|
[Sponsors] |
August 4, 2014, 18:18 |
SimpleFoam - instable simulation
|
#1 |
New Member
Join Date: Jul 2014
Posts: 17
Rep Power: 12 |
Hello,
I try to simulate a normal piple flow with one outlet and one inlet. The pressure at inlet is set to "zero gradient" the velocity at the inlet is set to 3 m/s. the pressure at the outlet is set to 0, the velocity at the outlet is set to zero gradient. in the middel part the velcity is set to uniform 0 and the pressure to "zero gradient". I use the solver simpleFoam. The turbulence model has been switched off in order to speed up the calculations time. Now the fundamental problem: At the beginning everything looks as expected. the solver works well. After 50 iterations the simluations starts getting unstabel. the velocity at the outlet rises enormously. after 100 iterations the simlations is broken! Do you have any idea, how the simlation can be stabilized? Which possibilities do i have to change the solver options? is it necessary to minimize the time steps? Is the wrong solver being used? Best regards! |
|
August 5, 2014, 09:09 |
|
#2 |
Senior Member
Laurent DASTUGUE
Join Date: May 2014
Location: Chartres, France
Posts: 122
Rep Power: 12 |
Hi Specialist,
you don't say enough about your case to be helped . Firstly : What is the value of your timestep ? In fact, this value is less important in SimpleFoam than in transient calculations but it is a good information to understand the job. Secondly : Which kind of thermophysical properties do you use ? Maybe you have some values which are uncompatible. I don't know if i can help you but i know that i need this informations to try... Have a good day. Laurent |
|
August 7, 2014, 06:07 |
|
#3 |
Senior Member
Philipp
Join Date: Jun 2011
Location: Germany
Posts: 1,297
Rep Power: 27 |
Hey specialist,
Firstly, the solver simpleFoam is a steady-state solver. The time-step in your controlDict is just used to name the directories where the current iteration is stored. It doesn't change anything at all in the solver. I normally set it to "1", so the names of the directories (1000, 2000, ...) match the value of the iteration. Secondly, what Reynolds number do you have? You write "Normal pipe" with 3m/s - that should be very turbulent for most cases. Generally nearly every pipe flow that is used in some factory / experiment is turbulent. That means, your simulation can crash just due to the fact that you solve the wrong equations. If this is the case, run the case with a turbulence model and see what happens. Also, If you need more help you need to post the log file and all boundary conditions.
__________________
The skeleton ran out of shampoo in the shower. |
|
August 7, 2014, 06:28 |
|
#4 |
New Member
Join Date: Jul 2014
Posts: 17
Rep Power: 12 |
That is a good point! Where is the Reynold number defined? Can you give me an estimated value of a Reynold number, I should use? So far, I have copied the parameters from a simplefoam tutoial.
Do I have to pay attenttion to any parameters if i use the turbulence modell? Otherwise I would copy the turbulence parameters from a simple foam tutorium. Is that possible? Regards! |
|
August 7, 2014, 07:17 |
|
#5 |
Senior Member
Philipp
Join Date: Jun 2011
Location: Germany
Posts: 1,297
Rep Power: 27 |
Re = D * v / nu
With nu the value you set in the constant/transportProperties file and "D" the diameter of the tube and "v" the averave velocity (3m/s). You will activate the turbulence model by setting "RASModel" in constant/turbulenceProperties and the name of the model (such as kOmegaSST) into constant/RASProperties. When starting simpleFoam it will ask for every missing file in the 0 directory.
__________________
The skeleton ran out of shampoo in the shower. |
|
August 8, 2014, 08:49 |
|
#6 |
New Member
Join Date: Jul 2014
Posts: 17
Rep Power: 12 |
okay,so far the simulations runs a way better, when I use the turbulence modell. Nevertheless it is not stable at all, when the velocity is to high. Do you know, which parameters have to be changed in order to come along with the 3 m/s? The problem is the same: the velocity at the outlet increases and makes the whole calculation instable.
I am not that familiar with the parameters in the turbulence modell. I use the kOmegaSST modle. i really hope, you can help me! |
|
August 8, 2014, 08:53 |
|
#7 |
Senior Member
Philipp
Join Date: Jun 2011
Location: Germany
Posts: 1,297
Rep Power: 27 |
1) How do you create your mesh?
2) Do you know about the y+ value? 3) What is the Reynolds number? 4) Please post the k and omega boundary conditions.
