Is it feasible to use ammonia as a refrigerant in centrifugal compressors?

Many use centrifuges to compress ammonia

Some people say that due to its low molecular weight, ammonia is less cost-effective to use in centrifuges compared to fluorine, in terms of COP. Is that the case?

Comparison and Discussion on the Performance of R717 and R22 in Large-Scale Commercial Refrigeration Systems. Publication date: 2014-7-2. Number of views: 2907. Comparison and Discussion on the Performance of R717 and R22 in Large-Scale Commercial Refrigeration Systems. By Huang Jinsong and Li Xiaohu from the National Institute of Trade Engineering Design; Liu Xin from Beijing Huashang Construction Supervision Co., Ltd. Abstract: By comparing the performance of R717 and R22, as well as considering the actual operational requirements and environmental protection needs of these two types of refrigeration systems, this paper further emphasizes the necessity of using R717 as a refrigerant in large-scale commercial refrigeration systems. Keywords: R717, R22, ozone layer, greenhouse effect. I. Introduction Since its introduction in 1936, R22 has been widely used in fields such as commercial refrigeration and air conditioning. As a medium-temperature refrigerant, R22 has an ODP of 0.055 and a GWP of 0.34; it belongs to the HCFCs category of refrigerants. The Vienna Conference in 1995 mandated that developed countries cease using it by 2020 (with the EU setting a earlier deadline of 2015) ； In developing countries, production was frozen in 2016, and its use will be discontinued by 2040. The significant early enforcement of the ban on HCFCs will make it extremely difficult for Chinese companies to phase out the production and maintenance of R22 in a few years. For the above reasons, the use of R22 was somewhat restricted compared to R717 in the previous period. II. New developments in China’s refrigeration industry in recent years: In recent years, with the emergence of multi-head parallel units, the use of R22 in large-scale commercial refrigeration systems has not been affected by restrictions; on the contrary, its usage has shown an upward trend year by year. There have also been debates in China regarding which refrigerant should be used in large-scale commercial refrigeration systems. This phenomenon should draw the attention of the refrigeration industry. Based on an analysis of this phenomenon, the main reasons are as follows: 1. Compared with R717 refrigeration systems, R22 systems have advantages such as simpler design, less space requirement, easier operation, and greater ease in achieving automatic control of the refrigeration system. 2. Some production facilities are located in urban areas, and for reasons related to production safety, R22 is also used as a refrigerant in large commercial refrigeration systems. 3. Thanks to its high degree of automation and the ease of implementing individual unit energy regulation, the refrigeration unit can operate under relatively economical conditions. 4. Low maintenance costs, etc. Taking all these factors into account, although R22 has a certain impact on the ozone layer and the greenhouse effect, it is only used as a temporary substitute fluid. Moreover, the initial investment for R22 refrigeration systems is about 30–40% higher than that for R717, but some companies still choose R22 as the refrigerant considering their own economic interests and return on investment. III. Performance comparison and theoretical calculations of R717 and R22. Ammonia (R717) is a substance that exists naturally; it has been used as a refrigerant for over 100 years. It is one of the most widely used medium-temperature refrigerants in China today. It does not damage the ozone layer, has no greenhouse effect, offers good thermal performance, is inexpensive, and helps save energy. Therefore, the advantages of ammonia as a refrigerant are unmatched by other refrigerants, and it is widely used both domestically and internationally. Its biggest drawback is the potential impact on people and goods due to leaks, but these problems can be significantly reduced through improvements to machinery and refrigeration systems. 1. Comparison of R717 and R22 in terms of environmental impact Table 1: ODP and GWP values of refrigerants. R717: 0.00; R22: 0.04–0.06 and 0.32–0.37. With current technological levels, refrigerants with an ODP value of less than or equal to 0.05 and a GWP value of less than or equal to 0.32 can be used as temporary substitutes. From an environmental perspective, ammonia is undoubtedly a natural refrigerant, and its use contributes to environmental protection. 