Background Image
Previous Page  11 / 15 Next Page
Information
Show Menu
Previous Page 11 / 15 Next Page
Page Background

Рис. 7. The leading edge of the blunt-nose

wedge, computation mesh detail

the ram air in the lateral direction

are presented in [10]). The Mach

number of the ram air is estimated

from the experimental data using

the angle of the shockwave leaving

the wedge in photos corresponding

to the quasi steady flow modes. As

a result the input parameters at the

inflow boundary in the numerical

simulation are the following:

M = 7.0 or 4.5 (two quasi

steady modes were regarded); the

temperature is 100 K; viscousity

coefficient is equal to the air

viscousity coefficient at the entry

temperature.

For two model gas dynamic ducts (with sharp and blunt-nose edges) two

different meshes are created. These sets are three-dimensional in design,

they contain only one layer of cells in

Z

axis direction, so these meshes

will be regarded as two-dimensional. The mesh for the sharp-edge model

contains 363 500 triangular or quandrangular cells with a minimal cell size

of 2

10

4

near hard surfaces. The Gmsh generator [20] was used to create

the mesh.

The enthropy layers generated near the decelerating point of the blunted

body can significantly affect the flow inside the ramjet air intake, so these

layers should have a good resolution. The resolution must be especially

high near the blunt-nose wedge leading edge. The mesh for the blunted

edge contains 1 200 000 cells with the minimal cell size of 0.01 mm near

the hard walls. Fig. 7 shows details of the mesh for the blunt-nose model

near the wedge leading edge.

All calculations were done in a complete calculation field without any

assumptions on the flow symmetry. On the one side it was used as an

additional verification of the solution quality, on the other hand furher

research is planned on the effects of the entry current non-uniform patters

and a small angle of attack.

The least squares method [13] with Michalak and Ollivier–Gooch

limiters [16] was used to calculate the physical variables gradients in the

cells centres. The calculations were done with Runge-Kutta second-order

method [19] with Courant number of 0.5.

The comparison of the calculation results with the experimental data is

presented in Fig. 8, 9. The upper part of each figure contains the Shlieren

image from the test, and the lower part — isolines of the calculated density

14 ISSN 0236-3941. HERALD of the BMSTU. Series “Mechanical Engineering”. 2015. No. 1