Purdue University
Purdue e-Pubs
International Compressor Engineering Conference School of Mechanical Engineering
2004
Noise and Vibration Reduction in Compressors for
Commercial Applications
Marcio Silveira
Embraco
Follow this and additional works at: http://docs.lib.purdue.edu/icec
Silveira, Marcio, "Noise and Vibration Reduction in Compressors for Commercial Applications" (2004). International Compressor
Engineering Conference. Paper 1662.
http://docs.lib.purdue.edu/icec/1662
This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact epubs@purdue.edu 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
C065, Page 1
NOISE AND VIBRATION REDUCTION IN COMPRESSORS FOR
COMMERCIAL APPLICATIONS
Márcio Silveira
Embraco – Empresa Brasileira de Compressores SA
Joinville, Santa Catarina, Brazil
Phone 0055 47 4412082, Fax 0055 47 4412730, Marcio_Silveira@embraco.com.br
ABSTRACT
One of the main sources of noise in commercial refrigeration systems is the compressor, whether due to the level of
noise directly radiated, or due to the interaction with the system. This interaction with the system also occurs as a
result of gas pulsation and vibration levels that excite the components connecting the compressor to the system.
This work presents studies carried out on reciprocating commercial compressors in the attempt to identify, quantify
and to determine the generating sources of the noise, the ways of transmission and the influence of the final
irradiator in the total noise and the vibration of the compressor through sensibility analysis of the product and
process. The results show how important it is for the design engineer to choose the right components when defining
new components for the development of quieter products or to modify existing products for the purpose of making
them quieter.
1. INTRODUCTION
The sound power level of a refrigeration system is basically formed by sound sources, energy propagation paths and
irradiation of several parts and components, with these excitation, transmission, and radiation features playing an
important role in refrigerator noise reduction.
The J compressor family is basically used in commercial application as cold store, air conditioning, refrigerating
chamber, etc. As it is a very old design, this compressor presents a sound power and vibration level above the limits
defined for the final customer. The need to reduce noise limits for this kind of application has become more and
more urgent over recent years.
Considering the noise limits, it is possible to verify that the J compressor has high total noise and the predominance
of noise peaks at low and high frequency which can be observed through subjective analysis.
The noise problem was supported based on the Six Sigma approach.
The job was divided into three steps to accurately characterize all noise problem solving phases:
• Identify predominant source through an analysis of main sources of compressor noise generation;
• Characterization of the main paths of vibratory energy transmission;
• Irradiation energy reduction through the redesign of the shell, this being the cause of irradiation.
The analyses were conducted considering the total noise and 1/3 octave band to enable sound quality improvements
impact strongly on customer perception.
The main objective of this work was to reduce 5.0 dB the total noise level of J compressor on average. Reduction of
vibration and noise levels in low frequency is also considered very important for improving the sound quality of the
refrigeration system for the final customers.
In order to reach the noise reduction goal an analysis was conducted consisting primarily of an investigation into the
main paths of vibratory energy flow in the compressor and the identification of the main points requiring
modification.
A detailed study of the main points responsible for the noise enabled a guided approach, which only concentrated on
modifications to the parts affecting the final result.
This approach starts with a road map defining each step in the noise analysis and the understanding of the sound
power level in the system.
International Compressor Engineering Conference at Purdue, July 12-15, 2004
C065, Page 2
2. ROAD MAP
What is the noise generator in a
commercial compressor?
Sources Transmission paths Irradiator
Compression Mechanism Motor Shell
System
Discharge Suction Suspension
What are the more important
components to noise?
Vibration Level Sound Power Level Pulsation Level
Brainstorming
Sensibility analysis Failure Tree
Product Map Noise Characterization Process Analysis
Simulation
Experimental DOE
FMEA Optimization RSD
Results
Conclusion
Figure 1: Road Map simplified
International Compressor Engineering Conference at Purdue, July 12-15, 2004
C065, Page 3
3. NOISE CHARACTERIZATION
The noise diagram in a commercial compressor can be divided as shown in figure 2 below. It is possible to observe
that the vibratory energy transmission in the final process will be perceptible by the customer as high noise level.
COMPRESSION DISCHARGE DISCHARGE
SYSTEM LINE
SUCTION KIT MOTOR
CAVITY SHELL SUSPENSION
COMPRESSOR
NOISE LEVEL
Structural Connections Air-borne Irradiation
REFRIGERATION SYSTEM SOUND POWER LEVEL
Figure 2: Energy vibratory flow of compressor
Sources
Paths
Irradiator
4. NOISE CONTRIBUTION
The sensibility analysis enables the contribution of each system to the frequency band to be identified and,
consequently, the total noise of compressor. Figure 3 shows the contribution in the frequency spectrum of the 5 most
important components in the compressor.
