CFD Online URL
[Sponsors]
Home > Forums > FLUENT

convergence problems by species transport

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Display Modes
Old   May 16, 2010, 19:15
Default convergence problems by species transport
  #1
New Member
 
Felipe B. Ch.
Join Date: May 2010
Posts: 14
Rep Power: 6
Felipeb is on a distinguished road
Hi, I hope any one can help me

I am simulating a free jet injection of hydrogen in two points in co-flow with air in a cylindrical mixing tube. First I run the simulation just with air and it has a good convergence. When I activate the option of species transport(using a Kinetic Mechanism from Chemkin), all the residuals go down except the residual of the Continuity Eqn. More than that it goes up and stays around 5e+01. I all ready tried changing the mesh(for a finer one) but it doesnt show any improvement .

Thank you

Felipe
Felipeb is offline   Reply With Quote

Old   May 16, 2010, 19:39
Default
  #2
New Member
 
aek
Join Date: Feb 2010
Posts: 5
Rep Power: 6
saif is on a distinguished road
hi felipb

i think u need to give more details for ur prb

why u use the Kinetic Mechanism from Chemkin if u focus to simulate only the jet dynamycs ?
i think also u dont need to make several changes in ur mesh if u say that they gave u a ggod convergence in ur first step



Quote:
Originally Posted by Felipeb View Post
Hi, I hope any one can help me

I am simulating a free jet injection of hydrogen in two points in co-flow with air in a cylindrical mixing tube. First I run the simulation just with air and it has a good convergence. When I activate the option of species transport(using a Kinetic Mechanism from Chemkin), all the residuals go down except the residual of the Continuity Eqn. More than that it goes up and stays around 5e+01. I all ready tried changing the mesh(for a finer one) but it doesnt show any improvement .

Thank you

Felipe
saif is offline   Reply With Quote

Old   May 16, 2010, 20:05
Default
  #3
New Member
 
Felipe B. Ch.
Join Date: May 2010
Posts: 14
Rep Power: 6
Felipeb is on a distinguished road
I am using a chemkin kinetic mechanism because I want to later on simulate the ignition by activating the volumetric reactions, but first I wanted to check for the mixing process(profile), thats why right now I am simulating without reactions.
On the order side if I change the composition of the fuel inlet (hydrogen) to for example O2 or N2. The residuals start going down but very slowly, thats why I thought in making the mesh finer to maybe improve simulation of the diffusion process of H2 but I increase its density in almost factor 3 and I dont see any change. By the way im simulating 3D with symmetry boundaries in a quarter of a cylinder(just a quarter because the of the h2 injection position).
Felipeb is offline   Reply With Quote

Old   May 19, 2010, 00:15
Default
  #4
New Member
 
Mohammad Ghazvini
Join Date: Mar 2010
Posts: 15
Rep Power: 6
mahmohammad is on a distinguished road
I had the same problem with continuity. I played with the under-relaxation factors and it worked. decrease all the under relaxation factors to 0.2 for example. Try that and wait until it converges.

Mohammad
mahmohammad is offline   Reply With Quote

Old   May 19, 2010, 03:18
Default
  #5
New Member
 
Felipe B. Ch.
Join Date: May 2010
Posts: 14
Rep Power: 6
Felipeb is on a distinguished road
Thanks!!!! mahmohammad I am going to try that.
Felipeb is offline   Reply With Quote

Old   May 28, 2010, 20:43
Thumbs up it did work
  #6
New Member
 
Felipe B. Ch.
Join Date: May 2010
Posts: 14
Rep Power: 6
Felipeb is on a distinguished road
For the people who saw my thread, is now solved:

I improved the mesh quality; in some areas the Aspect Ratio wasnt that good, (my geometry wasnt that simple). I increase a little bit the amount of cells close to injections points. I used the Pressure based solver with the coupled algorithm and I reduced the courant number to 40 at the beginning and later to 100 to increase the convergence speed. I also reduce the under-relaxation factors for the eqn of the species I was simulating from the default one to a value of 0.8 . The user manual of fluent does not recommend to use under-relaxation factors below 0.5.

Thanks Mohammad for your help

good luck to the others

Felipe
Felipeb is offline   Reply With Quote

Old   May 29, 2010, 15:19
Default
  #7
New Member
 
Mohammad Ghazvini
Join Date: Mar 2010
Posts: 15
Rep Power: 6
mahmohammad is on a distinguished road
Felipe

I happy that your problem is solved.
FYI, in some cases I had to start with underrelaxation factor of 0.01. You have to decrease it when there is sudden change in a variable. And is very common in reaction and species problems.

Mohammad
mahmohammad is offline   Reply With Quote

Old   June 3, 2010, 12:19
Default
  #8
New Member
 
Felipe B. Ch.
Join Date: May 2010
Posts: 14
Rep Power: 6
Felipeb is on a distinguished road
Ok, I will take it in account, because reactions is my direction =) ... =S.
Thanks a lot again

Felipe Bolanos
Felipeb is offline   Reply With Quote

Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Species transport Eisa FLUENT 0 May 17, 2006 05:40
Species transport eqn using UDS Vishal FLUENT 0 March 22, 2006 00:00
first step in species transport without reaction naus FLUENT 0 February 26, 2006 17:54
species transport equation daniel FLUENT 0 June 22, 2005 09:51
unsteady species transport solomon FLUENT 0 December 19, 2003 20:06


All times are GMT -4. The time now is 08:23.