Ivar KJELBERG
COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
27.04.2011, 15:21 GMT-4
Hi
I believe not yet (hope it will come in next release ;)
But you can do some searches in the Equation view (turn on via Preferences) look upt the "descriptions", then you can find many by doing an indexed search on the PDF of the help files.
The notation is "ns." for Navier Stokes, but I believe that is either a 3.5 translation or a 4.0 name, normall in natif 4.1 it should be spf (single phase flow), the physics identifier, "nr" "nz" for the normal component along r and z (so you are working in 2D-axi I understand, else it would have been nx, ny, nz. These are defined along an Edge = Boundary then. "ns.nr" is to be understood as a field of the type ns.nr(r,z) ,as as the edge scans over the r,z plane the nr value might change (if the line is not straight).
Attached to the coordinates r,z you have the displacements u,v (in fact in NS I believe its the velocities ? (sorry pls check). I cannot remember what is f0 that too you will have to search. But the equation you have is the weak expression of the scalar product of f0 with the normal component of (u,v) w.r.t the edge that is selected in the BC.
It's a beginning, but as a many names has changed from 3.5 and 4, and a few since the first 4.0 it's indeed not easy to follow ;)
--
Good luck
Ivar
Hi
I believe not yet (hope it will come in next release ;)
But you can do some searches in the Equation view (turn on via Preferences) look upt the "descriptions", then you can find many by doing an indexed search on the PDF of the help files.
The notation is "ns." for Navier Stokes, but I believe that is either a 3.5 translation or a 4.0 name, normall in natif 4.1 it should be spf (single phase flow), the physics identifier, "nr" "nz" for the normal component along r and z (so you are working in 2D-axi I understand, else it would have been nx, ny, nz. These are defined along an Edge = Boundary then. "ns.nr" is to be understood as a field of the type ns.nr(r,z) ,as as the edge scans over the r,z plane the nr value might change (if the line is not straight).
Attached to the coordinates r,z you have the displacements u,v (in fact in NS I believe its the velocities ? (sorry pls check). I cannot remember what is f0 that too you will have to search. But the equation you have is the weak expression of the scalar product of f0 with the normal component of (u,v) w.r.t the edge that is selected in the BC.
It's a beginning, but as a many names has changed from 3.5 and 4, and a few since the first 4.0 it's indeed not easy to follow ;)
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
28.04.2011, 03:06 GMT-4
Hi!
I found out, that ns.f0 is the normal stress, which can be defined in the 'open boundary' node.
Can you tell me, what the test-function does? Is it the same test function you use for weak differentiations?
Best regards,
Christian
Hi!
I found out, that ns.f0 is the normal stress, which can be defined in the 'open boundary' node.
Can you tell me, what the test-function does? Is it the same test function you use for weak differentiations?
Best regards,
Christian
Ivar KJELBERG
COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
28.04.2011, 05:41 GMT-4
Hi
the test functions are the weak formulation, that is too long to handle here, take a close look to the doc, and some of the books in the book reference
www.comsol.eu/support/books/
such as "Introduction to Computation and Modeling with Differential Equations" by Lennart Edsberg and
"The Finite Element Method Basic Concepts and Applications" by Darrell W. Pepper and Juan C. Heinrich
--
Good luck
Ivar
Hi
the test functions are the weak formulation, that is too long to handle here, take a close look to the doc, and some of the books in the book reference
http://www.comsol.eu/support/books/
such as "Introduction to Computation and Modeling with Differential Equations" by Lennart Edsberg and
"The Finite Element Method Basic Concepts and Applications" by Darrell W. Pepper and Juan C. Heinrich
--
Good luck
Ivar