1 Introduction

1.1 Background and objective

Since the refrigeration system was invented, human living standard has grown rapidly. The air conditioners have made people comfortable regardless of the outside circumstances and the refrigerators have extensively enhanced the quality of their meals. Nowadays, people can manage to stay in the deep sea, in the stratosphere, and even in the outer space due to the HVAC (heating, ventilating, and air conditioning) technology. Although the refrigeration system has improved the human life conditions, there also exist many counter effects. The refrigerants in the refrigeration system have destroyed the ozone layer in the stratosphere and have heated the earth. The increasing use of air conditioners and refrigerators has resulted in the great demand for electricity. As about 50% of electricity is generated from fossil fuel globally (https://www.export.gov/article?id=Korea-Energy-New-and-Renewable), it also causes global warming and environmental pollution.

Based on the knowledge that the refrigeration system has a negative impact on the environment, numerous efforts have been made to make it eco-friendly. An energy-efficient refrigeration system has emerged as one of the solutions. As the energy-efficient system requires less energy for the same degree of cooling or heating capacity, it lowers the electricity demand. Therefore, the improvement of the refrigeration system’s performance may be a valid approach to preserving the environment. There are three representative ways to enhance the performance of the refrigeration system, as follows:

  • Development of highly efficient components [1, 2];

  • Invention of a new refrigerant [3,4,5]; and

  • Proposing an alternative refrigeration cycle [6, 7].

The recent studies with regard to the search for more efficient components (compressor and heat exchanger) have concentrated on the development of new technologies to achieve better performance. The invention of a new type of refrigerant demands chemical knowledge to be able to determine its chemical characteristics and its effects on other materials, on the environment, etc. Alternative refrigeration cycles have been studied, and there are many promising options for increasing the thermodynamic efficiency compared to the traditional vapor compression cycle. Two of the most explored alternative refrigeration cycles at present are the ejector and injection refrigeration systems.

Between the foregoing, the vapor injection refrigerator is an advanced refrigeration system employing highly efficient components and an alternative refrigeration cycle. Its system configuration is similar to that of the two-stage vapor compression system, which uses two compressors for enhancing the system performance. The vapor injection system, however, is equipped with an adapted compressor called “vapor injection compressor.” The vapor injection compressor has three refrigerant ports: one for suction, one for discharge, one for injection. As the vapor injection refrigerator uses only one compressor, the size of the apparatus can be reduced in comparison with the conventional two-stage vapor compression system. The compressor is the refrigeration system component that consumes the most electricity; thus, to enhance the refrigeration system’s performance, it is important to reduce the power input of the compressor. Due to the aforementioned strengths of the vapor injection system, numerous studies on the system mentioned above have been conducted by many researchers ever since the concept of the vapor injection compressor was proposed [8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29]. Especially, the studies on the conditions of intermediate pressure, which is the pressure of the by-passed refrigerant after the application of the injection electronic expansion valve, have been most actively performed for the system optimization.

Oquendo et al. [8] characterized the vapor injection compressor. The by-passed mass flow rate was correlated to the characterization results. The model proposed in the study showed a less than ±5% deviation for the mass flow rate through the evaporator, by-passed mass flow rate, intermediate pressure, and compressor power input. Wang et al. [9] presented the optimization of the vapor injection system in terms of universal control. They developed a vapor injection system model to investigate the vapor injection effect and the component parameters. Based on the identified evaporator characteristics and the results of the thermodynamic analysis, they proposed a set of general principles of the vapor injection system. Baek et al. [10] conducted a performance analysis of a carbon dioxide (CO2) heat pump with vapor injection. The optimal intermediate pressure improved the heating capacity, but the coefficient of performance (COP) remained constant because the compression work rose altogether. From another study conducted by Heo et al. [11], the optimal by-passed mass ratio (the ratio of the by-passed mass flow to the total mass flow) increased with respect to the pressure ratio because the effect of the economizer on the compressor performance was enlarged. Baek et al. [12] also investigated a CO2 supercritical heat pump with vapor injection and concluded that the by-passed mass ratio and the water flow rate through the gas cooler, a condensing device for the supercritical system, are important control parameters in terms of the system performance. Torrella et al. [13] evaluated the inter-stage conditions for a two-stage refrigeration cycle. The inter-stage vapor injection was investigated, along with the inter-stage liquid injection. Intermediate pressures applying the arithmetical mean and the geometric mean were compared to each other, and based on the results, it was concluded that the arithmetical mean pressure is more appropriate from the point of view of energy. Wang et al. [14] experimentally investigated the vapor injection technique. Two scroll-type compressors were employed for the test benches. One had an injection port while the other had the same displacement but no injection port. The results showed that vapor injection increased the system performance in terms of both the cooling and heating applications. The cooling capacity and COP were enhanced by 14 and 4%, respectively, and the heating capacity and COP increased 30 and 20%, respectively. He et al. [15] performed experiments on the vapor injection heat pump, and the experiment results showed that the intermediate pressure had a large effect on the heating capacity and the liquid refrigerant temperature after the application of the economizer. In addition, many researches were conducted by Gosney [16], Odilio [17], Rasi [18], Czaplinski [19], Baumann and Blass [20], Threlkeld [21], Arora and Dhar [22], De Lepeleire [23], Domanski [24], Zubair et al. [25], Khan and Zubair [26, 27], Ouandha et al. [28], and Ratts and Brown [29] to present the optimal intermediate pressure for the two-stage vapor compression and vapor injection systems.