__________________
The skeleton ran out of shampoo in the shower. |
|
August 8, 2014, 09:15 |
|
#8 |
New Member
Join Date: Jul 2014
Posts: 17
Rep Power: 12 |
1) I used gridgen for meshing
2) no, sorry, what is the y+ value? 3) The diameter of the tube changes, so the reynoldnumer is not constant? Am I right? look at the picture attached /*--------------------------------*- C++ -*----------------------------------*\ | ========= | | | \\ / F ield | OpenFOAM: The Open Source CFD Toolbox | | \\ / O peration | Version: 2.2.2 | | \\ / A nd | Web: www.OpenFOAM.org | | \\/ M anipulation | | \*---------------------------------------------------------------------------*/ FoamFile { version 2.0; format ascii; class volScalarField; object k; } // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // #include "include/initialConditions" dimensions [0 2 -2 0 0 0 0]; internalField uniform $turbulentKE; boundaryField { outflow-14 { type inletOutlet; inletValue $internalField; value $internalField; } inflow-17 { type fixedValue; value $internalField; } Dreieck5-4 { type kqRWallFunction; value $internalField; } outflow_rechteck-5 { type kqRWallFunction; value $internalField; } rechteck54-6 { type kqRWallFunction; value $internalField; } rechteck37b-7 { type kqRWallFunction; value $internalField; } Dreieck7-8 { type kqRWallFunction; value $internalField; } Dreieck4-9 { type kqRWallFunction; value $internalField; } HK3-10 { type kqRWallFunction; value $internalField; } rechteck37a-11 { type kqRWallFunction; value $internalField; } HK2-12 { type kqRWallFunction; value $internalField; } HK1-13 { type kqRWallFunction; value $internalField; } Dreieck1-15 { type kqRWallFunction; value $internalField; } rechteck_inflow-16 { type kqRWallFunction; value $internalField; } rechteck18-18 { type kqRWallFunction; value $internalField; } Dreieck8-19 { type kqRWallFunction; value $internalField; } Dreieck2-20 { type kqRWallFunction; value $internalField; } rechteck63a-21 { type kqRWallFunction; value $internalField; } Dreieck6-22 { type kqRWallFunction; value $internalField; } dreieck3-23 { type kqRWallFunction; value $internalField; } rechteck63b-24 { type kqRWallFunction; value $internalField; } inflowOS1-25 { type kqRWallFunction; value $internalField; } inflowOS2-26 { type kqRWallFunction; value $internalField; } HKRE3-27 { type kqRWallFunction; value $internalField; } unterseite2-28 { type kqRWallFunction; value $internalField; } hkre-29 { type kqRWallFunction; value $internalField; } HKRE2-30 { type kqRWallFunction; value $internalField; } unterseite1-31 { type kqRWallFunction; value $internalField; } Inlet-9 { type fixedValue; value $internalField; } } // ************************************************** *********************** // *--------------------------------*- C++ -*----------------------------------*\ | ========= | | | \\ / F ield | OpenFOAM: The Open Source CFD Toolbox | | \\ / O peration | Version: 2.2.2 | | \\ / A nd | Web: www.OpenFOAM.org | | \\/ M anipulation | | \*---------------------------------------------------------------------------*/ FoamFile { version 2.0; format ascii; class volScalarField; object omega; } // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // #include "include/initialConditions" dimensions [0 0 -1 0 0 0 0]; internalField uniform $turbulentOmega; boundaryField { //- Set patchGroups for constraint patches #include "${WM_PROJECT_DIR}/etc/caseDicts/setConstraintTypes" //#include "include/fixedInlet" inflow-17 { type fixedValue; value $internalField; } outflow-14 { type inletOutlet; inletValue $internalField; value $internalField; } Dreieck5-4 { type omegaWallFunction; value $internalField; } outflow_rechteck-5 { type omegaWallFunction; value $internalField; } rechteck54-6 { type omegaWallFunction; value $internalField; } rechteck37b-7 { type omegaWallFunction; value $internalField; } Dreieck7-8 { type omegaWallFunction; value $internalField; } Dreieck4-9 { type omegaWallFunction; value $internalField; } HK3-10 { type omegaWallFunction; value $internalField; } rechteck37a-11 { type omegaWallFunction; value $internalField; } HK2-12 { type omegaWallFunction; value $internalField; } HK1-13 { type omegaWallFunction; value $internalField; } Dreieck1-15 { type omegaWallFunction; value $internalField; } rechteck_inflow-16 { type omegaWallFunction; value $internalField; } rechteck18-18 { type omegaWallFunction; value $internalField; } Dreieck8-19 { type omegaWallFunction; value $internalField; } Dreieck2-20 { type omegaWallFunction; value $internalField; } rechteck63a-21 { type omegaWallFunction; value $internalField; } Dreieck6-22 { type omegaWallFunction; value $internalField; } dreieck3-23 { type omegaWallFunction; value $internalField; } rechteck63b-24 { type omegaWallFunction; value $internalField; } inflowOS1-25 { type omegaWallFunction; value $internalField; } inflowOS2-26 { type omegaWallFunction; value $internalField; } HKRE3-27 { type omegaWallFunction; value $internalField; } unterseite2-28 { type omegaWallFunction; value $internalField; } hkre-29 { type omegaWallFunction; value $internalField; } HKRE2-30 { type omegaWallFunction; value $internalField; } unterseite1-31 { type omegaWallFunction; value $internalField; } //#include "include/frontBackUpperPatches" } // ************************************************** *********************** // |
|
August 8, 2014, 09:17 |
|
#9 |
New Member
Join Date: Jul 2014
Posts: 17
Rep Power: 12 |
I try to attache a pic later!
