Acoustic Horn Design – The Easy Way (Ath4)

[mabat]

Acoustic Horn Design – The Easy Way (Ath tool)

This is an associated thread for a waveguide generator called Ath (Advanced Transition Horn), version 4.
Ath is a piece of software for designing waveguides and horns. At the same time it makes it possible to easily simulate their acoustic behaviour by means of FEA (BEM) via ABEC/AKABAK tool.

The current Ath release is available at https://at-horns.eu

Thread milestones (last edited 26.12.2020, no longer maintained):

#233 Birth of the OS-SE formula
Acoustic Horn Design – The Easy Way (Ath4)
Acoustic Horn Design – The Easy Way (Ath4)

#375 Measuring driver's exit wavefront
Acoustic Horn Design – The Easy Way (Ath4)

#666 Some of the practical verifications:
Acoustic Horn Design – The Easy Way (Ath4)

#844 Ath 4.4.1 - Introducing superformula
Acoustic Horn Design – The Easy Way (Ath4)

#922 JBL M2 "How-to"
Acoustic Horn Design – The Easy Way (Ath4)

#933 JBL M2 - Ath clone
Acoustic Horn Design – The Easy Way (Ath4)
Acoustic Horn Design – The Easy Way (Ath4)

#1003 Comparison of an OS waveguide and a flat piston
Acoustic Horn Design – The Easy Way (Ath4)

#1117 Ath 4.4.3 released (Fusion 360 import)
Acoustic Horn Design – The Easy Way (Ath4)

#1533 Ath 4.5.0 released (scripts not backward compatible)
Acoustic Horn Design – The Easy Way (Ath4)

#1602 General phase-plug discussion
Acoustic Horn Design – The Easy Way (Ath4)

#2389 Joining profiles of different curvatures
Acoustic Horn Design – The Easy Way (Ath4)

#2494 Tritona Waveguide
Acoustic Horn Design – The Easy Way (Ath4)
Acoustic Horn Design – The Easy Way (Ath4)


Spherical-wave phase plug

#1706 Initial ideas
Acoustic Horn Design – The Easy Way (Ath4)

#1824 Kessito's input
Acoustic Horn Design – The Easy Way (Ath4)
Acoustic Horn Design – The Easy Way (Ath4)

#1892 Compression cavity modal analysis
Acoustic Horn Design – The Easy Way (Ath4)

#2016 Diaphragm suspension analysis
Acoustic Horn Design – The Easy Way (Ath4)


Free standing waveguides

#2600
Acoustic Horn Design – The Easy Way (Ath4)

#2936 Rolling back the profile
Acoustic Horn Design – The Easy Way (Ath4)

#3390 Employing axisymmetric BEM (finally)
Acoustic Horn Design – The Easy Way (Ath4)

#3427 Rollback added
Acoustic Horn Design – The Easy Way (Ath4)

#3532 Source amplitude shading
Acoustic Horn Design – The Easy Way (Ath4)

#3549 OS-SE formula extended: k
Acoustic Horn Design – The Easy Way (Ath4)

#3664 Effect of an enclosure and other parameters
Acoustic Horn Design – The Easy Way (Ath4)

#3822 Compression driver throat plug
Acoustic Horn Design – The Easy Way (Ath4)

#4186 Preferred in-room response and DI target (discussion)
Acoustic Horn Design – The Easy Way (Ath4)

#4503 Incorporating compression driver model
Acoustic Horn Design – The Easy Way (Ath4)

 

[FredrikC]

Amazing work mabat!

Unfortunately, I am busy writing my masters thesis so I will have to wait to explore this for a few days, but I promise to give some feedback and share any ideas and results when I have had a go.

I skimmed through your documentation, which looks great, and noticed that you mention that you need a way to export coordinates. Why is that?

With modern cad software like Fusion 360 or Solidworks, the STL mesh can be turned into a surface body, which can be thickened and manipulated with holes, driver flange etc etc;



If the horn is eventually ending up in a 3D printer it will always be converted back to an STL so there is not really any need for a perfectly smooth surface.

 

[NiToNi]

Brilliant, Marcel! I will also check it out as soon as time allows and give feedback.

 

[Dave Zan]

Hi everyone, as I promised...

That is very nice work, thank you.

Best wishes
David

 

[mabat]

... With modern cad software like Fusion 360 or Solidworks, the STL mesh can be turned into a surface body, which can be thickened and manipulated with holes, driver flange etc etc;

That would mean that sufficiently fine STL may be enough (?). As I said, I don't work with such software at all, never had. Until now everyone complaned about the STLs that it is not possible to work with it further. Maybe they just did it wrong

OK, STL is already an option, so who knows how, can proceed with that.

