How would one model this (see attachment) to understand the directivity vs baffle dimensions vs stuffing etc?
Its basically an open baffle with open sides rather than back and some stuffing in the void. Aim is to get good directivity into lower frequencies (lower than baffle step frequency)
Thanks
Its basically an open baffle with open sides rather than back and some stuffing in the void. Aim is to get good directivity into lower frequencies (lower than baffle step frequency)
Thanks
Attachments
Seems some cardioid to me. Modeling those is hard, as the stuffing properties are quite crucial and most materials aren't documented well enough. I'd build one, prepare various sets (densities) of stuffing and start the polar mapping of all.
That is more of a Cardioid dipole than a TL. You can get a general idea from ABC Dipole, The Rdge, or Google Passive Cardioid(some excellent threads here) for more info.
How much low end are you looking for?
How much low end are you looking for?
Last edited:
I can't remember but there is someone here who is well versed in building these. I think they come from Finland, maybe hang around with kimmosto.
I just remembered Grimm audio uses a 50cm or so wide baffle and that supports directivity to 250hz but that is a closed design. Too tired to think now. Night
I'm going off a long memory here but can you install the analysis toolkit.. eg File>Options>Add-Ins
Kimmo certainly did some cardioids himself 😉 and if we are going that way, I would not forget to mention Martijn Mensink, who's around here too sometimes. But I think you meant Juhazi.
[Edit]forget it. Next time try a dm to the cardioid experts here 😉 you know them.
Last edited:
I did do that but there was another problem. The .CHM file was not showing any text in the right pane so I couldn't see the install instructions but I spotted in the word document that the generated .TXT file needs to be in the same folder as the .XLS file. I fixed that issue and now it works fine.
This could be modeled in Akabak using three waveguides emanating from a common central round chamber the size of the driver cutout. The waveguides would be radial pie wedges out. One up and two to the sides. The exit apertures are modeled as radiators aimed out in appropriate directions and with appropriate sizes. Stuffing could be adjusted and polar plots viewed. Akabak 3 uses the ABEC GUI and may be better as you can model the structure with 3D mesh. Although I don’t think that’s necessary. A boundary element model with 3 waveguides and central chamber will work well.
Cool! Might download and play tomorrow. Steep learning curve? Have used horn resp years ago.
A boundary element method cannot model the absorption by the stuffing which is not on the boundary but distributed throughout some of the solution volume. It is possible to couple a boundary element method to another form of modelling to handle the absorption but that would be unusual. I would suggest checking if AKABAK has this capability before investing time learning how to use it.
A more appropriate form of modelling for your problem is likely to be one that solves acoustic equations in 3D space rather than just a surface. You will need to check if it includes terms to model absorption within the volume and not just at the boundaries with only pure propagation within the volume. Elmer might do what you want.
In Akabak, “Waveguide” elements can be specified to have stuffing distributed within the waveguide volume element. You can add stuffing in Akabak using the damping term and set damping as pressure loss in Pascal per unit area per distance (Pa/m^3). I find 5000 to 7500 Pa/m^3 corresponds well to typical polyfill stuffing. You can also look up tabulated pressure loss values for mesh screen and use mesh as well.