Chemistry and Technology of Fuels and Oils, Vol. 40, No. 1, 2004
EJECTOR SYSTEMS E. V. Karmanov, Yu. N. Lebedev, V. G. Chekmenev, and I. A. Aleksandrov
UDC 621.68
In practice, methods based on use of nomograms that correlate the ejection factor with the degree of expansion of the ejecting gas and degree of compression of the ejected gas are used in designing ejectors for oil refineries and petrochemical plants [1]. However, there are no references as to how this dependence was constructed in the descriptions of these methods. For this reason, it is not possible to assess the legitimacy of using it for systems with thermophysical properties different from the vaporair mixtures for which it is recommended. These methods recommend increasing the calculated flow rate of a high-pressure gas by some constant in designing ejectors for evacuating and compressing gases. This frequently leads to an unsubstantiated increase in the cost of the process or selection of ejector design dimensions which are far from optimum. In addition, it is frequently necessary to calculate ejectors that operate with vapor and gas streams with an extremely wide range of thermophysical properties: from hydrogen to heavy crude distillate vapors. Vapor and gas ejection processes with consideration of the real thermophysical properties have been calculated in many studies [2-4]. Not only the theoretical principles of the process at subsonic and supersonic flows but also selection of ejector design parameters for implementation of so-called calculation operating regimes characterized by the appearance of critical sections in different parts of the ejector are examined in these studies. However, serious problems arise in using the calculation equations in these studies for designing industrial ejector systems. One of the most serious problems is the necessity of calculating the function of state of complex mixtures of real gases in the wide ranges of pressure and temperature variations that take place in industrial ejectors. Simulation systems such as HYSYS and PRO II, widely used in designing oil refining processes and having a broad base of thermodynamic properties both for individual substances and for hypothetical components of continuous crude oil mixtures, can be used to solve this problem. These systems can be used to simulate not only the elementary but also the complex processes that take place with a change in the phase state or chemical transformations. Their elementary units are expanders, compressors, mixers, etc., that simulate expansion, compression, and mixing processes and combined with specially developed commercial functions and subroutines, they allow creating a model of relatively complicated processes, including the ejection process. A diagram of the experimental ejector model consisting of nozzle, mixing chamber, and diverging part is shown in Fig. 1. Figure 2 shows a diagram of the calculated model of the ejection process based on the HYSYS simulation system with material I, II and energy III-VII flows. Expansion of high-pressure (ejecting) gas I in the nozzle is simulated by two successive expanders 1 and 2 for calculating the flow parameters in the converging and diverging diffuser sections. High-pressure gas internal energy flows III and IV, converted into motive kinetic energy in expansion in the converging and diverging diffuser sections, are calculated. For calculation the area of the critical section (throat), the condition of equality of the flow rate in the throat to the sound velocity is used. ____________________________________________________________________________________________________ Kedr-89 SIC. Translated from Khimiya i Tekhnologiya Topliv i Masel, No. 1, pp. 62 64, January February, 2004. 80
0009-3092/04/40010080Ó2004Plenum Publishing Corporation
Fig. 1. Diagram of the ejector: 1, 2, 3) diffuser converging section, throat (critical section), and diverging section; 4) mixing chamber; 5, 6, 7) converging section, throat (cylindrical part of ejector), and diverging section. The ejector mixing chamber is simulated by mixer 3, where ejecting gas flow I , after coming out of the nozzle, and low-pressure (ejected) gas flow II enter. Heating of the mixed gas stream by loss of kinetic energy during mixing is simulated by compressor 4. The amount of energy released is determined by flow V. Compression in the converging and diverging diffuser sections are simulated by two successive compressors: compressor 5, for calculating compression of the mixture of ejecting and ejected flows in the section from the outlet of the mixing chamber to the throat of the diverging part of the ejector, and by compressor 6 in the section from the throat to exit of flow from the diffuser. Flows VI and VII of the kinetic energy of the mixture of gases converted into internal energy in compression in the converging diffuser and diffuser are also determined. The HYSYS simulation system does not allow calculating the gas-dynamic functions, critical section areas, and momentum balance, so that commercial module 7, which implements all missing functions and conveys the results of the calculation to standard objects of the simulation system, is included in the model of the ejection process. The processes that take place in the ejector can be considered adiabatic only in a first approximation. The heat exchange between the gas stream and ejector walls can be neglected with no large error, but the losses 81
