**3. Design and simulation**

#### **3.1. Design and simulation of the ejector**

For the ejector, in order to get a low suck pressure for the vapor generator, the spreading ratio (SR) defined as the ratio of the throat area to the tube area should be very small, and the velocity should be very high according to energy conservation. So selecting one optimized ejector to obtain a good performance of the vacuum heat pump system is very important; we designed three ejectors with different spreading ratios, of which the ratios were 0.0156, 0.0532, and 0.0946, respectively, and the throat diameters were 1.5, 3, and 4 mm, respectively, shown in **Table 1**.

(A) SR = 0.0156 and Dt = 1.5 mm; (B), SR = 0.0532 and Dt = 3 mm; and (C), SR = 0.0946 and Dt = 4 mm.

The performance of the above three ejectors were analyzed by FLUENT software. The fluid was the water, the inlet pressure was 0.6 MPa, and the inlet velocity is 1.6 m/s. The simulated results were shown in **Figure 3**.

**Table 1.** Physical structure of three different ejectors 3-a:SR=0.0156,Dt=1.5mm, 3-b:SR=0.0532,Dt=3mm, 3-c:SR=0.0946,Dt=4mm.

of the inner wall, and the pressure of the fluid will gradually decrease. And the flow of refrigerant in capillary tube can be divided into four stages: overcooling phase, single-phase metastable phase, gas–liquid two-phase metastable phase, and gas–liquid phase in thermal equilibrium phase. Therefore, we should choose different sizes of capillary tubes to measure the influence of different types of capillary tubes on the inlet overcooling and refrigerant liquid phase exit volume fraction. It is very important to select an optimized capillary tube to obtain a great system performance of heat pump distilled water, so we designed five capillary tubes with different sizes (inner diameter × length, unit mm), of which the sizes were 1.7 × 1700, 1.7 × 1500, 1.4 × 1500, 1.4 × 1400, and 1.4 × 1300. The inner diameter of the capillary tube used in the system is 0.5–2 mm, and the length is 1–4 m. The inner diameter and length of each capillary tube are different, but their materials are all copper tubes. After selecting the inner diameter and the length, the flow rate of the capillary tube depends on the difference between the cooling degree, the return air pressure, the suction pressure, and

Distilled Water Production by Vacuum Heat Pump http://dx.doi.org/10.5772/intechopen.76839 83

The VOF multiphase flow model is used in the five capillary flow simulations, and the performance is simulated and analyzed by FLUENT software. The fluid is the refrigerant R22, the inlet pressure is 1.8 Mpa, the outlet pressure is 0.6 Mpa, the inlet refrigerant temperature is

Capillary tube is a slender structure; the length is greater than the diameter, if only using unstructured grid and drawing the number of grid will be too much; it is easy to exceed the limits of computer processing, so here structured grids are used, internal for hexahedral grid

At the inlet pressure which is 1.8 Mpa, the saturation temperature of the refrigerant R22 is 47°C; before entering capillary refrigerant is supercooled. Of five kinds of capillary tube in experimental conditions, the coolant temperature in the entrance of the capillary tube is 41°C, namely, supercooling degree is 6°C; the five models of capillary throttling effect comparing

**1.** Inlet coolant temperature is 41°C, liquid phase distribution of refrigerant in different types

so on.

simulation diagram are:

**Figure 4.** Capillary grid division.

of capillary tubes is as follows:

314.15 K, and the outlet temperature is 279.16 K.

and external for tetrahedron, mesh model as shown in **Figure 4**.

**Figure 3.** Internal velocity distribution of ejector.

From **Figure 3**, it can be seen that the maximum speed of the ejector (A) throat is 110 m/s and the velocity of the water vapor injection is more than 50 m/s. Compared with the ejector (B), the maximum speed of the throat is 30 m/s, and the velocity of the steam injection is about 4 m/s. While as the ejector (C), there was a reverse flow in the suck line, which implied that the water vapor from the vapor generator cannot be sucked into the condensate absorber. This can be analyzed from the perspective of conservation of energy.

