当前位置:首页 >> 能源/化工 >>

A numerical study with FSI mode on the characteristics of pressure fluctuation


INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 11, No. 4, pp. 589-596 DOI: 10.1007/s12541-010-0068-2

AUGUST 2010 / 589

A Numerical Study with FSI Mode on the Characteristics of Pressure Fluctuation and Discharge Valve Motion in Rotary Compressors with Single and Dual Muffler
Hee Moon Chae1 and Chang Nyung Kim2,#
1 Department of Mechanical Engineering, Graduate School of Kyunghee University, Yongin, Korea, 446-701 2 Department of Mechanical Engineering, College of Engineering, Kyunghee University, Yongin, Korea, 446-701 # Corresponding Author / E-mail: cnkim@khu.ac.kr, TEL: +82-31-201-2578, FAX: +82-31-202-9715 KEYWORDS: Rotary compressor, Single muffler, Dual muffler, Discharge valve, Pressure fluctuations, FSI

Rotary compressors in air-conditioners have been considered for the efficiency enhancement and noise reduction. To perceive the characteristics of noise in a rotary compressor, the features of discharge valve motion and pressure fluctuation in compressors have to be examined. This study has been conducted to investigate the characteristics of discharge valve motion and pressure fluctuation in association with refrigerant flow in rotary compressors with single and dual muffler. The current study has been performed with the FSI mode since the discharge valve oscillates in association with periodic compression of refrigerants in the compressors. For the case of dual muffler, it has been observed that the displacement of discharge valve is smaller than that in the case of single muffler since the compressor with dual muffler has larger inner resistance to the refrigerant flow than that in the case of single muffler. Also, the standard deviation and the energy spectrum of the pressure fluctuation with the dual muffler are smaller than those with the single muffler. Therefore, the use of the dual muffler is expected to contribute to the noise reduction. To the contrary, it has been found that the efficiency of the compressor with dual muffler is smaller with the diminution of volume flow rate compared to the case of single muffler. This study may supply a basis for the design of rotary compressors with higher efficiency and lower noise.
Manuscript received: December 23, 2009 / Accepted: June 3, 2010

1. Introduction
Generally, compressors used in an air-conditioner can be classified into the reciprocal, scroll and rotary types based on the method for the refrigerant to be compressed. In a reciprocating compressor, the refrigerant is inhaled, compressed and discharged in a cylinder by the reciprocating motion of a piston. Moderate and large-size air-conditioners adopt this type of compressors having an outstanding capability of compression. In a scroll compressor, the refrigerant is inhaled, compressed and discharged simultaneously by a revolving space of crescent shape formed by the stationary and revolutionary scrolls. In a rotary compressor, a rotor compresses the refrigerant in a cylinder and this type of compressors are frequently used in a small-sized household air-conditioner. Much works have been performed to increase the efficiency and to decrease the noise for a rotary compressor. ? KSPE and Springer 2010

Three main sources of the noise in a rotary compressor can be considered as follows:1-3 (1) the noise in association with pressure fluctuation caused by the refrigerant compression in a cylinder, (2) the resonant noise in conjunction with the resonance observed in an inner space of a rotary compressor, (3) the mechanical noise caused by the friction and contact between the moving parts of a rotary compressor. Therefore, these noises may be reduced when a good care is taken of the design of discharge port of a cylinder, shape of compressor cavities, mufflers and resonators. In order to reduce the noise in a rotary compressor, attention has been paid to the resonators and mufflers. Sano4 showed that the noise in a rotary compressor could be greatly reduced by installing a resonator near the discharge valve and that resonant noise in the main body of a compressor can be reduced with the use of well-

