The cross-section area of the exhaust muffler must
exhaust temperature, the speed of sound would be
be large enough not to generate excessive back
about 1870 ft./see (from equation 21 in the N&V
pressure and muffler self-noise.
manual), which is about 1.7 times the speed of
sound in air at normal temperature, assuming the
(3) Table 3-3 offers a reasonable design for the
exhaust gases are made up largely of the normal
intake muffler: 4-in. -thick parallel baffles sepa-
contents of air. This means that the exhaust muf-
rated by 12-in. -wide air spaces. An 8-ft. length of
fler should be about 1.7 times longer than it would
such design will meet the desired insertion loss
have to be at normal temperature to produce the
values in all bands. This length will help the intake
same insertion loss.
stack qualify as a class 1 lined band (a 4-ft.-length
(2) Table 3-6 offers a reasonable design for the
muffler would not be long enough; fig. 3l); and
exhaust muffler: 8-in. -thick parallel baffles sepa-
the relatively large percent of open area will mini-
rated by 8-in. -wide air spaces. The 8-ft. length ex-
mize inlet pressure drop.
ceeds the insertion loss requirement in all the oc-
(4) Table 4-4 summarizes the sound power lev-
tave bands, but by only 1 dB in the 125-Hz band. A
els of the three engine components with these muf-
7-ft. length (at normal temperature) would very
flers installed. Comparison of the inlet and exhaust
nearly meet the 10-dB requirement at 125 Hz. For
PWLs of tables 43 and 44 (co1. 3 and 4) shows the
the elevated temperature, the length should be in-
amount of insertion loss assumed for the mufflers.
creased to about 12 ft.: (7 x 1.7 approximately).
were listed, wheras only the inlet and exhaust muf-
A l-dB excess of noise still appears in the 125-Hz
flers have been evaluated here. In a total study,
band, but the total design appears well balanced
the SPL inside the Engine Room should be esti-
over the 63- through 2000-Hz bands.
mated (Room Constant and engine casing PWL are
(5) The insertion loss values used in this study
required), and the PWL radiated by the external
and given in the chapter 3 tables are intended for
shell of the housing should be calculated (as in
information and guidance only. As stated in para-
para. 32). In the muffler analysis above, the noise
graph 34a, muffler manufacturers should be
reduction of the housing was merely estimated
cons.ulted on the design and performance of their
from its similarity with the type 5 enclosure of ta-
ble 2-7. The noise of the gear and generator in the
e. Other aspects of this sample problem.
Generator Room should also be estimated (from
above, several parts of the total noise problem
chap. 7 tables in the N&V manual), and the noise