Design and Development of a Variable Rotary Compressor

Purdue University Purdue e-pubs International Compressor Engineering Conference School of Mechanical Engineering 2000 Design and Development of a Variable Rotary Compressor S. Harte Y. Huang L. Sud Follow this and additional works at: http://docs.lib.purdue.edu/icec Harte, S.; Huang, Y.; and Sud, L., "Design and Development of a Variable Rotary Compressor" (2000). International Compressor Engineering Conference. Paper 1415. http://docs.lib.purdue.edu/icec/1415 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

Design and Development of a Variable Rotary Compressor Shane Harte, Yong Huang, Lavlesh Sud Climate Control Division Plymouth, MI 48170 ABSTRACT The following paper describes the design and development issues with a variable capacity control 2-stage rolling piston compressor. The control mechanism was based on a gas-bypass control strategy. During the development, design issues surfaced that were not significant in the case of the full capacity version of the compressor. When the capacity was reduced, loss generation mechanisms remained constant. These resulted in higher discharge temperatures and lower compressor efficiencies. Improvements were made to reduce losses in line with capacity reductions. 1.0 INTRODUCTION The development of a rotary variable capacity compressor is discussed in this paper. An initial concept was developed and tested. It was found at the time, that there was significant room for improvement. This paper discusses the modifications that were made to improve the functional performance of the compressor. Stage 1 vane with kinematically ---~v.,... ed suction port 2.0 BASE COMPRESSOR The base compressor was a fixed capacity two stage sliding vane compressor, as shown in Fig. 1. Compression to an intermediate pressure occurred in the outer compression chambers. After passing through an intermediate chamber, a second stage of compression was carried in the inner volumes. Suction ports in the two first stage vanes were kinematically timed to optimize flow into the suction cavity. Two discharge reed valves in the first stage allowed flow into an intermediate cavity. Ports in the two second stage vanes allowed flow into the second stage. Compression occurred and the flow was exhausted through discharge ports. The capacity of the base compressor was given by the following equation. It was dependant on the suction volume of the_first stage. mcomp = 1lvol Psuc f.tgi RPM ' \ \ \ Fig. 1 Two Stage Sliding Vane Compressor 3.0 VARIABLE MECHANISM The ratio of the first to second stage volumes was approximately 2.2. Given this ratio, it was determined that a 40% variable compressor could be developed using gas by-pass from the intermediate plenum. If full by-pass were accomplished, the compressor capacity would now be: mcomp = 1lvol Psuc Vstg2 RPM 409

As can be shown in Fig. 2, the effect of creating a flow path from the intermediate plenum to the suction chamber was to reduce the intermediate plenum pressure. This in turn limited the capacity of the compressor to the second stage suction volume. The valve that was used is a by-pass valve, the function of which is described in Mabe [1]. It regulates the bleed back based on the suction pressure seen by the A/C system. The compressor was built and tested without any modifications. The compressor displayed worrying efficiency trends at the lower capacities, see Fig. 3. Based on the discharge temperature measurements and COP, it was obvious that more work was required to improve the efficiency. The efficiency of the baseline compressor was competitive at 100% capacity, Huang [2]. The by-pass strategy at 40% capacity revealed glaring deficiencies in the compressor design that were otherwise not observed. By the end of the study, the performance of the compressor was improved significantly although it did not quite reach the performance of the competing technology. These improvements involved: a) improving gas flow within the compressor b) reducing friction sources 4.0 Flow Improvements A previous study of flow through the first stage discharge valve had shown that it was going to be a problem. Two transducers were used. They were placed axially above the discharge ports as shown in Fig. 4. Another transducer was placed in the intermediate plenum. Centre~ Housing Rear Plate ~ Pressure Transducers ~ l Discharge Port.. Fig. 4 Transducer Placement The results are shown in Fig. 5. As can be seen, the overcompression at the first stage valve was significant. The over-compression losses consisted of three phenomenon. 4.1 Throttling at valve: In the baseline configuration, pressures ran as high as 100 PSIG in the compression chamber, over-shooting the intermediate chamber pressures by 30 PSI. By raising the reed stop height by 50% and increasing the port area by 20%, the port throttling was substantially reduced. The overcompression was reduced to an average of 15 PSI. As shown in Fig. 3, changes resulted 1n a improvement in the efficiency. these marked 4.2 Stiction: The phenomenon of valve stiction can be seen graphically in Fig. 5. The valve was not instrumented for motion, however the motion can be inferred from the pressure measurements. At valve opening, there is a change in the two chamber pressures measured. This is to be expected, the valve nearest the port should have the lower pressure due to a Bernoulli effect as the flow adjusts. Efforts were made to reduce stiction. Increasing the port size helped in two ways. The total force 410

