Jake T <
[email protected]> wrote:
I recently discussed my coyote audio capture methods with a friend of
mine, who is an electrical engineer. He suggested using a large cone to accomplish the same task citing that it was less bulky and cumbersome
than the dish and might even perform better. I told him that I would be concerned with how it would affect the frequency response, but, on the
other hand, so does the parabolic system I made. Just out of curiosity,
any idea why cones are not used?
Assuming you mean something like a public address loudspeaker in
reverse:
The horn that would cover the frequency range you need would have a
mouth as large as a parabolic dish covering the lowest frequency, so you
would gain nothing in size. Conical horns have a very uneven frequency response which varies according to the angle of incidence and the
acoustic impedance of whatever is on the 'sharp' end, so they sound
nasty and are virtually impossible to equalise. Hyperbolic horns are
much worse.
Exponential horns are an interesting case: there is a mismatch between
the large end and the surrounding air which reflects sound waves when
they are used as a 'projector' of sound. This results in a significant
hump in the frequency response around the bass cutoff frequency which
makes the graphs of their performance look better but actually sounds
worse. For gramophones, this hump corrects for the bass cut during
recording (a sort of mechanical/acoustic RIAA playback equivalent) but
when driven by a good loudspeaker they sounded 'honky'.
The tractrix horn is the best-sounding, although its low frequency
response doesn't look as good on paper as an exponential of similar
size. The mouth flare is a good match to the surrounding air and, as
long as you don't terminate the sharp end with anything stupid, it will
give a good flat response to sounds arriving on-axis.
Most of the research was done on horns used as projectors, not as
listening devices, but the foregoing remarks apply to either. A lot of
the mathematical approaches dating from the 1920s are highly suspect and
were based on unproven assumptions about the wavefront progression
(there were patents on exponential horns and academics may have had a
vested interest in writing papers 'proving' that they were the perfect
shape, which helped sales).
If you really must use a horn, make it a tractrix; otherwise I suggest
you stick to a parabolic reflector.
--
~ Liz Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk
--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)