,,
, ,,,
,9
9
9,
9,,
Fig. 2. Diagram of calculation of an ejector in the HYSYS modeling program: 1, 2) expanders simulating expansion of high-pressure flow in a nozzle before and after the critical section; 3) mixer; 4) compressor simulating heating in loss of kinetic energy during mixing; 5, 6) compressors simulating compression of flow in the converging diffuser and diverging diffuser in the ejector; 7) module for calculating commercial equations and returning data to simulated objects; I) high-pressure gas; II) low-pressure gas; III, IV) internal energy of high-pressure gas converted into motive kinetic energy in expansion in converging diffuser and nozzle diffuser; V, V, VII) kinetic energy of mixture of gases converted into internal energy in loss during mixing, compression in converging diffuser and compression in diffuser. in kinetic energy near the ejector walls due to viscous friction forces cannot be neglected, since this can lead to important errors. For this reason, expansion and compression processes in expanders and compressors must be calculated with consideration of the so-called adiabaticity factor. This factor reflects the proportion of the kinetic energy not converted into heat, i.e., not lost in the process, and can only be determined experimentally, since its theoretical calculation is too complicated. Experimental correction factors were introduced in the equation for calculating the flow rates in an ejector in [2]. These factors insignificantly decrease the calculated rate in comparison to the gas flow rate in the condition of absence of losses. The squares of these factors characterize the proportion of kinetic energy not dissipated in the process. This proportion can be used as the adiabaticity factor. The standard mixing object in simulation systems allows calculating the flow mixing process in consideration of the laws of conservation of mass and internal energy but does not make it possible to describe the condition of conservation of momentum. Since calculation of the ejector mixing chamber is impossible without satisfaction of the condition of conservation of momentum and the law of conservation of energy with consideration of the kinetic energy of the gases, it is performed separately in special commercial module 7 of the program (see Fig. 2), where implementation of commercial functions and transfer of the results to other simulation objects are possible. For describing the condition of conservation of momentum, it is necessary to know the pressure along the wall of the ejector mixing chamber and more precisely, the pressure integral over the surface of the chamber. The equations for approximate calculation of the latter are reported in [2]. As for calculation of the kinetic energy of the flows with consideration of the law of conservation of energy, it is necessary to know the energy released in the expanders and consumed in the compressors for the different ejector sections in order to calculate the nozzle and diverging part. 82
This model can be used in designing ejectors with a broad spectrum of ejected and ejecting gases without having to resort to direct solution of unwieldy equations and calculation of functions of state of complex mixtures. The proposed method of calculation was tested in processing the data from bench tests of an ejector in the airair system at Penzkhimmash OJSC [5] and in processing the published data on operation of ejectors in heliumair and airhelium systems. The vapor-ejector vacuum system on the EDUAVT-2 unit at Rosneft Oil CompanyKomsomolsk Oil Refinery OJSC is operating successfully, guaranteeing the rated indexes with significant savings of steam in comparison to the results of calculating an ejector system with the All-Russian Scientific-Research Institute of Petroleum-Related Machine Engineering. The proposed method is also used in designing a; start-up ejector in a hydrotreating unit and the sulfur collector blowing ejector in the sulfur production unit in the exhaustive crude refining complex at Komsomolsk Oil Refinery. REFERENCES 1. V. N. Ramm, Vapor-Jet Vacuum Ejector Units [in Russian], Goskhimizdat, Moscow-Leningrad (1949). 2. E. Ya. Sokolov and N. M. Zinger, Jet Units [in Russian], 3 rd ed., Energiya, Moscow (1989). 3. G. N. Abramovich, Applied Gas Dynamics [in Russian], 4 th ed., Nauka, Moscow (1976). 4. M. E. Deich, Industrial Gas Dynamics [in Russian], Energiya, Moscow (1974). 5. E. V. Karmanov et al., in: Proceedings of the Scientific and Industrial Conference Current Problems in t h e St a t e a n d D e v e l o p m e n t o f t h e R u s s i a n O i l a n d G a s C o m p l e x , M o s c o w, January 23-24, 2003 [in Russian], I. M. Gubkin RGU Nefti i Gaza, Moscow (2003), p. 58.
83
