Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Defining magnetic potential boundary conditions - need const mag field btw current planes in 2d simulation
Posted 25.11.2010, 09:29 GMT-5 3 Replies
Please login with a confirmed email address before reporting spam
Dear All
I am new to COMSOL. I want to solve for current distribution in the windings of a toroidal inductor. As a very simplified case I have attached a file, in which I have 2 copper bars, carrying +1Amp and -1 Amp.
I need to define the boundary conditions in a way that the magnetic field comes from left side and gets out from right side between the two planes. Physically I should have a constant magnetic field between the planes (bars), and zero magnetic field outside those (on top and bottom regions).
This means I should physically have +constant magnetic vector potential (name it +AzConst) on the top boundary (boundary 11), and -AzConst on boundary 2 (the bottom boundary). And I need whatever function Az(y) has on the left side boundary (1 and 9) to be the same function (Az(y)) on the right side boundaries (12 and 16).
Do you know how can I define these boundary conditions?
Bests,
- Mohammad
I am new to COMSOL. I want to solve for current distribution in the windings of a toroidal inductor. As a very simplified case I have attached a file, in which I have 2 copper bars, carrying +1Amp and -1 Amp.
I need to define the boundary conditions in a way that the magnetic field comes from left side and gets out from right side between the two planes. Physically I should have a constant magnetic field between the planes (bars), and zero magnetic field outside those (on top and bottom regions).
This means I should physically have +constant magnetic vector potential (name it +AzConst) on the top boundary (boundary 11), and -AzConst on boundary 2 (the bottom boundary). And I need whatever function Az(y) has on the left side boundary (1 and 9) to be the same function (Az(y)) on the right side boundaries (12 and 16).
Do you know how can I define these boundary conditions?
Bests,
- Mohammad
Attachments:
3 Replies Last Post 27.11.2010, 04:35 GMT-5
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
it sounds like you are asking for a periodic boundary condition, or ?
there are also manual ways to do that I believe there is an exdample in the ACDC doc 3.5, but with the new V4 its not trivial to be a beginner and try to port a 3.5 explanation to v4. Try first to read the v4 doc, and the 4.1 if you have got it, it's more complete than the 4.0
--
Good luck
Ivar
it sounds like you are asking for a periodic boundary condition, or ?
there are also manual ways to do that I believe there is an exdample in the ACDC doc 3.5, but with the new V4 its not trivial to be a beginner and try to port a 3.5 explanation to v4. Try first to read the v4 doc, and the 4.1 if you have got it, it's more complete than the 4.0
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Dear Ivar,
Thank you very much for your message. Would you be able to direct me to the document you are speaking about?
Thank you again,
- Mohammad
Thank you very much for your message. Would you be able to direct me to the document you are speaking about?
Thank you again,
- Mohammad
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
your model 1.mph is for me a 2D model, that means that you have a depth or thickness in "z" of 1[m] (the default of COMSOL in this case
then you use a "single coil turn, but where does it turn ? Its, like you have set it up a plate with a 1A current flowing in the "z" direction (positive for the upper, negative how you have defined it for the lower one.
The simulation shows this correctly, appart that the left and side border are cut such that the field cannot loop around, but the default external magnetic insolation BC's makes a sort of symmetry but in such a way that B field is paralle to the boundary whic is in contradiction with the plates and current flow, see the dynamic help that is usefull, even if it means a lot of reading, but that you must accept (the reading) when you start doing "multiphysics"
To make a "coil" it is better to use the 2D axi mode: it means that you have a cylindrical system with an axis at r=0 (horizontal axis COMSOL remains the axis in 2D.axi for "r" (the horizontal and usual x direction) and "z" for the vertical axis (the ususal y axis)
inductors are a somewht delicate subject it requires already quite some knowledge of COMSOL, it is worth to do first resistor and capacitive exercices, before attacking coils
--
Good luck
Ivar
your model 1.mph is for me a 2D model, that means that you have a depth or thickness in "z" of 1[m] (the default of COMSOL in this case
then you use a "single coil turn, but where does it turn ? Its, like you have set it up a plate with a 1A current flowing in the "z" direction (positive for the upper, negative how you have defined it for the lower one.
The simulation shows this correctly, appart that the left and side border are cut such that the field cannot loop around, but the default external magnetic insolation BC's makes a sort of symmetry but in such a way that B field is paralle to the boundary whic is in contradiction with the plates and current flow, see the dynamic help that is usefull, even if it means a lot of reading, but that you must accept (the reading) when you start doing "multiphysics"
To make a "coil" it is better to use the 2D axi mode: it means that you have a cylindrical system with an axis at r=0 (horizontal axis COMSOL remains the axis in 2D.axi for "r" (the horizontal and usual x direction) and "z" for the vertical axis (the ususal y axis)
inductors are a somewht delicate subject it requires already quite some knowledge of COMSOL, it is worth to do first resistor and capacitive exercices, before attacking coils
--
Good luck
Ivar