From the literature investigation that was conducted in this study, it was found that improvements of the vapor injection system in terms of performance optimization have been achieved by numerous researchers. Many models of the vapor injection system have been established, and they showed reliable results, but most of the relevant researches focused on the heating performance of the vapor injection system because the improvement of the heating performance compared to the conventional single-stage vapor compression system was found to be larger than that of the cooling performance. In addition, few researches [30,31,32] have presented an equation for the optimal intermediate pressure. Some of them investigated the cooling performance, but most of them only employed it to compare its improvement to that of the heating performance.

The main purpose of the present study was to analyze the performance characteristics of the vapor injection refrigeration system, by focusing on the cooling capacity and COP. Experiments were performed for various conditions, including the operating conditions of the experiment setup, to confirm the influence of intermediate pressure on the performance characteristics of the vapor injection refrigeration system. As a result of the present study, two equations for calculating the optimal intermediate pressure were formulated as functions of the temperature and pressure, respectively.

1.2 Characteristics of the vapor injection refrigeration system

Schematic and P-h diagrams of the vapor injection system are shown in Figs. 1 and 2, respectively. In addition to the basic components of the conventional single-stage vapor compression refrigeration system, including the vapor injection compressor, the vapor injection system employs supplementary devices: one economizer, one injection electronic expansion valve. The liquid refrigerant from the condenser (#3) is by-passed. The larger-amount refrigerant goes directly into the economizer while the other is expanded by the injection electronic expansion valve (#4). The latter refrigerant then passes through the economizer to absorb the heat from the former refrigerant, and then vaporizes (#5). The refrigerant further cooled by the economizer (#6) is throttled by passing through the main electronic expansion valve (#7), and then vaporized in the evaporator (#1). The vapor injection compressor inhales the refrigerant gas from the evaporator and economizer, and pressurizes it. Generally, the refrigerant from the economizer is colder than that from the evaporator; thus, the injecting vapor cools the working fluid from the evaporator. As a result, the discharged refrigerant has a lower temperature than the conventional single-stage vapor compression system. The vapor refrigerant is sucked by the injection port during the compression process of the refrigerant from the evaporator; hence, the cooling effect of injecting vapor and the compression process happen simultaneously. The predicted vapor injection process is plotted with dotted lines in Fig. 2 (#2′, #2″).

Fig. 1
figure 1

Schematic diagram of the vapor injection refrigeration system

Fig. 2
figure 2

Pressure-enthalpy (P-h) diagram of the vapor injection refrigeration system

The main improvements of the vapor injection system are the extra-refrigerating effect and low discharge temperature. When the liquid refrigerant from the condenser was cooled, the enthalpy difference between the evaporator inlet and outlet (i.e., the “refrigerating effect”) increased. For this reason, the same degree of evaporation capacity can be achieved with a smaller refrigerant amount compared to that in the conventional refrigeration system. Fig. 3 shows how the vapor injection system lowers the discharge temperature. The discharged refrigerant is state 2′′′ if there is no injected fluid. The discharge temperature decreases from state 2, however, when an economizer is installed. A higher discharge temperature chemically decomposes the refrigerant and compressor oil, damages the compressor, and causes mechanical failure. Therefore, the vapor injection system can achieve better performance with higher stability by applying an economizer thereto.

