Wiggles in Cp plots over Store Fins
 

Hi all,
This is a strange experience I am having here.Here is what its all about:

1. Made Viscous mesh for Generic Wing Pylon with Finned Store model using Pointwise V17. Examined at various cuts, and the mesh seemed to be okay. Totally made 3 blocks, blk1-A Rectangular block with dense volume cells around Finned Store, blk2-A hemisphere with coarse volume cells around the entire wing-pylon-FinnedStore geometry, blk3-One more bigger hemisphere around the entire blk2 with more coarsen grid (shared pics).

2. Solved the grid using SU2 with approximately 20,000 iterations, and the Cl and Cd values seemed to be converging.

3. Post Processed the solution files using Tecplot 360. Took various slices on the Fins, and for every slice, plotted Cp V/s X/C plots.

Surprisingly, I see wiggles on the Cp curves. I dont understand why!!.

Can somebody help me understand why I am getting Wiggles!! I'm struggling for the past 2 weeks with this problem.

Attaching pics of Cp plots
 

Posting reply as the max pic upload in the original thread is 5, uploading Cp plot pic.

Black Delta -->Experimental Data
Pink curve -->Cp curve on Fin 1
At Slice 1 near the store body

What free stream conditions were you using in your simulation? What was your initial wall spacing around the store? SU2 doesn't support wall functions yet, so you must solve to the wall. Are you running a steady, or unsteady simulation? The ringing you see in the resulting metric may be due to a number of things: a CFL that is set to a value which is too large, inherent unsteady flow phenomena at the location of your cut plane, etc.

From what I can see of your mesh, which isn't too much, it looks okay--you may want to ensure that there are no jumps in edge spacing between blocks. There are a number of things to consider, but without reviewing the actual input files and mesh a lot of them are just guesses.

Configuration File
 

Thanks Davis for the reply,
Attached the configuration file, I think that answers all of your doubts :)

1. Initial wall spacing around the store=5.6422E-5
Note: as per the y+ formula obtained from the cfd-online.com website, the wall spacing was coming to be in E-8 which was not practical, hence after some R&D in Internet i reduced it to E-5.

2. Ensured that there are no jumps in edge spacing between blocks.

Kindly refer to it and reply with you inputs :)

Good day :)

Apologies for the delayed reply--I didn't seem to get a notification from the forum that a response had been posted. Your configuration file looks okay; though, I might recommend you switch from Green Gauss to Weighted Least Squares for spatial gradients given the grid elements in your mesh.

The other thing to note, that if your reference length is 1m (3.28ft), Reynold's number ~28e+06, Freestream Mach No. 1.2, gamma 1.4, and adiabatic wall temperatures, then I calculate that y+ = 5.0 (upper wall spacing limit) should be 2.28e-04in (~5.8e-06m). This is about an order of magnitude smaller than what you indicated the initial wall spacing was in your mesh.

A few other questions relating to your mesh--what does Pointwise say is the Max Included Angle and Max Centroid Skewness for your mesh? These can be determined by selecting the blocks of your mesh, and then selecting the corresponding metric from the Examine menu in Pointwise. Typically aim for Max Include Angles below 170 degrees at most and Max Centroid Skewness below 0.9. These are guidelines only, and mesh quality criteria vary considerably between solvers.

Lastly, I will reiterate that supersonic flow is inherently unsteady. The shock waves that form around your analysis model are moving in space which causes time varying changes in the air properties in their vicinity. Thus, it can be difficult to get a smooth solution using a Reynold's Averaging approach.