Subdomain=1; MeshFileAlias="M1"
1 Mesh Include SD1D1001

Elements "SD1D1002"
Subdomain=1; MeshFileAlias="M1"
1 Mesh Include SD1D1002

Elements "SD1D1003"
Subdomain=1; MeshFileAlias="M1"
1 Mesh Include SD1D1003

Driving "S1001" // horn driver
DrvGroup=1001; Direction=z
1 1 RefElements=SD1D1001 Weight=0.25 // outer surround
2 1 RefElements=SD1D1002 Weight=0.75 // inner suuround
3 1 RefElements=SD1D1003 Weight=1.0 // dome itself

observation.txt

Driving_Values
DrvType=Acceleration; Value=1.0
401 DrvGroup=1001

Radiation_Impedance
BodeType=Complex; GraphHeader="RadImp"
Range_min=0; Range_max=2; RadImpType=Normalized
404 1001 1001 ID=8001

 

[aragorus]

The casting technique looks pretty nice!
I've always been tempted to do a Jesmonite cast.

 

[jzagaja]

From concrete one can even make a bowling ball that works,

 

[mabat]

I wanted to avoid casting if possible. Applying concrete manually as a clay (the "press technique") attracted me immediately, as I think it can't be more simple overall. Not to mention the costs.