Fig. 3
figure 3

Comparison of the compression processes with vapor injection and without vapor injection

2 Experiment apparatus and method

2.1 Experiment apparatus

A vapor injection refrigeration system with a 9 kW cooling capacity at a 30 °C condensation temperature and a − 40 °C evaporation temperature was designed to conduct experiments using R404A. The schematic diagram of the experiment apparatus is presented in Fig. 4. The experiment setup consisted of three cycling loops using refrigerant, coolant, and brine, respectively. The refrigerant loop, which is the vapor injection refrigeration system, consists of a vapor injection compressor, a condenser, an evaporator, an economizer, and two electronic expansion valves (installed at the main circuit and at the by-passed circuit, respectively) with a step motor and a receiver tank. The coolant loop included a water pump, a volumetric flow meter, and thermostatic bath. The thermostatic bath consisted of a refrigerator and an electric heater to control the coolant temperature so as to make it constant. The brine loop consisted of a brine pump, a volumetric flow meter, and an electric heater unit.

Fig. 4
figure 4

Experiment apparatus for the vapor injection refrigeration system

The refrigerant discharged from the compressor flows through the condenser and receiver, then the liquefied refrigerant is by-passed, as described above. The injection electronic expansion valve in front of the economizer expands the by-passed refrigerant and conducts heat exchange with the refrigerant, which flows directly from the receiver to the economizer. Through this process, the refrigerant after the application of the economizer is further cooled and is expanded by the main electronic expansion valve. The two-phase refrigerant flows through the evaporator and becomes vapor by absorbing the heat from the brine. The vapor refrigerant is then inhaled by the compressor. The vapor exhausted from the economizer is then injected and merged with the refrigerant in the compression process. The discharged refrigerant goes to the condenser again, and the circulation is completed. Thermostatic bath is installed outside the vapor injection system for supplying coolant at a constant temperature. The brine chilled by the evaporator circulates the electric heater unit, and as a result, the brine’s temperature becomes constant, and such brine is produced through the evaporator.

The Emerson scroll-type compressor with an injection port and 35.2 m3/h displacement at 60 Hz was selected for the vapor injection compressor. Counter-flow brazed plate heat exchangers made of stainless steel with copper brazing were employed as the condenser, evaporator, and economizer. The condenser and evaporator were made by Alfa Laval, and the economizer was produced by Swep. The numbers of plates in the condenser, evaporator, and economizer were 66, 40, and 20, respectively, and the dimensions were 155 × 113 × 527, 187 × 111 × 526, and 76 × 48.8 × 317 mm. The refrigerant flow rates through the evaporator and economizer were controlled by the two electronic expansion valves of the Sporlan and Sanhua companies, respectively. The main electronic expansion valve had 500 steps and was controlled with a Dotec controller. The injection electronic expansion valve, on the other hand, had 400 steps and was adjusted with a controller manufactured by Emerson. Tables 1 and 2 show the detailed specifications of the main and accessory components, respectively.

Table 1 Specifications of the main components used in this study
Table 2 Specifications of the accessory components used in this study

The power inputs of the coolant and brine pump and of the compressor were measured with a power meter. The temperature and pressure of the refrigerant, coolant, and brine were measured using a T-type thermocouple and a pressure transducer, respectively. The flow rates of the refrigerant, coolant, and brine were measured using a volumetric flow meter. Table 3 presents the detailed specifications of the instruments that were used in this study.

Table 3 Specifications of the measuring instruments used in this study

2.2 Experiment conditions and method

Table 4 presents the test conditions for the experiment. As can be seen in Table 4, the refrigerant that was used in the experiment was R404A, and the brine and coolant were FC3283 and R718, respectively. The experiment was conducted at the 30 and 35 °C condensing temperatures and with 5 °C suction superheating, and the degree of liquid subcooling was 5 °C. The evaporating temperatures were set to −45, −40, and − 35 °C. The pressure value calculated using Baumann and Blass’s equation [20] was set as the standard intermediate pressure (SIP), and five pressure values (from −100 to +100 kPa of the standard medium pressure range), including the standard, were employed as the variables. The influence of the medium pressure on the system performance characteristic was analyzed for each test case.

