Handbook for Sound Engineers

498 Chapter 16

In both cases, force F 1 , the sound pressure on the
front of each diaphragm, is the same. Force F 2 is the
force on the back of the diaphragm when the micro-
phone is used at a distance from the sound source, and
F 0 is the resultant force. The force F 2 con the back of the
diaphragm is created by a close sound source. Laterally,
the vector sum F 0 c is considerably larger in magnitude
than F 0 and therefore produces greater output from the
microphone at low frequencies. This can be advanta-
geous or disadvantageous. It is particularly useful when
vocalists want to add low frequency to their voice or an
instrumentalist to add low frequencies to the instrument.
This is accomplished by varying the distance between
the microphone and the sound source, increasing bass as
the distance decreases.

16.2.3.1.1 Frequency Response

Frequency response is an important specification of
unidirectional microphones and must be carefully
analyzed and interpreted in terms of the way the micro-
phone is to be used. If a judgment as to the sound
quality of the microphone is made strictly from a single
on-axis response, the influence of the proximity effect
and off-axis response would probably be overlooked. A
comparison of frequency response as a function of
microphone-to-source distance will reveal that all unidi-
rectional microphones experience a certain amount of
proximity effect. In order to evaluate a microphone, this
variation with distance is quite important.
When using a unidirectional microphone^2 in a hand-
held or stand-mounted configuration, it is conceivable
that the performer will not always remain exactly on
axis. Variations of ±45° often occur, and so a knowledge
of the uniformity of response over such a range is
important. The nature of these response variations is
shown in Fig. 16-15. Response curves such as these
give a better indication of this type of off-axis perfor-
mance than polar response curves. The polar response
curves are limited in that they are usually given for only
a few frequencies, therefore, the complete spectrum is
difficult to visualize.
For applications involving feedback control or noise
rejection, the polar response or particular off-axis
response curves, such as at 135° or 180°, are important.
These curves can often be misleading due to the
acoustic conditions and excitation signals used. Such
measurements are usually made under anechoic condi-
tions at various distances with sine-wave excitation.
Looking solely at a rear response curve as a function of
frequency is misleading since such a curve does not
indicate the polar characteristic at any particular

frequency, but only the level at one angle. Such curves

also tend to give the impression of a rapidly fluctuating

high-frequency discrimination. This sort of performance

is to be expected since it is virtually impossible to

design a microphone of practical size with a constant

angle of best discrimination at high frequencies, Fig.

16-16. The principal factor influencing this variation in

rear response is diffraction, which is caused by the

physical presence of the microphone in the sound field.

This diffraction effect is frequency dependent and tends

to disrupt the ideal performance of the unidirectional

phase-shift elements.

To properly represent this high-frequency off-axis

performance, a polar response curve is of value, but it,

too, can be confusing at high frequencies. The reason for

this confusion can be seen in Fig. 16-17, where two

polar response curves only 20 Hz apart are shown. The

question that arises then is how can such performance be

properly analyzed? A possible solution is to run polar

response curves with bands of random noise such as

octaves of pink noise. Random noise is useful because of

its averaging ability and because its amplitude distribu-

tion closely resembles program material.

Anechoic measurements are only meaningful as long

as no large objects are in close proximity to the micro-

phone. The presence of the human head in front of a

Figure 16-15. Variations in front response versus angular

position. Note: Curves have been displaced by 2.5 dB for

comparison purposes.

Figure 16-16. Typical fluctuations in high-frequency rear

response for a cardioid microphone. Courtesy Shure

Incorporated.

20 50 100 1000 10K 20K

Frequency–Hz

Relative response–dB

5dB

Relative response–dB

0

10

20

30

20 50 100 1000 10K20K

Frequency–Hz

0 o

o

(^1) » 3