Purdue University

Purdue e-Pubs
International Compressor Engineering Conference School of Mechanical Engineering

1990

Suction Muffler Design for Noise Reduction on

High Efficiency Reciprocating Compressors
G. Lampugnani
Aspera s.r.l.

F. Peruzzi
Aspera s.r.l.

P. Vay
Aspera s.r.l.

Follow this and additional works at: https://docs.lib.purdue.edu/icec

Lampugnani, G.; Peruzzi, F.; and Vay, P., "Suction Muffler Design for Noise Reduction on High Efficiency Reciprocating Compressors"
(1990). International Compressor Engineering Conference. Paper 750.
https://docs.lib.purdue.edu/icec/750

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for
additional information.
Complete proceedings may be acquired in print and on CD-ROM directly from the Ray W. Herrick Laboratories at https://engineering.purdue.edu/
Herrick/Events/orderlit.html
 SUCTION MUFFLER DESIGN FOR NOISE REDUCTION ON
HIGH EFFICIENCY RECIPROCATING COMPRESSORS

G.Lampugnani,F.Peruzzi,P.Vay
Engineering Dept.
ASPERA s.r.l. Turin ITALY

ABS'l'RACT
The development of high efficiency reciprocating com-
pressors for LBP domestic applications to E.E.R. level of 5
BTU/hW and more, requires the optimization of various
components involved in compression cycle, friction losses,
electric motor efficiency,etc ••
In this paper is considered particularly the suction muffler
design and related eXperimental techniques developed to
optimize the noise attenuation factor and pressure losses.
The need of keeping the suction pressure drops as minimum as
possible, forces in oversizing the suction ports and ducts.
By this fact, some more noise is generated by the suction
process.
Correct sizing of suction muffler is therefore required to
reduce the suction noise emission.
An eXperimental technique was developed to verify many
suction muffler design, without testing all the configura-
tions on the compressor, avoiding the test.repeatibility and
the unavoidable instability of the noise source, i.e. the
compressor itself.

REACTIVE MUFFLERS

A "silencer" could be defined as an apparatus inserted

in a flow of sound wave energy to reduce the wave amplitude
between two fixed section. Reactive type mufflers are
generally sized as a succession of chambers of different
volumes and shape connected by ducts. These types of muf-
flers operate by reflecting part of the energy of the wave
incident on the muffler.
The study of the suction mufflers of the high efficiency
reciprocating compressors, with their pipe resonators and
chambers separating baffles, was carried out with simplify-
ing assumptions of plane wave propagation in a duct with
eXpanding and contracting section (fig 1).

584
 2

The size and the section variations provide impedance

mismatches for the sound wave traveling through the muffler.
This fact results in a partial reflection ,of the sound
energy back toward the source;in this way only a part of the
sound energy is transmitted. The contracting section pro-
vides the same acoustic effect of the expanding section and
could be calculated by the same relationship:

1 82 (2*PI*f*L)
T.L.=lO*log 10 (1 + - * ( •8en 2- - - - - [dB]
4 81 82 c
where:
f= frequency
L= length of expansion chamber
c~ speed of sound of fluid
PI= 3.14
Sl= smaller section; S2= larger section

SIDE RESONATOR

Some types of mufflers are composed by Helmholtz

resonators connected to the side of suction passage. These
resonators are formed by a cavity of volume V connected to
the duct by a passage having length Land sectionS (fig 2).
Through this passage the sound waves reach the cavity, that
acts like a spring while the connecting passage acts like a
mass. In other words, the resonator is equivalent to a
spring-mass system with its resonating frequency given by:

c s
J * ( [ Hz )
2*PI (L*V)
where:
S= passage section
L- passage length
V= volume of the cavity
PI= 3.14
C= speed of sound of fluid
When at the resonator passage is present an acoustic wave
with frequency coincident with the resonator frequency an
high attenuation of acoustic wave energy occurs.

