Specifying turbulence parameters for natural convection on vertical flat plate

abu250feldman · December 12, 2013, 13:22

Hi all,
I am simulating natural convection from vertical wall. the fluid is water.
I know this is very easy to calculate using simple correlations, however for some reason, I got more then 50% error between the manually calculated and CFD results. I have a wall 2.4 m high with Tinf=10 C and DT of 20 C. The Ra_L is turbulent (equal 5e12). I used two correlations for turbulent flow 1. 0.1*Ra^1/3 and 2. (0.825+0.387*Ra^(1/6)...)^2 (churchill and cho). I got h=450 W/m^K and h=534 W/m^K respectively.

In the CFD however, I get 340 W/m^K when using the "turbulence intensity and length" with default parameters. when I change to "Turbulence energy and dissipation" I get 400 W/m^K. Both results are way off...
I can get different results if I change the parameters but its is difficult to know the correct values.

My question is (finally): which model is better for natural convection (my case) and does anyone (Boris maybe?) has some guidelines on specifying the correct turbulence parameters?

Thanks in advance

Boris_M · December 20, 2013, 10:01

Hi Yoav,

this is a validation case the developers are considering with each version before shipment.
In fact I have the project setup and can send you the results if you provide me with your email address in a private message.

Regards and happy holidays,
Boris

enayath · May 5, 2014, 14:42

Boris,

Could you please share this information with me also?
I sent my email address in a private message.

Thank you,
Hooman

December 12, 2013, 13:22	Specifying turbulence parameters for natural convection on vertical flat plate	#1
abu250feldman New Member Yoav Join Date: Nov 2010 Posts: 4 Rep Power: 16	Hi all, I am simulating natural convection from vertical wall. the fluid is water. I know this is very easy to calculate using simple correlations, however for some reason, I got more then 50% error between the manually calculated and CFD results. I have a wall 2.4 m high with Tinf=10 C and DT of 20 C. The Ra_L is turbulent (equal 5e12). I used two correlations for turbulent flow 1. 0.1Ra^1/3 and 2. (0.825+0.387Ra^(1/6)...)^2 (churchill and cho). I got h=450 W/m^K and h=534 W/m^K respectively. In the CFD however, I get 340 W/m^K when using the "turbulence intensity and length" with default parameters. when I change to "Turbulence energy and dissipation" I get 400 W/m^K. Both results are way off... I can get different results if I change the parameters but its is difficult to know the correct values. My question is (finally): which model is better for natural convection (my case) and does anyone (Boris maybe?) has some guidelines on specifying the correct turbulence parameters? Thanks in advance

May 5, 2014, 14:42		#3
enayath Senior Member Joe Join Date: Feb 2012 Location: Canada Posts: 112 Rep Power: 14	Boris, Could you please share this information with me also? I sent my email address in a private message. Thank you, Hooman Last edited by enayath; May 23, 2014 at 12:40.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
drag of flat plate with cavity	Far	FLUENT	0	May 18, 2010 05:57
DNS for Impinging Jet on a Flat Plate	Chandra Shekhar	Main CFD Forum	3	January 18, 2010 06:03
Discussion: Reason of Turbulence!!	Wen Long	Main CFD Forum	3	May 15, 2009 10:52
Seperated Flow and Turbulence modelling flat plate	Simon Mizzi	Main CFD Forum	1	December 16, 2003 02:26
Turbulence dead? well... can somebody help me with a flat plate meanwhile?	Javier	Main CFD Forum	2	May 23, 2000 05:41

December 20, 2013, 10:01		#2
Boris_M Disabled Join Date: Jul 2009 Posts: 616 Rep Power: 24	Hi Yoav, this is a validation case the developers are considering with each version before shipment. In fact I have the project setup and can send you the results if you provide me with your email address in a private message. Regards and happy holidays, Boris