Turbulence model

Mr.Mister · May 26, 2020, 21:13

Hey guys. I'm writing a paper on two-phase air-water flow simulation in a vertical pipe using Ansys Fluent. The objectives of my work are to compare numerical results with experimental data (pressure loss along the pipe and the flow void fraction) and see if the flow pattern in the simulation matches the prediction of a flow pattern map.

I decided that i will use the VOF model for the simulation, since it allows me to see the flow pattern, but now i'm unceartain about the turbulence models.
A lot of papers on the subject uses the standard k-episilon model.

I have done some research about the models standard k-epsilon, realizable k-epsilon, standard k-omega and SST k-omega and learned that they have advantages and disadvantages near and far from the wall. I also learned that they have differences depending on high or low reynolds number (my Reynolds number is about 16000).

Do you guys know wich turbulence model would be best for my air-water simulation in a vertical pipe? i've read so many papers, but i'm still very confused.

PS: In a turbulent flow what is considered a low Reynolds number or a high Reynolds number?

vinerm · May 27, 2020, 04:05

It won't matter much, you can use $k-\varepsilon$ . The term low and high Re number in reference to turbulence models refer to local Re in the flow. Closer to boundaries, Re is always low, hence, low-Re models are more accurate, such as, $k-\omega$ , while away from the boundaries, $k-\varepsilon$ is more appropriate. Looking at your objective, $k-\varepsilon$ is better.

Mr.Mister · May 30, 2020, 16:59

Quote:

Originally Posted by vinerm

It won't matter much, you can use $k-\varepsilon$ . The term low and high Re number in reference to turbulence models refer to local Re in the flow. Closer to boundaries, Re is always low, hence, low-Re models are more accurate, such as, $k-\omega$ , while away from the boundaries, $k-\varepsilon$ is more appropriate. Looking at your objective, $k-\varepsilon$ is better.

Thank you very much for the answer Vinerm! Could you explain why for my case the k-epsilon is better?
Best regards.

vinerm · May 31, 2020, 05:54

$k-\varepsilon$ is better suited for your case due to two reasons

1. The interest lies in what is happening away from the walls.
2. Since flow is the core is of more importance, you can use coarser mesh at the walls and use standard-wall-function.

May 26, 2020, 21:13	Turbulence model	#1
Mr.Mister New Member Join Date: May 2020 Posts: 5 Rep Power: 5	Hey guys. I'm writing a paper on two-phase air-water flow simulation in a vertical pipe using Ansys Fluent. The objectives of my work are to compare numerical results with experimental data (pressure loss along the pipe and the flow void fraction) and see if the flow pattern in the simulation matches the prediction of a flow pattern map. I decided that i will use the VOF model for the simulation, since it allows me to see the flow pattern, but now i'm unceartain about the turbulence models. A lot of papers on the subject uses the standard k-episilon model. I have done some research about the models standard k-epsilon, realizable k-epsilon, standard k-omega and SST k-omega and learned that they have advantages and disadvantages near and far from the wall. I also learned that they have differences depending on high or low reynolds number (my Reynolds number is about 16000). Do you guys know wich turbulence model would be best for my air-water simulation in a vertical pipe? i've read so many papers, but i'm still very confused. PS: In a turbulent flow what is considered a low Reynolds number or a high Reynolds number?

May 27, 2020, 04:05	Turbulence Model	#2
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 35	It won't matter much, you can use $k-\varepsilon$ . The term low and high Re number in reference to turbulence models refer to local Re in the flow. Closer to boundaries, Re is always low, hence, low-Re models are more accurate, such as, $k-\omega$ , while away from the boundaries, $k-\varepsilon$ is more appropriate. Looking at your objective, $k-\varepsilon$ is better. Mr.Mister likes this. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

May 31, 2020, 05:54	Flow Structure	#4
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 35	$k-\varepsilon$ is better suited for your case due to two reasons 1. The interest lies in what is happening away from the walls. 2. Since flow is the core is of more importance, you can use coarser mesh at the walls and use standard-wall-function. Mr.Mister likes this. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Question about adaptive timestepping	Guille1811	CFX	25	November 12, 2017 17:38
Question about matching of solver and turbulence model	louistse	OpenFOAM Running, Solving & CFD	1	February 1, 2017 21:36
Overflow Error in Multiphase Modelling with Two Continuous Fluids	ashtonJ	CFX	6	August 11, 2014 14:32
An error has occurred in cfx5solve:	volo87	CFX	5	June 14, 2013 17:44
Wrong calculation of nut in the kOmegaSST turbulence model	FelixL	OpenFOAM Bugs	27	March 27, 2012 09:02