|
[Sponsors] |
August 6, 2013, 13:34 |
fixedGradient use
|
#1 |
Senior Member
James
Join Date: May 2013
Posts: 116
Rep Power: 12 |
Hi everyone,
In my setup, I have an inlet and an outlet. The only info is that P_in=0 Pa and P_out=-20 Pa, so can I use fixedGradient in this way? (p file) in----type fixedValue value 0 out---type fixedGradient gradient -20 I guess itīs something more complex than this. If so, how can I compute my pressure gradient with the information I have? Thanks. |
|
August 6, 2013, 16:39 |
|
#2 |
Senior Member
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,208
Rep Power: 26 |
Hi
set both to fixedValue if you have their values and set velocities to fixedGradient. temperature in inlet should be fixedValue and in outlet fixedGradient. |
|
August 7, 2013, 04:42 |
|
#3 |
Senior Member
James
Join Date: May 2013
Posts: 116
Rep Power: 12 |
Thanks for the reply Ehsan,
I am not sure if I understanding well. Do you mean : (p file) in----type fixedValue value uniform 0 out---type fixedValue value uniform -20 (u file) in----type fixedGradient gradient ??? out---type fixedGradient gradient ??? If I set fixedValue for p and zeroGradient for U the solution converges but without physical meaning. Which values should I put for gradient in the U file??? Thank you so much |
|
August 7, 2013, 05:14 |
|
#4 |
Senior Member
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,208
Rep Power: 26 |
U's should be zeroGradient when p value is specified.how you say its physically meaningless? you should specify p and T in inlet and only p at outlet.
|
|
August 7, 2013, 05:41 |
|
#5 |
Senior Member
James
Join Date: May 2013
Posts: 116
Rep Power: 12 |
Ehsan,
I attach the files of my last simulation. It has no physical meaning because I expect velocitites between 4-6 m/s and pressure values of about 0/-20 Pa, and the results starts to increase in magnitude until they reach values of about 1e10 in both variables. Is anyone understanding what's happening? I have used zeroGradient instead of pressureInletVelocity in a first stage, but nothing seems to change... P file /*--------------------------------*- C++ -*----------------------------------*\ | ========= | | | \\ / F ield | OpenFOAM: The Open Source CFD Toolbox | | \\ / O peration | Version: 2.1.x | | \\ / A nd | Web: www.OpenFOAM.org | | \\/ M anipulation | | \*---------------------------------------------------------------------------*/ FoamFile { version 2.0; format ascii; class volScalarField; object p; } // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // dimensions [0 2 -2 0 0 0 0]; internalField uniform 0; boundaryField { in_left { type fixedValue; value uniform 0; } in_right { type fixedValue; value uniform 0; } out { type fixedValue; value uniform -20; } wall { type zeroGradient; } } U file /*--------------------------------*- C++ -*----------------------------------*\ | ========= | | | \\ / F ield | OpenFOAM: The Open Source CFD Toolbox | | \\ / O peration | Version: 2.1.x | | \\ / A nd | Web: www.OpenFOAM.org | | \\/ M anipulation | | \*---------------------------------------------------------------------------*/ FoamFile { version 2.0; format ascii; class volVectorField; object U; } // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // dimensions [0 1 -1 0 0 0 0]; internalField uniform (0 0 0); boundaryField { in_left { type pressureInletVelocity; value uniform (0 0 0); } in_right { type pressureInletVelocity; value uniform (0 0 0); } out { type zeroGradient; } wall { type fixedValue; value uniform (0 0 0); } } // ************************************************** *********************** // T file /*--------------------------------*- C++ -*----------------------------------*\ | ========= | | | \\ / F ield | OpenFOAM: The Open Source CFD Toolbox | | \\ / O peration | Version: 1.6.x | | \\ / A nd | Web: www.OpenFOAM.org | | \\/ M anipulation | | \*---------------------------------------------------------------------------*/ FoamFile { version 2.0; format ascii; class volScalarField; location "0"; object T; } // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // dimensions [ 0 0 0 1 0 0 0 ]; internalField uniform 293; boundaryField { in_left { type fixedValue; value uniform 293; } in_right { type fixedValue; value uniform 293; } out { type zeroGradient; } wall { type fixedValue; value uniform 306; } } // ************************************************** *********************** // I am a beginner so if someone knows a better way to make the set up, I only have a channel (a little complex) with two inlets, an outlet and a wall. I know pressure at inlet and at oultlet, I have no info about verlocities and I know temperature values on the boundaries. Thanks for all your help, itś difficult to star with OF alone... |
|
August 7, 2013, 16:54 |
|
#6 |
Senior Member
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,208
Rep Power: 26 |
Hi
is your case incompresseble or compressible? its nice if you could send me your case so that I can have a look into it(delete unnecessary files if you want to send for little space)
__________________
Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King. To Be or Not To Be,Thats the Question! The Only Stupid Question Is the One that Goes Unasked. |
|
August 8, 2013, 05:56 |
|
#7 |
Senior Member
James
Join Date: May 2013
Posts: 116
Rep Power: 12 |
Hi Ehsan,
I am working with stationary, incompressible and laminar flow (low Re and M, and quasistationary state due to the values of Strouhal and Womersley numbers). I attach my files (excep mesh file, I am not allowed to share it). I am a beginner in OF. but I have tried several things realted to BCs, URFs, nonOrthogonalCorrectors...and always failed!I cannot understand why pressure driven flows gives so many problems. If you see something wrong or you have any suggestions Iĺl be very grateful (I have been working on this issue several weeks, and all threads related seems to have similar troubles) Thank you for your knoweldge Ehsan! (I read some of your posts, it will be useful for me to start using pressureTools, but first I need to solve this!)CASE_PRESSURE_DRIVEN.tar.gzCASE_PRESSURE_DRIVEN.tar.gz |
|
August 9, 2013, 15:30 |
|
#8 |
Senior Member
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,208
Rep Power: 26 |
Hi
set internalField to -.25 and test it and after that if not resolved set "value uniform 4" for velocity. why nNonOrthogonalCorrectors set to 8?is your mesh unstructured? its too high.it better be 3 for maximum for bad meshes.
