Solving Transport equation for additional variable using CFX

creddy_trddc · June 3, 2011, 08:44

Hi everyone,

I want to solve for transport equation for additional variable in CFX (Eqn. is attached with this post). Convective term of the equation has an additional velocity vector added to the fluid velocity, except this equation is in standard transport equation which can be solved using Ansys CFX.

Can Some one tell me how to incorporate this additional term in CFX solver.

Thnx in advance

baserinia · June 3, 2011, 13:36

Is the additional velocity divergence-free? If it is, all you need to do is add a source term like this to your transport equation: $\rho u_1 \cdot \nabla N^*$

creddy_trddc · June 6, 2011, 02:45

Hi Amir,

Thnx for your reply...additional velocity is the floatation velocity, it depends on the size of the particle it has only vertical component. So I think this additional velocity is not divergence free....

I have tried to simulate the way you told by taking the additional velocity as divergence and keeping this term as source but I m not getting the results..

Can you please tell me in CFX is it possible to caluculate the gradient of userdefined variable ( $\nabla\cdot\\\left(u_N^* N^*\right)\$ )
where

is the additional variable...and

is defined as function of particle size using USER CEL

ghorrocks · June 6, 2011, 08:42

What are you trying to model? Why are the built-in particle or eularian models inadequate?

creddy_trddc · June 6, 2011, 09:09

Dear Glenn,

I want solve the Transient-Transport equation for Number Density Distribution of Spherical Particles in the steady state flow field.

In the equation attached in my starting post...additional convective velocity component added to primary fluid velocity is due to the Floatation caused due to the density distribution between primary fluid and particles. The Source term includes the effects different collision mechanisms (Turbulent, Stokes, Adhesion to walls etc...) between the particles.

Can you plz tell me how to incorporate this Additional COnvective term.

Presently I am creating an Algebraic expression for

and accessing it's divergence in CFX PRE using user subroutine.

Can u plz tell me this approach is correct?

ghorrocks · June 6, 2011, 09:28

There are already built-in with CFX models for floatation, turbulent, viscous/stokes, adhesion to walls etc. Why are these not acceptable?

creddy_trddc · June 6, 2011, 09:32

Dear Glenn,

I am using CFX5 (ANSYS 11)...are this models available in this version...I am unable to find them....Can you please give some reference where I can get the details of these models in CFX.

Thanks a lot for your help.

creddy_trddc · June 6, 2011, 10:07

Dear Glenn,

Here I am giving the link for the Journal paper which I am refering for my work,

http://www.jstage.jst.go.jp/article/...0/11/1597/_pdf

on Page no: 1599 the details of transport equations were mentioned.

thanks.

ghorrocks · June 6, 2011, 20:19

I don't have time to read it in detail, but from a quick look it describes a eularian-eularian model where the equation you describe is implemented. So why don't you also do a eularian-eularian model in CFX and just add the extra models for agglomeration and the other unusual bits. This way you can use the default handling of the phase slip velocities, viscous effects, floatation etc and only need add a few minor extra models.

Zaktatir · June 16, 2011, 09:35

how can you get ride of the turbulent diffusion term.. i am trying to define a correct number density additional variable of separated droplets from water film but i am struggling with that term: it leads to very big number of droplets..

ghorrocks · June 16, 2011, 19:57

I am not an expert in the details of the model. I would contact CFX support for some examples and advice.

chivalryflamen · April 2, 2015, 22:07

Hi Creddy,

I know this is a old thread, but did you ever get to solve this problem? I'm having a similar term in my equation and it really bothers me... THanks,

Luke

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
SimpleFoam k and epsilon bounded	nedved	OpenFOAM Running, Solving & CFD	16	March 4, 2017 09:30
Extrusion with OpenFoam problem No. Iterations 0	Lord Kelvin	OpenFOAM Running, Solving & CFD	8	March 28, 2016 12:08
emag beta feature: charge density	charlotte	CFX	4	March 22, 2011 10:14
SimpleFoam k and epsilon bounded	nedved	OpenFOAM Running, Solving & CFD	1	November 25, 2008 21:21
Could anybody help me see this error and give help	liugx212	OpenFOAM Running, Solving & CFD	3	January 4, 2006 19:07

