Brass instruments function much differently than audio horns. First of all they are built extremely resonant by mismatching length and horn taper. The taper of the bell is there not for horn type efficiency, rather to change the acoustical length of the instrument depending on the played frequency. In addition, we get different notes out of the trumpet by "overblowing" the fundemental resonance. Our Pedal note is the fundemental - one wavelength in the instrument, let's call this note C, the next "natural note" is 2 wavelengths or one octave higher also a C. The next is 3 wave lengths (G), 4 wavelengths is a C again, 5=E, 6=G(octave higher than the 3 wavelength g), 7 is discordant but close to a Bb, 8 wavelengths are the octave again. This continues until the 17th overblown partial where the resonance of the instrument no longer is sufficient for control, the entire work is now performed by the lips.
It is interesting that the sound of the trumpet comes from acoustic mismatch. What we hear is what leaks out from the standing wave - NOT an amplified buzz of the lips! Throat and reactance matching on musical instruments is done with the construction of the mouthpiece
How does this apply to loudspeaker horns? Well, Romy has posted that we should NOT use the horns at the bottom of their calculated possible range. The reason is that they become resonant there like a musical horn. Overblowing is very much like harmonic distortion. The voice becomes that of the standing wave, not the driver. We need to use the horn in the range where we have pattern control, not resonant support.
The "feedback" that Klaus and Reinhard use actually could correspond to the braces used on brass instruments. They must be critically placed or the response and articulation is destroyed. Some of the braces on musical instruments are used for feedback, others short circuit resonance by damping it. |
|