CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CD-adapco (http://www.cfd-online.com/Forums/cd-adapco/)
-   -   Streamline plot (http://www.cfd-online.com/Forums/cd-adapco/54290-streamline-plot.html)

C.Bhasker January 8, 2005 04:59

Streamline plot
 
Can some one help to create stream line plot for the three dimensional problems solved in Star-CD. In Star-CD Post, no direct option is available to create streamline plot. Looking forward to receive the hints.

Jörn Beilke January 8, 2005 07:43

Re: Streamline plot
 
You have to look for "partickle tracks". Streamlines only exist in 2d.

Anton Lyaskin January 9, 2005 03:28

Re: Streamline plot
 
... and you can plot them as countours of stream function (becomes available when you choose Cell and Wall/Bound data)

C.Bhasker January 9, 2005 23:50

Re: Streamline plot
 
Mr Anton Lyaskin:

Your hint has successfully worked in creation of streamline plot for 3d problems. Thanks lot! I am finding some problem in creating the blocks using best fitted shapes of circular geometries. (precisely, block line forms straight instead of of following arc).t will be highly beneficial if you provide me an hint.

Alternatively I have successfully established the close link for the grids generated in Altair's Hypermesh to Star-CD and then to Tecplot.

Can you please clarify for import of gambit grids along with regions in star-CD which I am unable to do so ofcourse, if you worked.

Regards,


Anton Lyaskin January 10, 2005 04:05

Re: Streamline plot
 
Concerning the blocks, there are 2 possible reasons. 1. The corners of your block are not belong to a spline forming the arc (go to spline tool and check the numbers of the end vertices of the spline, then check which vertices are specified for the block). 2. It's a bug in displaying the blocks with spline edges (I've encountered it a couple of times, just try to generate mesh for the block).

For importing Gambit grids I have no experience, sorry.

C.Bhasker January 10, 2005 06:05

Re: Streamline plot
 
thanks!

C.Bhasker January 12, 2005 02:21

Re: Streamline plot
 
Mr Anton:

In some cases, under cell and wall data, stream is not getting and following message is displaced:

POST*>

getv,all,stream **NOTE - REALLOCATING MEMORY **NOTE - DENS IS CONSTANT (NOT SOLVED FOR) VECTOR COMPONENT U STORED VECTOR COMPONENT V STORED VECTOR COMPONENT W STORED SCALAR ITEM DENS STORED ***ERROR - THE MODEL IS NOT TWO DIMENSIONAL ** STREAM FUNCTION CANNOT BE CALCULATED DUE TO ABOVE ERRORS

in some cases above error is not displayed. I have compared this problem output with the other problem output and not found any difference in steup.

Kinldy know the possible hint to correct the above error.

bskr

POST*>

Anton Lyaskin January 12, 2005 05:15

Re: Streamline plot
 
Stream function can be plotted for 2D models only (i.e. those which has one layer of cells with symmetry boundaries). And sometimes it can't be plotted even for them :) It's some kind of a bug, I've couldn't figure out the reason

Brian January 12, 2005 05:22

Re: Streamline plot
 
The two-dimensional form of the Navier-Stokes equations can be re-written in the so-called stream function / vorticity formulation. This can NOT be done in three dimensions, so both vorticity and stream function are constructs that exist in two dimensions only. As a streamline is a contour of constant stream function, there is therefore no such thing as a streamline in three dimensions.

There is an expression for the definition of stream function in terms of velocity and other variables. It is this expression that is applied when you plot stream function (GETV STRE) in Prostar.

However, while calculating the expression, the code checks to see whether your mesh is truly two-dimensional. If it discovers that it is not, the error report you mention is issued.

So you need to look closely at your mesh - do you truly have only one cell thickness everywhere in the third dimension?

You can always "fake" three-dimensional streamlines by creating droplet or particle tracks, with zero particle mass (zero size). These particles will exactly follow the flow, and thus resemble streamlines, even though they are not really.

MG January 12, 2005 15:14

Re: Streamline plot
 
To import gambit mesh, first save the mesh as .NEU file in gambit. In starcd, type: gamb, FILENAME


C.Bhasker January 12, 2005 23:30

Re: Streamline plot
 
Thanks for your hint. I have already done this and successful in reading gambit grids in star CD and happy about it.

-bskr

C.Bhasker January 12, 2005 23:38

Re: Streamline plot
 
To test this, firstly I have created 2d sketch in xy plane and later extruded to z dimension (uniformly) I have created the grid using best shaped blocks. I have set up this and solved in post when loaded cell/wall bound data stream is successufully loaded.

In similar way I have tested another geometry of slightly different shape and solved. This time in post under data, under cell/wall bound data, stream is not getting loaded. I think support guys of Star-CD should look at this and respond for correction.

Your suggestion for plotting streamline with zero particle mass is excellent. I will try and come back.

Thanks for your response.

bskr

C.Bhasker January 12, 2005 23:44

Re: Streamline plot
 
Mr Anton:

Thanks for your response.

I hope support guys at Star-CD may looking at these messages and some correction to this bug follows in coming days.

I have successfully read the gambit grids in star-CD files as neu extension in star-CD and happy about it.

-Bskr


All times are GMT -4. The time now is 07:16.