[erncyc]

Hello friends,

Currently the setfields utility in OpenFOAM allows for the input of constant data in the setfieldDict dictionary. For instance when using cylinderToCell, it is required that one provides the values p1 and p2 which are the coordinates of the cylinder centre. In this case the coordinates are real values. I would like a situation where i can define a varying coordinate system ie p1 (x1, y1, z1) and p2 (x2, y2, z2). Anybody with an idea of how I might be able to do this. I checked the cylinderToCell.H class file and noticed that p1 and p2 are declared as "const vector". I appreciate any suggestion. Thanks.

[gschaider]

Originally Posted by erncyc

Hello friends,

Currently the setfields utility in OpenFOAM allows for the input of constant data in the setfieldDict dictionary. For instance when using cylinderToCell, it is required that one provides the values p1 and p2 which are the coordinates of the cylinder centre. In this case the coordinates are real values. I would like a situation where i can define a varying coordinate system ie p1 (x1, y1, z1) and p2 (x2, y2, z2). Anybody with an idea of how I might be able to do this. I checked the cylinderToCell.H class file and noticed that p1 and p2 are declared as "const vector". I appreciate any suggestion. Thanks.

I'm not quite sure what you mean with "a varying coordinate system". I'd like to understand the problem before I make my usual suggestion: use funkySetFields

[erncyc]

Originally Posted by gschaider

I'm not quite sure what you mean with "a varying coordinate system". I'd like to understand the problem before I make my usual suggestion: use funkySetFields

Hello Bernhard, firstly I really want you to know I appreciate your responding to my query, thanks a bunch. My problem involves the use of setFields (or a similar utility) that can set a 2D cylindrical region with centres that change with time. I wish to be able to provide new cylinder centres at every time step. Currently the cylinderToCell option allows for provision of a static coordinate (defined in setFields as p1 and p2). If I don't use funkysetfields, is there a way I might be able to do this by modifying the "cylinderToCell.H" class. Thank you again for your help.

[Bernhard]

Do I understand correctly, that you want to set this during runtime? setFields and funkySetFields are designed for setting initial conditions. If you want to have a moving cylinder, you should put it in the solver itself, and not with a utility. You have much more freedom if you set the value each timestep in a loop over all cells.

[erncyc]

Originally Posted by Bernhard

Do I understand correctly, that you want to set this during runtime? setFields and funkySetFields are designed for setting initial conditions. If you want to have a moving cylinder, you should put it in the solver itself, and not with a utility. You have much more freedom if you set the value each timestep in a loop over all cells.

Thanks again Bernhard. Yes you are right, I wish to set the region during runtime. My problem is (since I am using an immersed boundary method), I need to use a static mesh and find a way of connecting the displacement of the cylinder axis with the mesh coordinate system. I believe yes you are right i should use the solver (currently i am on interFoam), but this is where my problems begin...can't figure out how. Thanks

[gschaider]

Originally Posted by erncyc

Thanks again Bernhard. Yes you are right, I wish to set the region during runtime. My problem is (since I am using an immersed boundary method), I need to use a static mesh and find a way of connecting the displacement of the cylinder axis with the mesh coordinate system. I believe yes you are right i should use the solver (currently i am on interFoam), but this is where my problems begin...can't figure out how. Thanks

And you need a source term in that cylinder or what? swak4Foam gives you a number of options (built on the machinery that drives funkySetFields):

- the manipulateField functionObject allows you to set an existing field according to an expression (which may depend on the time)
- there is a class expressionSource which you can incorporate into your solver to add a source term according to your needs (there is also a class for fixing the values in selected cells)
- in the development version for 2.x (mercurial repository) there is also a class based on basicSource which allows you to have expression-dependent sources in solvers that use the sourcesProperties-dictionary (no need to modify the solver)

[erncyc]

Originally Posted by gschaider

And you need a source term in that cylinder or what? swak4Foam gives you a number of options (built on the machinery that drives funkySetFields):

- the manipulateField functionObject allows you to set an existing field according to an expression (which may depend on the time)
- there is a class expressionSource which you can incorporate into your solver to add a source term according to your needs (there is also a class for fixing the values in selected cells)
- in the development version for 2.x (mercurial repository) there is also a class based on basicSource which allows you to have expression-dependent sources in solvers that use the sourcesProperties-dictionary (no need to modify the solver)

Yes I need a source term. I have never used swakFoam...how could I get and install it? I currently have OF 2.01...thanks

[gschaider]

Originally Posted by erncyc

Yes I need a source term. I have never used swakFoam...how could I get and install it? I currently have OF 2.01...thanks

You find it under http://openfoamwiki.net/index.php/Contrib/swak4Foam.

The basicSource-source-term currently is only in the development version. expressionSource in the release. Search the forum there is an explanation for it.

There may be an issue with compilation on 2.0.1 (the 2.x-branch is currently only tested on 2.1.x)

[erncyc]

Originally Posted by gschaider

You find it under http://openfoamwiki.net/index.php/Contrib/swak4Foam.

The basicSource-source-term currently is only in the development version. expressionSource in the release. Search the forum there is an explanation for it.

There may be an issue with compilation on 2.0.1 (the 2.x-branch is currently only tested on 2.1.x)

Thank you again. I tried to install but I got the error message below:

ernest@cfdlab-desktop:~/OpenFOAM/swak4Foam$ nice ./Allwmake
No 'swakConfiguration'. Python etc won't work
Checking swak4Foam-version and generating file
No 'bison' found. This is absolutely essential for swak4Foam. Can't go on
Requirements for Library not satisfied. I see no sense in going on

[Bernhard]

Originally Posted by erncyc

Thank you again. I tried to install but I got the error message below:

ernest@cfdlab-desktop:~/OpenFOAM/swak4Foam$ nice ./Allwmake
No 'swakConfiguration'. Python etc won't work
Checking swak4Foam-version and generating file
No 'bison' found. This is absolutely essential for swak4Foam. Can't go on
Requirements for Library not satisfied. I see no sense in going on

Did you read the prequisites in the README? You need bison to install it, and according to the error message, you should install that one first.

[erncyc]

Originally Posted by Bernhard

Did you read the prequisites in the README? You need bison to install it, and according to the error message, you should install that one first.

oh ok...I will read README and try again thanks.

[gschaider]

Originally Posted by Bernhard

Did you read the prequisites in the README? You need bison to install it, and according to the error message, you should install that one first.

Got nothing to add.

And while you're at it install flex ....

[Kummi]

Hello everyone,
My topic of research is on coke oven process using 1D solver. Heat transfer is considered to be one-dimensional for the lump coal (placed inside the domain).

To define geometrical regions in the form of sphere (coal) inside my domain, I adopted setFieldsDict file ~ to initialize fields. But unfortunately, the set field directory gives shapes of sphere for 2D or 3D modeling in better way - whereas not in the case of 1D modeling.

I am hereby including my blockMeshDict file (only blocks) and setFieldsDict file.

blockMeshDict file:

convertToMeters 0.01;
vertices
(
(0 0 0) (15 0 0) (15 1 0) (0 1 0) (0 0 7) (15 0 7) (15 1 7) (0 1 7)
);
blocks
(
hex (0 1 2 3 4 5 6 7) (1250 1 1) simpleGrading (1 1 1)
);
edges
(
);

setFieldsDict file:

defaultFieldValues

(

volScalarFieldValue C 0

);

regions

(

sphereToCell
{
centre (0.03 0 0.03);
radius 0.02;

fieldValues
(
volScalarFieldValue C 800
);
}

);

As a result, the initialized field in paraFoam [Attachment] ~ is not in the form of sphere because the set field dimensions is influenced only in X- directional for 1D mesh.

Dear Foamers, I want to know the possibility of creating fields in the form of sphere inside my domain for 1D solver using setFieldsDict.

If not setfieldDict, is there any relevant options to initialize fields for 1D solver ?

Kindly share your ideas plzz..

Thank you

[gkarlsen]

Deleted Deleted

[cyss38]

I am not sure to well understand...



But :

in 1D mesh (your case), sphere doesn't exist... because there is only one cell in the y and z directions.


So no solution to make a sphere.