590 / AUGUST 2010

INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 11, No. 4

designed mufflers. Also, Ahn and Kim5 studied acoustic characteristics associated with the mufflers and resonators in consideration of the turbulent flow passing through the compressors. Jang et al.6 and Kim et al.7 investigated the noise characteristics depending on the positions of mufflers, the change of flow area inside a compressor, and the shape of discharge valves. Most of the previous studies to reduce the noise in a rotary compressor have focused on acoustic features. However, in order to clearly examine the noise in a rotary compressor studies on the refrigerant flow inside a compressor are needed and the FSI (FluidStructure Interaction) method is to be introduced to investigate the interaction of the refrigerant flow and the discharge valve motion. Generally, the magnitude of the pressure fluctuation of the refrigerant discharged from the cylinder of a rotary compressor is fairly large and the noise generated when the refrigerant passes the discharge valve increases with an increase in the magnitude of the above mentioned pressure fluctuation. Fairly large pressure fluctuation of the refrigerant is observed in the compression process in the cylinder. In order to reduce the pressure fluctuation of the refrigerant and, as a result of this, to reduce the noise, mufflers with one or more refrigerant passages can be installed downstream of the discharge valve in a rotary cylinder. In the present study, the flow field of the refrigerant and the behavior of the discharge valve are numerically analyzed with Fluid-Structure Interaction (FSI) method for rotary compressors of eccentric–rotor type with single muffler and dual mufflers. The unsteady pressures of the refrigerant at several key points in a rotary compressor are examined, which are basic data for analyzing the characteristics of noise in a rotary compressor. This study would support the design and development of a low-noise rotary pump.

Fig. 1 Structure of the rotary compressor

2.2 Governing equations
The refrigerant compressed in the cylinder periodically pushes the discharge valve and passes it, where the interaction of the refrigerant and the discharge valve can be observed. To analyze this interaction, FSI (Fluid-Structure Interaction) method has been used. The refrigerant flow has been numerically analyzed with the finite volume method. Since the fluid domain is changed with time periodically due to the motion of the discharged valve, the meshes for fluid domain are re-generated with time. The behavior of the discharge valve has been calculated with the finite element method. For the refrigerant flow the continuity, the momentum equation, the energy equation and the equation of state are considered and for the discharge valve motion the equation of structural dynamics is used. The governing equations adopted are given in the below. The continuity
?ρ + ? ? ( ρ v) = 0 ?t

(1)

2. Problem Formulation
2.1 Refrigerant flow in a rotary compressor
Usually, a rotary compressor includes an accumulator, a cylinder, a discharge valve, a stopper, a muffler as shown in Fig. 1. The accumulator is a device which allows only evaporated refrigerant gas to get into a rotary compressor, keeping liquid refrigerant from coming into a rotary compressor. In the cylinder the refrigerant from the accumulator is compressed by the interaction of the rotor and stator. The discharge valve partly contributes to the refrigerant compression implemented in the cylinder and partly allows the refrigerant to get out of the cylinder through periodically repeated openings when the refrigerant pressure in the cylinder is higher than that in the main body of a compressor. The stopper prevents the discharge valve from opening excessively and the muffler is used to reduce the noise by suppression of pressure fluctuation of the refrigerant coming out of the discharge valve. The refrigerant that passed the muffler is supposed to flow through the main body of the compressor and, after passing the discharge tube, to reach the condenser. In the current study, the compression of the refrigerant in the cylinder is repeated periodically sixty times per a second ( f = 60 Hz).

The momentum equation

? ( ρ v) + ? ? ( ρ vv ? τ ) = f B ?t
The energy equation ?( ρ E ) + ? ? ( ρ vE ? τ ? v + q ) = f B ? v + qB ?t 1 where E = v?v + e 2 and τ = (? p + λ? ? v) I + 2? ev The equation of state

(2)

(3)

ρ = ρ ( p, T )
The equation of structural dynamics

(4)

MU

t

+?t

+ CU

t

+?t

+ KU = R +? ? F
t t t

t

(5)

At the interface of the fluid and solid the normal velocity of the fluid and solid are the same and the surface forces exerted on the fluid and solid are in the opposite direction with the same magnitude, which can be written as follows;
n?d = n?d
f s

(6)

INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 11, No. 4

AUGUST 2010 / 591

τ = ?τ
f

s

(7)