International Compressor Engineering Conference at Purdue, July 12-15, 2004
C065, Page 4
Noise Spectrum Distribution
100%
90%
80%
SWL [dB] 70% Mechanism
60% Discharge
50% Shell
40% Suspension
30%
Suction
20%
10%
0%
100 200 400 800 1600 3150 6300 Total
Frequency [Hz]
Figure 3: Noise contribution of the components in the compressor
5. DEVELOPMENT
With the contributions defined, the new J compressor project was defined with the follow options:
A - New Shell B - New Muffler
Figure 4: Shell Modal analysis Simulation Figure 5: Muffler Modal analysis Simulation
C - New Suspension D- Discharge Modified
[dB/1.00u m/s²] 1/3 Octave Synthesis(Ac fundo)
Working : normal com pressao acc tubo com amorteciment0 : Input : FFT Analyzer
150
145
140
135
130
125
120
115
110
105
100
200 500 1k 2k 5k 10k
[Hz]
Figure 6: Spring Response Simulation Figure 7: Discharge Damping
International Compressor Engineering Conference at Purdue, July 12-15, 2004
C065, Page 5
This proposed J compressor received a new name - NJ
In figures 8 and 9 it is possible observe all the proposed modifications and to comp are them with the old one.
Shell
Discharge
Suction
Muffler
Suspension
Figure 8: NJ compressor
Figure 9: Current J compressor
International Compressor Engineering Conference at Purdue, July 12-15, 2004
C065, Page 6
6. RESULTS
6.1 Compressor Noise
The noise results of some models are shown in the table below. The frequency spectrum is shown in figure 10 for
model NJ2192GK.
Table 1: NJ vs. J noise comparison
Refrigerant Sound Power Level [dBA]
NJ Current J ∆ Reduction
R404 65.0 71.0 6.0
R22 61.5 67.0 5.5
R134a 59.0 65.0 6.0
Sound Power Level
J2192GK
NJ2192GK
SWL [dBA]
100 160 250 400 630 1000 1600 2500 4000 6300 10000
Frequency [Hz]
Figure 10: Noise Comparison NJ vs. Current J compressor
6.2 System Noise
Figure 11 presents the noise impact in a refrigeration system when the NJ compressor is applied in the refrigeration
system.
Noise Refrigeration System with NJ x J
sys. J2192GK
Subjective analysis
sys. NJ2192GK
Compressor noise
Fan noise
SWL[dBA]
100 160 250 400 630 1000 1600 2500 4000 6300 10000
Frequency [Hz]
Figure 11: Noise Comparison in the Refrigeration System
International Compressor Engineering Conference at Purdue, July 12-15, 2004
C065, Page 7
6.3 Compressor Vibration
The vibration levels present a reduction of more than 30% as shown in the table below:
Table 2: NJ vs. J vibration comparison
Model Vibration Level
NJ Current J ∆ Reduction
R404 3.20 6.73 6.0
R22 3.40 5.45 5.5
R134a 3.90 6.00 6.0
7
6
5
4
J
3 NJ
2
1
0
R404 R22 R134a
Figure 12: Vibration Comp arison NJ vs. Current J compressor
6.4 Jury Analysis
The jury analysis system comparing the same system with current J and NJ is shown in figure 13:
Jury analysis - Sound quality Jury analysis - Subjective loudness
8
8
7 7
6 6
5 5
4
4
3
3
2
2
1
1
0
system 1 more system 2 more the same 0
enjoyable enjoyable more quiet system 1 more quiet system 2 the same
Figure 13: Jury Analysis Comparison NJ x Current J in a system
International Compressor Engineering Conference at Purdue, July 12-15, 2004
C065, Page 8
7. CONCLUSIONS
With the proposed modification a significant total noise reduction was gained in the NJ compressor as well as a
reduction in the frequency bands, improving the sound quality, which could be perceived by the final customer. The
vibration reduction was considerably higher than 30% helping reduce the sound impact in the refrigeration system.
The main object was reached considering the total noise reduction in the compressor and the gain to the customer
when using the compressor in the refrigeration system.
REFERENCES
Silveira, M., 2003, Compressor J Total Noise Reduction , Engineering Report., Embraco.
Lenzi, A., 2002, Acoustics and Vibration , Mechanical Engineering Department, UFSC,
Werkema, M. C., 2000, 6 Sigma: Black Belt Training, FDG
International Compressor Engineering Conference at Purdue, July 12-15, 2004
