Post processing of Turbine Blade case

Moderators: davetaflin, salter, steve, jprenaud, straton, Chris Idso

screech1987
Newbie
Posts: 7
Joined: Fri Apr 01, 2011 7:39 am
Location: United Kingdom

Post processing of Turbine Blade case

Postby screech1987 » Tue Apr 05, 2011 5:36 am

Ladies and Gents,

Currently I am looking to post process a batch of results utilising the same geometry and need to attain the a way of outputting the following, using data I have extracted from StarCCM+:

1) boundary layer thickness (delta), Momentum thickness (theta) and shape factor (H).
Essentially this as far as i can see requires me to know the boundary layer thickness, and the rest can be interpolated from the corresponding location for other parameters. I have an approximation for delta being roughly equal to 1.936ymax (the ymax being essentially a velocity gradiant based on wall distance).

Wall distance, velocity are all outputable from Star but I am at a loss how to script Tecplot to search for the boundary layer edge.

2) Plots of coefficients (CP,Cf etc) over a blade relative to the displacement along the blade suction surface. -Is the way to do this by using a duplicate surface which is altered so exists in one plane? (e.g. Z=0)


Hope the above makes some sense, and welcome any ideas. even if it just provides food for thought.

Thanks for any help in advance,

Martin

User avatar
salter
Moderator
Posts: 188
Joined: Thu Dec 21, 2006 10:57 am
Location: NASA Langley Research Center
Contact:

Re: Post processing of Turbine Blade case

Postby salter » Mon Apr 11, 2011 1:52 pm

Martin,

I've written a number of macros for Tecplot to do this, but it does require some additional information. Are you using structured or unstructured grids? Is this 2D or 3D? Is there an enthalpy variable or other variable besides y+ in it for detecting the boundary layer edge? Can you acquire surface normals in the dataset?

All of these pieces of information I have used to construct a macro that marches from the wall to the boundary layer limit. Finding the thickness is not too difficult. Performing the integration is a bit more involved for integrated quantities but possible nonetheless.
Steve...

Aerospace Research Engineer
Aerothermodynamics Branch
NASA Langley Research Center

screech1987
Newbie
Posts: 7
Joined: Fri Apr 01, 2011 7:39 am
Location: United Kingdom

Re: Post processing of Turbine Blade case

Postby screech1987 » Tue Apr 12, 2011 2:04 am

Steve,

I have two setups, one is 3D and the other is a slice mesh one cell thick. However both are structured. On both I am comparing experimental data that has been recorded at one span wise poisition, so usually create a slice across the domain out at this point and reload the slice data so the same script can be used on both.

From the metthod I am using its just the wall distance and velocity that are used.

Thanks
Martin

User avatar
salter
Moderator
Posts: 188
Joined: Thu Dec 21, 2006 10:57 am
Location: NASA Langley Research Center
Contact:

Re: Post processing of Turbine Blade case

Postby salter » Tue Apr 12, 2011 10:29 am

Martin,

The determination of boundary layer thickness can be done with a macro where you compute the parameter you want to use for determining boundary layer thickness everywhere then search along a ray that is orthogonal to the wall. Compute the normal vector as you would with the unit normal vector in I, J, or K depending on the orientation of the grid that provides the computational direction orthogonal to the wall. Then, using the local cell sizes, march along a vector from a point on the wall until you find where the criteria is satisfied. Chances are that you will simply hit a point where the criteria changes from true to false, and not an exact point. To find the point, a simple linear interpolation is sufficient to find a more exact point. Store each point into a zone that can later be used for integration. To do this, you need to know the number of points which means you either do this computation twice (once to find the number of points and a second time to store them from one zone to another), or you set the number of points and interpolate the points you create with the known data. Once a line is extracted, all remaining integrated quantities can be computed.

Good Luck.
Steve...



Aerospace Research Engineer

Aerothermodynamics Branch

NASA Langley Research Center


Return to “Analyzer”

Who is online

Users browsing this forum: No registered users and 1 guest