Propagation of flame whirl at combustion of lean natural gas charge in a chamber of cylindrical shape

Transcription

1 Bronisław SENDYKA* Marcin NOGA Cracow University of Technology, Poland Propagation of flame whirl at combustion of lean natural gas charge in a chamber of cylindrical shape The article presents the results of tests on lean mixtures of high methane natural gas GZ-50 air composition λ=1.4 carried out in a test stand cylindrical combustion chamber. The bases of the chamber were made in form of quartz glass windows. It permitted observation and recording of there occurring combustion process. The combustion process was filmed by use of a video camera of system MiniDV. During investigations a relatively strong whirl of the burning and not combusted charge in the chamber was observed, though, before testing the mixture in the chamber was not in motion. A proper treatment and analysis of obtained pictures permitted to determine the angle of whirl of the flame front and angular velocity of this whirl. Its volume changes with displacement of the flame front to chamber fragments more distant from the spark plug. The formulated conclusions indicate at the influence of ignition plug location initiating the combustion process in the chamber on the degree of gas whirl and at some inconveniences resulting from the applied method of recording of the picture of flame propagation. Key words: Flame, natural gas, lean burn, alternative fuels, combustion INTRODUCTION Investigations of the combustion process of a charge of lean natural gas air mixture was carried out at the test stand presented in fig. 1. The mixture used for studies was prepared in a steel bottle of a high methane natural gas GZ-50 and compressed air. It was characterized by a relative air - fuel ratio λ=1.4. The mixture combustion was initiated by a spark from the spark plug of a car engine of an ignition system of direct type, characterized by increased energy of spark discharge and integration of the coil and power transistor in the one housing. Fig.1. View of the test stand. 1. Pipe supplying the mixture to the combustion chamber. 2. System controlling ignition in the chamber. 3. Combustion chamber with a system of valves permitting filling, pressure control and combustion product discharge from the chamber. 4. Ignition coil with a built in power transistor. 5. Manometer used for control of charge pressure in the chamber The mixture of such and not another composition was not a random choice. Available mixtures of richer composition ( λ=1.0 and λ=1.2 ) combust at much higher rate, whereas, combustion of a composition λ=1.6 gives a much less visible contrast. The applied recording method by use of a digital video camera creates big problems in recording combustion processes of a mixture of another composition than λ=1.4. In the case of richer mixtures a too low filming rate is a problem, whereas for a leaner mixture the recorded picture is less readable and due to it, hard to analysis. The envelope of the cylindrical combustion chamber of 80 mm diameter is made of carbon steel whereas both its bases made of quartz glass 20 mm thick. This permits observation and recording of process occurring inside the bomb. The ignition plug was 424

2 Spalanie/Combustion screwed into an opening made in the envelope of the cylinder. In investigations recording was carried by use of a digital video camera of system MiniDV. The objective was placed at a distance of 150 mm from the plane of the chamber. Its axis was running with the axis of the chamber. The structure and situation of the recorder of the picture is shown in fig. 2. Fig. 2. Scheme of the test chamber structure with marked situation of the camera recording the combustion process INVESTIGATION METHODS Before testing, the chamber was filled with a natural gas-air mixture up to the pressure 0.5 MPa from a bottle in which the pressure reached the value of about 8 MPa. As it has been mentioned, the combustion process was recorded by use of the system MiniDV. This equipment permits to record the picture at a speed of 25 frames of resolution 720 per 576 points per second. The camera was, of course, switched on a few second before initiation of ignition. RESULTS OF CARRIED OUT STUDIES The results of the performed test was a film in which the process of flame propagation took 8 frames. These were presented in fig The structure of the camera determines recording of the picture every 40 ms. The combustion velocity of the natural gas air mixture of composition λ=1.4 is high enough to permit a fairly precise analysis of the obtained picture. Mixtures of a lower coefficient of air excess combust considerably faster and recording of this process by use of a standard video camera does not give expected results. In fig. 3 the moment of spark discharge between the electrodes of a spark plug is visible. Fig. 4 presents the very beginning of the combustion process where the blue flame [2] propagates in spherical form. Analyzing the consequent shots it can be stated that from the moment of about 80 ms following ignition fig. 5 a tendency for asymmetry of front flame propagation occurs. The deflection angle of the center of the flame front from the horizontal direction can be adapted as the measure of this irregularity. The center of the flame front was defined on the basis of the blue image of pre-flame phenomena. This is shown in a schematic form in fig 11. At the beginning ( fig. 4 ) the center line of the flame front is identical with horizontal direction. Subsequently, after about 80 ms from the moment of spark discharge this line shows evident upward deflection fig. 5 and then changes its deflection downward in relation to the horizontal line. This is visualized in figs So, considering the above, one can speak of a whirl in propagation of the flame front of the not whirled (stabile) charge. In the following step we associate the increment of the deflection angle of the center line of the flame front with time interval between consequent shots is reflected by the rate defined as: ω = α t where: ω momentary angular velocity of direction change of the center line of the flame α change of angular deflection of the center line of the flame front from the horizontal level t time interval between consequent shots SILNIKI SPALINOWE/Combustion Engines 2007-SC SC2-176 (P07-C176) 425

