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
Mar 18, 2013, 2:14 p.m. EDT
Hi
try useing a Definition - Selection node where you define each bondary and then refer to that Selection rather to the individual items.
Also it would perhaps (just a guess) work better if you define the line a s a polylineof 2 segments and not just an added "point". It probably has to do with the element id numbering algorithm, you point is displacing such that COMSOL identifies a topology change, and then it get lost in the element renumbering. The selection often (but not alwys) helps.
--
Good luck
Ivar
Hi
try useing a Definition - Selection node where you define each bondary and then refer to that Selection rather to the individual items.
Also it would perhaps (just a guess) work better if you define the line a s a polylineof 2 segments and not just an added "point". It probably has to do with the element id numbering algorithm, you point is displacing such that COMSOL identifies a topology change, and then it get lost in the element renumbering. The selection often (but not alwys) helps.
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Mar 22, 2013, 4:14 a.m. EDT
Hi Ivar,
Thanks for your quick answer.
I tried all of your recommendation to fix the problem but unfortunately nothing worked.
My problem is a bit complicated so I tried to do the same Parameter Sweep with a really simple heat transfer problem; include rectangle that one of its edges has temperature B.C and convective cooling B.C on part of other edge, all the other boundaries have the default condition.
Again, when I changed the size of the parameter that controlled the lengths of the B.C one of the condition expend all over the edge. One thing I notice, it take the B.C of the boundary numbered 1(the other part of the edge numbered 3) no matter what B.C defined for both part.
I tried to play in this simple problem with a different parameter but nothing solves the problem.
Do you have any other suggestion to fix this problem?
Thank you again,
Rama
Hi Ivar,
Thanks for your quick answer.
I tried all of your recommendation to fix the problem but unfortunately nothing worked.
My problem is a bit complicated so I tried to do the same Parameter Sweep with a really simple heat transfer problem; include rectangle that one of its edges has temperature B.C and convective cooling B.C on part of other edge, all the other boundaries have the default condition.
Again, when I changed the size of the parameter that controlled the lengths of the B.C one of the condition expend all over the edge. One thing I notice, it take the B.C of the boundary numbered 1(the other part of the edge numbered 3) no matter what B.C defined for both part.
I tried to play in this simple problem with a different parameter but nothing solves the problem.
Do you have any other suggestion to fix this problem?
Thank you again,
Rama
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Mar 22, 2013, 9:33 a.m. EDT
Hi Rama,
please have a look at the attached model. It does not have any physical meaning and the values of properties are randomaly chosen. It is just meant to serve as a quick demonstration.
In this model positions of one point on the left and one point on the right boundary are changed by a parametric sweep (the parameter is "y_pos").
The selections of the boundary conditions "Atmosphere/Gauge 1" and "Hydraulic Head 1" are done using the selections "Box 1" and "Box 2" (Model 1 node -> Definitions, as Ivar suggested), which both depend on the "y_pos" parameter.
In this example model it works.
Maybe you can use a similar approach for your model?
Best regards,
Juergen
Hi Rama,
please have a look at the attached model. It does not have any physical meaning and the values of properties are randomaly chosen. It is just meant to serve as a quick demonstration.
In this model positions of one point on the left and one point on the right boundary are changed by a parametric sweep (the parameter is "y_pos").
The selections of the boundary conditions "Atmosphere/Gauge 1" and "Hydraulic Head 1" are done using the selections "Box 1" and "Box 2" (Model 1 node -> Definitions, as Ivar suggested), which both depend on the "y_pos" parameter.
In this example model it works.
Maybe you can use a similar approach for your model?
Best regards,
Juergen