
[Sponsors] 
November 1, 2006, 03:17 
VOF&TRANSIENT AND CONVERGE

#1 
Guest
Posts: n/a

I am trying to model annular flow as 2D axisymmetric under zero gravity with VOF model. There are two velocity inlet as vapor and liquid in my horizontal pipe model. There is a liqud, gas and interaction domain in my model. I described two pressure outlet with same static pressure. Time step is very low and residuals seems normal but I think I need to wait for a long time. How can I understand this solution will converge or not ? thanks already for your answers


November 1, 2006, 04:57 
Re: VOF&TRANSIENT AND CONVERGE

#2 
Guest
Posts: n/a

hi mark, i have a question to u, im trying to model stratified flow in a horizontal pipe using vof method,i also want to specify two phases liquid and vapour at inlet of the pipe, and in between there will be interaction in the domain and i want to calculate the volume fraction of gas formed to liquid heating. can u give me idea regarding this,please. i have the fluent tutorial of vof method in that they have taken domain dimensions i,e horizontal and vertical as according to taylorraleigh instability parameter, do u have any idea regarding this,and also in that they have used the vertical flow,can it be used to horizontal flow like which we are doing. please reply, and one more clarification , In stratified flow already some liquid will be there with vapour, and in your annular case also how did u specify the volume fraction of liquid in that? i will be eagerly waiting for ur reply


November 1, 2006, 05:49 
Re: VOF&TRANSIENT AND CONVERGE

#3 
Guest
Posts: n/a

Firstly, I didnt take into account liquid in the gas phase because of the model limitation. I seperated gas and liquid as velocity inlet. Liquid inlet thickness must be equal liquid film tickness ( I have experimental results )If you want to calculate effect of heat transfer you will need a UDF for this purpose ( if you consider phase change ). You need to define phase change rate in the mass conservation equation and latent heat in the energy equation but certainly this is not a simple thing. I hope I could help you.


November 1, 2006, 23:11 
Re: VOF&TRANSIENT AND CONVERGE

#4 
Guest
Posts: n/a

hi mark thanks for your reply, if we are writing udf for source and energy terms, can we track the volume fraction of liquid in the domain? for your problem how you fixed the horizontal domain and vertical domain, and how much did you take the initial mesh resolution? 2. do u have any idea about taylorraleigh instability parameter? waiting for your reply.


November 2, 2006, 08:35 
Re: VOF&TRANSIENT AND CONVERGE

#5 
Guest
Posts: n/a

In fact I could not reach converged solution. I am still working on...I tried to write a UDF for condensation but this is not a simple problem. We can find volume fraction of liquid domain with VOF. I have not got enough informaiton about taylorraleigh instability parameter except fluent tutorial which you told before.


November 3, 2006, 00:00 
Re: VOF&TRANSIENT AND CONVERGE

#6 
Guest
Posts: n/a

hi mark thanks for reply, it well help me but have u seen the fluent tutorial on vof in fluent. in that he has calculated the domain (i,e horizontal nad vertical dimensions of grid) of the problem based on the rt instability parameter. how did u fix your domain dimensions i,e horizontal and vertical size? waiting for your reply


November 3, 2006, 02:31 
Re: VOF&TRANSIENT AND CONVERGE

#7 
Guest
Posts: n/a

I think you mean how can we separete domains? I meshed liquid and vapor domains seperately in my 2D model and (horizontal pipe) and described different velocities for vapor and liquid phase at pipe inlet.


November 3, 2006, 04:35 
Re: VOF&TRANSIENT AND CONVERGE

#8 
Guest
Posts: n/a

Mark no, what i mean is how did you fix the domain, means for suppose 2D horizontal pipe mesh, there will be horizontal dimension of the grid and also vertical dimension, and also how did u define the distance between the nodes. i think u have fluent tutorial on vof method, in that tutorial it is mentioned that the horizontal and vertical dimension of the grid is fixed on the taylorraleigh instability parameter. 2.thats fine we can define different velocities for vapour and liquid at inlet. but how have you meshed the liquid and vapour domains separately? can u tell me the process behind that meshing of different domains seperately? waiting for ur reply


November 3, 2006, 08:44 
Re: VOF&TRANSIENT AND CONVERGE

#9 
Guest
Posts: n/a

I didnt use a special parameter to define mesh dimension. I choosed possible mesh scheme according to pipe dimension. My model is 2D and rectangular. Right side of this model inlet and left side outlet. There is two domain with interaction as liquid and vapor in this rectangular geometry. Liquid domain is represented by a other rectangular and that of vapor. Consequenlty,I meshed this subdomains seperately like below.
.................................................. . liquid domain velocity inlet . .................................................. . . . vapor domain velocity inlet . . . . . .................................................. 

November 6, 2006, 03:34 
Re: VOF&TRANSIENT AND CONVERGE

#10 
Guest
Posts: n/a

hi mark, did u consider taylorraleigh instability parameter for your problem? in your problem do u have any change of phase of liquid to vapour? can u please once again explain how did u mesh the vapour and liquid domain seperately which u have mentioned in your earlier mail, i did not get it clearly. waiting for your response.


November 11, 2006, 04:35 
Re: VOF&TRANSIENT AND CONVERGE

#11 
Guest
Posts: n/a


November 12, 2006, 23:18 
Re: VOF&TRANSIENT AND CONVERGE

#12 
Guest
Posts: n/a

hi mark you told me that, u have created two rectangles one for liquid and one for vapour. your inlet for liquid and and vapour , are they different? u told that u did not use any special parameter for mesh resolution, then during your interaction of liquid and vapour phase did u get the volume feraction of vapour evaporated from liquid due to mass transfer? i have inlet of both vapour and liquid at same section, then can i go for the type of meshing in the similar way which u have done. waiting for your reply, as early as possible.


November 15, 2006, 02:34 
Re: VOF&TRANSIENT AND CONVERGE

#13 
Guest
Posts: n/a

hi,mark you said in your mail that u have created two rectangles for liquid and vapour.have u taken interaction as one of your choice as boundary conditions. please send me what is the dimension u have selected for these two rectangles? expecting reply.


Thread Tools  
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Laminar doesn't converge; Turbulent models do?  Amit  FLUENT  11  April 23, 2015 22:55 
HELP !In relaxtion factor converge is taken or not  MANOJ KUMAR  FLUENT  5  September 22, 2005 04:16 
Converge problem for multiphase flow  Jen  FLUENT  2  September 8, 2005 08:47 
Converge problem for multiphase flow  Jen  FLUENT  4  July 20, 2005 16:52 
converge problem  Li Jingping  FLUENT  2  July 8, 2003 02:54 