Page 694 - Automotive Engineering Powertrain Chassis System and Vehicle Body

Interior noise: Assessment and control C HAPTER 21.1

Around TDC there is a rapid reversal in side force power. However, it is difficult to measure (see Section
produced by the slider-crank mechanism. This produces 21.1.3).
piston slap as the piston impacts on the cylinder-liner. An alternative scheme is to measure sound pressure
Piston slap is normally the dominant source of mechan- level at specified locations around the engine and use this
ical noise in the diesel engine (Lalor et al., 1980). There is for rating engine noise. The most commonly used stan-
side force throughout the cycle, along with other force dard method of this kind is detailed in SAE J1074. The
reversals but the one at TDC yields the highest rate of important information in J1074 is as follows:
change of side force. Piston slap noise increases with The engine is tested either outdoors in a flat, open
engine speed. It also increases with turbocharging. It is space or in an acoustically treated test cell that rep-
mostly controlled by reducing clearance between the licates the outdoor environment (commonly a semi-
piston and the cylinder-liner. anechoic cell with large sound-absorbing wedges on
In gasoline engines, piston/liner clearances are rela- the walls and ceiling and a flat concrete floor).
tively small, and mechanical noise tends to be dominated The engine is either tested in its bare state (with just
by impacts in the crankshaft bearings made through the enough equipment to run – pumps and manifolds are
oil film (Lalor et al., 1980). At low engine speeds these fitted but the intake/exhaust noise is ducted away) or
are magnified by increasing engine load. At high engine in its fully equipped state (everything fitted including
speeds, the inertia effects of the crank mechanism ancillaries and sometimes full intake and exhaust
dominate so there is little load dependency. systems).
Other sources of mechanical noise include:
The engine is tested at the maximum power point, at
timing drive; the maximum torque point, at the point of maxi-
valve train; mum speed but minimum load and also at idle.
fuel injection equipment. Sound pressure levels (slow response, both ‘A’- and
‘C’-weightings) are measured at three positions for
each engine operating condition. These are at 1.0 m
21.1.4.4 The effects of engine speed
from the longitudinal centres of the vertical planes
and load on noise forming the smallest rectangular box which com-
pletely encloses the bare engine. The measuring
The total noise emission (combustion and mechanical) points are on both sides and in front of the engine at
for the DI diesel changes only slightly over the normal the height of the exhaust manifold and at least 1 m
operating speed range. For the NA engine the slope is off the ground.
around 30 dB per decade and for the turbocharged engine The noise levels at the three specified locations are
around 20 dB per decade (Priede, 1975). There is reported. Octave band results are also reported for
modest load dependency for the NA-DI diesel engine the location with the highest ‘A’-weighted level.
(4–5 dB) and little for the turbo DI diesel unless excessive A survey is made of ‘A’-weighted sound pressure level
mechanical noise occurs due to the boost pressure. at the same height and distance from the box as the
Small high-speed NA-IDI diesel engines with smooth
combustion show little load dependence and a greater specified locations. If the survey reveals readings more
than 3 dB above the highest reading at the specified
speed dependence (around 40 dB per decade) than the locations, then the survey readings are also reported.
DI diesel engines.
Gasoline engines have two sets of noise characteris- The reported results should be the averaged results
1
tics. At low speeds (up to say 2500 rev min ) they have of two or more test results within 2 dB of each other.
modest load dependence (around 5 dB increase in noise If a semi-anechoic cell is used it must be large enough to
level due to increasing load) and slight speed dependence undertake the measurements and each microphone
(20 dB per decade). At higher speeds there is little load should be at least one-quarter wavelength from the walls
dependence and greater speed dependence (50 dB per and from the ceiling to avoid the near-acoustic fields of
decade) due to the effects of inertial forces on the me- these absorbing surfaces (see Section 21.1.3.1).
chanical noise. This explains the sudden onset of roaring
engine noise commonly experienced as gasoline engines
are revved hard. 21.1.4.6 Engine noise source ranking

Various noise source ranking techniques are discussed in
21.1.4.5 Measuring engine noise Section 22.1.2. All are in common use for engine noise
source ranking.
The most universal parameter for quantifying the noise At higher frequencies (<300 Hz say) the shielding
emission from any source (including the engine) is sound technique generally gives reliable results and is easy to

705