|
|
August 8, 2014, 09:20 |
|
#10 |
New Member
Join Date: Jul 2014
Posts: 17
Rep Power: 12 |
|
|
August 8, 2014, 09:22 |
|
#11 |
Senior Member
Philipp
Join Date: Jun 2011
Location: Germany
Posts: 1,297
Rep Power: 27 |
I see the picture but where is the pipe? I thought it is supposed to be a pipe flow...
__________________
The skeleton ran out of shampoo in the shower. |
|
August 10, 2014, 05:25 |
|
#12 | |
Member
Join Date: May 2013
Posts: 51
Rep Power: 13 |
Quote:
2. Those informations you are looking for. 3. In case of your next questions Cheers, seav |
||
August 11, 2014, 16:22 |
|
#13 |
New Member
Join Date: Jul 2014
Posts: 17
Rep Power: 12 |
ok, thanks!
But I still do not know how to define the Reynoldnumber in OF. If my calculations are right, are I have a Reynoldnumber of 12000. Re=(3 m/s * 0.06m)/(1.5*10^-7) Where and how do I define this paramter in the files? |
|
August 11, 2014, 17:14 |
|
#14 |
Member
Join Date: Feb 2014
Posts: 63
Rep Power: 12 |
Hi,
I hope you need a steady state solution and that is why you use simpleFoam. There is a tutorial for a pipe flow problem using cyclic boundary conditions, to get a fully developed pipe flow solution. what it does is basically map the outlet to your inlet and iterate until a steady solution is obtained. But tutorial is on pisoFoam or pimpleFoam but it could help you. |
|
August 11, 2014, 17:31 |
|
#15 |
New Member
Join Date: Jul 2014
Posts: 17
Rep Power: 12 |
thanks alot! I will think about that!
Do you have an answer to the question before? How do I define the Reynoldnumber in the files? |
|
August 11, 2014, 18:47 |
|
#16 |
Member
Join Date: Feb 2014
Posts: 63
Rep Power: 12 |
Hi,
Why do you need to specify Reynolds number ? To solve k-omega you only to specify k & omega at boundaries. Check this link. http://www.cfd-online.com/Wiki/Turbu...ary_conditions |
|
August 12, 2014, 03:54 |
|
#17 |
Senior Member
Philipp
Join Date: Jun 2011
Location: Germany
Posts: 1,297
Rep Power: 27 |
Specialist, you can not define the Reynolds number in OpenFOAM. But you define all other values that define the Reynolds number: Inlet velocity, size of your geometry, kinem. visosity!
__________________
The skeleton ran out of shampoo in the shower. |
|
August 12, 2014, 05:52 |
|
#18 |
Member
Join Date: May 2013
Posts: 51
Rep Power: 13 |
Tubulence intensity needs Re number.
Set proper Boundary Condition from here. You can achive what you need. seav |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
interFoam vs. simpleFoam channel flow comparison | DanM | OpenFOAM Running, Solving & CFD | 12 | January 31, 2020 16:26 |
Requiring some help regarding tornado simulation | amarjogot | Main CFD Forum | 0 | June 20, 2013 13:26 |
simulation around an hydrofoil using simpleFoam - SA model | mesnardo | OpenFOAM Running, Solving & CFD | 1 | May 29, 2013 07:47 |
Cyclone Simulation (simpleFoam) | erncyc | OpenFOAM | 4 | January 28, 2011 10:40 |
Continuous vs interrupted simulation | sega | OpenFOAM Running, Solving & CFD | 4 | November 3, 2008 15:29 |