Investigation of the Discharge Flow of a Reciprocating Compressor Using PIV

Transcription

1 Purdue University Purdue e-pubs International Compressor Engineering Conference School of Mechanical Engineering 2000 Investigation of the Discharge Flow of a Reciprocating Compressor Using PIV H. J. Myung LG Electronics Inc. I. S. Lee LG Electronics Inc. Follow this and additional works at: Myung, H. J. and Lee, I. S., "Investigation of the Discharge Flow of a Reciprocating Compressor Using PIV" (2000). International Compressor Engineering Conference. Paper 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 Herrick/Events/orderlit.html

2 INVESTIGATION OF THE DISCHARGE FLOW OF A RECIPROCATING COMPRESSOR USING PIV Hwan Joo Myung and In Seop Lee Digital Appliance Research Laboratory, LG Electronics Inc, Korea ABSTRACT PIV(Particle Image Velocimetry) was used to visualize the discharge flow field in a model reciprocating compressor. As the first step to understand the flows in a reciprocating compressor, in which various complex flow phenomena exist, this experiment focused on the behavior of the circular jet flow and basic interaction phenomena of discharge flow with the valve. Fundamental phenomena related to the discharge port and valve were clearly shown in terms of velocity as well as vorticity field. Basic noise and efficiency characteristics which have to be considered in the design process can be estimated from these results. INTRODUCTION There have been various researches on reciprocating compressors for the purpose of increasing efficiency and decreasing noise 1-4. Related to the refrigerant flow, due to the difficulties of understanding the fundamental phenomena, the improvement of performance usually depends on limited knowledge. It is very difficult to understand the flow inside reciprocating compressor by using either experimental or numerical method because of the complex geometry, existence of moving parts, small size of each part, high pressure, high turbulence and unsteady pulsating characteristics. However, in order to maximize the performance of reciprocating compressor, it is necessary to understand highly the fundamental phenomena of the complex flows inside of it. In this study, as the first step to investigate the flows in a reciprocating compressor, PIV technique was used to measure the discharge flow 'of a simplified discharge model. When the compressed refrigerant exits the cylinder, interacting with the discharge port and valve, it becomes the sources of noise, vibration and inefficiency. Various parameters such as port shape, port area, valve shape, valve height are related to the discharge flow characteristics. The most fundamental phenomena of the discharge flow are the discharge jet flow and interaction of the discharge flow with the valve. The size of the discharge part of the commercial reciprocating compressor is very small and the measurement section is limited due to the valve or spring. So in this experiment, to understand discharge flows comprehensively, the discharge part of the reciprocating compressor was simplified and enlarged. PIV technique having outstanding feature of field measurement is appropriate to investigate the flow in the compressor which has highly unsteady flow characteristics. Following the leading researches 5-7, with the recent development of accuracy and development of electronic techniques, PIV is being used widely including the area of turbulence, high speed flow and design process of products 8 Applying PIV technique effectively to the development of commercial products is one of the purposes of this study. 391

3 EXPERIMENTAL EQUIPMENT Test model The test model of the discharge part of a reciprocating compressor and PIV system are shown in Fig. 1. The size of the discharge part is enlarged by three times that of a commercial compressor. The flow is steadily supplied by the compressed air system. Honeycomb and mesh are used to straighten the flow and to eliminate turbulence components. The detailed configuration of the test section, discharge port and valve system are shown in Fig. 2. Passing the nozzle, the flow enters the cylinder and then through the discharge port, it enters the discharge plenum and finally it exits through the exit hole. The three test cases are 1) the jet flow, 2) interaction of discharge flow with a valve for large valve height and 3) interaction of discharge flow with a valve for small valve height corresponding to the Fig. 2(b ), (c) and (d) respectively. PIV System PIV system consists of ND:YAG laser, CCD camera having 1Kx1K resolution, synchronizer, Pentium PC and software. Olive oil was used as particles, which were generated by an atomizer, being seeded into the upper part of the chamber and then mixed with the air. Velocity vector was obtained using two frame cross correlation technique and the interrogation window size was 32x32 pixels. The size of the interrogation window is directly related to the spatial resolution. The optimum size of the interrogation window is the case that it contains 10 ~ 14 particles when the particle size is about two pixels and that the length of the window is over the four times of the particle displacement. Laser Power Sr:11chronizer Fig. 1 Experimental equipment 392

