الفهرس 
AcknowledgementOnly 14 pages are availabe for public view
 |  | from | 236 |  |  |
| | | from | 236 | | |
Abstract
The pulsed ejector is used in pumping and thrust augmentation applications. It employs the unsteady, intermittent (pulsed) primary flow to generate repeated pressure exchange cycles in the augmenter tube in order to augment both the primary mass flow rate and thrust. Unlike steady flow ejectors, which augment the primary flow by virtue of viscous mixing, the pulsed ejector depends on wave action that becomes reversible under ideal conditions and, therefore, appears to offer higher efficiency. Also pulsed ejectors possess a unique quality which is the capacity to tremendously augment the primary flow at exceptionally small geometries and consequently weights. This quality is not existent in steady flow ejectors. The design and development of the pulsed ejectors has proceeded largely by trial and error, a method that is time consuming, costly and does not garantee an optimum design. Therefore, this study probes in depth towards a complete understanding of the phenomena intrinsic in the pulsed ejector tubes.
In the theoretical part of this study, a complete one dimensional numerical model is developed. The model is based on the modified method of characteristics of Spalding. The method has been selected after careful and thorough consideration of various numerical techniques applied to a physical initial value problem. For the first time, the proposed numerical model is developed to include such influential parameters as the entropy interfaces viscous and heat transfer effects. Therefore it is constructed with sufficient generality to include all significant processes which occur in the pulsed ejector. It is also incorporated with adequate
boundary conditions for the two flow fields of both the primary and augmenter tubes. In the experimental part of this work an experimental setup is constructed with the intention of providing experimental results for companson with the corresponding numerical predictions. It is also used to help understand, and therefore numerically describe, many of the phenomena within the pulsed ejector. Excellent correlation and agreement is demonstrated between the experimental measurements and the numerical results. Consequently it is concluded that the model is capable of interpreting the phenomena within the ejector tube with great accuracy. The proposed model is finally used to perform a systematic theoretical investigation of various geometric and non- geometric (performance) design parameters along with their effects on the performance of the unsteady pulsed ejector. The investigation shows that the performance of the pulsed ejector IS dependent on both the primary flow configurations, which is responsible for the generation of strong pressure wave action in the augmenter tube and also on the capacity of the augmenter tube itself to allow for the propagation of the generated pressure waves in a tuned sequence in order to augment the primary flow to a maximum. The augmenter length to diameter ratio is shown to vary according to the primary flow frequency. While an optimum augmenter to primary tube area ratio of 4.15 is obtained, from a thrust augmenting view point, no optimum area ratio could be reached on a mass flow rate ratio basis. The augmenter divergence angle could not be fully investigated due to departures from one dimensional flow behaviour.