A computer simulation program was developed assuming plane

wave sound propagation. By defining the"geometry of the
muffler", the physical properties of the fluid, the fre-
quency range, the attenuation curve is obtained.

SBS
 3

INSERTION LOSS MEASUREMENT METHOD

Transmission Loss is one of the most widely used

parameters to describe the muffler characteristics, from a
theoretical point of view.
The Transmission Loss is defined as:
Wi
T.L.= 10 log 10 [----;;-J
where:
wi sound power incident on the muffler [W]
Wt = sound power transmitted through the muffler [W]
The attenuation characteristics of the mufflers, are gener-
ally shown as curves of transmission loss vs. frequency.
From an experimental point of view it is more usual to
define the attenuation characteristics of the mufflers as
"Insertion Loss" curves.
The Insertion Loss is defined as the difference, in sound
pressure level, measured at one point with and without the
muffler inserted between the noise source and the measuring
point:
Insertion Loss = L2-Ll [dB]
where:
Ll= sound pressure level with muffler
L2~ sound pressure level without muffler

The Insertion Loss measurement involves the· evaluation of

two quantities:
1) static insertion loss
2) self-generated noise level
The evaluation of self-generated noise is necessary when
fluid flows through the muffler, but to simplify the test
procedure we measured the static insertion loss measure-
ments, without gas flow.
The muffler under test was connected at the end of a duct,
acoustically insulated. The noise source was a loudspeaker
with a good frequency response in the 300- 3000 Hz. range.
This source was connected at the opposite end of the duet
and powered by a "white noise" generator.
All the test equipment was assembled in a semi-anechoic room
(fig.3). The average sound pressure level was measured by
five microphones with the noise generator working at the
fixed power.The same measurement was performed without the
muffler connected to the duct.
The sound pressure level was recorded by a third-octave
real time analyzer and the Insertion Loss was obtained for

586
 4

each third-octave band from 40 to 16000 Hz.

CHECK OF liEAS1JREMENT RELIABILITY

The assumed simplification to carry on the measures of

the Insertion Loss without gas flow, has been checked
measuring the sound pressure emitted by a compressor with
and without muffler. Both air and refrigerant R-12 were used
as working fluids.
In fig.4 is shown the attenuation due to the muffler mea-
sured on the compressor, in comparison with the one measured
on the test apparatus: the workinq fluid of the compressor
was air. In fig. 5 the working fluid of the compressor was
R-12.
It's interesting to observe that the difference in attenua-
tion between air and R-12 is corrected by a simple frequency
shift in the ratio of the speeds of sound of the fluids;
c .
F Rl2= F • ......!..!.!:
air

where:
F frequency
C ~ speed of sound,air or R 12

MOFFLER CONFIGURATIONS AND RELATED NOISE ATTENUATION VALUES

Some solutions have been developed with reference to

schematic drawings of fig.6
The insertion loss values are reported in the tested con-
fiqurations as follows:
Solution 1 Attenuation [dB(A)]
a) -with inlet pipe and baffle 20.5

"b) -with inlet pipe

-without baffle 11.3

c) -without inlet pipe

-with baffle 11.0

d)
-without inlet pipe and
baffle 6.4

Solution 2
It differs from solution 1 for the substitution of inlet
pipe with an elastic bellow inserted between the muffler and
shell suction tube.

587
The attenuation values are:
a) with baffle and bellow 18.0 dB(A)
b) with perforated baffle and bellow 13.5 "
c) with bellow and without baffle 11.3
d) without baffle and without bellow 3.5

COMMEHTS:

The attenuation effect of bellow in solution 2 a) is

mainly in the 1/3 octave band of 400-500 and. 630 Hz. The
bellow reduces the compressor cavity excitation in the range
of typical suction gas noise spectrum.
Similar effect is obtained by the inlet pipe (solution 1a).

