Turbulent kinetic energy profile Openfoam analysis

massakimi · January 3, 2018, 14:31

I have run simulations by increasing the number of cells near the wall and plotted K vs r graph. As the cells are refined near the wall the turbulent kinetic energy(TKE) maximum value is increasing.

But the velocity profile remains almost same near the wall. According to the formula K = 3/2*( Uavg * turbulent intensity) ^2.

Capture.jpg . The figure shows the TKE for 2 different meshes

So why does the maximum value of TKE increases with mesh refinement?

Santiago · January 4, 2018, 20:43

Judging by the unphysical bump in the outer region of the boundary layers, i guess the black line is the coarse grid. The formula you show is irrelevant in this case. In fact, the tke profile is self-similar in wall boundary layers.

massakimi · January 5, 2018, 13:08

Thanks a lot for the response. So you mean that the tke profile is similar near the wall boundaries irrespective of the mesh refinement ?

Santiago · January 5, 2018, 13:18

What im saying is that your profiles are unphysical, and one reason is grid coarseness. For instance, Is it physical to have the max TKE at the walls?

massakimi · January 5, 2018, 13:48

No . I think the profile should somewhat look like this physically. am i right ?

kinetic.png.

So i need to refine the mesh a lot to get the. accurate profile near the wall.

Santiago · January 5, 2018, 16:35

Refer to the paper of Kim, Moin, and Moser (1987) for details on how to set up such cases and, more importantly, on the physics. I assume you're trying DNS/LES.

massakimi · January 5, 2018, 17:28

. Thank you for the reference. I am solving for k-epsilon model

January 3, 2018, 14:31	Turbulent kinetic energy profile Openfoam analysis	#1
massakimi New Member Arjun Join Date: Feb 2014 Posts: 20 Rep Power: 12	I have run simulations by increasing the number of cells near the wall and plotted K vs r graph. As the cells are refined near the wall the turbulent kinetic energy(TKE) maximum value is increasing. But the velocity profile remains almost same near the wall. According to the formula K = 3/2( Uavg turbulent intensity) ^2. Capture.jpg . The figure shows the TKE for 2 different meshes So why does the maximum value of TKE increases with mesh refinement?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
question about turbulent kinetic energy	junker4236	Main CFD Forum	19	April 19, 2017 04:46
Turbulent kinetic energy in k-epsilon model	khunyeu	FLUENT	0	October 7, 2013 01:31
Contours of turbulent kinetic energy in LES using ANSYS FLUENT 13	khosrow	FLUENT	5	July 29, 2013 05:28
LES and total turbulent kinetic energy	Boerge	FLUENT	1	September 8, 2012 11:41
ATTENTION! Reliability problems in CFX 5.7	Joseph	CFX	14	April 20, 2010 15:45

January 4, 2018, 20:43		#2
Santiago Senior Member Santiago Lopez Castano Join Date: Nov 2012 Posts: 354 Rep Power: 15	Judging by the unphysical bump in the outer region of the boundary layers, i guess the black line is the coarse grid. The formula you show is irrelevant in this case. In fact, the tke profile is self-similar in wall boundary layers.

January 5, 2018, 13:08		#3
massakimi New Member Arjun Join Date: Feb 2014 Posts: 20 Rep Power: 12	Thanks a lot for the response. So you mean that the tke profile is similar near the wall boundaries irrespective of the mesh refinement ?

January 5, 2018, 13:18		#4
Santiago Senior Member Santiago Lopez Castano Join Date: Nov 2012 Posts: 354 Rep Power: 15	What im saying is that your profiles are unphysical, and one reason is grid coarseness. For instance, Is it physical to have the max TKE at the walls?

January 5, 2018, 13:48		#5
massakimi New Member Arjun Join Date: Feb 2014 Posts: 20 Rep Power: 12	No . I think the profile should somewhat look like this physically. am i right ? kinetic.png. So i need to refine the mesh a lot to get the. accurate profile near the wall.

January 5, 2018, 16:35		#6
Santiago Senior Member Santiago Lopez Castano Join Date: Nov 2012 Posts: 354 Rep Power: 15	Refer to the paper of Kim, Moin, and Moser (1987) for details on how to set up such cases and, more importantly, on the physics. I assume you're trying DNS/LES.

January 5, 2018, 17:28		#7
massakimi New Member Arjun Join Date: Feb 2014 Posts: 20 Rep Power: 12	. Thank you for the reference. I am solving for k-epsilon model