2. Comparison of the latent heat of vaporization between R717 and R22 Table 2: Refrigerant temperature, R717 (KJ/kg), R22 (KJ/kg); Ratio. -10°C: 1295 vs 213.1, ratio of 6.07; -30°C: 1358 vs 227.76, ratio of 5.96; -40°C: 1387 vs 234.00, ratio of 5.93; -45°C: 1401 vs 237.15, ratio of 5.9. As can be seen from Table 2, the latent heat of vaporization of R717 is more than 5 times that of R22. To produce the same amount of cooling, the amount of ammonia required for circulation is less than 1/5 of that needed for R22. Due to the large circulation volume of R22, the motor power consumed by its refrigerant transfer pump is greater than that of R717. 3. Comparison of the gas specific weight of R717 and R22 Table 3: Refrigerant temperature, R717 (kg/m3), R22 (kg/m3), and ratio -10°C: 2.397, 15.29, 6.37 -20°C: 1.605, 10.76, 16.7 -30°C: 1.039, 7.35, 47 -45°C: 0.498, 9.388, 7.8 As can be seen from Table 3, under the same operating conditions, the gas specific weight of R22 is more than 6 times that of R717; therefore, with a compressor of the same model, the circulation volume for R22 is more than 6 times that for R717. Since R22 is heavier than R717, it experiences greater resistance loss in the return gas flow, and its suction temperature is also lower. 4. In terms of heat transfer performance, R717 has a heat release coefficient outside the tubes that is about 4 times higher than that of R22, and its condensation heat transfer coefficient inside the tubes is about 3 times higher. Therefore, under the same system cooling conditions, the heat transfer coefficient of the condenser for R22 is lower than that for R717. 5. Comparison of various parameters for R717 and R22 when using refrigeration compressors of the same model under the same operating conditions: The following shows the comparison of several parameters for an open-type 16-screw compressor with an economizer of the same model, at a condensing temperature of 40°C under different operating conditions: Table 4 Evaporation temperature (°C) R717 R22 Cooling capacity (KW) Q0 Output power (KW) Ne Coefficient of performance ε Circulation rate (kg/h) Cooling capacity (KW) Q0 Output power (KW) Ne Coefficient of performance ε Circulation rate (kg/h) -35 10 4.87 6.8 1.36 52 88.01 21.78 0.41 1.51 32 166.0 -30 138.78 4.11 1.64 93 84.21 55.68 8.21 1.76 42 785.1 -25 180.39 1.81 1.96 35 01.81 95.49 5.62 0.04 43 517.6 -20 229.39 8.92 2.31 96 44.12 40.11 01.62 2.36 34 376.9 -15 287.11 05.32 2.72 78 14.32 91.21 06.62 2.73 15 377.1 -10 306.71 04.22 2.94 51 013 291.51 02.22 85 36 516.3 -5 380.51 08.63 3.50 51 250 355.41 05.43 3.37 37 845.4 6. Comparison of several parameters for R717 and R22 in two-stage compression: The following shows the comparison of operating parameters for the two refrigeration systems when using 3 open-type 16-screw compressors as the low-pressure stage and 1 open-type 16-screw compressor as the high-pressure stage: Table 5 Evaporation temperature (°C) R717 R22 Cooling capacity Q0 (KW) Output power Ne Coefficient of performance ε Circulation rate (kg/h) Cooling capacity (KW) Q0 Output power (KW) Ne Coefficient of performance ε Circulation rate (kg/h) High-pressure stage Low-pressure stage High-pressure stage Low-pressure stage -28 537.31 11.54 0.32 2.31 21 630.95 47.09 99.05 4.02 0.96 11 1144.9 -30 493.91 112.037.32 2.20 51 491.05 09.11 01.25 0.92 0.05 10 293.7 -35 397.51 111.430.51 1.96 15 20.64 23.61 05.24 3.31 1.80 84 05 -38 347.41 09.826.61 1.83 33 1024.43 77.61 06.439.01 1.69 02 410.4 -40 316.91 08.424.31 1.74 99 29.33 76.81 06.836.41 1.61 46 997.7 -45 248.81 03.619.31 1.54 12 34.82 93.71 06.530.41 1.43 75 440.6 The above results are obtained by using the calculation programs provided by the equipment manufacturers to determine the operating parameters of R717 and R22. By analyzing the data presented in Tables 1–5, it can be seen that when t0 < -15°C, using a single-stage screw compressor with an economizer results in a higher coefficient of performance for R22 compared to R717; whereas when t0 > -15°C, R717 has a higher coefficient of performance than R22. And when a two-stage compression is required because the evaporation temperature is below -28°C, the coefficient of performance of R717 in two-stage compression is higher than that of R22. The difference in their coefficient of performance ranges from 2% to 13%. The performance of the two