Member
Joined 2003
Paul W said:
Its fairly high down the throat, but this is a small distance. At any rate, there is nothing that you can really do about the boundary flow, laminar or not, so there is not much point in worrying about it. In general no one has ever hyposthesized that boundary layer problems were significant in compression drivers and they have been modeled in extreme detail. If the effect is insignificant there then it has to be even less in the waveguide.
I take it that nobody was ever interested in actual testing of this either. I, however, find this vey interesting, and have conducted some listening tests with attemps to disrupt the boudary flow. At a later time, some measurements will be conducted.gedlee said:
I do recall someone did some listening tests using horns of different material. The results were, as long as the surface was very smooth, all horn of same shape sounded the same.
Member
Joined 2003
When you start talking about boundary layers please consider that there is very little relationship between what would be considered a conventional boundary layer in aerodynamics or fluids engineering and the boundary layer generated by the passage of an acoustic wave with a highly time dependent frequency spectrum and amplitude.
Paul W said:
The aircraft guys are able to manipulate boundary flow to minimize turbulence.
I'm just trying to noodle the optimum WG throat to match with a "throatless" compression driver having this internal construction. Suggestions?
Hi
Here is what JBL uses in the S9800 to match the compression driver's mouth in the vertical plan (the 435Be driver layout is identical to the 2435)
Attachments
gedlee said:
Semantics yes, but everyone's perspective is different and to be able to fully communicate some times requires a "shift in perspective".
Pertinent points of interest (IMO) to readers left "hanging":
1. use of the term "second derivative" was most certainly a term of art - and likely made that post at best only partially comprehended by most readers.
2. non-discussion of diffraction at the mouth in conjunction with the overall waveguide profile and diffraction. (..sure, *I* understood what you are doing here to *reduce* this problem - but I don't think others fully comprehend this yet. Moreover it *is* a REDUCTION, not an elimination of the problem, and it may well be the most significant factor for readers,(conceptually and perhaps audibly), on this forum.)
3. the emphasis on modest improvements in diffraction relating to HOM's may well be moot for most readers *if* most readers fit your definition of listening at "low spl's" (..and I'd bet that it does).
Note that none of the 3 issues is nit-picking, because each issue is overwhelmingly relevant to most readers here. (..also note that none of this is intended to be harsh criticism, rather instead as cautionary device for future replies.)
gedlee said:
1. You managed to describe everything else in a "lay" fashion, is it not possible with "second derivative" ?
2. I can see your point here (..limiting to "profile"). As a practical matter beyond the simply theoretical, there is going to be a curvature transition at that waveguide exit of some sort, "divorcing" this in this context is not only pointless, but also counter-productive.
3. I'm not suggestion that you "doll out" a conclusion based on that presumption, but rather that you provide a caveat for clarification to give all readers the opportunity to come to their own informed conclusions. (..and the same suggestion for #2.)
No need to reply to the question posed here.. again, its just a cautionary device.
Really, the only reason why I initially responded was that:
A. It was interesting information. (Thank you.)
B. I found it was difficult to read.. If I found it was difficult to read I'm pretty sure that at least a few others did as well. (..sure its a presumption, but no an unreasonable one.)
ScottG said:
Try this: If you take a stiff wire and hold it at 0 degrees and at the throat radius and then take the other end and hold it at the design angle, say 45 degrees, it will form the shape of an OS waveguide. The wires stiffness will minimize the second derivative (the change in slope).
Problem is that doesn't prove anything unless you realize that it is a change in slope (the second derivative) that causes the diffraction.
"If you take a stiff wire and hold it at 0 degrees and at the throat radius and then take the other end and hold it at the design angle, say 45 degrees, it will form the shape of an OS waveguide. The wires stiffness will minimize the second derivative (the change in slope)."
It's worth noting that the rate of change of slope is curvature, and that a constant curvature in 2D is simply a segment of a circle.
Presumably the "spheroid" in OS comes from the fact that revolving a portion of a circle produces a spherical surface.
It's worth noting that the rate of change of slope is curvature, and that a constant curvature in 2D is simply a segment of a circle.
Presumably the "spheroid" in OS comes from the fact that revolving a portion of a circle produces a spherical surface.
gedlee said:
..that's all I needed.
" Diffraction is created by a change in the slope of the waveguides bounding surface. The amount of difraction therefor depends on the second derivative of this bounding curve."
I never "got" that you were (again) just explaining diffraction with the second sentence - trying to reinforce the first sentence. To me it appeared that the second sentence was adding something new via "second derivative".
If the second sentence was: " The amount of diffraction therefor depends on THIS second derivative of THE bounding curve." It would have made more sense to me.
I think "bounding" is also important here, and not defined. (..though its absence didn't occur to me until just now re-reading it again.)
(For others: Wavelength/frequency *length* equal to or greater than the opening at any point in the waveguide equals "bounding".) I think.. correct me if I'm wrong.