From the above, in consideration of the no-slip condition at the interface the fluid velocity at the interface can be written as

compressor. The refrigerant used in the current rotary compressor is R-22 (CHClF2 ), whose properties at P = 2.0 MPa and T = 55°C is given in the Table 1. The properties of the discharge valve made of Sandvik 20C steel are given in the Table 2. Table 1 Properties of R-22 (CHClF2 , 2.0MPa, 55°C ) Variable Density Viscosity Thermal conductivity Internal energy Enthalpy Thermal expansion coefficient Table 2 Properties of Sandvik 20C steel Variable Young's modulus Poisson's ratio Thermal expansion coefficient Thermal conductivity Value 85.9 kg / m3 0.3 10.4 1/ K 49 W / m ? K Value 85.9 kg / m3

n ? v = n ? ds

(8)

With the use of Eqs. (1), (2), (3) and (4), the flow field is to be calculated. Based on the information obtained from the above calculations, Eq. (5) is integrated to obtain the displacement of the solid materials. This process is repeated until the solutions are converged at a given time step and is moved to the next time step, which allows one to solve unsteady flow field and solid motion.

1.522 × 10?5 kg / m ? s 1.474 × 10?2 W / m ? K 3.984 × 105 J / kg
4.216 × 105 J / kg 8.622 × 10?3 1/ K

2.3 Models for numerical analysis
A 48-frame rotary compressor (with single muffler and dual mufflers) is the model for the present study and the model geometries of the compressor including the whole shape, discharge valve and stopper, single muffler and dual mufflers are shown in Fig. 2.

2.4 Boundary conditions and initial conditions
The measured transient pressure of refrigerant at the exit of the cylinder is given as the inlet boundary condition, which is shown in Fig. 4. In the pressure variation as a function of the rotation angle of the rotor, two peaks are observed in a cycle. The pressure values of Pmin , Ppeak1 and Ppeak 2 in the figure are specified in the Table 3. The difference in the pressures of the maximum and the minimum is about 1.85 MPa. A numerical model with a pipe connected to the exit of the compressor is considered to take into account the pressure variation

(a) whole geometry

(b) discharge valve and stopper

Bottom-muffler

Top-muffler

(c) single muffler

(d) dual muffler

Fig. 2 Model geometry of the rotary compressor

(a) flow for the single muffler

(b) flow for the dual mufflers

Fig. 3 Flow pattern in the two models In the case of single muffler, a round plate with two refrigerant passages is installed in the compressor, while in the case of dual mufflers the bottom plate with one refrigerant passage and the top plate with two refrigerant passages are set in the compressor. Schematic flow patterns of the two cases are given in Fig. 3. Here, the compressed refrigerant in the cylinder pushes and passes through the discharge valve periodically, passes through the refrigerant passages of the muffler, flows through the region between the compressor casing and the inner block of a compressor (including stator, coil and shaft), and gets to the outlet of the

Fig. 4 Input pressure variation in a cycle (Period=0.016667s, 60Hz) Table 3 Pressure values at several points of the input pressure wave Position P_min P_peak1 P_peak2 Value 0.5984 MPa 2.3446 MPa 2.4525 MPa

592 / AUGUST 2010

INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 11, No. 4

with time and position in the compressor(refer to Fig. 3). The length and inner diameter of the pipe are 0.5m and 0.16m, respectively, and the pressure given at the end of the pipe in the calculation is 2.1MPa. The original and computational models for the compressor are indicated in Fig. 5.

3. Results and Discussion
3.1 The behavior and the natural frequency of the discharge valve
In order to obtain reliable numerical results of this problem, it is crucial to numerically obtain the natural frequency of the discharge valve accurately so that numerically obtained natural frequency of the discharge valve can reflect the real natural frequency of the discharge valve. Therefore, care has been taken of the mesh shape and the node number in the simulation of the motion of the discharge valve. The obtained natural frequency with 10 nodes for each cell in the current numerical calculation yields 191.5 Hz, similar to the real natural frequency, 200 Hz, of the discharge valve. The behavior of the discharge valve with the rotation angle of the rotor has been simulated. In Fig. 6, the displacements of the discharged valve for the cases of single muffler and dual mufflers have been predicted under the condition of the given input pressure given in Fig. 4. The discharge valve begins to open when the rotor angle is around 175°. Two peaks of the displacement of the discharge valve are observed and the displacements are larger at the peak 2 than at the peak 1 for both the two cases.