#### **3.2. Design and simulation of the capillary**

The capillary tube is a small tube with small inner diameter. Due to the small inner diameter, when the fluid flows through the capillary tube, it will be greatly frictional resistance

**Figure 4.** Capillary grid division.

From **Figure 3**, it can be seen that the maximum speed of the ejector (A) throat is 110 m/s and the velocity of the water vapor injection is more than 50 m/s. Compared with the ejector (B), the maximum speed of the throat is 30 m/s, and the velocity of the steam injection is about 4 m/s. While as the ejector (C), there was a reverse flow in the suck line, which implied that the water vapor from the vapor generator cannot be sucked into the condensate absorber. This

The capillary tube is a small tube with small inner diameter. Due to the small inner diameter, when the fluid flows through the capillary tube, it will be greatly frictional resistance

can be analyzed from the perspective of conservation of energy.

**3.2. Design and simulation of the capillary**

**Figure 3.** Internal velocity distribution of ejector.

82 Desalination and Water Treatment

of the inner wall, and the pressure of the fluid will gradually decrease. And the flow of refrigerant in capillary tube can be divided into four stages: overcooling phase, single-phase metastable phase, gas–liquid two-phase metastable phase, and gas–liquid phase in thermal equilibrium phase. Therefore, we should choose different sizes of capillary tubes to measure the influence of different types of capillary tubes on the inlet overcooling and refrigerant liquid phase exit volume fraction. It is very important to select an optimized capillary tube to obtain a great system performance of heat pump distilled water, so we designed five capillary tubes with different sizes (inner diameter × length, unit mm), of which the sizes were 1.7 × 1700, 1.7 × 1500, 1.4 × 1500, 1.4 × 1400, and 1.4 × 1300. The inner diameter of the capillary tube used in the system is 0.5–2 mm, and the length is 1–4 m. The inner diameter and length of each capillary tube are different, but their materials are all copper tubes. After selecting the inner diameter and the length, the flow rate of the capillary tube depends on the difference between the cooling degree, the return air pressure, the suction pressure, and so on.

The VOF multiphase flow model is used in the five capillary flow simulations, and the performance is simulated and analyzed by FLUENT software. The fluid is the refrigerant R22, the inlet pressure is 1.8 Mpa, the outlet pressure is 0.6 Mpa, the inlet refrigerant temperature is 314.15 K, and the outlet temperature is 279.16 K.

Capillary tube is a slender structure; the length is greater than the diameter, if only using unstructured grid and drawing the number of grid will be too much; it is easy to exceed the limits of computer processing, so here structured grids are used, internal for hexahedral grid and external for tetrahedron, mesh model as shown in **Figure 4**.

At the inlet pressure which is 1.8 Mpa, the saturation temperature of the refrigerant R22 is 47°C; before entering capillary refrigerant is supercooled. Of five kinds of capillary tube in experimental conditions, the coolant temperature in the entrance of the capillary tube is 41°C, namely, supercooling degree is 6°C; the five models of capillary throttling effect comparing simulation diagram are:

**1.** Inlet coolant temperature is 41°C, liquid phase distribution of refrigerant in different types of capillary tubes is as follows:

As you can see from the figure above, the shape of the liquid phase change of the five types of capillary tube is similar; the refrigerant gasification rate is faster in the first half of the tube, with the gradual reduction of the refrigerant in the liquid phase; the amount of heat added to refrigerant gasification is also decreasing; this leads to a gradual decrease in the gasification rate of the refrigerant. Obviously, the volume fractions of the liquid phase of the refrigerant after the throttling are, respectively, 5, 4, 2, 3, and 1. The larger the volume fraction of the liquid in the capillary tube, the less the flash gas caused by throttling, the better the system.

**Figure 5.** The liquid phase distribution of the refrigerant in capillary tube No. 1 was different.

Distilled Water Production by Vacuum Heat Pump http://dx.doi.org/10.5772/intechopen.76839 85

**Figure 5.** The liquid phase distribution of the refrigerant in capillary tube No. 1 was different.

As you can see from the figure above, the shape of the liquid phase change of the five types of capillary tube is similar; the refrigerant gasification rate is faster in the first half of the tube, with the gradual reduction of the refrigerant in the liquid phase; the amount of heat added to refrigerant gasification is also decreasing; this leads to a gradual decrease in the gasification rate of the refrigerant. Obviously, the volume fractions of the liquid phase of the refrigerant after the throttling are, respectively, 5, 4, 2, 3, and 1. The larger the volume fraction of the liquid in the capillary tube, the less the flash gas caused by throttling, the better the system.