3 Fig. 3. Frame 1; t = 0 ms Fig. 4. Frame 2; t = 40 ms Fig. 5. Frame 3; t = 80 ms Fig. 6. Frame 4; t = 120 ms Fig. 7. Frame 5; t = 160 ms Fig. 8. Frame 6; t = 200 ms Fig. 9. Frame 7; t = 240 ms Fig. 10. Frame 8; t = 280 ms 426

4 Spalanie/Combustion Fig. 11 Angle of deflection of the flame front center from the level. Yellow line marks the border between the blue flame and not combusted charge. Red lines shows respectively: perpendicular to the line of the flame front drawn from its center and the horizontal level. α angle between these lines Calculation results of the value of the deflection angle from the center of the flame front in function of time interval from the moment of spark discharge on the spark plug is shown in fig angle angular velocity 400 Angle of deflection of the line of flame front center from the horizontal level α [deg] ,04 0,08 0,12 0,16 0,2 0,24 0, Time interval between the moment of spark discharge on the spark plug elestrodes t [s] Momentary angular velocity of direction change of the line of the flame front center ω [ [deg/s] Fig. 12 Values of the angle of deflection of the direction of the line of the flame front center from the level α and momentary angular velocity of direction change of the center line of the flame ω in function of time from the moment of spark discharge on the ignition plug electrodes CONCLUSIONS FROM INVESTIGATIONS AND CALCULATIONS Analyzing the above presented results it should be stated that combustion of load which wasn t in motion generates whirl of considerable intensity. Most probably it occurs, among others, in consequence of not equal temperature and density of the fresh mot combusted charge of the mixture which is subjected to the combustion process and of gases which are the combustion products. At such a dynamic process as combustion a gradient of pressure is probable to occur in the, actually, closed chamber [5], this also may be cause of the observed whirl moment of the gases. This state may be also influenced by placing the ignition plug at the side of the combustion chamber. In future a possibility of another plug location in the chamber should be considered. It s planed to locate the ignition plug at the top of the chamber and make such a preparation of the plug as to make it possible to initiate ignition from inside of the combustion chamber [1], [3]. Great discount encountered in the carried test is created by low recording efficiency of video cameras. As far as resolution may be regarded satisfactory for such tests, so the rate of filming equaling 25 frames per second limits the cognitive possibilities of investigations. SILNIKI SPALINOWE/Combustion Engines 2007-SC SC2-176 (P07-C176) 427

5 In close future one may expect similar investigations to be carried registering the picture in Schlieren method which gives incomparably better results in studies of fast changing processes which combustion of a gas-air charge in a closed chamber doubtlessly is. REFERENCES [1] Elia, M., Moore, P., Ulinski, M., and Metghalchi, M., Laminar Burning Velocity of Methane-Air-Oxygen-Argon (CH 4 -O 2 -Ar) Mixtures. Proceedings of the ASME Internal Combustion Engine Division, Columbus, Indiana, ICE- Vol. 32-3, [2] Kowalewicz, A., Tworzenie mieszanki i spalanie w silnikach o zapłonie iskrowym. Wydawnictwa Komunikacji i Łączności, Warszawa [3] Rahim, F., Ulinski, M. and Metghalchi, M, Burning Velocity Measurements in Spherical and Cylindrical Vessels. Proceedings of the ISME Conference, 2001 [4] Rychter, T., Teodorczyk, A., Modelowanie matematyczne roboczego cyklu silnika tłokowego. Państwowe Wydawnictwo Naukowe, Warszawa [5] Sendyka, B., Noga, M., Flame Propagation Velocity of Natural Gas Air Mixtures in Combustion Chamber, Proceedings of the FISITA 2006 World Automotive Congress, Yokohama,