Hi All,

I am doing a secondary flow analysis through a labyrinth seal. I need to find out the force acting on the rotating wall. I have swithched on the Wall Force option. I am unable to understand the contents of .erd file. Can any one tell me what is S, N and T (x,y andz) forces as given in .erd file?

Is there any way we can find the forces with respect to a cylindrical coordinate system?

Expecting an early reply...

Thanks in advance, AP

Look at the on-line help for EDGRAPH - it tells you what the components are. The force on a boundary is split into 6 components: shear (S) and normal (N) in each of the 3 components of the global Cartesian coordinate system. In each direction, there is also the total (T=N+S).

You might find it easier to work with the .erd file by loading data into pro-STAR using EDGRAPH, or the graph panels in the gui.

I think you would have to convert to a cylindrical system by hand - I don't think STAR will do it for you.

Thanks Mr. Richard,

I have few more doubts (I am relatively new to Star - CD)... What force value is given in the erd file. Are they the average value of all the forces on the wall for which we are monitoring or are they just the algebraic sum...?

I tried "edgraph,casename,Txforce," and I got the following error... Can you please tell me what could be the problem...

DEFINE FRAME 1 WITH DATA OF VARIABLE TXFORCE ON FILE casename.ecd.

STORE DATA OF SETS OR REGIONS 1 TO 1 BY 1 STARTING WITH GRAPH REGISTER 1.

**ERROR - FILE casename.ecd CANNOT BE OPENED ************************************************** **************************** DRAWING FRAME 1 : **ERROR - X-AXIS REGISTER(S) NOT DEFINED FOR FRAME 1 **ERROR - Y-AXIS REGISTER(S) NOT DEFINED FOR FRAME 1

FRAME 1 CANNOT BE DRAWN

Thanks Again, AP

The forces are obtained by summing stress*area over all the boundary faces.

The problem with your edgraph, I think, is that you have taken the online help too literally. You should type your casename (e.g. star), not "casename". Or you can just type ,, (two commas) and that will leave the field blank, which will in turn mean that pro-STAR uses the current case name.

Hi Mr. Richard,

Thanks for your help.... In fact I have given the actual case name only (not just casename.erd), still I am getting that error. What could be the problem here...?

If I plot any one component of the force (say Tx) and if I do the command 'summa cset' on that, should I get the same Tx value as in the .erd file....? But I am getting different values... Why is this...?

Thanks again, AP.

Another problem with your edgr command could be that you are typing "edgr,casename,Txforce" so it is adding the default ".ecd" extension to "casename" - hence the error message about this file not existing. Try "edgr,casename.erd,Txforce".

To compare with forces from the pst file, you have to be careful to choose the p/pstat and rela/abso when you do getwall to be consistent with the forces in the .erd file. I don't know what the right choice is here. You also have to be careful to choose the same surface for summation. There is also the possibility that the erd file will be more accurate because star can work in double precision whereas pro-STAR is single precision.

Hi Mr. Richard,

Thanks a lot... It is working.....

Regards, AP