June 3, 2011, 13:36		#2
baserinia New Member A.R. Baserinia Join Date: Jan 2010 Location: Canada Posts: 24 Rep Power: 16	Is the additional velocity divergence-free? If it is, all you need to do is add a source term like this to your transport equation: $\rho u_1 \cdot \nabla N^*$

June 6, 2011, 02:45		#3
creddy_trddc New Member creddy Join Date: Aug 2010 Location: Kanpur/UP/India Posts: 25 Rep Power: 16	Hi Amir, Thnx for your reply...additional velocity is the floatation velocity, it depends on the size of the particle it has only vertical component. So I think this additional velocity is not divergence free.... I have tried to simulate the way you told by taking the additional velocity as divergence and keeping this term as source but I m not getting the results.. Can you please tell me in CFX is it possible to caluculate the gradient of userdefined variable ( $\nabla\cdot\\\left(u_N^* N^*\right)\$ ) where is the additional variable...and is defined as function of particle size using USER CEL

June 6, 2011, 08:42		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	What are you trying to model? Why are the built-in particle or eularian models inadequate?

June 6, 2011, 09:09		#5
creddy_trddc New Member creddy Join Date: Aug 2010 Location: Kanpur/UP/India Posts: 25 Rep Power: 16	Dear Glenn, I want solve the Transient-Transport equation for Number Density Distribution of Spherical Particles in the steady state flow field. In the equation attached in my starting post...additional convective velocity component added to primary fluid velocity is due to the Floatation caused due to the density distribution between primary fluid and particles. The Source term includes the effects different collision mechanisms (Turbulent, Stokes, Adhesion to walls etc...) between the particles. Can you plz tell me how to incorporate this Additional COnvective term. Presently I am creating an Algebraic expression for and accessing it's divergence in CFX PRE using user subroutine. Can u plz tell me this approach is correct?

June 6, 2011, 09:28		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	There are already built-in with CFX models for floatation, turbulent, viscous/stokes, adhesion to walls etc. Why are these not acceptable?

June 6, 2011, 09:32		#7
creddy_trddc New Member creddy Join Date: Aug 2010 Location: Kanpur/UP/India Posts: 25 Rep Power: 16	Dear Glenn, I am using CFX5 (ANSYS 11)...are this models available in this version...I am unable to find them....Can you please give some reference where I can get the details of these models in CFX. Thanks a lot for your help.

June 6, 2011, 10:07		#8
creddy_trddc New Member creddy Join Date: Aug 2010 Location: Kanpur/UP/India Posts: 25 Rep Power: 16	Dear Glenn, Here I am giving the link for the Journal paper which I am refering for my work, http://www.jstage.jst.go.jp/article/...0/11/1597/_pdf on Page no: 1599 the details of transport equations were mentioned. thanks.

June 6, 2011, 20:19		#9
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	I don't have time to read it in detail, but from a quick look it describes a eularian-eularian model where the equation you describe is implemented. So why don't you also do a eularian-eularian model in CFX and just add the extra models for agglomeration and the other unusual bits. This way you can use the default handling of the phase slip velocities, viscous effects, floatation etc and only need add a few minor extra models.

June 16, 2011, 09:35		#10
Zaktatir Senior Member Join Date: Jan 2010 Location: Germany Posts: 268 Rep Power: 17	how can you get ride of the turbulent diffusion term.. i am trying to define a correct number density additional variable of separated droplets from water film but i am struggling with that term: it leads to very big number of droplets..

June 16, 2011, 19:57		#11
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	I am not an expert in the details of the model. I would contact CFX support for some examples and advice.

April 2, 2015, 22:07		#12
chivalryflamen New Member Luke Xu Join Date: Dec 2014 Posts: 1 Rep Power: 0	Hi Creddy, I know this is a old thread, but did you ever get to solve this problem? I'm having a similar term in my equation and it really bothers me... THanks, Luke