"Buffet" having nan value in su2
 

Greetings everyone
I am new user of SU2. I am trying to run a cfg file for mesh generated through pointwise for SD8020 airfoil with a zero thickness flap at 0.6c with flap legth 0.3c.
On running simulation in su2 its history file is showing nan for all rows of "Buffet" column causing divergence in the simualtion. I am not able to understand that what is this "Buffet" and how can i remove this bug.
please help
regards
chandan

Hi Chandan,

This post may be better suited for the main forum.
As for your issue, I have doubts the "Buffet" value is causing divergence as it is an aerodynamic coefficient. Could you please your config file/ log output?

Best,
Wally

BUFFET DIVERGENCE WITH Nan
 

Dear Wally
attached is the cfg file of pertaining issue. i will wait for your reply
thanks and regards

Hi Chandan11,

Your config file looks fine. I suggest running first order (MUSCL_FLOW=NO) for a bit to see if your issue resolves. Additionally, you could run explicitly with a lower CFL number.

Can you post your log outputs or mesh? What does the flow solution look like when it begins to diverge?
This could possibly be a mesh issue as well.

Wally

Su2 divergence due to "buffet" having nan
 

Dear Wally
greetings
thanks for addressing my issue.
I did as per your suggestion with respect to MUSCL_FLOW , however, still problem persists.
I am attaching my mesh file along with cfg.
regards and thanks in anticipation
chandan

Interesting, I am more convinced it is a mesh issue.

On that note, it looks like the pointwise file does not want to be un-compressed. I am unable to unzip the folder, on any multiple machines strangely. Can you reupload the pw file, or maybe try the .su2 file?

Thanks

Dear Wally
soryy, i am not able to upload py or su2 file. can you please give me your email where i can send the same
thanks

After looking at your mesh, I suspect you have a few issues. First, there is no boundary layer to resolve the viscous phenomena, this will will cause issues for your heatflux walls.
Additionally, the one-dimensional flap geometry will also probably cause issues, at least I would expect it to. I would recommend looking at other viscous meshes on the su2 website to get an idea of the BL meshes.

thanks Wally
i'll look into it again