Acoustic Horn Design – The Easy Way (Ath4)

... I'd appreciate to get a *.step-file as well, which would ease up the transfer to a CAD program to finally print the horn. Any possibility?
Hi Christoph,

that's not at all easy. To get a STEP or IGES file I would have to use some surface splines (NURBS) to get the same results as I have now and I'm not sure I'm willing to do that. I agree that STL is probably not really the best option but it's all I have for now - it's just very easy to do. I could make the STL as a volume instead of just a sheet - that may ease the processing, I don't know. But as I understand, there are ways to do that in CADs.

I will add an option of a "point cloud" output. That may also help some.
 
Last edited:
Can this program be use to make full range wood horns? Or mid/high wood horns?
You would have to explain what do you mean by that. The program uses some basic set of parameters to get a horn. It depends on your needs if this is sufficient or not. Basically you can combine hyperbola and a clothoid in various ways along the contour, in any dimensions you like.
 
Last edited:
I thought about the dome tweeters. The biggest problem I see might be that the real dome (especially a soft one) will not move coherently as a one body at the higher frequencies where the simulation would be the most interesting. I will probably implement it, as it is not difficult, but I would be hesistant to make much of it. Maybe for hard domes it will give more reliable results. But hey, it's similar to compression drivers flat wavefront assumption.
 
Hi guys

I know that you are already couple of steps forward but need some help with ath4 application.
I followed instructions, extract ath4, copy in same directory gmsh-4.3.dll, made cmd shortcut and copy file path in "start in". When I double click cmd shortcut nothing happen, actually new cmd window blink for a second and disappear. I rename ath4 to ath4.exe and tried again, same thing.
After that I've tried to go regular procedure: open cmd, type cd (file path), enter, type start and application name (both ath4 and ath4.exe) and same thing happens.
After first fail I deleted all files and download and repeat everything again on two computers with same results.
So please is there any known solution ...
 

Attachments

  • cmd.png
    cmd.png
    182.1 KB · Views: 880
Last edited:
I thought about the dome tweeters. The biggest problem I see might be that the real dome (especially a soft one) will not move coherently as a one body at the higher frequencies where the simulation would be the most interesting. I will probably implement it, as it is not difficult, but I would be hesistant to make much of it. Maybe for hard domes it will give more reliable results. But hey, it's similar to compression drivers flat wavefront assumption.

Hi Marcel,

I've found that 3/4" dome tweeters work really nicely on horns and waveguides. The SB Acoustics SB19 works on virtually every waveguide I've come across.

I agree that a lot of dome tweeters are useless on horns and waveguides, but there are a handful that work nicely.
 
What I meant was the simulation trustworthiness. I know that domes may actually work fine (although with much less headroom than CD).

I can easily make the throat driving element in a shape of a dome tweeter. I'm just not so sure about how the acoustic results will match the reality. Anyway, let's try. I guess I will just use some weighting for the surround output contribution.
 
Well...
Have you built anything based on the Ath4 software to validate the modelling thus far?
...when it comes to conformance of modelling and measurement of the final built, then ABEC has to be discussed, not Ath4. Ath4 just makes it easier to built an ABEC model, i.e. helps with the input to ABEC. Here based on an OS-waveguide.

When it comes to AEBC: This software can do very precise modelling, depending on the level of detail of the model and some experience...
 
Has anyone had success creating a high quality CAD solid part from an Ath4 project stl file?



So far I've tried using Meshmixer to convert a hi-resolution mesh to a solid but always end up having to sacrifice detail in order for Meshmixer to be able to successfully convert it into a solid.


Any tips or tricks would be great!
 
Have you built anything based on the Ath4 software to validate the modelling thus far?

From the previous experience with different OS waveguides I can see a lot of resemblence, especially regarding the axial ripple and diffraction issues. That I would quite believe but I yet have to build a horn based on ath4 exactly 1:1.

This is one existing horn built based on ath4 algorithm:
hifi.slovanet.sk :: Zobrazi? t?mu - Projekt STH12 (2p - 12"+1"@STH zvukovod by MaBat) (see measurements down the page)

So when ABEC shows nice and smooth polars, I would believe it couldn't get much better...
 
That is great, Patrick (or John). The first step, i.e. converting STL to solid, can also be easily automated. I'll definitely do that, to generate the STL as solid right away, I just have been quite busy for the last several months so not much work has been done on ath4.

What also seems needed is to have an increasing number of points as approaching the mouth to have the STL smooth enough everywhere. It "only" requires a complete rewriting of the code that generates the mesh - something I saw from the start but tried to avoid for as long as possible. I didn't consider STL as a format suitable for 3D modeling, but it seems a lot of people will want to use it directly as it is. It should also help with the BEM however.