Originally Posted by hjasak (Post 185688)

4) If this is not enough for you (there is another possibility), I have written a newDirectionMixedFvPatchField boundary condition, with 4 components. Unlike the above one, this b.c. allows both the normal and tangential condition to be mixed separately between the fixedValue and fixedGradient b.c. with two blending factors. You need this kind of thing for contact problems with friction and large deformation and similar. In my version, the stuff lives in:
OpenFOAM-1.2/applications/solvers/newStressAnalysis/materialModels/newDirectionM ixed

Originally Posted by Sherlock_1812 (Post 460937)

Hi Philip,

I have been trying to re-write valueFraction for each face of my patch as symm(n*n). I have a very basic doubt.
Is it better to write a new BC (derived from directionMixed (maybe?)) where this definition of valueFraction is changed as symm(n*n) while looping over all faces of the patch? If yes, where do I make the changes? In the 'updateCoeffs' section?

Thanks.

Originally Posted by bigphil (Post 328702)

Hi Nicolas,

Yes it is possible, using directionMixed, to give every face on a boundary patch a different gradient in the tangential direction. I would do something like this in a header file at the start of a time-step (although it might be possible to use groovyBC but I am not sure):

By the way, "U" is my solution variable above.

Hope this helps,
Philip

Originally Posted by nlc (Post 672025)

Hi Behzad,

OpenFOAM je 3D n'y défaut 1D and 2D case are achieve by setting special BC. The equation you share seem to be 2D. You need first to specify your equation on 3D then figure ought what type of BC you need. Fixed, gradient, mixed !! Then you'll know what to look for.

It seems to me that you should start a new thread. Wright the hole set of equations for your problem with detail on the boundary condition you want to use. Or a link in an article relevant to your case.

It is better to use vector notation then component details as OF use vector and tensor notation. Add me on your new tread please as it seems interesting.

NLC

Originally Posted by behzad-cfd (Post 672026)

Hi Nicolas,

Thanks for kind reply. as you mentioned OF operates in 3D my case is 2D and I'll set front and back plane as empty and that's not the problem. I showed component details for clarifying the variable relations and boundary conditions. any way I'll open new thread and add you there.

Kind regards,
Behzad

Originally Posted by Pavithra (Post 735945)

Hello Everyone

I am unable to understand the mixed boundary condition clearly.

I have a vertical wall and I want to set the x-component of velocity to be u = 0 and y-component of velocity to be set as dv/dx = 0.

In order to achieve this, I have implemented the following

sideWall
{
type directionMixed;
refValue uniform (0 0 0);
refGradient uniform (0 0 0);
valueFraction uniform (0 0 0 0 0 1);
}

but, when I plot I see non-zero values for the x-component.

Kindly, please help me.

With Thanks & Regards,
Pavithra.

Originally Posted by bigphil (Post 736070)

Hi Pavithra,

To set the normal component to fixedValue and the two tangential components to fixedGradient, the valueFraction should be:
where 'n' are the unit normals.

Alternatively (your case), to set the normal component to fixedGradient and the two tangential components to fixedValue, the valueFraction should be:
In your case, n = (0 1 0), so the valueFraction should be:
Philip

Originally Posted by bigphil (Post 736088)

Hi Pavithra,

It would be helpful if you could give the unit normal of this "vertical wall", as it is unclear what exactly you are referring to.

Philip

Originally Posted by Pavithra (Post 736090)

Sir,



My wall is in yz-plane and the unit normal vector is (1 0 0)


Thank You.

Originally Posted by bigphil (Post 736119)

In that case, n*n = (1 0 0 0 0 0), where the componets are (XX XY XZ YY YZ ZZ).

This will set Ux to zero and set dUy/dx and dUz/dx to zero.

Philip

Originally Posted by Pavithra (Post 745995)

Hello Everyone,

Back again here :D:D:D:D
I want to implement the following boundary condition on the horizontal walls of a 2-D square cavity.








The value of varies from 0 to 1 in steps of 0.25.

I understand that when , the boundary condition becomes and .

Also, when , the boundary condition becomes and .

I am able to achieve the above two cases using directionMixed boundary condition.

But I am unable to figure out how to set the boundary condition when takes a value such as 0.25, 0.5 or 0.75.

For instance, when = 0.25, it becomes .

Kindly please give me some directions to implement this boundary condition.

Thank You.

With Thanks
Pavithra.

Originally Posted by shiyu (Post 791458)

Hi Pavithra,

I faced a very similar question, like yours. Have you figured out how to implement the boundary condition below?

Originally Posted by bigphil (Post 316165)

Hi G. Paulo,


I occasionally use directionMixed for stress simulations where I want a fixedValue (displacement) in the patch normal direction and some fixedGradient (traction) in the patch tangential direction.