__________________
Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King. To Be or Not To Be,Thats the Question! The Only Stupid Question Is the One that Goes Unasked. |
|
August 10, 2013, 05:54 |
|
#9 |
Senior Member
James
Join Date: May 2013
Posts: 116
Rep Power: 12 |
Thanks Ehsan, but I am not sure if I am understanding well:
You say "set internalField to -.25 and test it and after that if not resolved set "value uniform 4" for velocity" 1. Do you mean that I have to write in p file: internalField uniform -0.25 or -25 for the initial condition?? 2. If I have no success, you suggest that I write: internalField uniform (0 4 0) or this magnitude you propose is related to any BC in the U file? Why have you made this choice? Are these values arbitrary or you have used any theoreticall results? Please give me some light over this... Thank you so much for your help, I hope I can return to you some day... |
|
August 10, 2013, 07:41 |
|
#10 |
Senior Member
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,208
Rep Power: 26 |
Hi James
you have put -0.25 in out: out { type fixedValue; value uniform -0.25; } it should be -0.25 or -25? I say one time set internalField value same as outlet(-25 or -.25 ?) instead of inlet value(0), set that value for velocity on inlet patches,that value only used for first iteration and may be better that it has some velocity instead of starting from rest,and 4 is a guess only test it for 1 or 6 too. your welcome.hope this resolved.
__________________
Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King. To Be or Not To Be,Thats the Question! The Only Stupid Question Is the One that Goes Unasked. |
|
August 12, 2013, 07:13 |
|
#11 |
Senior Member
James
Join Date: May 2013
Posts: 116
Rep Power: 12 |
Hi Ehsan,
Unfortunately, I have no success with only establishing the internalField of pressure file the same as the outle value. At first (maybe the first 30 iterations) all looks good, but then suddenly residuals started to rise up and values of pressure and velocity explodes. I am not understanding how to use the other advise. If I set a value for velocity in patches I guess I am changing the problem (or at least overconstraining it, because at these boundaries I have established info for pressure). Do you think it is a good practise?? For example: p file: internalField uniform -20 in---fixedValue 0 out--fixedValue -20 wall---zeroGradient U file internalField uniform 1 (this value is a guess as you said?) in---pIOV (0 0 0) (or a zeroGRadient??) out---zeroGradient (or a fixedValue, the same as I put in internalField 1??) wall---fixedValue uniform (0 0 0) I am disappointed, I see many degrees of freedom (maybe the problem is the mesh, but I have tried three or four different mesh configurations, or perharps the problem is turbulence, because I am on limit of laminar-turbulent flow, or even the sharp geometry makes solovers go crazy...) Thank you so much for all your recommendations! |
|
August 12, 2013, 08:30 |
|
#12 |
Senior Member
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,208
Rep Power: 26 |
Hi
its good if you send me your mesh so that I can run it to see what happen.
__________________
Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King. To Be or Not To Be,Thats the Question! The Only Stupid Question Is the One that Goes Unasked. |
|
August 13, 2013, 04:21 |
|
#13 |
Senior Member
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,208
Rep Power: 26 |
__________________
Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King. To Be or Not To Be,Thats the Question! The Only Stupid Question Is the One that Goes Unasked. |
|
August 13, 2013, 11:31 |
|
#14 |
Senior Member
James
Join Date: May 2013
Posts: 116
Rep Power: 12 |
Ehsan,
All looks great!I am starting simulation and residuals are falling and magnitude of variables remain bounded as expected! At first it looks like OF diverges a little with respect to FLUENT values, I need to do a grid independency study. But nothing could be possible without your last post! Thank you so much, your help was awesome! I only have one doubt. If I want to establish a pressure gradient of 20 Pa, how can I set the static pressure at the outlet? I have P_total_in=101325 Pa, P_total_out=101305 Pa, but P_static_out=P_total_out-0.5*rho*v^2, but I donīt have info about velocity values at the outlet. Is it enough to set a fixedValue of 101305) I guess I have to compute first dynamic pressure, but I donīt know how without v values. Once again THANK YOU VERY MUCH FOR SHARING KNWOLEDGE! |
|
August 13, 2013, 11:59 |
|
#15 |
Senior Member
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,208
Rep Power: 26 |
I'm glad your problem resolved.could you send me the corrected case?
you should set something lower than 101325 in outlet and let solver compute the velocity there after that look that if p0 is 101325 or is lower or upper and adjust pressure,but I don't understand,if total pressures are the same then what makes the flow of fluid? and could you do something for me? please click on balance button in one of my posts you like and approve that.so my rep power can be increased and correct! thanks
__________________
Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King. To Be or Not To Be,Thats the Question! The Only Stupid Question Is the One that Goes Unasked. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
fixedGradient BC type with the channelOodles solver | Bedotto | OpenFOAM Running, Solving & CFD | 5 | April 15, 2019 09:49 |
fixedGradient BC | immortality | OpenFOAM Pre-Processing | 0 | July 5, 2013 02:17 |
fixedGradient BC w/ non-uniform wall spacing | brent_craven | OpenFOAM | 5 | September 19, 2012 12:12 |
wall temperature fixedGradient in buoyantBoussinesqSimpleFoam | David_010 | OpenFOAM | 0 | February 8, 2010 10:05 |
fixedGradient boundary condition | skabilan | OpenFOAM Running, Solving & CFD | 2 | December 3, 2009 13:55 |