(a) original model

(b) numerical model

Fig. 5 Comparison between the original and numerical model Initial pressure and velocity of the refrigerant in the compressor are 2.1MPa and 0m/s, respectively. In the use of the FSI (FluidStructure Interaction), the surfaces of the discharge valve (except for the surface in contact with the floor of the compressor) are regarded to be FSI interfaces.

2.5 Numerical method
In the current calculation, unstructured tetrahedral grids for the refrigerant domain and structured hexahedral grids for the discharge valve are used. To examine the grid size dependency, four different grid systems with 91,200 grids, 102,800 grids, 110,300 grids and 124,700 grids are examined. The numerical results with the 91,200 grids and 102,800 grids are slight different from that with the 110,300 grids, but the results with the 110,300 grids and 124,700 grids are of little difference. Therefore, the numerical analysis in the current study is performed with the grid system of 110,300 grids. To investigate the time step size dependency, three different time step sizes of 1.042 × 10?4 s (160 time steps in a periodic time of 0.016667s), 8.333 × 10?5 s (200 time steps) and 6.944 × 10?5 s (240 time steps) are considered. Basically the numerical results with the second and third time step size are of negligible difference. Therefore, the time step size of 8.333 × 10?5 s has been chosen in the current analysis. However, because of a considerable pressure variation in the cylinder for the range of rotation angle of 342° ~ 360°, the time step size of 4.1666 × 10?5 s is used. In the current calculation, a commercial software package ADINA 8.4 has been used, where the SIMPLEC algorism has been employed. For the simulation of a cycle it takes around 72 hours with Pentium-4 (Dual) 2.8 GHz personal computer and the current calculation has been done for 10 cycles. Fig. 6 Motion of the discharge valve with time (in terms of rotation angle) The displacement for the case of dual mufflers is smaller than that for the single muffler case since the flow resistance in the compressor in the case of dual mufflers is larger compared with the case of single muffler. The rotational angles of the rotor and the displacements of the discharge valve at the two peaks in the case of single muffler and dual mufflers are given in Table 4, where it is seen that the maximum displacement in the case of single muffler is larger than that in the case of dual mufflers around by 0.42 mm. Table 4 Displacement of the discharge valve at peak points Rotation angle Displacement Peak1 2.00 mm 257° Single Muffler Peak2 3.16 mm 338° Peak1 1.71 mm 256° Dual Muffler Peak2 2.74 mm 338°

3.2 Pressure waves
The pressure distributions in the two cases of the compressor have been obtained against the rotation angle of the cylinder. The pressure monitoring points in the two cases have been depicted in

INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 11, No. 4

AUGUST 2010 / 593

Fig. 7. In the case of the single muffler, the pressures at the flow passage of the muffler and at the outlet of the compressor are shown in Fig. 8 in comparison with the input pressure observed just below the discharge valve. Also, in the case of the dual mufflers, the pressures at the flow passages of the top and bottom muffler and at the outlet of the compressor are denoted in Fig. 9. These figures show the pressure variations around for 2.4 cycles and, here, it is seen that the pressure behavior for the second period is almost the same as that for the first period. Therefore, no calculation has been carried out for the third and fourth period. As for the second period, the discharge valve is closed around for the rotation angle ranging from 360° to 540° since the pressure inside the compressor is larger than the input pressure (that is, the pressure just below the discharge valve). Around when θ = 540° the discharge valves start to open and the pressure in the compressor have two peaks in association with the input pressure depending on the rotation of the cylinder (Fig. 8(a) and Fig. 9(a)). When θ > 330°, the pressures in

the compressor decrease drastically in association with the steep decrease in the input pressure. The pressures at the previously mentioned positions (Fig. 7) show that pressure decreases along the flow direction of the refrigerant (Fig. 8(b) and Fig. 9(b)).