84 Desalination and Water Treatment

**1.** The volume distribution of refrigerant liquid phase in the same type of capillary tube:

of increasing with the increase of the supercooling degree. Therefore, under certain conditions, the higher the degree of supercooling, the less flash gas produced by throttling, the higher the volume fraction of the liquid component of refrigerant. Combined with the experiment, the entry refrigerants 41°C under 1, 2, 3, 4, and 5 capillary outlet refrigerant liquid volume fractions are, respectively, 0.599, 0.646, 0.649, 0.646, and 0.649; it has already satisfied the requirements of heat pump system. Comprehensive to the practical situation and design experience of heat pump system, this system selects the supercooling degree at 6°C, and the temperature of the refrigerant before the throttling is 41°C as the temperature

Distilled Water Production by Vacuum Heat Pump http://dx.doi.org/10.5772/intechopen.76839 87

A set of experimental transposition was designed to verify the possibility of producing distilled water by the ejector. The main equipment include compressor (2R11B225A), pump (DP-35), condensation absorber (diameter 100 mm and height 300 mm), water generator (diameter 100 mm and height 300 mm), and capillary (length of 400 mm and diameter 2 mm). The above three kinds of forms of ejector were tested. Ejector C failed to form steam ejector function, so

From **Figure 6**, it can be seen that the minimum pressure of ejector A can reach −0.085 Pa, the corresponding water vapor generator temperature at 50°C, and it can produce very good water vapor ejector effect, meeting the temperature requirements of the condenser of the refrigeration system. The lowest pressure can reach −0.034 Pa, the corresponding water vapor generator temperature at 73°C, but at this temperature, the efficiency of the refrigeration sys-

From **Figure 7**, it can be seen that there is a mixed fluid of water and vapor in the ejectors A and B, while the ejector C produces the backflow, which cannot form an effective water vapor ejector effect. The ejector A is selected as a system unit, and the three different powers of the compressor were used in the production of distilled water. **Table 3** is the amount of distilled

**Figure 6** shows throat pressure of the ejectors (A) and (B) versus time.

water produced by three experiments.

tem will be very low. **Figure 8** is the fluid state of the ejector in the experiments.

of the capillary inlet refrigerant.

**4. Results and discussion of the experiment**

**4.1. Experimental results and discussion of ejector**

In order to study the effect of the cooling degree on the volume fraction of the liquid phase of the capillary export refrigerant, this simulation simulated the distribution of the liquid phase of the refrigerant under five degrees of supercooling for each type of capillary. In this system, before throttlinh the refrigerant saturation temperature is 47°C. And every 3°C, select a temperature value for degree of supercooling and the selected temperatures arerespectively 44, 41, 38, 35, 32, as shown in **Figures 5**. And we only focus on the experiment diagram of capillary No. 1, and the phase volume fraction distribution of refrigerant in the process of capillary throttling of the other four capillary tubes is uniformly expressed in **Table 2**.

According to the results of comprehensive simulation analysis, the volume fraction of the liquid phase of the liquid phase of the five types of capillary tubes shows the trend


**Table 2.** The maximum and minimum position with their volume fractions of other capillaries at different temperatures.

of increasing with the increase of the supercooling degree. Therefore, under certain conditions, the higher the degree of supercooling, the less flash gas produced by throttling, the higher the volume fraction of the liquid component of refrigerant. Combined with the experiment, the entry refrigerants 41°C under 1, 2, 3, 4, and 5 capillary outlet refrigerant liquid volume fractions are, respectively, 0.599, 0.646, 0.649, 0.646, and 0.649; it has already satisfied the requirements of heat pump system. Comprehensive to the practical situation and design experience of heat pump system, this system selects the supercooling degree at 6°C, and the temperature of the refrigerant before the throttling is 41°C as the temperature of the capillary inlet refrigerant.
