Greetings to all enthusiasts
Tried to import the results into akabak - naturally failed. On the akabak website it is written that the program does not accept the results of gsmh version 4. "The new mesh-file of version 4 cannot be read at the moment. Work-arounds:
Use GMSH version 3 (see GMSH-web-site at Downloads/All Versions-Binaries, GMSH3 saves mesh-file of version 2.2).
Use GMSH version 4 and save mesh-file with the help of menu File/Export+File-Type=msh (eventually dialog opens where you can select mesh-file of version 2.2)." ATH does not work with GMSH version 3.0.6 giving errors. Is it possible in the ATH configuration file to specify the format of the output *.msh file to version 2.2 when it is generated?
Tried to import the results into akabak - naturally failed. On the akabak website it is written that the program does not accept the results of gsmh version 4. "The new mesh-file of version 4 cannot be read at the moment. Work-arounds:
Use GMSH version 3 (see GMSH-web-site at Downloads/All Versions-Binaries, GMSH3 saves mesh-file of version 2.2).
Use GMSH version 4 and save mesh-file with the help of menu File/Export+File-Type=msh (eventually dialog opens where you can select mesh-file of version 2.2)." ATH does not work with GMSH version 3.0.6 giving errors. Is it possible in the ATH configuration file to specify the format of the output *.msh file to version 2.2 when it is generated?
Hi,
that's not the problem as the version of MSH is always 2.2, regardless of gmsh version used - there's always "Mesh.MshFileVersion = 2.2;" at the beginning of each .geo file that is generated (I use gmsh 4.6.0 at the moment).
The real problem is that the scripts used by the old ABEC are not directly importable into AKABAK anymore - some constructs apparently changed. It's a pity but that's how it is.
Personally, I don't like the GUI approach chosen for AKABAK, I was perfectly happy with the scripting approach - and that's still what I need, so I still use ABEC. Can't really help at the moment.
that's not the problem as the version of MSH is always 2.2, regardless of gmsh version used - there's always "Mesh.MshFileVersion = 2.2;" at the beginning of each .geo file that is generated (I use gmsh 4.6.0 at the moment).
The real problem is that the scripts used by the old ABEC are not directly importable into AKABAK anymore - some constructs apparently changed. It's a pity but that's how it is.
Personally, I don't like the GUI approach chosen for AKABAK, I was perfectly happy with the scripting approach - and that's still what I need, so I still use ABEC. Can't really help at the moment.
Perhaps the simplest way how to get the sim into AKABAK would be just to use the generated mesh (*.msh) file alone (which has internal tags for all that's needed) and set the rest via the AKABAK GUI.
The *.msh file alone should be definitely readable by AKABAK/ABEC, as it's always format 2.2, as I wrote (and is has been always like that).
The *.msh file alone should be definitely readable by AKABAK/ABEC, as it's always format 2.2, as I wrote (and is has been always like that).
Wow that looks awesome @jzagaja , I like walnut but too much money. I was thinking of just doing layers of baltic birch plywood.
Maybe try wood that after machinig looks good e.g. Meranti used in windows industry.
NoMabat, have you got the sand horns you got printed already running?
^^Mainly interested about the manufacturing technique and how it worked out. Sound is not as interesting since it will be good if the complete system is good![]()
I suspect differences in DI and loading between horns will be noticeable, albeit subtle.
Perhaps aragorus could also share his experiences with the axisymmetrical and bi-radial waveguides that he showed previously.
I'll start by saying that my results with Ath are waaaay more consistent than my experience lofting various "handmade" contours in Fusion360.
I have only used the BMS4550 driver in the various horns I've printed, but I'm pretty happy with the results!
It usually measures better than it simulates
The response is not flat (as the driver's response is not flat) but a little low Q notch filter around 2.5k makes it pretty flat.
I make active speakers and normally just use DSP to correct, but lately I've been playing around with a passive filter that does it just as good.
I'm not much into subjective opinions about speakers, but I can say that my results are not "honky" and the sound is spacious and present.
Crossed at 6-700Hz to 12" B&C 12ndl88 reflex boxes tuned just over 30hz
Hope that answers the question
Here are some pictures of prototypes I've been printing:
![20210422_070328648_iOS.jpg 20210422_070328648_iOS.jpg](https://www.diyaudio.com/community/data/attachments/866/866551-b8057258cf892062bf67045e3241d0f2.jpg)
![20210422_070342298_iOS.jpg 20210422_070342298_iOS.jpg](https://www.diyaudio.com/community/data/attachments/866/866557-3828a635e609c8e6933b5136358514a7.jpg)
![20210416_070201296_iOS.jpg 20210416_070201296_iOS.jpg](https://www.diyaudio.com/community/data/attachments/866/866563-912863f02e41d679394a87ca75346c89.jpg)
![20210513_132238709_iOS.jpg 20210513_132238709_iOS.jpg](https://www.diyaudio.com/community/data/attachments/866/866571-4a5c701f2e934696a7a1d4c63992a676.jpg)
There is an invisible cage protecting them acoustic creatures from these light eaters and I think it looks marvelous! Also would like to understand the 2-way system with parameters and so.for the love of (insert deity name here) take that damn plant off those albums....
Here's the horizontal polars at 1 meter. Not looking very good.
Any gating/windowing applied?
- Home
- Loudspeakers
- Multi-Way
- Acoustic Horn Design – The Easy Way (Ath4)