(a) pressure waves at the monitoring points

(a) with single muffler

(b) with dual muffler

Fig. 7 Positions of the monitored pressure in the two models (b) pressure variations in detail of the part B Fig. 9 Pressure waves in the case of dual muffler Table 5 Root mean squares of the pressures at different positions Position RMS Input pressure 1.679 MPa Single Muffler Muffler passage 2.043 MPa Compressor outlet 2.035 MPa Input pressure 1.679 MPa Bottom muffler passage 2.046 MPa Dual Muffler Top muffler passage 2.039 MPa Compressor outlet 2.028 MPa (a) pressure waves at the monitoring points The root mean square (RMS) values of the pressure for a period at several specified points in the two models are given in Table 5. The RMS value of the pressure at the refrigerant passage of the top muffler in the case of dual mufflers is smaller than that at the refrigerant passage of the muffler in the case of single muffler around by 4,000 Pa and the RMS value at the compressor outlet in the dual mufflers case is also smaller than that of the single muffler case around by 7,000 Pa. The pressure variations at the refrigerant passage of the muffler in the single muffler case and at the refrigerant passage of the top muffler of the dual mufflers case are given in Fig. 10 and it is shown that the pressure at the refrigerant passage of the top muffler of the dual mufflers case is generally smaller than that at the refrigerant passage of the muffler of the single muffler case. The

(b) pressure variations in detail of the part A Fig. 8 Pressure waves in the case of single muffler

594 / AUGUST 2010

INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 11, No. 4

Fig. 10 Comparison of the pressure waves at muffler passages for the two models pressure variations at the compressor outlet for the two models are shown in Fig. 11 and the pressure at the compressor outlet of the dual mufflers case is smaller than that of the single muffler case, where the pressure difference observed in this figure is smaller than

Fig. 11 Comparison of the pressure waves at the compressor outlet for the two models that seen in Fig. 10. In the current numerical simulation, the boundary condition for pressure at the end of pipe attached to the compressor is 2.1MPa. Therefore, the pressure of the refrigerant observed at the outlet of the compressor can be influenced more or

Table 6 Comparison of the fast Fourier transforms of the pressure in the two models Frequency 60Hz 120Hz 180Hz 240Hz Position (a)Input pressure 703,230 122,287 207,158 60,687 246,701 218,284 234,188 163,099 Single (b) Muffler flow passage muffler (c) Compressor outlet 169,997 146,665 157,745 102,130 (b1) Muffler flow passage 227,013 202,509 219,372 157,174 (Bottom muffler) Dual (b2) Muffler flow passage 216,726 191,497 208,260 145,875 muffler (Top muffler) (c) Compressor outlet 146,737 125,892 138,216 89,159 Frequency Position (a) Input pressure Single (b) Muffler flow passage muffler (c) Compressor outlet (b1) Muffler flow passage (Bottom muffler) Dual (b2) Muffler flow passage muffler (Top muffler) (c) Compressor outlet Frequency Position (a) Input pressure (b) Muffler flow passage (c) Compressor outlet (b1) Muffler flow passage (Bottom muffler) (b2) Muffler flow passage (Top muffler) (c) Compressor outlet Frequency Position (a) Input pressure (b) Muffler flow passage (c) Compressor outlet (b1) Muffler flow passage (Bottom muffler) (b2) Muffler flow passage (Top muffler) (c) Compressor outlet 540Hz 13,491 28,729 14,060 25,634 23,054 16,269 1020Hz 1,428 15,802 11,382 16,027 14,628 11,504 1500Hz 2,271 15,031 9,652 13,238 11,735 9,788 600Hz 14,316 29,613 22,755 25,316 24,155 29,288 1080Hz 1,212 18,969 14,391 18,215 17,127 14,059 1560Hz 2,524 14,853 10,284 14,054 12,968 10,206 660Hz 10,250 35,043 26,030 33,173 31,344 23,127 1140Hz 1,190 15,240 11,441 15,404 14,142 11,386 1620Hz 1,481 14,399 9,604 14,410 13,196 10,422 720Hz 7,052 28,676 19,845 25,657 25,886 18,930 1200Hz 2,218 19,973 13,984 17,674 16,999 13,858 1680Hz 2,705 12,660 8,272 13,532 12,484 10,084

