Performance Measurements Of A Semi-Hermetic Carbon Dioxide Compressor

This paper describes the performance measurements of a prototype carbon dioxide compressor using a compressor load stand. The compressor load stand was specifically constructed for this purpose and is based on a hot-gas bypass design. The compressor is a semi-hermetic, two-piston, single-stage, reciprocating compressor with an estimated cooling capacity of 3 tons of refrigeration. Compressor tests were conducted for varying suction temperatures and pressures, and discharge pressures. For each test, the compressors mass flow rate, power consumption, and temperatures and pressures at each state point were recorded. In addition, the volumetric and overall isentropic efficiencies were reported. The results show volumetric efficiencies between 0.8 and 0.5 and overall isentropic efficiencies of up to 0.55 for pressure ratios between 1.5 and 6.5. The efficiencies are only slightly affected by different superheats. INTRODUCTION The transcritical cycle technology using carbon dioxide as the refrigerant has recently received increased attention as a possible replacement for vapor compression cycle technology using fluorocarbon-based refrigerants. In particular, three applications of carbon dioxide systems can be identified that show comparable or better performance as well as economic feasibility compared to vapor compression systems. The most prominent of these applications is automotive air conditioning. By now, most of the major automobile manufacturers have carbon dioxide prototype systems and several new and innovative designs for heat exchangers, compressors, and valves have emerged from studies in this area. The second application are environmental control units (ECU), which are packaged air-to-air air conditioners that are used in cooling of mission critical electronics and personnel. The U.S. Army currently maintains roughly 22,000 units of varying capacity either in service, storage, or on order that use HCFC-22 as their refrigerant and need to be replaced by 2010. The third application that shows great promise for transcritical carbon dioxide systems is the one of heat pump water heaters. In fact, the first commercial product has been introduced on the Japanese market. While the compressor development for automotive air conditioning application has excelled over the last several years, relative little information is available in the literature with respect to hermetic or semihermetic compressors that need to be used in the later two applications specified above. Therefore, a research effort is currently underway at the Herrick Laboratories, which specifically focuses on measuring the performance of hermetic and semi-hermetic carbon dioxide compressors. CO2-COMPRESSOR BACKGROUND Most of the early investigations on hermetic-type CO2 compressors focused on the design issues associated with the use of CO2 (Fagerli 1996a; Fagerli 1997) or the modification of existing HCFC-22 compressors to use with carbon dioxide (Adolph 1995; Fagerli 1996b; Koehler et al. 1997 and 1998; Hwang and Radermacher 1998). In more recent studies, prototype designs of hermetic compressors for use with carbon dioxide have been built and analyzed. Tadano et al. (2000) developed a prototype hermetic two-stage rolling piston compressor with a cooling capacity of 750 W. This compressor was considered a “first cut” device and will provide the basis to develop larger compressors for heat pump water heating, refrigerating, or airconditioning applications. The compressor was designed to operate between a low pressure of 3 to 4 MPa and a high pressure of 10 MPa. The diameter of the compressor shell was 117.2 mm and the height was 244.3 mm. The compressor displacement was 2.633 cm. A two-stage compression with two rolling pistons was chosen to maintain small pressure differences across each compression stage. The intermediate pressure was selected to be 5 to 6 MPa. The inside of the hermetic shell was at intermediate pressure to minimize the gas leakage between the compressing chambers and the inner space of the shell. Several tests were performed as a function of motor frequency, superheat, and pressure ratio. The authors reported isentropic efficiencies of up to 88%. However, these efficiencies did not include motor and shell losses. A durability test over 1000 hours was also performed. Neksa et al. (2000) reported on the development of a series of semi-hermetic reciprocating compressors with swept volumes in the range of 0.5 to 12.6 m/h. The compressor series consists of singleand twostage compressors with two cylinders, running at nominal speeds of 2900 rpm (50 Hz). This corresponds to cooling capacities in the range of 0.6 to 15 kW at -35°C evaporating temperature. The measurements of compressor efficiencies as a function of pressure ratio were presented for a two-stage compressor in the 4-pole assembly (1450 rpm), which was in its early stage of development. A volumetric efficiency of up to 80% and an isentropic efficiency of up to 60% were reported for the compressor. Also, some preliminary measurements of a newer compressor operating at 2900 rpm were presented. In summary, all investigations with respect to carbon dioxide compressors have focused on developing a prototype compressor or a better understanding of the fundamental concepts of carbon dioxide compression. However, much of the information associated with these studies is not necessarily available to the public and detailed performance data of CO2 compressors is still difficult to obtain. This information however, is needed to be able to evaluate the performance potential of the transcritical carbon dioxide technology on a system level. LOAD STAND For the purpose of measuring the performance of carbon dioxide compressors, a new compressor load stand was designed and built. The load stand consists of three units: (1) the compressor unit including a high voltage power supply, compressor controls, and an oil cooler, (2) the load stand unit, hosting the main sections of the cycle, and (3) the monitoring unit with the data acquisition system and the personal computer. The load stand is designed to provide maximum flexibility with respect to changing compressors and testing hermetic or open-drive compressors. The load stand characteristics are as follows: capable to measure cooling loads from 5 to 20 kW, compressor power supply delivering a power of up to 15 kW, compressor speed adjustment by inverter in a range of 30 to 60 Hz, depending on the compressor type, and determination of the refrigerant mass flows rate of up to 10 kg/min. The compressor load stand is based on the hot gas bypass concept. The idea behind this concept is to anchor the intermediate pressure below the critical pressure in the two-phase region by condensing a fraction of the refrigerant flow. Using this stable anchoring pressure, the suction and discharge pressures are controlled by using appropriate metering valves in the discharge line and bypass line. Figure 1 illustrates the ideal process cycle in a logarithmic enthalpy-pressure diagram. The compressor discharges high pressure, high temperature carbon dioxide at state point 2, which is throttled to the intermediate pressure at state 2a. Figure 1: Process cycle shown in a logarithmic pressure – enthalpy diagram The oil is separated from the CO2 using an oil separator. This is done to reduce the oil concentration in the CO2 flow rate before passing the CO2 through the main flow meter so that only the refrigerant mass flow rate is measured. After passing through the main flow meter, the CO2 flow is split. Most of the flow goes through the bypass loop, while the remaining flow enters the primary loop. The bypass loop includes the bypass expansion valve, where the fluid is throttled to the suction pressure (state point 5). The primary loop condenses the CO2 in the water-cooled condenser. Subcooled liquid at state point 3 exits the condenser and is throttled through the primary expansion valve to the suction pressure (state point 4). The two fluid streams are then combined just before the mixing chamber and exit the mixing chamber at state point 1. A schematic of the load stand indicating all significant features is shown in Figure 2. All control valves are based on manual operated metering valves, since automatic control valves are not yet available for the given application. All piping is stainless steel and the whole test stand was designed to withstand pressures of up to 135 bar (2000 psia). The test stand reaches steady state operating conditions within approximately 30 minutes. An electrical heater in combination with a PI-controller fine-tunes the suction temperature. The load stand is equipped with several safety features to protect the system and the environment from failures. Pressure relief valves in each section protect the load stand from uncontrolled pressure increases. A compressor winding temperature controller protects the compressor motor. A 1 Steady state: maximum 0.5 Kelvin standard deviation for the temperature, maximum 0.1 % standard deviation of the mean value for the pressure and 1 % standard deviation of the mean value for the flowand power meters measured over a 10 minute period. 1 2