I wonder which devices would give a planar axisymmetric wavefront except for compression drivers? Ribbons (but I guess it can't be rectangular)? Other speakers? What does a dome tweeter give?

I haven't taken any courses in acoustics (but is M.sc in EE), so I don't understand most of the things in the paper Anyway, it's interesting to read and try to understand.

/Mattias

Glad that you could get something out of my paper. The book, which is updated and written ten years after the paper, does a far better job at explaining things.

You are pretty much correct that a compression driver can actually create a flat planar wavefront. The first source that we tried however was a flat honeycomb piston rom Panasonic. Worked just as expected. A conical diapragm does not give a flat wavefront at the troat and as such the waveguide deos not work well without a phase plug. A dome is an axisymmetric source, but not planar, as such it will not work as well as the flat piston.

When the source is not planar then the performance is very hard to predict since the non-planar aspects will excite the detrimental higher order modes.

I should also point out that current phase plug designs do not really give flat planar wavefronts either. Thats because of an error in the design asumptions made by Bob Smith from whose work the current phase plug designs originate. I have a patent pending on a correction to the design which would achive a lat wavefront. I also have a patent pending on the use of interchangable phase plugs in a compression driver.

I've been doing a lot of work in this area in the last couple of years. I think that my speakers can attest to the benifits of that work.

Earl

Thanks for taking your time and answering!

Mattias

Cheers,
Mattias

Quit the contrary they are the most common sources for waveguides - compression drivers.

"A compressiondriver has like 105dB sensitivity while the 15" bass lies in the 96-99dB region. I'm thinking the extra sensitivity of the compressiondriver isn't needed since it'll have to be attenuated in the crossover anyway"

All of which is a good thing. Sensitivity is relatively unimportant - unless its really low - its Max SPL that matters. And a compression driver has lots of that. Sure a 1" tweeter would work on a waveguide, they work fine, but they suffer from limited LF capability and really loose it at higher SPL's. I also use the high sensitivity to advantage in my passive crossovers with the resistor pad for lower the sensitivity. This makes the driver look almost like a purely resistive load, which is ideal for passive crossovers.

"Is it possible to improve the compressiondriver (more linear?) but sacrificing some sensitivity?"

My studies have shown that compression drivers have no perceivable nonlinear distortions at any level - even 124 dB. So I would say that a compression driver is just about the ideal HF source. And, of course, this is what I use. It is not a coincidence.

But compression drivers on horns have a deserved reputation for poor sound quality. I fixed this problem and now consider them to be the best available source for HF response - even for audiophile loudspeakers.

Regards,
Mattias

In discussions of this sort it is essential that we delineate linear distortions from non-linear distortions. Frequency response aberations are linear.

The horn or waveguide dominates the frequency response of the system, although there are a couple of driver effects. One is the resonance of air arround the voice coil - this needs to be controlled - and at higher frequencies there are diaphragm resonances. But basically compression drivers are far flatter devices than piston sources. This is because of the high damping both acoustically and electro-mechanically.