Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 80
PROCEEDINGS OF THE FOURTH INTERNATIONAL CONFERENCE ON ENGINEERING COMPUTATIONAL TECHNOLOGY
Edited by: B.H.V. Topping and C.A. Mota Soares
Paper 84

Axi-Symmetric Supersonic Ejector for High Exit Pressure

N. Kameda, E. Spiegler and M. Wolfshtein

Faculty of Aerospace Engineering, Technion, Israel Institute of Technology, Haifa, Israel

Full Bibliographic Reference for this paper
N. Kameda, E. Spiegler, M. Wolfshtein, "Axi-Symmetric Supersonic Ejector for High Exit Pressure", in B.H.V. Topping, C.A. Mota Soares, (Editors), "Proceedings of the Fourth International Conference on Engineering Computational Technology", Civil-Comp Press, Stirlingshire, UK, Paper 84, 2004. doi:10.4203/ccp.80.84
Keywords: supersonic turbulent flow, ejector pump, hysteresis, flow conditions.

Summary
The paper is concerned with supersonic turbulent flow inside an ejector pump. Usually such devices are used to pump large fluid volumes at low pressure-ratio between the inlet and outlet of the ejector. In the present study the opposite case of very high total outlet pressure relative to the secondary inlet stagnation pressure and very low flow rates is examined. Such an ejector may be used to generate very high vacuum. The motivation for this research came from a study of the generation of hydrogen by dissociation of very hot water vapor under very high vacuum into hydrogen and oxygen, where the ejector is the only device which can pump the very hot fluids used in the process.

The goals of the study are to (i) To understand the structure of the flow; (ii) To identify the important mechanisms affecting this flow; (iii) To seek the upper limits of the pressure ratio possible in such devices.

The flow inside a supersonic ejector is rather complicated, with systems of shock waves, expansion waves and separation bubbles prevailing inside the ejector. A separation bubble near the exit from the primary nozzle grows when the pressure ratio increases until it completely blocks the ejector. Thus conditions of zero flow rate or even reverse flow rate can occur. In some cases unsteady flow develops, which may be harmful to structural elements.

The problem was attacked using a commercial flow solver (INCA). The procedure chosen was to apply a time-like iterative procedure until a steady state was reached. Usually the solution of a previous run was chosen as an initial condition. It turned out that the direction of change of the parameters had a significant effect on the results for some values of the parameters. In such cases the choice of high initial primary stagnation pressure gave different result from a choice of low initial primary stagnation pressure. Thus the problem appears to have non-unique solutions, and the process exhibits hysteresis. Therefore the problem is of interest to the theoretician who is interested in the conditions generating this phenomenon, and to the engineer who needs to ensure certain performance of the device.

The results suggest two mechanisms for physical instability of this flow:

  1. When two hysteresis solutions exist for the same boundary conditions it is possible that small disturbances may cause fluctuations of the solution between these two solutions.
  2. At certain flow conditions no steady solution was obtained. An unsteady test- run demonstrated a periodic solution in which the flow rate fluctuates continuously between a positive and a negative value.

Both these cases may pose great difficulties to the mechanical designers due to periodic fluctuations of the stresses inside the structure. On the other hand, the development of any of theses solutions is highly sensitive to boundary conditions, and the reliability of the solution depends on very careful numerical treatment. It may be pointed out that reliability of the solution in such cases relies heavily on the numerical accuracy and on the validity of the turbulence model chosen.

The major conclusions are

  1. It is essential to perform detailed numerical studies of such flows, as it is extremely expensive in both time and money to perform the required parametric studies experimentally, in order to choose a reliable working design.
  2. The high sensitivity of such flows requires a very careful numerical investigation, and in particular attention should be given to numerical accuracy and to situations of more than one solution.
  3. Any such solution requires the application of certain physical models (turbulence model in the present case, two phase flow or combustion models in other cases). For practical reasons it does not appear possible to eliminate modeling. Yet such an investigation calls for great care in the choice and application of such models.
  4. A combination of careful numerical and experimental procedures is required to produce reliable and accurate results.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £95 +P&P)