300Hz 56,859 95,272 56,036 93,106 85,496 48,206 780Hz 8,270 24,480 17,505 25,258 23,494 17,109 1260Hz 1,531 14,774 11,020 15,071 14,420 11,663 1740Hz 1,900 11,300 6,659 12,621 11,634 9,320

360Hz 26,365 55,780 28,377 57,435 51,410 24,809 840Hz 2,003 19,736 14,439 19,264 18,243 13,434 1320Hz 1,789 14,003 10,413 14,614 14,360 11,111 1800Hz 3,880 8,597 4,231 10,187 9,464 7,689

420Hz 20,385 42,547 22,128 42,129 37,017 18,095 900Hz 3,742 16,611 11,702 17,773 16,512 11,984 1380Hz 2,812 12,538 8,209 12,199 11,023 8,997 1860Hz 2,328 8,513 3,084 7,907 7,056 5,280

480Hz 16,622 18,497 5,006 13,946 15,441 2,616 960Hz 1,198 19,311 14,369 18,505 16,855 13,500 1440Hz 2,535 12,759 8,639 11,449 10,616 8,471 1920Hz 3.065 9,763 3,832 7,611 6,520 4,278

Single muffler Dual muffler

Single muffler Dual muffler

INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 11, No. 4

AUGUST 2010 / 595

less by the fixed given pressure at the end of the pipe. This means that the pressure difference observed at the two refrigerant passages (Fig. 10) and at the compressor outlet (Fig. 11) of the two different cases may be underpredicted compared with a real situation. In Table 6, energy spectra of the pressure variation with time at several points in the two cases have been obtained in terms of the frequency with the use of Fast Fourier Transformation. As expected, the energy spectrum for the input pressure of the refrigerant has a peak at 60 Hz in association with the rotor’s RPM (refer to Fig. 4). However, the energy spectra values of the refrigerant pressure at the refrigerant passage of the muffler(s) at f = 60 Hz are decreased as the refrigerant passes the discharge valve and muffler(s) and the magnitude of these values are comparable with those at f = 120 Hz, 180 Hz and 240 Hz. It can be noted that for higher frequencies (say, for f = 1,080 Hz ) the energy spectra values of the refrigerant pressure at the refrigerant passage of muffler(s) and the compressor outlet are quite larger than those at the compressor inlet. However, it is seen that as the refrigerant flows to the compressor outlet the energy spectra values are decreasing for a given frequency (say, for f = 1,080 Hz ). In general, the energy spectra values of the compressor in the dual mufflers case are lower than those in the single muffler case, though the trend is not obviously apparent. In the dual mufflers case, the flow resistance is larger so that larger pressure drop is expected in the region of the top and bottom muffler, compared with the single muffler case. It is expected that this larger pressure drop in the dual mufflers case is related with lower energy spectra values of the compressor in the dual mufflers case.

measurement and the numerical analysis are given in Table 8. The predicted flow rate for the case of single muffler is larger than the measured flow rate for the case of single muffler by 11.5 % and the predicted flow rate for the dual mufflers case is smaller than the measured flow rate for the single muffler case by 11.2 %. The smaller flow rate in the dual mufflers case is attributed to the fact that the dual mufflers case has more flow resistance in the compressor compared with the single muffler case.