4 Discharge Plenum 135 ~ =~ ~ Measurement Area,.,.. "./ " /':Valve Seat (b) Case 1 Valve Measurement...,.._...,. Area (c) Case 2 (a) Test section (d) Case 3 Fig. 2 Test section and measurement area RESULTS & DISCUSSIONS Fig.3 shows mean velocities and vorticities for each three cases. Mean velocity for jet flow shows distinctive feature of the near field jet flow. The mixing region increases as the flow moves downstream. The mean vorticity also shows the symmetric feature of the jet flow. Detailed jet flow characteristics according to the port shape are referred to the reference by present authors 9, from which the effective area of discharge flow as well as velocity characteristics affecting the valve can be comprehensively investigated. In the case of interaction of discharge flow with valve, the mean velocity exiting the discharge port shows different characteristics from those of jet flows. The velocity of the edge region above the discharge port is larger than that of core region. The core region of the discharge flow was suppressed due to the valve and the deflection region exiting the region between the valve and valve seat dominates the whole flow field. Due to the valve the separation region at the port edge region moves outward direction of the port. There exist distinct differences between case 2 and case 3. In case 2 with larger valve height, there is large separation of flow at the port edge region, while in case 3 separation is confined only in the small region. These characteristics will be shown more clearly from the instantaneous velocity figures. The mean vorticity in case 2 is larger than in case 2, which shows more severe separation at the edge region of both port and valve. It can be estimated that the port shape affects the vorticity characteristics severely thus affects the noise and vibration characteristics to a large extent. Instantaneous velocity and vorticity field are shown in Fig. 4. Vortex shedding occurs at the edge region and the vorticity becomes weak moving downstream. From the results of reference by present authors 9, this vortex flow depends comparably on the port shape. In case 2, larger 393

5 vortex exists than in case 3. It is identified that the vortex flow originating from the port and valve edge characterizes the discharge flow. In some cases of reciprocating compressors, this highly vortical flow interacts with the edge of the some objects thus can be a source of noise and vibration. From these results it is identified that flow at the edge of the port affects the entire discharge flow field, so it can be estimated easily that the port shape affect severely the noise characteristics. Valve height also affect severely the discharge flow field. The fact that the case of larger valve height has larger vorticity means that it can be possible that noise can be larger in the case of larger maximum valve displacement. This noise characteristic is confirmed from the noise measurement of the commercial reciprocating compressor. In order to improve the performance of reciprocating compressors, It IS needed to understand more comprehensively the flows in compressors. Study on the discharge part as well as the suction part of reciprocating compressors using PIV can be effective to the development of compressors. CONCLUSIONS The fundamental characteristics of the discharge flow were investigated using PIV measurement. The test model was enlarged and simplified in order to investigate the basic phenomena of the discharge flow. Vortical flow at the edge of the discharge port was clearly observed both in the jet flow and interaction flow of discharge flow with valve. In the case of larger valve height, vortex size and strength are larger than the case of smaller valve height, so it is identified that the valve height can affect the noise and efficiency characteristics of the compressor. For further development of compressors, PIV technique can be used effectively to investigate the unsteady flows in compressors. REFERENCES 1. Deschamps, C.J. and Ferreira, R.T.S., "Turbulent Flow Through Valves of Reciprocating Compressors", PlJ!due Univ., pp , Zhou, W. and Kim, J., "Optimal Sizing of the Discharge port Area of a Reciprocating Compressor Utilizing Computer Simulation Technique", pp.37-41, Purdue Univ, Cavallini, A., Del Col, D. and Doretti, L., "Unsteady-State Analysis of the Compression Cycle of a Hermetic Reciprocating Compressor", pp , Purdue Univ, Wambsganss, JR.M.W., "Mathematical Modelling and Design Evaluation of High-Speed Reciprocating Compressors", Ph.D Thesis, Purdue Univ., Adrian, R. J., "Statistical Properties of Particle Image Velocimetry Measurements in turbulent Flow, Laser Anemometry in Fluid Mechanics", pp , Adrian, R. J., "Particle-imaging Techniques for Experiments Fluid Mechanics, Ann. Rev. Fluid Mech.", Vol.23, pp , Westerweel, J., "Digital Particle Image Velocimetry-Theory and Application", Ph.D Thesis, Delft University of technology", The Netherlands, Lee, I.S., "A Study on the Design of Home Appliances for Energy Saving Using PIV", Ph.D. Thesis, Osaka Univ., Myung, H.J. and Lee, I.S., "Visualization of Jet Flows Using PIV", 9 1 h International Symposium on Flow Visualization, to be held in August,

6 -3m/s... "l''''tll11lllllff Iff fllllllllllllfflf111rrrrrrrl ~, "1 1 ttt!lllllllfl I If IIIII //// f / rr,,, 1"... uuhttl!llll111llllllllflfff///fflllllllllffl111lfltttll u...,1,, f f Iff Iff If rr, 1, ,,,, nnlllllll r rr r r rrr r r 11111," ~I I r I, It!llll r 111 r r rrrrrlltlllllllll',, h. F-~-----~ ''lllllffffffffffffffllllllllllllllr, 1 --~-~---~ ffllfll/11 Velocity (a) Discharge jet flow (b) Interaction of discharge flow with a valve for large valve height Vorloclly: (c) Interaction of discharge flow with a valve for small valve height Fig. 3 Mean velocity and vorticity 395

7 Fig. 4 Instantaneous velocity and vorticity 396 Vonicity m/s (a) Discharge jet flow -!Ornls (b) Interaction of discharge flow with a valve for large vaive height (c) Interaction of discharge flow with a valve for small valve height