 # Questions about the boundary constraint

Hello, @jgiven100
I noticed that there is the boundary constraint with “apply_friction_constraints(dt);” and “apply_velocity_constraints();”. I know that I can fix the boundary nodes as force = 0 or momentum=0 (or fix one direction). How do I understand this function using the fiction? is there any documents I can follow.
Thanks a lot~

Hi @fengzk

Here is the link to documenation for all boundary conditions, which includes friction constraints.

This sliding block example shows one of the ways to implement friction along a surface.

Hello, @jgiven100
I followed your suggestion and tested some examples. I have some new understanding of the BC setting and the rotate background grid. Thanks again.
But I am still a little bit confused about the friction boundary. I know that the boundary (node) looks like the FEM boundary? and we can define the FEM boundary as the fixed or sliding(only fix one direction or two direction). So I think we can use the apply_velocity_constraints() function to set the node as acc=0 and vel=0 in some direction.
And I didn’t see that we can set the boundary as a friction surface in the FEM, which theory or method we followed in this function apply_friction_constraints(dt)? or how can I understand it easily.
Thanks a lot~

Hi @fengzk, the frictional boundary is modeled assuming the Coulomb friction theory. So instead of applying a non-slip (v_tangent=0) or slip BC (no constraint in v_tangent), we apply a v_tangent increment with magnitude according to the friction coefficient and the a_normal.

Hello, @bodhinandach
Thanks for your answer. As I know, we can define contact algorithm and use Coulomb friction theory with friction coefficient. I always thought that the boundary setting is similar to FEM and there is only fixed or slip boundary in the FEM.
Thanks again.