Solutign :,
This solution differs from solution 2 only for splitting the
internal volume in two volumes v 1 and v 2 and connecting them
with two openings s 1 and ·s 2 to the gas flow duct.
The attenuation value is 18.3 dB(A).

PRESSURE LOSSES OF VARIOUS SOLUTIONS

To compare the gas pressure losses of different solu-
tions, the "steady" air flow, corresponding to a fixed
pressure drop of 0.981 kPa, was measured. The air flow and
related attenuation data are:
Solution nr. Ai3 flow Attenuation
(dm jsec) dB(A)
-ld) 1.2833*10:; 6.4
-2d) 1.1111*10_3 3.5
-2a) 0.9864*10_3 18.0
-lc) 0.9611*10_3 11.0
-ld) 0.7917*10_3 11.3
-3 ) 0.7444*10_3 18.3
-la) 0.7083*10 20.5

COMMENTS:

As expected, the noise attenuation is about inversely

proportional to the "steady flow" pressure loss of the
muffler. The choice among different solution is based on the
compromise between these two opposite ·effects and the E.E.R.
targets of different compressor series.

588
 6

TYPICAL REACTIVE MUFFLER

Fig.1

TYPICAL SIDE RESONATOR

~ ~
~

Ag. 2

589
 7

c:
..2
.i:::s
10
[i]
.5
en ;
u
::I 10
a;.... ..
0
:::s
0
u
m c
Q. 0
"'
~
u
Q.
ca

10
Ql
c:
0
~
a.
e
.2
e
.. ..
0'1
.s;
:::s
~ "
IV
CD
CD
IV
Ql ~
.!! a.
10
0 "CC
c: :::s
.! 0
...I
:c
3:

590
 INSERTION LOSS CURVES INSERTION LOSS CURVES
WORKING FLUID AIR COMPRESSOR WORKING FLUID:R 12
Solution 2 case C Solution 2 case C
30

I. L.muaured oot••t appuatus

.. I.L. measwed ooteat app11atua
LL m•••ured on compressor tLm•lsLJr..ci on comp:~tU:SDr

... ..
iii
!!. 10
' ··......,
iD
!!. 10
/
/ ·"' ~ ·- .-··"' ·- ·-......··,
~ I t5 ' I ··.......
"' ~ ~ \....... .
~ :> :1
z
s ~
IU

-··-··-··-··- ....... ...... ___ _ i

--- -· -
~
/.· ~

·10 ...

4G 6J 100 11i0 25.0 400 6311J. 1000 11600 2500 .. 000 UOO 1 OOOB 11i00il
... 11110 .fil
I L

100
J

l6G
j_ J -L

250
J L

400
l .L-l

i:W
I I 1 l_ L [_ L- L _J_l_J

1000 HiOO iii!500 4000 i:IOO 10000 16000

_L_ L j

THIRD OCTAVE BAND [Hz[ THIRD OCTAVE BAND [Hz[

Fig.4 Fig. 5

<XI
 9

SUCTION MUFFLER SOLUTIONS

OUTLET
SECTION
I SOLUTION

INTERNAL
BAFFLE

OUTLET
SECTION
I SOLUTION 2

INTERNAL RUBBER
BAFFLE BELLOY

INLET
SECTION

COMPRESSOR
SHELL
OUTLET
SECTION
I SOLUTION 3

INTERNAL RUBBER
BAFFLES BELLOY

INLET
SECTION-

COMPRESSOR
SHELL

592
 10

REFERENCE

Hamilton,J.F.
Measurement and control of compressor noise
Ray w. Herrick Laboratories, Purdue University (1988)
Handbook of noise control
Edited by Harris C.M.
McGraw Hill Book Company (3rd ed.)
Sharland I.
L'attenuazione del rumore
Woods Italiana (1980)
Peruzzi F., Bacci v., scandurra G.
E.E.R. Improvement on a reciprocating hermetic compressor
Proceedings of the 1980 Purdue conference

593