Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Abstract The ejector diffuser systems are used in High Altitude Test Facility to measure the thrust developed by motors which operate at very high altitude. In order to simulate the low pressure atmospheric condition ejector diffuser is used in the test facility thus simulating the low pressure environment and to recover the pressure of the exhaust gases before being let out to the atmosphere at ground level. The basic function of the ejector is to pressure recovery and sustaining the vacuum condition inside the test chamber. When the diffuser attains started condition, the momentum of the rocket motor exhaust creates sufficient suction to maintain low vacuum level in the test chamber: also the shock cells occurring in the diffuser duct seal the vacuum environment against backflow. The starting condition of the diffuser itself depends on the operational and geometric parameters of the diffuser and motor.
the rocket exhaust plume may enter into the initially evacuated test chamber and thereby result in an increase of nozzle exit pressure. At high back pressure, the exhaust plume can separate in the nozzle divergent portion and consequently, the full thrust of the rocket motor will not be realized. In order to operate the rocket motor without any flow separation, the low pressure environment corresponding to the flight situation has to be created initially and maintained during the testing, with the help of an external ejector device. However, after the achievement of supersonic flow at the diffuser inlet (known as the "diffuser- started" condition), a properly designed ejectordiffuser system can maintain the vacuum level by utilizing the momentum of the motor exhaust plume itself, without assistance from the external ejector. The "starting" criterion for an ejector- diffuser depends on various factors such as the geometries of the diffuser duct and the rocket nozzle, and the operational parameters for the rocket motor. The theory behind different configurations is being considered for their performance. The performance criterion for an ejector diffuser is early starting of the diffuser. The starting problem of a supersonic diffuser is that the deceleration of supersonic flow in a converging passage is not easy to establish. The initial shock has to be swallowed by manipulation of various geometric aspects of the diffuser. There are three types of ejector diffuser configuration possible. We shall look in to the various aspects of flow through these diffusers for the following cases. 1. 2. 3. Straight Diffuser Second Throat Diffuser Center Body Second Throat diffuser
FFUSERS find application in many fluid systems such as aircrafts, gas turbines, ramjets, wind tunnels, ejector diffusers with a second throat are employed in the testing of large expansion ratio rocket motors in High Altitude Test (HAT) facilities, to recover pressure from a low vacuum condition to the atmospheric value. An ejector- diffuser maintains the required vacuum for testing the motor, by preventing back flow from the atmosphere into the test chamber using the momentum of the rocket exhaust. The demand to enhance the performance of rocket motors at high altitudes has led to the development of cryogenic motors with large expansion ratio. Such motors involve a large degree of jet expansion resulting in a low back pressure (a few tens of milli bars) at the nozzle exit. From the ground testing point of view, a complexity associated with cryogenic rocket motors is that their stagnation pressure build up is very slow and the start up phase takes several seconds. Therefore, during the initial period, a part of
First let us consider the diffusers used in the wind tunnel. If we have to design a supersonic wind tunnel with a test section Mach number of 3 (see fig 1).Some immediate information about the nozzle is obtained from gas tables; at
M=3 Ae / A* = 4.23 and p0 / pe=36.7. Assuming the wind tunnel exhausts to the atmosphere the value of po of the reservoir to drive the tunnel is dependent upon how the nozzle exhausts the gases in to the atmosphere. There are several possibilities. The first is to simply exhaust the nozzle directly to the atmosphere as sketched in fig 2.
second throat diffusers. The first throat being the nozzle throat and that the second throat is the diffuser throat. Here the test section flow mach number Me and static pressure pe is slowed down through a series of oblique shock waves initiated by a compression corner at the inlet of the diffuser, further slowed by a weak normal shock wave at the end of the constant area section, and then subsonically compressed by a divergent section which exhausts to the atmosphere. At the diffuser exit, the static pressure is pd , which for subsonic flow at exit is equal to p. This oblique shock diffuser should provide greater pressure recovery (smaller loss in total pressure) than a normal shock diffuser. At2 / At1 = p01 / p02 Since the total pressure always decreases across shock waves and with in boundary layers, p02 will always be less than p01. Thus from the above equation the second throat must always be larger than the first throat. Indeed if we know the total pressure at two throats then the above equation tells precisely how large to make the second throat. If the second throat At2 is made smaller than demanded by the equation the mass flow through the tunnel chokes and supersonic flow in the test section is not possible. Up to this stage in the discussion, the most serious problem with diffuser has not yet been mentioned, the starting problem. The starting Problem Internal supersonic deceleration in a converging passage is not easy to establish. In fact, design condition can not be achieved with out momentarily over speeding the inlet air or varying the diffuser geometry. This difficulty is due to shock that arise during deceleration process, and it need not be related to boundary layer behavior. Therefore let us neglect boundary layer effects for the moment, while examining the starting behavior of converging diverging diffuser that is one dimensional and isentropic except for losses that occur in normal shocks may be present. This simplified analysis contains the essential features of the phenomena, and it could be a valid representation of a real flow from which the wall boundary layer is carefully removed by suction though porous walls. The fig 6 illustrates successive steps of a fixed-geometry converging-diverging inlet. To isolate the inlet behavior from that of the rest of the propulsion device, we assume that whatever is attached to the diffuser exit is capable of ingesting the entire mass flow. Thus the mass flow rate is limited only by choking at the minimum diffuser area A i. Condition (a) illustrates low speed subsonic operation, for which the inlet is not choked. In this case the air flow through the inlet, and hence the upstream capture area Aa\ At is determined by the conditions downstream of the inlet. In condition (b), though the velocity is still subsonic, the flow is assumed to be accelerated to sonic velocity at the minimum area At and the inlet mass flow rate is limited by the choking condition at At. Since the flow is assumed to isentropic, then At = A*
Figure 6 Successive steps in the acceleration and over speeding of one dimensional supersonic inlet
Thus for these conditions the capture area Aa must be less than the inlet area Ai, and there fore spillage will occur around the inlet. For sonic and supersonic flight speeds the spillage mechanism is necessarily non isentropic. That is, in order to sense the presence of the inlet and flow around it, the spilled air must be reduced to subsonic velocity upstream of the inlet plane. The mechanism for this deceleration is detached bow wave that stands sufficiently far upstream to allow the required spillage. One can imagine the process of establishing the detached shock wave as follows. Suppose that when the air reached supersonic velocity there was no shock. Then flow would enter, with out deviation, the entire inlet area. A i. act as capture area Aa. But for low supersonic Mach numbers (see fig. 7), the allowable capture area is limited by choking at At
principle is easily seen in terms of simple one dimensional analysis. Suppose the inlet is accelerated to the design Mach number MD with the starting shock present, as at (d) in fig.6 and fig 8.
