k-e, RNG or K-w

ReCr · March 12, 2015, 13:29

I cant decide which type of turbulence model I should be using. what do you think would most accurately model high speed flow separation off a flat plate, but could reasonably predict the mixing with the free stream of air above it?

fluid23 · March 12, 2015, 14:41

RNG is just a special form of k-e that tries to approximate all turbulent length scales rather than just one (as with k-e). RNG is probably a little more applicable than just a standard k-e approach. However, k-w tends to do a little better with pressure gradients (which are what drive separation).

That being said, if accurately modeling the flow separation is your goal you probably need to use a DNS approach, not RANS (maybe you can get away with LES). Of course, the term accurate is highly subjective. Most people can justify a RANS analysis with 2-equation model.

Hamidzoka · March 14, 2015, 13:41

I think you can get better results if you use low-Re turbulence models like SST-K Omega. High-Re turbulence models like standard k-epsilon generally gives poor prediction of separated flows.

bshkoj · March 18, 2015, 11:59

Hi. can every one help?
I want to simulate weir using ansys cfx code and i used RNG turbulence model. but the results are not good and i can not get the ventilation region in water profile.

March 12, 2015, 13:29	k-e, RNG or K-w	#1
ReCr New Member Adam Join Date: Nov 2014 Posts: 6 Rep Power: 11	I cant decide which type of turbulence model I should be using. what do you think would most accurately model high speed flow separation off a flat plate, but could reasonably predict the mixing with the free stream of air above it?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
RNG k-epsilon model implementation	romant	OpenFOAM Running, Solving & CFD	2	June 20, 2018 08:33
why RNG k-e is better & what does RNG really mean?	Bonnie White	FLUENT	1	August 8, 2006 09:14
HELP!! Prandtl numbers in RNG k-e turbulence model	maoasis	FLUENT	0	April 24, 2006 10:51
RNG k-e equation	Mahshid	Fidelity CFD	0	October 24, 2005 09:51
RNG with Schemit number	zwdi	FLUENT	0	August 2, 2004 11:54

March 12, 2015, 14:41		#2
fluid23 Senior Member Matt Join Date: Aug 2014 Posts: 947 Rep Power: 17	RNG is just a special form of k-e that tries to approximate all turbulent length scales rather than just one (as with k-e). RNG is probably a little more applicable than just a standard k-e approach. However, k-w tends to do a little better with pressure gradients (which are what drive separation). That being said, if accurately modeling the flow separation is your goal you probably need to use a DNS approach, not RANS (maybe you can get away with LES). Of course, the term accurate is highly subjective. Most people can justify a RANS analysis with 2-equation model.

March 14, 2015, 13:41		#3
Hamidzoka Senior Member Hamid Zoka Join Date: Nov 2009 Posts: 282 Rep Power: 18	I think you can get better results if you use low-Re turbulence models like SST-K Omega. High-Re turbulence models like standard k-epsilon generally gives poor prediction of separated flows.