wall y+ below one

hebeldebel · August 14, 2014, 05:08

Hi,

I am trying to simulate the heat transfer over a boundary with environment option. I checked wall y+ at it ranges from 0.00058 to 0.279.

I used K-Epsilon Turbulence , Realizable K-Epsilon Two-Layer and Two-Layer All y+ Wall Treatment.

As far as I can see, this should pose no problem, as I used All y+ Treatment and I am also below y+ << 1, therefore it should always use the u+ = y+ condition, right?

I do not use prism layers, is there a need to use them, when I am already well below y+ = 1?

Thanks for your answers

edit: In case I get y+ values between 5 and 30, all y+ will not compensate for it, right? It only switches between the logarithmic term and "u+ = y+", right?

npoutre · August 14, 2014, 12:30

If you have a Y+ << 1 and are using "All Y+ Wall Treatment" than the solver is integrating to the wall and is not actually using any wall functions. All Y+ Wall Treatment is nice if you have some cells on the wall that are in the low, buffer, and high Y+ regions. It compensates for all.

There's nothing special about prism layer cells compared to other volume cells except you have more control over them. They are nice because you can control the surface growth rate and thickness, with more fidelity, to ensure your boundary layer is sufficiently refined.

Your Y+ is very low so you probably have decent boundary layer resolution, but just make sure your growth rate isn't too large.

August 14, 2014, 05:08	wall y+ below one	#1
hebeldebel Member Join Date: Jan 2014 Posts: 40 Rep Power: 12	Hi, I am trying to simulate the heat transfer over a boundary with environment option. I checked wall y+ at it ranges from 0.00058 to 0.279. I used K-Epsilon Turbulence , Realizable K-Epsilon Two-Layer and Two-Layer All y+ Wall Treatment. As far as I can see, this should pose no problem, as I used All y+ Treatment and I am also below y+ << 1, therefore it should always use the u+ = y+ condition, right? I do not use prism layers, is there a need to use them, when I am already well below y+ = 1? Thanks for your answers edit: In case I get y+ values between 5 and 30, all y+ will not compensate for it, right? It only switches between the logarithmic term and "u+ = y+", right?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Thin Wall Heat Transfer BC for rhoSimpleFoam	swahono	OpenFOAM Running, Solving & CFD	12	October 4, 2013 11:49
[Commercial meshers] tmerge utility creates unwanted interface/walls comes in the final mesh	Shoonya	OpenFOAM Meshing & Mesh Conversion	11	January 20, 2012 06:23
Patches for OpenFOAM 1.7 on MacOS X	gschaider	OpenFOAM Installation	101	September 21, 2011 05:37
UDF for wall slipping	HFLUENT	Fluent UDF and Scheme Programming	0	April 27, 2011 12:03
Quick Question - Wall Function	D.Tandra	Main CFD Forum	2	March 16, 2004 04:29

August 14, 2014, 12:30		#2
npoutre New Member Noah Join Date: Aug 2014 Location: Seattle Posts: 2 Rep Power: 0	If you have a Y+ << 1 and are using "All Y+ Wall Treatment" than the solver is integrating to the wall and is not actually using any wall functions. All Y+ Wall Treatment is nice if you have some cells on the wall that are in the low, buffer, and high Y+ regions. It compensates for all. There's nothing special about prism layer cells compared to other volume cells except you have more control over them. They are nice because you can control the surface growth rate and thickness, with more fidelity, to ensure your boundary layer is sufficiently refined. Your Y+ is very low so you probably have decent boundary layer resolution, but just make sure your growth rate isn't too large.