Nice 
Do you have an idea how to simulate that? "cheap" additive manufacturing of today would enable objects like that be built easily but shouldn't there be some notions of system like that from the old days if it worked even though it would have been tedious to manufacture back in the golden age of audio? Is this what was showed few pages ago, some new phase plug or something that had many horns inside horns in the picture? Something new nevertheless. Would be nice if it worked. I predict it will beam the high frequencies (the axial modes from the driver).
Do you have an idea how to simulate that? "cheap" additive manufacturing of today would enable objects like that be built easily but shouldn't there be some notions of system like that from the old days if it worked even though it would have been tedious to manufacture back in the golden age of audio? Is this what was showed few pages ago, some new phase plug or something that had many horns inside horns in the picture? Something new nevertheless. Would be nice if it worked. I predict it will beam the high frequencies (the axial modes from the driver).
Last edited:
I think that it would work. It just seems to me to be problematic with construction. The HOM cut-ins would be much higher, but if this isn't important then it likely wouldn't be much of an improvement.
A single internal vane might work well for 2" waveguides. I could see stamping the inner cone in some metal and suspending it inside with foam. A very tough cut of the foam however.
Last edited:
Give me a few days, I'll make it an Ath feature. I'm woking on just that, it's not difficult.
Yeah, that's the reason for my doubts. I don't know.
After all it's not a problem to simulate driver axial modes as well.
Isn't it that the path lengths will be different on average inside each section? In that case I would assume a pretty wild acoustic summing at the end.
Edit: Hmm, but maybe it doesn't make a difference. It's difficult for me to imagine what's happening there.
Last edited:
reflection
In a CD the phase plug starts close to the diaphragm so that there are no unwanted cavities with reflections. Unless you can make the vanes thin, even blade-like, or otherwise minimize reflections, I would be concerned that the reflections back into the CD will be problematic, at least for some designs where the cavity is too big.
As soon as you mentioned this, I wondered if there's a CD with suitable phase-plug termination to use the ends of the phase plug rings as the start of your vanes - allowing any cavities to be minimized. Otherwise, unless I'm mistaken, this - nice idea - is going to be rather hit and miss with different CDs.
For example, many 2" CDs are made with a long front section, and I will be surprised if that works well with finite-thickness vane(s).
It's a great way to build on the 1P nature of OS.
Ken
In a CD the phase plug starts close to the diaphragm so that there are no unwanted cavities with reflections. Unless you can make the vanes thin, even blade-like, or otherwise minimize reflections, I would be concerned that the reflections back into the CD will be problematic, at least for some designs where the cavity is too big.
As soon as you mentioned this, I wondered if there's a CD with suitable phase-plug termination to use the ends of the phase plug rings as the start of your vanes - allowing any cavities to be minimized. Otherwise, unless I'm mistaken, this - nice idea - is going to be rather hit and miss with different CDs.
For example, many 2" CDs are made with a long front section, and I will be surprised if that works well with finite-thickness vane(s).
It's a great way to build on the 1P nature of OS.
Ken
If you stop the inner vanes and continue the outer boundary it is not the same result as running all boundaries directly to the end of the line...
If you take 3" diameter line...split it in half, from beginning to end the tuning is not the same of take the same, but the division wall is terminated before the end of the whole 3" line.
The path length difference is what creates a spherical wavefront. In an OS waveguide, the 0th mode travels along the radial coordinate and normal to ALL angular coordinates. This means that insertion of vanes that are angular coordinates themselves will always be normal to the wavefront and if the vanes are thin enough, then they will do nothing to the 0th mode. They will raise the cut-in for any HOM however, quite possibly pushing them out-of-band.
The thickness is an issue as the vanes need to be thin.
How did you generate the profile and the vanes? It would be interesting to see the mesh. (Beware that it probably must be a real OS, i.e. with k = 1.)
Haven't you tried the vanes longer?
Last edited: