[ehsan]

Hi,

1- Use Gambit, it is more easier and comfortable.
2- Use structured one, it can captures flow features near the wall much better.
3- See some papers of Boyd and coworkers on sphere flow simulations with DSMC. They used very fine grids.

Regards

[hyperion]

Hi Ehsan - First of all, many thanks to you and your collaborators for dsmcFOAM. Could you explain your use of Bird's eqn. 4.79 in the VHS model. It seems that your equation for sigmaT in birds notation would actually be:

sigmaT = pi*dref(pq)*d(pq) rather than sigmaT = pi*d(pq)*d(pq).

The code I'm talking about is lines 90-96 of variableHardSphere.C.

I assume the discrepancy has something to do with the fact that the function returns sigmaT*cR. Is this interpretation of the code incorrect?

Also, would you mind addressing my question in the thread:

http://www.cfd-online.com/Forums/ope...teraction.html


Many Thanks

[hyperion]

Hi Ehsan - I have yet another question for you. I didn't see any use of viscosityCoefficient from the properties file in the code. I used the command
"find . | xargs grep viscosityCoefficient" and no results were obtained. I apologize for asking so many questions of you, but I must at least try.

Thanks Again

[ehsan]

Hello,

Sorry for my delays. In fact, I am quite busy these days and could not come back to source codes, but you could easily find the solution of your questions if you first observe and run a simple DSMC code like Bird FORTRAN codes, available on the web, and then go to the dsmcFoam.

Regards
Ehsan

[gautcho]

Hey Claus, hey Marta,

thanks to your help via email I made a lot of progress understanding dsmcFoam and I can give you some updates to the most important questions (important note: I use version 1.7.1, I dont no anything about 2.0.0):


1. This is a very important one: zeroGradient doesn't work at least at an open boundary. Instead if you use zeroGradient, dsmcFoam takes the internalField-temperature to insert new molecules, which is really not a good idea! /E: Bad news (at least for me): It also doesn't work on walls. So every time you have got a MaxwellianThermal-wall-collision the temperature used to calculate the post-collision-velocity is based on the internalField-temperature.
2. Claus on your question about VHS: I also thought the implementation was wrong first, but I think, I found out, that d_pq in the implementation = d_pq_ref in Bird or something like this. To put it in a nutshell: Everything is fine, I think.
3. The famous pFA-Eq. from Bird is based on an hypothetical equilibrium on the other side of the cell. I changed it, so you can insert a massflow:

pFA[i] += mag(patch.faceAreas())*numberDensities_[i]*deltaT;

4. Also I implemented a mean free path- and a Kn-field.


I think, we definitily should discuss point 1.

[dark lancer]

hi
I have a problem that decide to solve it with dsmc and use Air for the gas.but I don't know about the numerdensity of Air?
the density of Argon is 1e20(where?) ?I see in thermodynamic and the density of Argon in the SI and US not this value?
what are these(rhoM &rhoN)?and is nessecary for difne the value for rhoM or RhoN?because I use zerogradiant for these parameters.

[dark lancer]

hi
I try to run a problem with dsmc and mesh my geometry in gambit then convert to OF.
I excute the checkMesh All of them its Ok but at the end line :

moosaie@:~/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical> checkMesh
/*---------------------------------------------------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: 2.1.1 |
| \\ / A nd | Web: www.OpenFOAM.org |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
Build : 2.1.1-221db2718bbb
Exec : checkMesh
Date : Nov 17 2012
Time : 12:53:01
Host : ".site"
PID : 6467
Case : /home/moosaie/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Disallowing user-supplied system call operations

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Create polyMesh for time = 0

Time = 0

Mesh stats
points: 1489344
faces: 4387940
internal faces: 4308820
cells: 1449460
boundary patches: 5
point zones: 0
face zones: 1
cell zones: 1

Overall number of cells of each type:
hexahedra: 1449460
prisms: 0
wedges: 0
pyramids: 0
tet wedges: 0
tetrahedra: 0
polyhedra: 0

Checking topology...
Boundary definition OK.
Cell to face addressing OK.
Point usage OK.
Upper triangular ordering OK.
Face vertices OK.
Number of regions: 1 (OK).

Checking patch topology for multiply connected surfaces ...
Patch Faces Points Surface topology
body 17480 17694 ok (non-closed singly connected)
symmetry 34040 34576 ok (non-closed singly connected)
outlet_from_side 13524 13764 ok (non-closed singly connected)
outlet 10902 11079 ok (non-closed singly connected)
inlet 3174 3267 ok (non-closed singly connected)

Checking geometry...
Overall domain bounding box (-262.5 -262.5 -0.4142099593) (1375 262.5 262.5)
Mesh (non-empty, non-wedge) directions (1 1 1)
Mesh (non-empty) directions (1 1 1)
Boundary openness (1.756589626e-16 1.121895293e-15 5.86943899e-15) OK.
Max cell openness = 1.916326799e-15 OK.
Max aspect ratio = 61.10879384 OK.
Minumum face area = 0.01039895189. Maximum face area = 280.93608. Face area magnitudes OK.
Min volume = 0.001605026005. Max volume = 2372.742765. Total volume = 167688127. Cell volumes OK.
Mesh non-orthogonality Max: 86.44406365 average: 27.94667529
*Number of severely non-orthogonal faces: 20108.
Non-orthogonality check OK.
<<Writing 20108 non-orthogonal faces to set nonOrthoFaces
Face pyramids OK.
***Max skewness = 4.752768679, 879 highly skew faces detected which may impair the quality of the results
<<Writing 879 skew faces to set skewFaces
Coupled point location match (average 0) OK.

Failed 1 mesh checks.

End

and I use dsmcFoam I see this Error:

moosaie@:~/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical> dsmcFoam
/*---------------------------------------------------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: 2.1.1 |
| \\ / A nd | Web: www.OpenFOAM.org |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
Build : 2.1.1-221db2718bbb
Exec : dsmcFoam
Date : Nov 17 2012
Time : 12:37:32
Host : ".site"
PID : 6370
Case : /home/moosaie/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Disallowing user-supplied system call operations

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Create mesh for time = 0


Constructing dsmcCloud


--> FOAM FATAL IO ERROR:
cannot find file

file: /home/moosaie/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical/0/dsmcSigmaTcRMax at line 0.

From function regIOobject::readStream()
in file db/regIOobject/regIOobjectRead.C at line 73.

FOAM exiting

moosaie@:~/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical>

I use this order:

./Allrun

I see this error:

moosaie@:~/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical> ./Allrun
Running blockMesh on /home/moosaie/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical
Running decomposePar on /home/moosaie/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical
Running dsmcInitialise in parallel on /home/moosaie/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical using 4 processes
Running dsmcFoam in parallel on /home/moosaie/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical using 4 processes
Running reconstructPar on /home/moosaie/OpenFOAM/OpenFOAM-2.1.1/tutorials/discreteMethods/dsmcFoam/hemispherical

and I don't have the blockMesh.

[murali]

--> FOAM FATAL ERROR:
Zero boundary temperature detected, check boundaryT condition.



From function Foam::FreeStream<CloudType>::inflow()
in file lnInclude/FreeStream.C at line 171.

FOAM aborting

#0 Foam::error:rintStack(Foam::Ostream&) in "/opt/openfoam201/platforms/linux64GccDPOpt/lib/libOpenFOAM.so"
#1 Foam::error::abort() in "/opt/openfoam201/platforms/linux64GccDPOpt/lib/libOpenFOAM.so"
#2 Foam::FreeStream<Foam:smcCloud<Foam:smcParcel< Foam:article> > >::inflow() in "/opt/openfoam201/platforms/linux64GccDPOpt/lib/libdsmc.so"
#3
in "/opt/openfoam201/platforms/linux64GccDPOpt/bin/dsmcFoam"
#4
in "/opt/openfoam201/platforms/linux64GccDPOpt/bin/dsmcFoam"
#5 __libc_start_main in "/lib/libc.so.6"
#6
in "/opt/openfoam201/platforms/linux64GccDPOpt/bin/dsmcFoam"
Aborted

[javad1404]

Hello

I'm a Master student. I'm going to use dsmcFoam in my thesis. But unfortunately I'm a beginner in both linux and OpenFOAM utilization. Anyway, I made up my mind to start with tutorials of dsmcFoam. And then I found out dsmcFieldsCalc doesn't calculate U and T in all of cases in tutorial. This is my problem. I don't know how I should solve this problem. By the way I'm using OpenFOAM 2.2.1 and Ubuntu 13.04. please help me. I need your help very much. Thank you very much in advance

[ehsan]

Hi

it writes data as Umean and OverallTmean in the output. If you use Tecplot or paraFoam, you could read these U and T.


Regards

[javad1404]

Hi Ehsan
I appreciate you very much for your reply. But unfortunately my problems are not solved yet. After running dsmcFieldsCalc for case supersonicCorner, UMean and overallT are not calculated. I'm just witness of writing following sentences on terminal.

Small value (min(mag(rhoNMean)) [0 -3 0 0 0 0 0] 1e-300) found in rhoNMean field. Not calculating dsmcFields to avoid division by zero.

For other cases it calculates UMean and overallT but when i want to calculate Mach number it writes "Missing U or T" on terminal and not solve Mach. Please help me. Thank you very much for your attention and help.

[ehsan]

It is a special case where I had not similar experiences, try changing nEqui or your mesh, maybe it works.

[javad1404]

Sorry, but would you please explain about nEqui more? What's this? How shoul I change it? Should I increase it or decrease it?

And also I have another request. Would you please send one of your true simple case to me?

[rejirv]

Hi Javad...

1. Just replace all occurrence of -300 to -30... and execute "dsmcFieldCalc". (only to check.. not recommended)
2. It is related to number of particles in cell..(insufficient number of particle).
3. Either you can increase number of "simulated" particle or increase the time of run.
4. nEquivalentParticles XXX; line in dsmcProperties.... is the core idea of DSMC (that means instead of simulating ""X number density"" we are simulated "X number density"/nEquivalentParticles )....
So if nEquivalentParticles=1; its purely molecular dynamics.

***** I am still studying dsmcFoam so not depend on my replay. Better read "GA Bird's book" or dsmcFoam paper in Journal of CAF.

[phys]

You gave the following as the entries for the dsmcProperties dict:

Originally Posted by archymedes

inOutflowCoeffs
{
inlet_patch
{
numberDensities
{
Ar 1e20;
O2 1.21e20;
}
}

outlet_patch
{
numberDensities
{
Ar 1e5;
O2 1.21e5;
}
}
}

When I try to enter that, it gives me a FOAM FATAL IO ERROR: keyword is undefined in dictionary. Usually such an error tells you what the keyword is, but in this case the keyword appears to be a blank space, or empty. Have you ever experienced this?

I am using v3.0+ but have upgraded your code to compile with the current version.

Thanks