Table 4 Experiment conditions

Through the preliminary test, the 13.80 kg optimal charge amount was determined. The compressor frequency was fixed at 60 Hz during the experiment. The cooling capacity was varied to produce a constant superheating degree at the evaporator outlet, and it was adjusted using the heat load controller of the heater unit. The thermostatic bath controlled the coolant inlet temperature to keep the condensation pressure constant. To achieve constant evaporation and intermediate pressure, the opening of the main and injection electronic expansion valves was controlled by PID, respectively. The valve openings were fixed when a steady state was reached. The temperature, pressure, mass flow rate, and power input of the refrigeration system were monitored using a data acquisition system. All the test data were recorded continuously for 40 min, with 2-s intervals. Each test was repeated three times to confirm the repeatability and reliability of the obtained data.

The uncertainty caused by the varying level of the sensors’ accuracy can be calculated using the Kline and McClintock method [33]. The analysis results for the uncertainties of the experiment parameters are listed in Table 5.

Table 5 Summary of the estimated uncertainties

2.3 Data reduction

The thermal properties of the refrigerant, brine, and coolant that were used in this study were calculated using Engineering Equation Solver (EES) [34]. To investigate the influence of intermediate pressure on the system performance, the COP was calculated using the following equations. In addition, the condensation capacity was computed to confirm the heat balance of the experiment apparatus, and the by-passed mass ratio, which affects the evaporation capacity, was defined.

The COP [−] is the performance indicator of the refrigeration system. It is calculated using the equation below:

$$ \mathrm{COP}=\frac{Q_e}{Aw} $$
(1)

where Qe [kW] is the evaporation capacity and Aw [kW] is the compression power input measured using the power meter.

The evaporation capacity[kW] is derived from the equation below:

$$ {Q}_e={G}_b\cdot {c}_{p,b}\cdot \left({T}_{b, in}-{T}_{b, out}\right) $$
(2)

where Gb [kg/s] is the mass flow rate of the brine flowing through the evaporator, cp,b [kJ/kg·°C] is the specific heat of the brine, and Tb,in and Tbout [°C] are the brine temperatures of the evaporator inlet and outlet, respectively. As the volumetric flow meter was installed in the experiment apparatus, the mass flow rate of brine was calculated using the following equation:

$$ {G}_b={\rho}_b\cdot {V}_b $$
(3)

where ρb [kg/m3] is the mass density of the brine and Vb [m3/s] is the volumetric flow rate.

The condensation capacity Qc [kW] is given as follows:

$$ {Q}_c={G}_w\cdot {c}_{p,w}\cdot \left({T}_{w, out}-{T}_{w, in}\right) $$
(4)

where Gw [kg/s] is the mass flow rate of the coolant passing through the condenser, cp,w [kJ/kg·°C] is the specific heat of the coolant, and Tw,in and Tw,out [°C] are the coolant temperatures in the condenser inlet and outlet, respectively. As the volumetric flow meter was included in the experiment setup, the mass flow rate of coolant Gw [kg/s] was computed using the following equation:

$$ {G}_w={\rho}_w\cdot {V}_w $$
(5)

where ρw [kg/m3] is the mass density of the coolant and Vw [m3/s] is the volumetric flow rate.

The heat balance of the refrigeration system means that the condensation capacity is the sum of the compression work and the evaporation capacity by the second law of thermodynamics, and is shown by the equation below:

$$ {Q}_c={Q}_e+ Aw $$
(6)

The by-passed mass ratio α [−] is defined as the ratio of the mass flow rate of the by-passed refrigerant to the total mass flow rate of the refrigerant passing through the condenser, and is expressed by the following equation:

$$ \alpha ={G}_{bp}/{G}_c $$
(7)

where Gbp [kg/s] is the mass flow rate of the by-passed refrigerant and Gc [kg/s] is the total mass flow rate of the refrigerant passing through the condenser.

3 Results and discussion

To achieve reliable and optimal operation at a high compression ratio, vapor injections were applied to the refrigeration system. Thus, the main aim of this experiment was to analyze the effect of intermediate pressure on the compression work and evaporation capacity of the vapor injection refrigeration system, and then to find the optimum intermediate pressure that maximizes the COP.