refrigerants does not differ much. IV. Comparison and Analysis of Actual Operating Conditions Some domestic equipment manufacturers conducted freezing tests on R22 refrigeration systems that use multiple compressors in parallel, as well as on ammonia refrigeration systems operated manually in some older factories. The test results showed that, with similar temperatures during the loading and unloading of meat, the electricity consumption per ton of meat frozen was approximately 90 kWh for the R22 refrigeration system and around 135 kWh for the R717 refrigeration system. Given the performance differences between these two refrigerants, such a large difference in energy consumption should not exist; the root cause lies in certain design differences between the two refrigeration systems being tested: 1. The R22 system is a fully automatic refrigeration system equipped with automatic energy regulation. The R717 refrigeration system tested employed manual operation. The R22 system can operate under relatively economical conditions at all times, as required by the design, whereas the R717 system tested could only be operated relying on the experience and skills of the workers. 2. The R22 refrigeration system uses multi-head parallel units, with loading and unloading being based on the number of units; in contrast, the R717 refrigeration system under evaluation relies on cylinders or slide valves for energy regulation. The reduction in cooling capacity and power consumption resulting from machine unloading is not proportional. 3. The air coolers in the R22 system library use imported axial flow fans. The R717 refrigeration system under test uses domestic motors. For the same 25t/day freezing chamber, the R717 refrigeration system is equipped with domestic motors having a power of 33KW, whereas the R22 refrigeration system uses imported axial flow fans supplied by the manufacturer based on factors such as cooling capacity, range, and pressure loss; these fans have a motor power of only 11KW. In each freezing room, the R717 refrigeration system causes the axial flow motor to consume an additional 22 KW of electricity per hour. This also adds 22 KW of heat to the interior of the warehouse. In addition, the refrigeration unit needs to generate an extra 22 KW of cooling capacity, resulting in a total power consumption of 16.2 KW. Therefore, 22 KW/h + 16.2 KW/h = 38.2 KW/h. Multiplying this by 20 hours gives 764 KW. Thus, each ton of frozen meat requires an additional 764 ÷ 25 = 30.5 kWh of electricity. 4. In the R22 refrigeration system tested, the manufacturer set the difference Δt between the storage temperature and the evaporation temperature of the refrigerant at 7°C, while in the R717 system under test, Δt was 10°C. During the operation of a refrigeration system, for every 1°C decrease in the evaporation temperature, energy consumption increases by 3% to 6%; in low-temperature conditions, a 1°C decrease in the evaporation temperature occurs. The cooling capacity has decreased by more than 6%. At 5,000 kcal per ton of meat for freezing (5.81 KW/h), 25 tons require a total of 145.3 KW. 145.3 KW × 6% = 8.718 KW. If freezing takes 20 hours each time, and the evaporation temperature is 3°C lower, the additional electricity consumption per ton of frozen meat is 3 × 111.7 KW ÷ 25 tons = 13.4 KW/h. 5. In the R717 refrigeration system under test, the temperature of the frozen products at different locations in the freezing chamber was highly uneven, with the maximum temperature difference ranging from -1.5°C to -25°C. This indicates that there are serious problems with the air flow organization in the freezing chamber of this design, and the freezing time varies depending on the location of the measurement points. Based on the analysis of the two refrigeration systems outlined above, due to factors such as the configuration of the cooling fan motor, the evaporation temperature, and automatic energy regulation, the additional electricity consumption required by the R717 refrigeration system for freezing 1 ton of meat is: 30.5 KW + 13.4 KW = 43.9 KW. Based on the above analysis, the tested R22 refrigeration system and R717 refrigeration system differ by more than 45 kWh in power consumption per ton of meat frozen; the issue lies not in the performance of the two refrigerants, but rather in the differences in the design of the two refrigeration systems. Similarly, if R22 refrigeration systems are not properly designed and managed, their power consumption