So to achieve a fixedValue in the normal direction, and a fixedGradient in the tangential direction on a patch, then the valueFraction (which is a symmTensor, where the components of a symmTensor are (xx xy xz yy yz zz) ) should be set:
valueFraction = symm(n*n)
where n is the patch normal.

If your patch faces all have the same normal then you can calculate symm(n*n) by hand and put it in your initial conditions, eg:
for a patch with n = ( 1 0 0 )
then the valueFraction = symm( (1 0 0)(1 0 0) ) = ( 1 0 0 0 0 0)
then refValue will be enforced in the patch normal direction and refGradient will be enforced in the patch tangential direction.

If your patch faces have all different normals then you will have to set the patch valueFraction in the solver (using refCast etc.) to be symm(n*n).

So to sum up, if you want a fixedValue in the patch normal direction and a fixedGradient in the patch tangential direction then it looks like this:
refValue uniform (fixed_value_you_want)
refGradient uniform (shear_gradient_you_want)
valueFraction uniform (symm(n*n))

directionMixed can also be used for other situations:
it can act just like a fixedValue BC if valueFraction = (1 0 0 1 0 1),

Or, it can act just like a fixedGradient BC if valueFraction = (0 0 0 0 0 0).

If you look at the directionMixed BC source code, you can see it calculates it.


Hope it helps,
Philip

Originally Posted by bigphil (Post 801447)

Hi tomtony,

Yes that's correct, assuming you want Ux and Uy to be zero.

By the way, solvers (or paraview) may also want "value" to be define in the boundary condition, which corresponds to the current value on the patch e.g. the initial condition; you could just set this to the same value as refValue if you are not sure.

Philip

Originally Posted by bigphil (Post 818410)

Hi Mohammad,

If I understand correctly, then yes your proposed valueFraction is correct. It will use fixedValue in the X and Z directions and fixedGradient in the Y direction.

As for "How it detects the normal direction?", I am not sure I understand your question; the normal directions are just taken directly from the mesh geometry. The valueFraction here is independent of the normals and it is up to you to define it appropriately. For example, if the normals were non-uniform on the patch then you may need to define the valueFraction field as non-uniform in order to enforce what ever condition you want.

Philip

Originally Posted by bigphil (Post 316165)

Hi G. Paulo,


I occasionally use directionMixed for stress simulations where I want a fixedValue (displacement) in the patch normal direction and some fixedGradient (traction) in the patch tangential direction.

So to achieve a fixedValue in the normal direction, and a fixedGradient in the tangential direction on a patch, then the valueFraction (which is a symmTensor, where the components of a symmTensor are (xx xy xz yy yz zz) ) should be set:
valueFraction = symm(n*n)
where n is the patch normal.

If your patch faces all have the same normal then you can calculate symm(n*n) by hand and put it in your initial conditions, eg:
for a patch with n = ( 1 0 0 )
then the valueFraction = symm( (1 0 0)(1 0 0) ) = ( 1 0 0 0 0 0)
then refValue will be enforced in the patch normal direction and refGradient will be enforced in the patch tangential direction.

If your patch faces have all different normals then you will have to set the patch valueFraction in the solver (using refCast etc.) to be symm(n*n).

So to sum up, if you want a fixedValue in the patch normal direction and a fixedGradient in the patch tangential direction then it looks like this:
refValue uniform (fixed_value_you_want)
refGradient uniform (shear_gradient_you_want)
valueFraction uniform (symm(n*n))

directionMixed can also be used for other situations:
it can act just like a fixedValue BC if valueFraction = (1 0 0 1 0 1),

Or, it can act just like a fixedGradient BC if valueFraction = (0 0 0 0 0 0).

If you look at the directionMixed BC source code, you can see it calculates it.


Hope it helps,
Philip

Originally Posted by qingqingliu (Post 847578)

Hi Philip,
I saw you have answered many questions about directionMixed. Based on my understanding of directMixed condition, I set up it for my case. I think you are the right person to ask if my setup is correct or not.
Here is my case:
Normal vector: (0 1 0)
Velocity vector: (Vx, Vy, Vz)
I want to set Vx = 0, dVy/dn = 0, and dVz/dn = 0. Below is my setup, do you think it is right?

type directionMixed;
refValue uniform (0 0 0);
refGradient uniform (0 0 0);
valueFraction uniform (1 0 0 0 0 0);

Thanks for your help!!!

Originally Posted by bigphil (Post 847629)

Yes, that looks correct.

After running the model, I suggest you check the boundary conditions are enforced as you expect, e.g. check Vx on the patch, and plot Vy and Vz away from the wall to check the gradient is zero.