3.1 Heat balance of the experiment apparatus

The refrigeration system absorbs heat from the low-temperature reservoir and releases it to the high-temperature reservoir. Electric power input to the compressor is required during the process. Hence, theoretically, the sum of the heat capacity of the evaporator and compressor power is equal to the heat capacity of the condenser.

Fig. 5 presents the heat balance of the experiment apparatus for all the test conditions. As plotted in Fig. 5, the condensation capacity was always lower than the sum of the evaporation capacity and compression work. This is due to the heat loss during the system operation. The heat loss occurred through the compressor body; thus, a system output lower than the system input resulted. The deviations of the heat balance were within ±10% for all the test conditions. A low heat balance deviation indicates the high accuracy of the experiment. Therefore, all the raw data acquired through the experiment in this study had sufficient reliability to validate the experiment results.

Fig. 5
figure 5

Heat balance of the experiment apparatus

3.2 Compression work

Fig. 6 shows the compression works with respect to the intermediate pressures. The highest compression power input was achieved at the 35 °C condensation temperature and − 35 °C evaporation temperature. The high evaporation temperature increased the compression power input due to the high mass flow rate through the evaporator. The higher condensation temperature also compounded the compression work due to the higher discharge pressure. The higher discharge pressure of the compressor increased the pressure difference between the condenser and the evaporator, and as a result, the compression work increased.

Fig. 6
figure 6

Compression works with respect to the intermediate pressures

The compression work consistently increased by up to 24.48% with respect to the intermediate pressure. Similar tenancies can be found in the researchers conducted by Oquendo et al. [8], Wang et al. [9], Wang et al. [14], Cao et al. [15], and Wang et al. [35]. The reason for this can be known through Fig. 7, which plots the by-pass mass ratios with respect to the intermediate pressures. As the intermediate pressure rose, the by-pass mass ratio increased altogether. As the mass flow through the evaporator did not depend on the intermediate pressure, the growth of the by-passed mass flow ratio indicates the increase of the total mass flow rate of the vapor injection system. Consequently, the mass flow rate through the compressor rose. Therefore, the compression power input rises with the increase of the intermediate pressure.

Fig. 7
figure 7

By-passed mass ratios with respect to the intermediate pressures

3.3 Evaporation capacity

Fig. 8 presents the evaporation capacities with respect to the intermediate pressures. The highest evaporation capacity appeared at the Tc = 30 °C and Te = −35 °C conditions. Each evaporation capacity of the identical evaporation temperature was located within the less than ±0.58 kW deviation range. The evaporation temperature had a large effect on the evaporation capacity. The increased mass flow rate through the evaporator due to the rising of the evaporation temperature enhanced the evaporation capacity. On the contrary, the condensation temperature affected the evaporation capacity in a reverse manner. Due to the rising condensation temperature, the refrigerant temperature of the condenser outlet increased, and as a result, the liquid refrigerant with a higher temperature was generated after the application of the economizer. Hence, the enthalpy difference between the evaporator inlet and outlet (the refrigerating effect) decreased, and the evaporation capacity degenerated.

Fig. 8
figure 8

Evaporation capacities with respect to the intermediate pressures

Based on the aforementioned results, it can be said that there was an intermediate pressure at which the evaporation capacity became maximum under each experiment condition. Analogous trends were presented in the studies that were carried out by Wang et al. [9], Wang et al. [14], and Wang et al. [35]. The mass flow rate of the by-passed refrigerant rose with the increase of the intermediate pressure. For the low medium pressure range, the rising of the by-passed mass flow rate enhanced the economizer capacity; thus, the enthalpy of the evaporator inlet decreased. Nevertheless, the enthalpy of the evaporator inlet started increasing when the intermediate pressure kept growing. As shown in Fig. 9, the enthalpy of the evaporator inlet increased with respect to the medium pressure after the lowest value was achieved. This was due to the saturation temperature growth of the by-passed refrigerant. With the rising of the intermediate pressure, the saturation temperature of the by-passed refrigerant increased altogether, and it resulted in the high temperature of the by-passed refrigerant at the economizer inlet, as plotted in Fig. 9. The higher temperature of the by-passed refrigerant in the economizer inlet limited the subcooling degree of the liquid refrigerant on the other side of the economizer. As a result, a higher liquid refrigerant enthalpy was produced by the increase of the intermediate pressure after the lowest value was accomplished, and it had a negative effect on the evaporation capacity. Based on the foregoing reasons, the evaporation capacity became maximum with respect to the intermediate pressures at each experiment condition.

