2D aerofoil meshing (sharp trailing edge) How to avoid high aspect ratio

pchoopanya · March 15, 2018, 10:30

Dear folks,

I am meshing a 2D aerofoil and need your suggestion to improve my mesh quality. Especially the orthogonality (plus angle) and cell aspect ratio.

My aerofoil has a sharp trailing edge.

I am using ICEM CFD as my mesher and all hexahedral element utilising a blocking strategy.

My questions are...

1. As I need to refine the mesh at the aerofoil surface, hence the first element is very thin. With the strategy I use, I end up getting elements with very large aspect ratio. Especially when you approach the outlet section. I believe once solution to this is to add more element in the streamwise direction (end up getting even more elements unnecessarily). But I believe there should be a more effective way to deal with this.

2. At the leading edge, I cannot avoid having cells with very sharp edges (very little angle value). What should be a more clever mesh strategy so I can improve the mesh quality and at the same time maintain low cell count?

NITHIN BALAJI · June 29, 2018, 06:45

hi man, i am currently doing a project on sensitivity analysis of airfoils for different NACA profiles. So i think i would be able to answer this question.

In order to capture the flow physics accurately you should maintain keep the Y+ value as 1, but as you do this you will end up with high aspect ratio's in the outlet. High aspect ratio's leads to divergence in the solution. To avoid this should increase the number of nodes behind the trailing edge and also on the airfoil surface. Doing this increases the computation cost and time. Make sure that you maintain aspect ratio less than atleast 150. the ideal value is 20-40, but even aspect ratio's upto 170 worked fine for me.

There is always a compromise between accuracy and time in your simulation. start from keeping the y+ value from 5 and then reduce it to 1. if you dont have that much computation facilities then keep the y+ value from 15-30 and use Enhanced Wall Treatment(EWT) in your simulation.

colinda1 · July 18, 2018, 05:24

I see you are using a structured grid with a C-topology where the clustering at the trailing edge is maintained all the way down to the outlet. I am not an expert of ICEM CFD but isn't it possible to somehow relax this clustering at the outlet while maintaining it at the trailing edge?

For the sharp edges at the leading edges I don't see any reason to be honest and the pictures don't give me any idea. Maybe someone else has an idea on this part?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Meshing near leading and trailing edge of propeller	surajp92	STAR-CCM+	1	June 1, 2017 01:40
[ICEM] Meshing Wing with low Y+ value and sharp trailing edge	sonic109	ANSYS Meshing & Geometry	0	July 27, 2015 09:29
[ANSYS Meshing] high aspect ratio	mhdmk	ANSYS Meshing & Geometry	1	October 12, 2014 19:54
Meshing trailing edge of aerofoil	CC	Main CFD Forum	7	December 10, 2008 20:22
Structured mesh of a 3d sharp trailing edge	Gilles	Main CFD Forum	2	May 23, 2008 13:23

June 29, 2018, 06:45		#2
NITHIN BALAJI New Member Nithin Balaji Join Date: May 2018 Posts: 1 Rep Power: 0	hi man, i am currently doing a project on sensitivity analysis of airfoils for different NACA profiles. So i think i would be able to answer this question. In order to capture the flow physics accurately you should maintain keep the Y+ value as 1, but as you do this you will end up with high aspect ratio's in the outlet. High aspect ratio's leads to divergence in the solution. To avoid this should increase the number of nodes behind the trailing edge and also on the airfoil surface. Doing this increases the computation cost and time. Make sure that you maintain aspect ratio less than atleast 150. the ideal value is 20-40, but even aspect ratio's upto 170 worked fine for me. There is always a compromise between accuracy and time in your simulation. start from keeping the y+ value from 5 and then reduce it to 1. if you dont have that much computation facilities then keep the y+ value from 15-30 and use Enhanced Wall Treatment(EWT) in your simulation.

July 18, 2018, 05:24		#3
colinda1 Senior Member Colinda Join Date: May 2012 Location: Brussels Posts: 152 Rep Power: 14	I see you are using a structured grid with a C-topology where the clustering at the trailing edge is maintained all the way down to the outlet. I am not an expert of ICEM CFD but isn't it possible to somehow relax this clustering at the outlet while maintaining it at the trailing edge? For the sharp edges at the leading edges I don't see any reason to be honest and the pictures don't give me any idea. Maybe someone else has an idea on this part?