CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   how to use CFD code for selas (

d murali mohan September 3, 1999 05:56

how to use CFD code for selas
I am doing a project on labyrinth seals for a space shuttle engine,my main aim is to find rotor dynamic coefficients of seals (K,C,M) an seal leeakage characterstics,pleaseee advice me as i am geetting problem in solving thee problem using FLUENT

John C. Chien September 3, 1999 12:15

Re: how to use CFD code for selas
(1). Recently, I heard that the space shuttle fleet was grounded for the wiring problem. And, a couple of holes had developed in the nozzle in the recent flight. So,....(2). I had used the code myself to model the 3-D radial turbine passage + turbine side cavity + 2 knife edge seals + radial compressor side cavity flows all in one. (3). The way to do is to divide the flow field into many sub-blocks, so that the mesh can be controlled even through the tip of the knife edge seal. The flow through the knife-edge seal must be computed with adequate mesh or cells through it. (4). In the steady state, one can specify the prssure boundary conditions at the compressor side of the cavity (entrance of the leakage), and the flow field throughout the cavity can be obtained including the mass flow rate. From there, you can extract information you need. (5). You did not define your rotor dynamic coefficient of seal, so, I don't know how to answer the question in that area. By the way, the time when I used the code, it had only unstructured tet mesh at the wall. Extremely fine mesh near the wall was required to obtain useful answers. (a word of caution, this problem is reserved for the professional.)

Janna J. September 3, 1999 13:56

Re: how to use CFD code for selas
need to define very fine b.c mesh (hex ). at least 3 layers. try Gambit then refine in Tgrid

regards from

unprofessional try

John C. Chien September 5, 1999 19:38

Re: how to use CFD code for selas
(1). It took me several weeks to model it and obtain a solution. (2). There was no attempt to conduct a second calculation or look at the mesh effect, turbulence model effect. (3). There were flow separations everywhere and no comparison was made with test data. The Prebfc was used to model the geometry. (4). In other study in a very simple cavity flow problem, considerable difficulties was encountered in comparison with test data. In that case, various mesh refinement, numerical schemes, turbulence models were tried. So, as a training tool for complex geometry flow problems, yes. As a design tool to provide design data, no. (this was my experience with the older version of the code.)

All times are GMT -4. The time now is 21:53.