Including turbulence damping at the interface

g_bochio · September 4, 2018, 08:53

Hello, everyone! I would like some help to add a source to the $\omega$ equation, so that I can supress turbulence generation near the interface in a two-phase flow.

This link shows the source that I want to add:
https://www.sharcnet.ca/Software/Ans...turb_damp.html

My main questions are:

1) Is it possible to implement this directly to the fvOptions dict?

2) How could I calculate $\Delta n$ ?

Thanks!

ano · September 4, 2018, 10:36

1) I didn't find a fvOptions term included in the omega equation of the kOmega models, so putting something into the fvOptions dict would not work. I guess it is easiest to put your term directly into the k-omega model.

2) There are two wall distance functions used in the turbulence models, both give you an volScalarField:
a) All turbulence models inherit from "turbulenceModel". In turbulenceModel a nearWallDist y_ is defined (as a private member), which would represent your $\Delta n$ .
The nearWallDist gives you the "Distance calculation for cells with face on a wall.", which should give you for all cells next to the boundary the distance between the face and the cell centre (and for the other cells it gives zero). From within your turbulence model, e.g. kkLOmega you can call it with

Code:

this->y()

to obtain $\Delta n .$ . (From within the solver you could call it with "turbulence->y()".)

b) The wallDist gives you the "Calculation of distance to nearest wall for all cells and boundary.". It is defined in some, but not all models, e.g. the kkLOmega. Unfortunately someone gave it also the variable name y_. So be careful that you don't mix them.

g_bochio · September 4, 2018, 13:10

Hey Ano, thanks for the reply.

$\Delta n$ in my case is not the distance from a wall. It is the distance from the interface between two fluids. Any idea how to calculate this?

ano · September 4, 2018, 15:07

Hey Bochio,

I am really sorry for not reading your message well and that my answer missed totally your question.

No, unfortunately I don't know how to calculate it.

g_bochio · September 5, 2018, 09:41

No problem!

mrishi · September 4, 2020, 09:58

Sorry to bump this, but did you find the solution to this problem?

Wenyuan · September 7, 2020, 04:22

Hi,

I have an implementation for the interFoam (interIsoFoam) family in the following repository
https://github.com/wenyuan-fan/varRhoTurbVOF_2

You can also find a manual and a tutorial on how to use it.

Phicau · September 7, 2020, 05:50

Hi Gustavo,
I think you will find this useful: https://olaflow.github.io/blog/turbu...ons-revisited/
The code is now compatible with OF8 and 2006+

https://github.com/phicau/olaFlow_su...enceMultiphase

Best,
Pablo

jayson · December 30, 2020, 02:37

Quote:

Originally Posted by Wenyuan

Hi,

I have an implementation for the interFoam (interIsoFoam) family in the following repository
https://github.com/wenyuan-fan/varRhoTurbVOF_2

You can also find a manual and a tutorial on how to use it.

hi,

I want to change your turbulence damping code for openFoam 2.2.0. How can I do?
thank you

Wenyuan · December 30, 2020, 03:56

Hi,

In OpenFOAM 2.2.0, turbulence models don't support fvOptions. There are two possible ways to use the turbulence damping model in that version.

1. You can modify the original turbulence models to allow them to use fvOptions, which can be achieved by adding several lines as you can find in new turbulence model implementations. Then you can use the turbulence damping fvOptions to add source terms.

2. You can put the source term directly into omega/epsilon equations, since you need to create new turbulence models anyway.

In either case, you may encounter warnings/errors. Unfortunately, I don't know exactly what they will be since I don't have that version installed.

jayson · December 30, 2020, 08:45

Quote:

Originally Posted by Wenyuan

Hi,

In OpenFOAM 2.2.0, turbulence models don't support fvOptions. There are two possible ways to use the turbulence damping model in that version.

1. You can modify the original turbulence models to allow them to use fvOptions, which can be achieved by adding several lines as you can find in new turbulence model implementations. Then you can use the turbulence damping fvOptions to add source terms.

2. You can put the source term directly into omega/epsilon equations, since you need to create new turbulence models anyway.

In either case, you may encounter warnings/errors. Unfortunately, I don't know exactly what they will be since I don't have that version installed.

Thank you! I will try it!!

hamdani · May 20, 2021, 05:49

Hi,

Do you have a plan to update the turbulence damping library to OpenFOAM-dev?
It seems that fvOption.H is not found in OpenFOAM-dev.

It would be very nice if you can update the library.
Many thanks

hamdani · May 20, 2021, 05:50

Quote:

Originally Posted by Wenyuan

Hi,

I have an implementation for the interFoam (interIsoFoam) family in the following repository
https://github.com/wenyuan-fan/varRhoTurbVOF_2

You can also find a manual and a tutorial on how to use it.

Hi,

Do you have a plan to update the turbulence damping library to OpenFOAM-dev?
It seems that fvOption.H is not found in OpenFOAM-dev.