I'm glad that you made it to actually run. Was not sure about all the libraries needed - I just gave it a try and released as it is.

 

[FredrikC]

That would mean that sufficiently fine STL may be enough (?). As I said, I don't work with such software at all, never had. Until now everyone complaned about the STLs that it is not possible to work with it further. Maybe they just did it wrong

OK, STL is already an option, so who knows how, can proceed with that.

I'm glad that you made it to actually run. Was not sure about all the libraries needed - I just gave it a try and released as it is.

Marcel, I have not run your software yet. That was an STL file that you posted in a different thread, which I just imported to solidworks as a surface body, thickened and extruded a flange to show that this is doable.

Best, Fredrik

 

[mabat]

Ah, OK. Thanks for the clarification.

 

[FredrikC]

Marcel, there is no gmesh.dll in my gmesh folder, just the exe. Where is it?

 

[mabat]

It's in the Software Development Kit (SDK), the third item: Gmsh: a three-dimensional finite element mesh generator with built-in pre- and post-processing facilities
Just extract the DLL to the same folder as the ath4 program.

 

[Patrick Bateman]

Marcel, there is no gmesh.dll in my gmesh folder, just the exe. Where is it?

Full disclosure:

I haven't read one word of the manual, because I'm horribly impatient.

But I was able to get it to work by doing the following:

1) downloaded OP's software

2) downloaded the gmsh SDK from here : Gmsh: a three-dimensional finite element mesh generator with
built-in pre- and post-processing facilities

3) I copied the DLL file from step 2 to the same directory as OP's software

From what I can see, OP's software depends on the DLL from the gmsh software development kit. Note that installing gmsh by itself won't satisfy the dependency, you have to get the SDK.

EDIT: as soon as I hit "submit", I see OP answered your question. Thanks OP!

 

[FredrikC]

Great, just tried the ATH15 example and it generates the STL and the Abec project files, but ABEC freezes when I try to open the project file. The STL looks fine.

Does it matter that I use Abec Pro?

Where to find the dll should be in the documentation or perhaps distrubte the .dll with your software if gmesh allow it.


EDIT: My bad, it just took some time to load because the mesh was so fine. Much finer than the STL??

 

[mabat]

I have no problem opening it, on my PC it's like three seconds. The density should be the same in fact.


For the dll, the manual says:

What you will need
...
- Gmsh SDK (available at Gmsh: a three-dimensional finite element mesh generator with
built-in pre- and post-processing facilities – you need just the file gmsh-4.3.dll)

What more do you need?
OK, maybe I could be even more clear on this but what's the use when people don't read it.

 

[FredrikC]

My bad Marcel, I overlooked the SDK part. Now that it is documented here, people will probably not repeat my mistake.

No idea why the STL and mesh file is so different in resolution yet.

Do you have any such issue Patrick Bateman?

 

[mabat]

Could you please upload the files? Did you change anything in the script?

- Oh, now I see. It won't be the same but I wouldn't say it could differ as much.

 

[Patrick Bateman]

My bad Marcel, I overlooked the SDK part. Now that it is documented here, people will probably not repeat my mistake.

No idea why the STL and mesh file is so different in resolution yet.

Do you have any such issue Patrick Bateman?

I'm seeing the following:

In the STL, I have 1344 vertices and 2592 faces.

In the MSH, I have 849 nodes and 1588 triangles.

 

[mabat]

To be precise: For the ath15 horn, three files are written: STL+MSH (exactly the same because these are from the same dataset) and ABEC mesh, which is made separately and that may be different. You can set the density of the ABEC one with the Mesh.* parameters.

 

[mabat]

FredrikC: I got the files, thanks. Now I can see you trouble - the density in your files is about five times higher than what I get. Now I don't have a clue how's that possible

 

[mabat]

This is my ABEC mesh -

 

[FredrikC]

Indeed, yours is much much more coarse. I have no idea why.

I tried downloading it all again, putting it on my c drive, load the example again. Same result. I'm running Win 10 on a eight core AMD Ryzen 2700. Is there any multi-threading involved in this process?

 

[mabat]

Patrick, how did you get your numbers? When I open my ath15.msh (not the ABEC one) in Gmsh I get these counts -
I will try to figure out, it must be some Gmsh thing. Like it took some smaller values for the mesh size at each point. But why?