Fig. 9
figure 9

Temperatures of the by-passed refrigerant in the economizer inlet and enthalpies of the evaporator inlet with respect to the intermediate pressures

3.4 Coefficient of performance

Fig. 10 shows the system COPs with respect to the intermediate pressures. In Fig. 10, the extension lines from the lowest intermediate pressures represent the COPs calculated via regression analysis [36]. At all the experiment conditions, the COP increased to the optimum intermediate pressure, and then decreased again. The optimum intermediate pressure is defined as the pressure at which the COP becomes maximum. As mentioned in the previous section, the evaporation heat was maximum, and the compression work increased; as such, the COP increased and then decreased at each experiment condition. Based on the foregoing results, it can be said that there was an optimum intermediate pressure for the maximum COP under each experiment condition.

Fig. 10
figure 10

System COPs with respect to the intermediate pressures

Similar tendencies of the COP can be found in the investigations by Wang et al. [9], Wang et al. [14], and Wang et al. [35]. The maximum COP was anticipated at around the lowest tested medium pressure for each experiment condition.

From Fig. 6-10, it can be known that there is an optimal intermediate pressure in terms of the vapor injection refrigeration system performance. The optimal intermediate pressure Pm,opt [kPa] can be calculated via regression analysis [36], and can be expressed as follows:

$$ {P}_{m, opt}=\sqrt{P_c\cdot {P}_e}-99.08 $$
(8)

where Pc and Pe [kPa] are the condensation and evaporation pressures of the vapor injection refrigeration system, respectively. Table 6 presents the calculation basis and data for eq. (8).

Table 6 Calculation basis and data for Eq. (8)

Also, the optimal intermediate pressure can be calculated by a function of the temperatures. The optimal inter-stage temperature ratio, θopt [−], is given as:

$$ {\theta}_{opt}=\frac{T_{bp, in}-{T}_e}{T_c-{T}_e}=0.32 $$
(9)

where Tbp,in, Tc, and Te [°C] are the temperature of the by-passed refrigerant in the economizer inlet and the condensation and evaporation temperatures of the vapor injection refrigeration system, respectively. The temperature of the by-passed refrigerant in the economizer inlet is the arithmetical mean of the condensation and evaporation temperatures when the θ value is 0.5. Therefore, eq. (9) implies that the temperature of the by-passed refrigerant at the optimal medium pressure is lower than the average condensation and evaporation temperatures. The calculation basis and data for eq. (9) are shown in Table 7.

Table 7 Calculation basis and data for Eq. (9)

4 Conclusions

The vapor injection refrigeration system was researched on in this study. The basic principles of the vapor injection system are discussed herein, and an experiment was carried out. The performance characteristics of the vapor injection refrigeration system were verified by the experiment results. The influence of the intermediate pressure on the vapor injection refrigeration system can be summarized as shown below.

  1. 1)

    The highest compression power input is accomplished at the 35 °C condensation temperature and at the −35 °C evaporation temperature. The compression work consistently increases with respect to the intermediate pressure due to the increase of the total mass flow rate through the compressor.

  2. 2)

    For the low medium pressure range, the growth of the by-passed mass flow rate increases the economizer capacity. In the high intermediate pressure range, however, the improvement by the increase of the by-passed mass flow rate is offset by the limited subcooling degree of the liquid refrigerant since the rising of the by-passed refrigerant saturation temperature. Hence, it can be seen that in this study, there was an intermediate pressure at which the evaporation capacity became maximum under each experiment condition.

  3. 3)

    The vapor injection refrigeration system’s coefficient of performance (COP) has the optimal intermediate pressure. For the low intermediate pressure range, the increasing degree of evaporation capacity is higher than that of compression work. The effect of the compression work increase, however, becomes larger in the case of the high medium pressure range. Consequently, the optimal intermediate pressure is observed in terms of the system COP.

  4. 4)

    Two functions employing the temperature and pressure as the inputs, respectively, are presented as the results of this study for the calculation of the optimal intermediate pressure.