per ton of meat frozen can exceed 135 kWh. V. Performance comparison of R717 and R22 in the same freezing system. The following is a comparison of the performance of R22 and R717 in two identical refrigeration systems. Both systems are equipped with exactly the same spiral freezers and refrigeration units; the only difference is that one system uses R717 as the refrigerant while the other uses R22. The comparison results show that the compressor in the system using R22 consumes about 14% more shaft power per unit of work compared to the R717 system. The main reasons for this are various issues associated with the use of R22, such as problems related to heat exchange efficiency, compressor efficiency, and pressure drops in the piping. Since the machinery in both systems is identical, the difference in performance is directly attributed to the refrigerant used. Below are the comparison results of two refrigeration systems: Table 6 compares the amount of freezing capacity and the cooling power required. System R717 R22: Freezing capacity in kg/h – 60, 2610; Total shaft power of the compressor in KW – 57.4, 68.5; Shaft power required per ton of frozen product in KW – 95, 112. Table 7 compares the performance of the evaporators in the freezers. System R717 R22: Average evaporation temperature at the evaporator return pipe in °C – -41.6, -43.6; Air inlet temperature in °C – -38.6, -39.2; Average refrigerant circulation rate required per ton of frozen product – 7.3:1, 14.0:1; Temperature drop caused by the average refrigerant pressure drop in °C – 0.6, 0.9; Cooling power in KW – 85.7, 86.3. Table 8 compares the saturated temperatures in the low-pressure receiver. System R717 R22: Saturated temperature in the low-pressure receiver in °C – -42.3, -44.9; Temperature drop in the wet vapor return pipe in K – 0.7, 1.3. Table 9 compares the temperature drops in the return piping. System R717 R22: Mass flow rate in kg/s – 0.0686, 0.467; Vapor flow velocity in the pipes in m/s – 14.4, 14.0; Temperature drop in the suction pipe in K – 0.1, 0.5; Total temperature drop in the suction piping from the evaporator to the compressor – 0.8, 1.8. Table 10 compares the performance of the evaporative condensers. System R717 R22: Heat transfer rate in KW – 140.2, 151.4; Wet-bulb temperature in °C – 25, 25; Condensation temperature in °C – 33, 35. Table 11 shows the average operating conditions of the compressors. System R717 R22: Compressor suction temperature in °C – -42.4, -45.4; Intermediate cooling temperature in °C – -12, -9; Compressor discharge temperature in °C – 76, 63; Cooling power in KW – 85.7, 86.3; Shaft power of the first-stage compressor in KW – 26.1, 33.7; Shaft power of the second-stage compressor in KW – 31.3, 34.8; Total shaft power of the compressor in KW – 57.4, 68.5; Coefficient of performance (COP) – 1.49, 1.26. From the comparisons in Tables 6 through 11, it can be seen that: 1. The significant advantage of R717 over R22 lies not only in the different efficiency levels of their compressors but also in the fact that R22 has poorer thermal properties and heat transfer capabilities compared to R717, resulting in lower performance of other key components in the system compared to those using ammonia. (See Table 6) 2. At the evaporator return air header, the temperature drop caused by the pressure drop is approximately 2K. This reduces the cooling capacity of the compressor. 3. The high power consumption of R22 systems increases the heat transfer load on the condenser. Moreover, as can be seen from Table 10, the efficiency of R22 evaporative condensers is lower than that of R717. All these drawbacks result in the evaporator efficiency of R22 being lower than that of ammonia. 4. When the machines of the two systems operate under exactly the same conditions, the coefficient of performance of the R22 units is about 18% lower than that of the ammonia units. VI. Recommendations Based on a comparison of the properties of R717 and R22, as well as a theoretical analysis of the operating parameters of compressors and a practical comparison of the performance of these two refrigeration systems, we believe that: 1. Without considering the impact on the atmosphere, both R717 and R22 are excellent medium-temperature refrigerants that are widely used worldwide. However, in an era where countries around the world are focusing on protecting the ozone layer, for the sake of future generations, we believe that R717 should be promoted as a refrigerant in large-scale commercial refrigeration systems, free from constraints imposed by the manufacturing environment and other conditions. 