Once the shock is established, the flow entering the inlet is no longer isentropic. Hence when the design Mach number of the air flow is reached, as at (d) in fig 6 the mass flow cannot pass through the throat area At This follows from the fact that that the choked mass flow through a given area At is Aa proportional to p0 , and from the fact that the fluid suffers stagnation pressure loss in traversing the shock. The result plotted in fig.7 , indicates that both the inlet area Ai will remain too large and spillage will continue even beyond the design Mach number MD ,unless the inlet can be over speeded to a Mach number MO . At this Mach number, the inlet is capable of ingesting the entire incident mass flow without spillage. The shock position will be just on the lip of the inlet, as in fig 6 (e), and a slight increase in speed, as to (e), will cause the shock to enter the convergence. Since the shock cannot attain a stable position within the convergence, it will move quickly downstream to come to rest within the divergence. at apposition determined by downstream condition. The low to the throat the is now isentropic, and the area ratio Aa /A* would be given by (e) in fig.7. The incoming flow is decelerated from Ai. to At, where upon it is re accelerated supersonically in the divergence. Having thus attained isentropic flow in the inlet, the Mach number may be reduced from MO to MD as at fig 6 (f). At exactly the design speed, the throat Mach number would be just unity and isentropic deceleration from supersonic to subsonic flow would exist. Even for this simplified model, however, this condition would be unstable. A slight decrease of flight speed or increase of back pressure would require spillage, and a shock would move rapidly out through the inlet to reestablish condition (d). Thus it might be better to maintain the throat Mach number slightly greater than unity while reaching subsonic velocities through a very weak shock just downstream of the throat. This simplified description contains the essential feature of the starting problem associated with an internally contracting passage. That is, an inlet having Ai. / At greater than 1 will always require spillage upon reaching supersonic air velocities, since Aa / At will always pass through a minimum of 1 just as sonic flight velocity is attained. It is necessary to perform some operation other than simply accelerating to the design speed in order to swallow the starting shock and establish isentropic flow. Over speeding is one such operation which can be utilized in wind tunnels. But not feasible to implement in the case diffuser ejectors used in High Altitude test facility for testing of rocket motors in low pressure environment. If over speeding is not feasible, it might be possible to swallow the shock by variation of geometry at constant gas velocity. This
Figure 8.
Now, if the actual area ratio can be decreased from Ai. / At to the value that can ingest the entire inlet flow behind the shock. The shock will be swallowed to take up a position downstream of the throat. This variation would normally involve a momentary increase
of the throat area from At to a new value At see fig 8. Having thus achieved isentropic flow with in convergence, the throat Mach number Mt is greater than second Throat Ejector Diffuser1, and a relatively strong shock occurs farther downstream. Completely isentropic flow can then be achieved by returning the area ratio to its original value, while operating conditions move from (d) to (f). A geometric variation such as shown in fig 8 would be difficult, mechanically for axis symmetric flow. However geometries that permit the axial motion of a center plug can be used. A schematic of a center body type Second throat Ejector diffuser is shown below.
REFERENCES
[1] Mechanics and Thermodynamics of Propulsion, Philip G. Hill university of British Columbia, Carl R. Peterson, Massachusetts Institute of Technology Manikanda Kumanran. R.Vivekanand, P.K.Sundarrajan., T.Balasubramanian., and Raja Manohar., Numerical Simulation of Ejector Performance in a High Altitude Test Facility 34th National Conference on Fluid Mechanics and Fluid Power, National Society of Fluid Mechanics and Fluid Power, Mumbai, India Dec.2007,pp.402410 Manikanda Kumanran.R.Vivekanand, .K.Sundarrajan,K.Kumaresan and Raja Manohar Optimisation of Second Throat Ejectors for High Altitude Facility, Journal of Propulsion and Power Vol 25,No.3, May June 2009 Annamalai. K., K.Sankaranarayanan, A K.Sriramulu, and Bhaskaran K. A Development of Design Methods for short cylindrical Supersonic exhaust Diffuser Experiments in Fluids, Vol.29, No.4.Oct 2000. pp.305-308 Modern Compressible Flow Jhon. D Anderson Jr Professor of Aerospace Engineering, University of Maryland College
[2]
[3]
[4]
[5]