on the valve (=pressure x area) increased, while the contact area (= valve area - port area) was reduced. In addition to this, miniature slots were machined under the valve to help reduce stiction, similar to Sabha [3]. In this case however, the function of the slot was purely to reduce stiction. The slot groove depth was on the order of a few microns deep. Outline of valve Port Slot Fig. 6 Stiction Reducing Port As can be seen in Fig 5, all the efforts to reduce stiction did not result in significant improvements in the performance. Neither the time to open nor the overcompression were substantially reduced. Khalifa and Liu [4] discussed in depth the stiction phenomenon. It would have been beneficial to analyse the effects of oil viscosity. However oil viscosity reduction was limited by lubrication and sealing requirements at other operating conditions (i.e. high pressure/low speed operation. Lower contact area between the valve and the port reduced marginally the valve time opening. 4. 3 Squeeze: As the aspect ratio increases in the compression cylinder, there is a noticeable separation in the pressure profile. With discharge ports mounted in the rear of the compressor for design efficiency, this was always an inherent design problem. As can be seen however in Fig. 5, the total inefficiency due to squeeze was not substantial. It can be seen, that the problem occurred in both the baseline and improved designs. Neiter et al [5] discuss this phenomenon in detail. tapered cutout to compression due to effect. 5.0 Friction Reduction They suggest a reduce over the squeeze Friction was known to be a significant contributor to the performance of the compressor. Previous studies had shown friction to account for 15% of power consumption at 100% capacity. Based on the initial test results, there was evidence that this component of losses was not reducing in line with the flow reductions. 5.1 Vane Tip: The baseline vane tip saw several contact modes, ranging from point contact, linear contact to sharp counterformal contact as shown in Fig 7a. It was known that this scenario was not good for the vane friction or vane following, Kawahara et al [6]. (a) (b) Fig. 7 Shape of Vane Tip The vane tip was re-shaped so that at every angle of rotation, a consistent counterformal contact was achieved, Fig. 7b. The advantages included: -The largest counterformal contact that is geometrically possible, improving the ability to form a lubrication layer -More even wear across the entire tip surface and more even surface velocities 5.2 Vane Side: The slot and vane width were precision machined to avoid cocking of the vane in the slot. This tight tolerance however created unnecessary high friction loads due to oil shear. 411

A large groove was machined into the side of the vane on the low load side of the vane. This detail, although only a few microns deep would double the gap width between the vane and slot wall, thus reducing the overall oil shear friction. In this paper, we saw several instances where slight design modifications had a significant impact on the efficiency of a compressor when operating at part load (variable mode). The design steps that made the compressor significantly more competitive were clearly defined. ACKNOWLEDGEMENTS Low Load The authors would like to thank the following for their important contributions. Ron Young, Paul Sterling, Scott Stewart, Gary Williams, Gus Strikis and Vi pen Khetarpal all of Visteon. REFERENCES Milled Pocket on low load side of vane 1] Mabe, A. et al, 'The Scroll Type Variable Capacity Compressor for Automotive Air Conditioners',SAE- 870037, Intl Congress and Exposition, Feb 23-27, 1987. 2] Huang, Y. et al, 'A Novel Automotive Two Stage A/C Compressor', to be published, 2000 Purdue Compressor Conference 3] Sabha, N.G., Compressor reed valve with valve plate channel', US Patent #5,672,053, Sep. 1997. Fig. 8 Friction Reducing Vanes The groove was machined a few microns deep to minimise any associated re-expansion volume, but at a scale where oil shear was effectively reduced. At the same time it was relatively easy to incorporate from a manufacturing standpoint. It required only a small programming change. As can be seen in Fig 3., when both of these friction reducing concepts were applied to the compressor a noticeable reduction in the compressor performance was achieved. 6.0 CONCLUSIONS As discussed in Harte et al [7], the volumetric efficiency has a significant effect on a compressor's efficiency when thermodynamic losses do not reduce in tandem. 4] Khalifa, H.E. and Liu, Xin, 'Analysis of Stiction Effect on the Dynamics of Compressor Suction Valve', Proc. International Compressor Engineering Conference at Purdue, pp. 87-93, 1998. 5] Neiter, J.J. et al, 'Analysis of Clearance Volume Equalization and Secondary Pressure Pulse in Rolling Piston Compressors, Proc. International Compressor Engineering Conference at Purdue, pp. 527-533, 1994. 6] Kawahara, K. et al, 'Tribologic-al Evaluation of Rotary Compressor with HFC Refrigerants', Proc. International Compressor Engineering Conference at Purdue, pp. 413-418, 1996. 7] Harte, S. et al, 'A Weighting Method to Determine the Impact of Volumetric Efficiency on the Thermodynamic Efficiency of a Compressor', Proc. International Compressor Engineering Conference at Purdue, pp. 61-68, 1996. 412

res sure Compressor turned on Cyc Sw Press Sue Press lnt Press / Valve starts to ~~-~'-.!.L' open, Pi drops Fully open valve, Pi ~-...---- "-----t,_~~----- and Ps drop together ---,\---------- Suction hits control press @ ---~-~---~------~ Control Press Clutch cycles off Time Fig.2 Valve Operation and Effect on Intermediate Pressure 1.351~ 1.3 ~ 1.25 ~ 1.2 ~ 1.15 ~ 1.05 1.1 t=========:::::::~~====j 1 +---------------------------------------------------------------~ 0.95 +-----------------------------------------------------------------~ 0.9 +----------------+----------------r---------------~--------------~ (]) c a> (J) ro al "0 (]) c. (]) 0 - (/) I.2 (J) L- (]) u.. c o ro _.> o:: Cl>c Cl - 0 me o.c E.s::: 0 (.) ()~ Fig. 3 Compressor Discharge Temp at 40% Capacity Vs. Baseline Technology 413

-(!) (i) 100 95 90 85 ~ 80 ~ I- ::J Ill Ill 75 ~ I- 70 0... 65 60 55 Point of valve opening 0 5 10 15 20 25 Suction Volume (cc} --Base-D1 --Base-D2 -o- Base-lnt --lmprov-d1 --lmprov-d2 --lmprov-lnt Fig. 5 Baseline Versus Improved PV Chart for Discharge Port 414