4. Conclusions
In the present study, the characteristics of the discharge valve behavior and the refrigerant flow field in the 48-frame rotary compressors with single muffler and dual mufflers have been analyzed with the Fluid-Structure Interaction method. In association with the two peaks of the inlet pressure with time, the two peaks in the displacement of the discharge valve are observed. The maximum value of the discharge valve displacement falls in the second peak, and is smaller in the case of the dual mufflers compared with the case of the single muffler. The RMS (Root Mean Square) value of the pressure at the compressor outlet and the refrigerant flow rate in the case of dual mufflers is smaller compared with the case of single muffler. Energy spectra of pressure with frequency have been obtained. The energy spectra at the compressor inlet have a sharp peak at f = 60 Hz, but the energy spectra at the discharge valve and at the refrigerant passage of the muffler for f = 60 Hz are attenuated, while the energy spectra for higher frequencies at the above positions are more or less increased. However, in general the values of the energy spectra are decreased as the refrigerant flows to the downstream. Also, the values of the energy spectra in the dual mufflers case are generally smaller compared with the single muffler case, which, together with the smaller standard deviation of the pressure at the compressor outlet, may lead to a smaller noise level in the dual mufflers case. However, the smaller volume flow rate of the refrigerant in the dual mufflers case can lead to a lower efficiency of the compressor, yielding lower capability of air conditioning. The current study supports fundamental understanding on the effect of several key parameters influencing the noise and the performance of a rotary compressor.

3.3 Pressure variation and mass flow rate of the refrigerant
It can be reasoned that the pressure variation with time is closely related with the noise characteristics of rotary compressors. Based on the time series of the pressure at the compressor exit in the two cases, the standard deviations of the pressure variation have been obtained based on the following equation and the values are given in the Table 7. Table 7 Pressure fluctuations Standard deviation ( σ ) Compressor outlet with single muffler Compressor outlet with dual muffler Table 8 Volume flow rates Volume flow rates Experiment with single muffler ADINA-FSI with single muffler ADINA-FSI with dual muffler

2.86 × 105 Pa 2.52 × 105 Pa

REFERENCES
30.66 × 10?6 m3 / rev 34.19 × 10?6 m3 / rev 27.22 × 10?6 m3 / rev
1. Sano, K., “Analysis of hermetic rolling piston type compressor noise and counter measurements,” Proc. of 1984 International Compressor Conference at Purdue, pp. 242-250, 1984. 2. Kawaguchi, S., “Noise reduction of rolling piston type rotary compressor,” Proc. of 1986 International Compressor Conference at Purdue, pp. 550-565, 1986. 3. Shige, N., “Prediction & visualization of a three dimensional sound field to reduce the noise of rotary compressor,” Proc. of 1986 International Compressor Conference at Purdue, pp. 591-

The standard deviation of the pressure at the outlet of the compressor in the case of dual mufflers is smaller than that in the case of single muffler around by 0.34 × 105 Pa, which may mean that the noise in the case of dual mufflers is smaller. The volume flow rate of the refrigerants obtained from the

596 / AUGUST 2010

INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 11, No. 4

601, 1986. 4. Sano, K. and Noguchi, M., “Cavity resonance and noise reduction in a rotary compressor,” IEEE Transaction on Industry Applications, Vol. IA-19, No. 6, pp. 1118-1123, 1983. 5. Ahn, B. H. and Kim, Y. S., “A study on noise reduction of rotary compressor,” Journal of the Korea Society for Power System Engineering, Vol. 3, No. 3, pp. 60-69, 1999. 6. Jarng, I. S., Kim, B. J., Youn, Y., Sung, C. M. and Lee, S. K., “Development of a low noise and high efficiency rotary compressor with a new muffler,” Journal of Fluid Machinery, Vol. 8, No. 2, pp. 23-30, 2005. 7. Kim, B. J., Youn, Y., Jung, C. H. and Lee, S. G., “Noise reduction of a rotary compressor by a new muffler,” Proc. of the Korean Society for Noise and Vibration Engineering Autumn Conference, pp. 141-145, 2003.


相关文章:
分流叶片_图文.pdf
structuralcharacteristics,itwasshownthattheresultsofpressurefluctuationandtheradical...fatigue.Keywords:splitterblade;moltensaltpump;FSI;unsteadyflow;structuraldynamic...