1
$\begingroup$

can you help me? I have a fluid simulation in 2D, where I compute by finite differences on the grid the velocity field and pressure. So this is my output from the program. Now I have to show how the vertical component of traction on the boundary looks like. But I am not sure what the word traction means, what I have to compute. I can compute the stress from the velocity field as

$$ \sigma_{xy}=\eta (\frac{dv_x}{dy}+\frac{dv_y}{dx}) $$

$$ \sigma_{xx}=2\eta \frac{dv_x}{dx} $$

$$ \sigma_{yy}=2\eta \frac{dv_y}{dy} $$

So this I would know how to compute from my results - by finite differences. What does the vertical traction mathematically means, what I have to compute?
What I about know is, that the traction in vertical (y-axis) could be: $$ \vec{t_y} = \sigma_{xy}\vec{e_x} + \sigma_{yy}\vec{e_y} $$ Am I right? Many thanks

Thsi is about traction:
"Note a convention that we have implicitly established: the sense of the force per unit area $t$ across the oblique surface is that the fluid penetrated by the unit normal $n$ acts on the fluid on the other side of the surface. This force per unit area is called the traction across the surface. Note further that $t_x$ indicates the traction across the y − z plane, not the x component of a traction. Representation of components requires a second subscript: $T_{xy}$ is the y component of the traction $t_x$ across the y − z face of the tetrahedron, whereas $T_{xy}$ is the x component of the traction ty across the x − z face. Thus, the first subscript represents the vector component of the force and the second subscript indicates the face on which the force is acting. " But how to compute it?

$\endgroup$

1 Answer 1

4
$\begingroup$

First, the stress in fluid dynamics must also include the pressure. In fact, it is $$ \sigma = 2 \eta \varepsilon(\mathbf v) + p I $$ where $\varepsilon(\mathbf v)$ is the symmetric gradient of the velocity $\mathbf v$. Then, the traction is the normal component of the stress at the boundary: $$ \mathbf t = \sigma \mathbf n $$ If you want you can further decompose the traction into the normal force $f = \mathbf n \cdot \mathbf t = \mathbf n \cdot \sigma \mathbf n=2\eta \mathbf n \cdot \varepsilon(\mathbf v) \mathbf n + p$ and the tangential friction force $\mathbf t - f \mathbf n$ for which it is easy to show that it doesn't depend on the pressure any more.

$\endgroup$
0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.