About dimensionless wall distance

josip76 · May 9, 2011, 04:26

Hey Pete thanks for your reply.

In furtherance to my question I am wondering if I could use a multi-block grid to increase the point spacing ratio outside the boundary layer and therefore reduce the computational time to reach solution without hindering the accuracy of the solution?

I understand that in general a multi-block grid is used to run a case in parallel on a multi-processor system or when there is a tricky portion of the geometry that is better treated with a series of smaller grids rather than one big grid. I was thinking treatment of the boundary layer would fit into the heading of a tricky region that could be treated with a separate smaller grid.

For example in examining a transonic airfoil I found the initial off the wall spacing using y+=1 to be about 4E-6m and then found the total boundary layer height at the trailing edge (using the law of the wake) to be about 0.01257m. So now i’m thinking I can split the mesh into a 2-block grid with the inside block extending 0.1257 meters (i.e. multiplied 0.01257 by 10 just to be certain I have the entire boundary layer in the block) outward from the airfoil and the rate of increase in vertical point spacing being 1.1 in this block and then increasing the point spacing ratio to 1.5 or even 2 in the outside block.

Thanks for any feedback on this

borhanalhoseini · November 24, 2013, 17:14

Dear Friends,
I have a series of data (velocity and normal distance u, y) and I want to find the u* so that I can find y+ and u+. free stream velocity is 2.7 m/s and characteristic length is 6...does anybody know how I can find these values?
Any help in this respect is highly appreciated

jyothishkumar · November 24, 2013, 18:15

Hi,

Try to find the value of utou(friction velocity - sqrt(wall shearstress(tou_w)/ro). Once you know these you can easily findout y+ and u+ easily.

regards

Quote:

Originally Posted by borhanalhoseini

Dear Friends,
I have a series of data (velocity and normal distance u, y) and I want to find the u* so that I can find y+ and u+. free stream velocity is 2.7 m/s and characteristic length is 6...does anybody know how I can find these values?
Any help in this respect is highly appreciated

borhanalhoseini · November 24, 2013, 18:26

Dear jyothishkumar,
Thanks for your email
I know that I have to find u_taw. Yet I do not know how to find it with these experimental data.
Do you have any idea?
Thanks again,

derz · June 9, 2014, 09:45

Nice discussion here, thanks guys.

I have a quick question: just realised, I always thought the y+<1 requirement was always referring to the height of the first layer, but some people here are using the actual centroid height of the first layer - i.e. half of the layer height.

Which is correct? I did a whole thesis assuming it was the actual layer height, not the height of the centre of the cell! Lol, maybe I was wrong...luckily I had the yplus down to about 3, so if it was in fact 1.5 all that time, then I still didn't over-resolve the near wall turbulence - I came close though!

May 9, 2011, 04:26	Multi-block grid usage	#21
josip76 New Member Mike Join Date: Apr 2011 Posts: 22 Rep Power: 15	Hey Pete thanks for your reply. In furtherance to my question I am wondering if I could use a multi-block grid to increase the point spacing ratio outside the boundary layer and therefore reduce the computational time to reach solution without hindering the accuracy of the solution? I understand that in general a multi-block grid is used to run a case in parallel on a multi-processor system or when there is a tricky portion of the geometry that is better treated with a series of smaller grids rather than one big grid. I was thinking treatment of the boundary layer would fit into the heading of a tricky region that could be treated with a separate smaller grid. For example in examining a transonic airfoil I found the initial off the wall spacing using y+=1 to be about 4E-6m and then found the total boundary layer height at the trailing edge (using the law of the wake) to be about 0.01257m. So now i’m thinking I can split the mesh into a 2-block grid with the inside block extending 0.1257 meters (i.e. multiplied 0.01257 by 10 just to be certain I have the entire boundary layer in the block) outward from the airfoil and the rate of increase in vertical point spacing being 1.1 in this block and then increasing the point spacing ratio to 1.5 or even 2 in the outside block. Thanks for any feedback on this

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[Commercial meshers] Fluent3DMeshToFoam	simvun	OpenFOAM Meshing & Mesh Conversion	50	January 19, 2020 16:33
Errors running allwmake in OpenFOAM141dev with WM_COMPILE_OPTION%3ddebug	unoder	OpenFOAM Installation	11	January 30, 2008 21:30
DIMENSIONLESS WALL DISTANCE	kuba	Main CFD Forum	0	December 12, 2006 13:42
MACRO FOR DIMENSIONLESS WALL DISTANCE Y+ ?	Wael	FLUENT	7	October 6, 2005 05:24
Multicomponent fluid	Andrea	CFX	2	October 11, 2004 06:12

November 24, 2013, 17:14		#22
borhanalhoseini New Member borhan Join Date: May 2011 Posts: 5 Rep Power: 15	Dear Friends, I have a series of data (velocity and normal distance u, y) and I want to find the u* so that I can find y+ and u+. free stream velocity is 2.7 m/s and characteristic length is 6...does anybody know how I can find these values? Any help in this respect is highly appreciated

November 24, 2013, 18:26		#24
borhanalhoseini New Member borhan Join Date: May 2011 Posts: 5 Rep Power: 15	Dear jyothishkumar, Thanks for your email I know that I have to find u_taw. Yet I do not know how to find it with these experimental data. Do you have any idea? Thanks again,

June 9, 2014, 09:45		#25
derz Member Derwin Parkin Join Date: Feb 2010 Posts: 35 Rep Power: 16	Nice discussion here, thanks guys. I have a quick question: just realised, I always thought the y+<1 requirement was always referring to the height of the first layer, but some people here are using the actual centroid height of the first layer - i.e. half of the layer height. Which is correct? I did a whole thesis assuming it was the actual layer height, not the height of the centre of the cell! Lol, maybe I was wrong...luckily I had the yplus down to about 3, so if it was in fact 1.5 all that time, then I still didn't over-resolve the near wall turbulence - I came close though!