|
[Sponsors] |
Code behaves strangely when the geometry is changed |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
January 7, 2009, 16:04 |
Hi All,,
With reference to
|
#1 |
Senior Member
Vishal Nandigana
Join Date: Mar 2009
Location: Champaign, Illinois, U.S.A
Posts: 208
Rep Power: 18 |
Hi All,,
With reference to my previous message, I am solving an electro osmotic flow in a simple square channel. Here is a paper from that, but right now I am neglecting the velocity term.. I finished coding it but had some convergence issues because of the large values of the transport properties and hence I non dimensionalised them and now the code is converging, my concentration values at inlet and outlet boundary conditions are same (=1). Hence the result should be 1 throughout right.. but stangely I am not getting 1.. instead I am getting some strange values... as mentioned in the earlier mesage, I would like to know whether I have coded them correctly or not.. Kindly do the needful... Thanks coupledFoam.tar.gz controlDict |
|
January 7, 2009, 16:26 |
Sorry I couldnt upload the pap
|
#2 |
Senior Member
Vishal Nandigana
Join Date: Mar 2009
Location: Champaign, Illinois, U.S.A
Posts: 208
Rep Power: 18 |
Sorry I couldnt upload the paper because of the file size limit... I am attaching the page where all the equations are given..
D1 = 1.96e-9 m2/s D2 = 0.87e-9 m2/s Z1 = 1 Z2 = -1 omega1 = D1/RT, R = 8.314J/molK, T = 297K omega2 = D2/RT |
|
January 7, 2009, 16:27 |
Sorry I couldnt upload the pap
|
#3 |
Senior Member
Vishal Nandigana
Join Date: Mar 2009
Location: Champaign, Illinois, U.S.A
Posts: 208
Rep Power: 18 |
Sorry I couldnt upload the paper because of the file size limit... I am attaching the page where all the equations are given..
D1 = 1.96e-9 m2/s D2 = 0.87e-9 m2/s Z1 = 1 Z2 = -1 omega1 = D1/RT, R = 8.314J/molK, T = 297K omega2 = D2/RT I have normalised all the equations and proceeded the coding and I am getting some strange results as mentioned previously.. Kindly give your valuable suggestions... Thanks Regards Vishal |
|
January 7, 2009, 16:37 |
Sorry I couldnt upload the pap
|
#4 |
Senior Member
Vishal Nandigana
Join Date: Mar 2009
Location: Champaign, Illinois, U.S.A
Posts: 208
Rep Power: 18 |
Sorry I couldnt upload the paper because of the file size limit...Here are the equations
dC1/dt = div(D1*gradC1)+ div(omega1*Z1*F*C1*grad(Phi)) dC2/dt = div(D2*gradC2)+ div(omega2*Z2*F*C2*grad(Phi)) div(grad(Phi)) = (-F*Z1*C1-F*Z2*C2)/e0*er where, Phi = electric field F - faradays constant e0 = permittivity of vacuum er = permittivity of water = 80 D - diffusion coefficient D1 = 1.96e-9 m2/s D2 = 0.87e-9 m2/s Z = valence number Z1 = 1 Z2 = -1 omega1 = D1/RT, R = 8.314J/molK, T = 297K omega2 = D2/RT I have normalised all the equations and proceeded the coding and I am getting some strange results as mentioned previously.. Kindly give your valuable suggestions... Thanks Regards Vishal |
|
January 8, 2009, 03:17 |
hi Vishal,
I am not very sure
|
#5 |
Member
Sachin Kanetkar
Join Date: Mar 2009
Posts: 57
Rep Power: 17 |
hi Vishal,
I am not very sure but i am thinking this is because the grid size. I have strong doubts that the convergence also fails because of the big grid size, maybe there are issues with time step and Courant no also. Since your details never mentioned about these things. maybe that is missing. This are just suggestions havent gone thru those long details. regards Sachin |
|
January 8, 2009, 13:06 |
Thanks Sachin.. I had figured
|
#6 |
Senior Member
Vishal Nandigana
Join Date: Mar 2009
Location: Champaign, Illinois, U.S.A
Posts: 208
Rep Power: 18 |
Thanks Sachin.. I had figured out the problem and now the code is working fine... Thanks for the help.
|
|
January 8, 2009, 17:29 |
Hi,
As mentioned earlier my
|
#7 |
Senior Member
Vishal Nandigana
Join Date: Mar 2009
Location: Champaign, Illinois, U.S.A
Posts: 208
Rep Power: 18 |
Hi,
As mentioned earlier my code is working properly and I get the desired results. However, I have some queries. I have four variables C1,C2 and Phi and U. I obtain the values of these variables at each time step. Now I would like to calculate another parameter T1 = -D1*grad(C1) - K1 * C1 * grad (Phi) + C1 * U T2 = -D2*grad(C2) - K2 * C2 * grad (Phi) + C2 * U Where D1,D2,K1,K2 are constants... Then I would like to calculate I by integrating T1 and T2 over the cross section given by the formula I = integral over S {(Z1*F*T1)+(Z2*F*T2)} dS where S is the cross sectional area of the channel.. Kindly let me know how to proceed further.. Thanks Regards Vishal |
|
January 14, 2009, 15:35 |
Regards
Vishal
|
#8 |
Senior Member
Vishal Nandigana
Join Date: Mar 2009
Location: Champaign, Illinois, U.S.A
Posts: 208
Rep Power: 18 |
Regards
Vishal |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
has someone changed the prepdf's thermo.db | yinyue | FLUENT | 1 | May 19, 2017 03:16 |
esqr - definition changed from version 9.2 to 9.3? | Andreas | FLOW-3D | 1 | February 8, 2009 23:23 |
Math rendering default changed to always PNG | Jonas Larsson | CFD-Wiki | 0 | December 10, 2005 10:00 |
Wiki URL changed | Jonas Larsson | CFD-Wiki | 0 | November 7, 2005 16:56 |
Can the changed mesh file be used? | Kim | FLUENT | 5 | September 8, 2004 09:49 |