This frequency
f
c
is called the cutoff frequency because, for frequencies lower than this, no power will be transmitted down the horn, i.e., the impedance at all positions along the horn is purely reactive. The throat impedance of an infinite hyperbolic horn is plotted in
Fig. 9.9.
In
Fig. 9.9, the throat impedances for the parabolic, conical, exponential, and hyperbolic horn types are shown. At very high frequencies, all these types behave about alike. At low frequencies, however, there are considerable differences. These differences can be shown by comparison of the throat impedances for the conical and hyperbolic horns with that for the exponential horn.
For all horns, the throat resistance is very low, or zero, below the cutoff frequency. Above the cutoff frequency, the specific throat resistance rises rapidly to its ultimate value of
ρ
0
c for those cases where the rate of taper is
small near the throat of the horn. For
example, the specific throat resistance for the hyperbolic horn reaches
ρ
0
c at about one-twentieth the frequency at which the specific throat resistance for the conical horn reaches this value. Similarly for the hyperbolic horn, the specific throat resistance approaches unity at about one-third the frequency for the exponential horn.
It would seem that for best loading conditions on the horn drive unit over the frequency range above the cutoff frequency, one should use the hyperbolic horn. However, it should also be remembered that the nonlinear distortion will be higher for the hyperbolic horn because the wave travels further in the horn before the pressure drops off owing to area increase than is the case for the other horns. For minimum distortion at given power per unit area, the conical horn is obviously the best of the three. The exponential horn is usually a satisfactory compromise in design because it falls between these two extremes.