The whole point is that the conversion from flat to spherical is done by means of the narrow channels where HOMs don't apply (are pushed out of the audioband). The wavefront doesn't even know it's being converted 
It's still not perfect however, I'm still chasing some minor issues...
It's still not perfect however, I'm still chasing some minor issues...
Last edited:
Looks good!
I received the .2mm nozzle and getting my printer to print with it atm.
Meanwhile I was thinking that a similar technique could be used to make a multicell horn, but instead of revolving - lofting could be used to do one segment at a time.
It would take significantly longer to simulate in 3d tho..
I think multicell horns are so awesome, and it's a pitty they have the issues they do
I received the .2mm nozzle and getting my printer to print with it atm.
Meanwhile I was thinking that a similar technique could be used to make a multicell horn, but instead of revolving - lofting could be used to do one segment at a time.
It would take significantly longer to simulate in 3d tho..
I think multicell horns are so awesome, and it's a pitty they have the issues they do
That will take a few days
- That's the beauty of SLA printing, it takes the same amount of time no matter how big or complex shape is printed, only the height is what counts. The printer I have is too small for the whole 2" plug, I would have to print it in several parts (not sure I want to do that, maybe I will try).
Last edited:
Once we have a spherical wave we can have constant directivity for an arbitrarily big source. Flat wavefront will beam on high frequencies, depending on its diameter. With the approach presented here we can produce a spherical wavefront as a result of an exponential expansion (i.e. with high efficiency), which would be a remarkable feat.
I've been saying that for more than a decade. The driver manufacturers can do all this stuff, they just aren't interested.
I looked at this problem a bit and concluded that the shorter the phase plug "throat" the less of a problem that it was. That's why I settled on the B&C DE500, because it has the shortest "throat" of any driver that I had looked at.
I looked at this problem a bit and concluded that the shorter the phase plug "throat" the less of a problem that it was. That's why I settled on the B&C DE500, because it has the shortest "throat" of any driver that I had looked at.
A portion of a spherical surface, large compared to the wavelength and vibrating radially, emits uniform sound radiation over a solid angle subtended by the surface at the centre of curvature.
Assuming that it proves possible to produce a spherical wavefront where the radial air motion has the same phase and amplitude over the cap surface at the exit from the phasing channels (despite some of the channels having bends), if the main horn profile flares at a rate greater than that defined by the spherical cone, then the wavefronts will not remain truly spherical across the entire horn cross-section as they progress along the axis towards the mouth. This will be particularly so if a Le Cléac'h type fold-back is employed at the mouth.
Interestingly, in his profile-generating algorithm Jean-Michel made provision for having either a curved or a plane wavefront at the horn throat.
I don't think there is a past in a horn/wg that don't matter for the rest of the journey to the mouth. The whole thing is an organisms from start to end and "everything matters" - seems like David has some insight here that might be better digested if taken along with a small piece of humble pie dessert!? ;-D
//
//
So why do you have one on your horn / waveguide?
Jean-Michel was the first person, to my knowledge, to take the flare tangent angle beyond 90 degrees.
Attachments