It would be very nice if you can update the library.
Many thanks

September 4, 2018, 08:53	Including turbulence damping at the interface	#1
g_bochio New Member Gustavo Bochio Join Date: Mar 2012 Posts: 3 Rep Power: 14	Hello, everyone! I would like some help to add a source to the $\omega$ equation, so that I can supress turbulence generation near the interface in a two-phase flow. This link shows the source that I want to add: https://www.sharcnet.ca/Software/Ans...turb_damp.html My main questions are: 1) Is it possible to implement this directly to the fvOptions dict? 2) How could I calculate $\Delta n$ ? Thanks!

September 4, 2018, 10:36	nearWallDist and wallDist	#2
ano Member ano Join Date: Jan 2017 Location: Delft Posts: 58 Rep Power: 10	1) I didn't find a fvOptions term included in the omega equation of the kOmega models, so putting something into the fvOptions dict would not work. I guess it is easiest to put your term directly into the k-omega model. 2) There are two wall distance functions used in the turbulence models, both give you an volScalarField: a) All turbulence models inherit from "turbulenceModel". In turbulenceModel a nearWallDist y_ is defined (as a private member), which would represent your $\Delta n$ . The nearWallDist gives you the "Distance calculation for cells with face on a wall.", which should give you for all cells next to the boundary the distance between the face and the cell centre (and for the other cells it gives zero). From within your turbulence model, e.g. kkLOmega you can call it with Code: this->y() to obtain $\Delta n .$ . (From within the solver you could call it with "turbulence->y()".) b) The wallDist gives you the "Calculation of distance to nearest wall for all cells and boundary.". It is defined in some, but not all models, e.g. the kkLOmega. Unfortunately someone gave it also the variable name y_. So be careful that you don't mix them.

September 7, 2020, 05:50		#8
Phicau Senior Member Pablo Higuera Join Date: Jan 2011 Location: Auckland Posts: 627 Rep Power: 19	Hi Gustavo, I think you will find this useful: https://olaflow.github.io/blog/turbu...ons-revisited/ The code is now compatible with OF8 and 2006+ https://github.com/phicau/olaFlow_su...enceMultiphase Best, Pablo __________________ Check out my new project: olaFlow --> The olaFlow Support Thread

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How to use the CFX periodic interface	zhihuawan	CFX	61	January 15, 2018 17:20
Error - Solar absorber - Solar Thermal Radiation	MichaelK	CFX	12	September 1, 2016 06:15
Overflow Error in Multiphase Modelling with Two Continuous Fluids	ashtonJ	CFX	6	August 11, 2014 15:32
Radiation interface	hinca	CFX	15	January 26, 2014 18:11
Turbulence damping at fluid interface	Eric	Siemens	0	October 4, 2002 15:32

September 4, 2018, 13:10		#3
g_bochio New Member Gustavo Bochio Join Date: Mar 2012 Posts: 3 Rep Power: 14	Hey Ano, thanks for the reply. $\Delta n$ in my case is not the distance from a wall. It is the distance from the interface between two fluids. Any idea how to calculate this?

September 4, 2018, 15:07		#4
ano Member ano Join Date: Jan 2017 Location: Delft Posts: 58 Rep Power: 10	Hey Bochio, I am really sorry for not reading your message well and that my answer missed totally your question. No, unfortunately I don't know how to calculate it.

September 5, 2018, 09:41		#5
g_bochio New Member Gustavo Bochio Join Date: Mar 2012 Posts: 3 Rep Power: 14	No problem!

September 4, 2020, 09:58		#6
mrishi Member Rishikesh Join Date: Apr 2016 Posts: 63 Rep Power: 10	Sorry to bump this, but did you find the solution to this problem?

September 7, 2020, 04:22		#7
Wenyuan New Member Wenyuan Fan Join Date: Mar 2017 Posts: 27 Rep Power: 9	Hi, I have an implementation for the interFoam (interIsoFoam) family in the following repository https://github.com/wenyuan-fan/varRhoTurbVOF_2 You can also find a manual and a tutorial on how to use it.

December 30, 2020, 03:56		#10
Wenyuan New Member Wenyuan Fan Join Date: Mar 2017 Posts: 27 Rep Power: 9	Hi, In OpenFOAM 2.2.0, turbulence models don't support fvOptions. There are two possible ways to use the turbulence damping model in that version. 1. You can modify the original turbulence models to allow them to use fvOptions, which can be achieved by adding several lines as you can find in new turbulence model implementations. Then you can use the turbulence damping fvOptions to add source terms. 2. You can put the source term directly into omega/epsilon equations, since you need to create new turbulence models anyway. In either case, you may encounter warnings/errors. Unfortunately, I don't know exactly what they will be since I don't have that version installed.

May 20, 2021, 05:49		#12
hamdani New Member Hamdani Join Date: Nov 2011 Posts: 5 Rep Power: 15	Hi, Do you have a plan to update the turbulence damping library to OpenFOAM-dev? It seems that fvOption.H is not found in OpenFOAM-dev. It would be very nice if you can update the library. Many thanks