I'm too afraid of not being able to reassemble it to its original performance. But I am really 
These where my first ever CD. I'm not totally sure about its SQ. I just can't make up my mind / figure out if it has a somewhat harsh side or is it that so many recordings are not so good. I have some data that points to the latter as it can be absolutely wonderful... On the other hand I would need some more experiance with other CDs to really ***** the situation. It's not a recent design - I see it being mentioned in threads from 2008 - surely we should have come some way since then?
I'm planning for a 1,4" CD - probably from SB and a larger ATH derived WG.... the yearly speaker project...
//
These where my first ever CD. I'm not totally sure about its SQ. I just can't make up my mind / figure out if it has a somewhat harsh side or is it that so many recordings are not so good. I have some data that points to the latter as it can be absolutely wonderful... On the other hand I would need some more experiance with other CDs to really ***** the situation. It's not a recent design - I see it being mentioned in threads from 2008 - surely we should have come some way since then? I'm planning for a 1,4" CD - probably from SB and a larger ATH derived WG.... the yearly speaker project...
//
This is what 18sound tells about it:
"The Proprietary Phase Plug (3P) technology identifies a combination of radial and tangerine slot geometric design. With its short openings and high flare rate value, 3P technology assures low distortion in the mid-high frequency range, providing a smooth coherent wavefront at the horn entrance."
It's yet something else, apparently some variation on radial design...
"The Proprietary Phase Plug (3P) technology identifies a combination of radial and tangerine slot geometric design. With its short openings and high flare rate value, 3P technology assures low distortion in the mid-high frequency range, providing a smooth coherent wavefront at the horn entrance."
It's yet something else, apparently some variation on radial design...
An other problem is that I made a design that is very cumbersome to pick apart - something I will make sure not happens again. The new ones will be a 2 part thing joined with screws for easy maintenance / tweaking... but basically still the same architecture as the HornFlower.
Whats a "short" opening?
One thing is for sure and I tried a lot... it cant be played below 1,4k for hifi - despite XO steepness... because I can't imagine the ST260 being responsible for that...
//
Whats a "short" opening?
One thing is for sure and I tried a lot... it cant be played below 1,4k for hifi - despite XO steepness... because I can't imagine the ST260 being responsible for that...
//
Conventional BEM codes become inaccurate with "thin shapes" due to the growth of numerical errors associated with integrating singular functions. The waveguide passages look like they might be "thin" w.r.t. this issue. Do you have an estimate of the size of the numerical error from this source in your simulations? Perhaps it is small enough to not significantly alter what is wanted from the simulations but if it is not considered it is likely to be a source of doubt in the results.
I don't wish to be discouraging since it's good to see challenging projects like this being tackled within the hobby. Indeed having a go and perhaps concluding geometries are bit too thin would still be positive useful information for the hobby. As you are likely aware a project like this would normally be tackled with a FEM acoustic code rather than a BEM one.
I don't wish to be discouraging since it's good to see challenging projects like this being tackled within the hobby. Indeed having a go and perhaps concluding geometries are bit too thin would still be positive useful information for the hobby. As you are likely aware a project like this would normally be tackled with a FEM acoustic code rather than a BEM one.
I tried to replicate the results from the work referenced below.
FEM / Fig.4 from https://www.researchgate.net/public...ers_with_narrow_channels_and_rigid_diaphragms
BEM via ABEC3:
This was in axial symmetry mode. It's pretty sensitive to the exact geometry defined.
I created this manually. With the "plotter" tool I'll be able to easily test both axial symmetry and full-3D mesh.
I guess a more proper way would be to compare a sim to some "simple" analytical solution, but I don't have any at hand.
FEM / Fig.4 from https://www.researchgate.net/public...ers_with_narrow_channels_and_rigid_diaphragms
BEM via ABEC3:
This was in axial symmetry mode. It's pretty sensitive to the exact geometry defined.
I created this manually. With the "plotter" tool I'll be able to easily test both axial symmetry and full-3D mesh.
I guess a more proper way would be to compare a sim to some "simple" analytical solution, but I don't have any at hand.
That looks promising for a geometry I would have expected to cause issues for conventional BEM. There are differences though which it may be wise to understand and quantify. Of course the published FEM results may not be fully grid independent though it should be straightforward to achieve in 2D. A fully resolved FEM simulation of a representative geometry is likely more useful than an analytical solution of a less representative geometry given the nature of the error.
Good stuff.
Good stuff.