2. In the design of the R717 refrigeration system, some of the advantages of the tested R22 refrigeration system design should be incorporated. Necessary measures should be taken in terms of automation level, automatic energy regulation, selection of the motor power for the cooling fan, and setting of the evaporation temperature, so as to fully utilize the advantages of R717. 3. Due to the use of cylinders or slide valves for unloading, the reduction in cooling capacity of the compressor is not proportional to the energy savings achieved by the motor. In particular, when screw compressors are operated at less than 80% load, the decrease in their output current is minimal. Tables 12 and Figure 1 show the records of the unloading level and output current of the JZLG20 screw compressor during unloading operation, as part of the automatic control system of the R717 refrigeration system used in the Wangkui project by our institute. This system is single-stage with an economizer; the temperature t0 is -35°C, the motor power is N=200 KW, and the rated current is 370 A. As can be seen from Table 12 and Figure 1, when the machine operates at a load of less than 80%, the current changes very little; when the load is below 40%, the current remains essentially constant. Therefore, multi-head parallel units suitable for R717 refrigeration systems should be developed to improve the energy-saving efficiency of such systems during automatic energy regulation. An ammonia two-stage compression refrigeration system with multiple parallel compressors allows for the separate control of the number of compressors operating in the high-pressure and low-pressure stages, based on the pressures in the low-pressure cycle tank and the intercooler. To enable it to operate within a relatively economical pressure range. Table 12: Energy regulation level %, Output current (A). Energy regulation level %, Output current (A): 100, 224, 69.9144; 94.5, 160, 61.2138; 91.3, 160, 52.8130; 86.5, 160, 40.3112; 80.1, 144, 32.8112; 76.9, 144, 28.2112. Figure 14: In the design of freezing chambers, it is necessary to organize the airflow within the chamber according to the freezing method, in order to ensure uniform temperature throughout the chamber and avoid differences in freezing time due to varying locations of measurement points. VII. Conclusion: It is more logical than ever to advocate the use of a refrigerant that is completely harmless to the environment and possesses good thermal properties. Ammonia has been used in industrial refrigeration for over a hundred years, and most of its drawbacks are already known; therefore, there are no issues regarding the safety and reliability of ammonia-based equipment. Wider use of ammonia as a refrigerant also requires a change in people’s perception of it, along with improvements to its applications and equipment. These improvements will increase costs, but they are far lower than the costs associated with finding an alternative refrigerant. Moreover, the lifespan of such refrigerant substitutes may be quite short. The situation is different for ammonia, which is a fluid that exists naturally; it does not damage the ozone layer nor exacerbates the greenhouse effect. It has good thermal properties, and its price is several times lower than that of any HCFC, CFC, or HFC refrigerants. By comparing the two refrigerants and conducting data analysis, this article provides strong evidence, both from an economic and a technical perspective, in favor of using R717 rather than R22 as a refrigerant.

Thank you so much for sharing； However, for centrifugal compressors, is there any impact from the molecular weight of the refrigerant? In other words, for those with a low molecular weight, the effect of centrifugation is relatively less noticeable?

Thank you so much for sharing； However, for centrifugal compressors, is there any impact from the molecular weight of the refrigerant? In other words, for those with a low molecular weight, the effect of centrifugation is relatively less noticeable?

Ammonia compressors in large-scale ammonia synthesis plants

After several incidents involving ammonia as a refrigerant, it was used less often in new projects. As for the so-called evaluations, those who took part in making decisions probably resorted to tactics of shirking responsibility, simply rejecting such uses in a crude manner.