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.
Electromagnetic actuator
Posted 22 févr. 2013, 07:19 UTC−5 Low-Frequency Electromagnetics 2 Replies
Please login with a confirmed email address before reporting spam
I have problem with coupling the magnetic that generated by 2D Coil with permanent magnate to actuate the cantilever beam. Also, I have faced problem to specify the boundary of moving mesh that is necessary to calculate the the displacement of cantilever. Also, the current flowing through the coil was set at zero, but the result still shows that there is a force of attraction between the permanent magnet and the coil, which is not suppose to be so. There should'nt have been any interacting forces between the coli and the magnet. However, I have attached my file.
Attachments:
2 Replies Last Post 22 févr. 2013, 12:24 UTC−5
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
hi
I would appreciate any help
Khaled
I would appreciate any help
Khaled
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Khaled,
The main problem is likely to be one of mesh refinement. There is also an issue in the setup of the moving mesh boundary conditions.
The 'force calculation' domain condition is useful in that it is a very general way of calculating magnetic force. However, this particular method can be very sensitive to the mesh on the boundaries of the chosen domain if the field at the domain boundaries is high. Because you are computing the force on a permanent magnet (which has a considerable field) this method will likely require an extremely fine mesh on the surfaces of your permanent magnet to yield accurate results.
Rather than performing a mesh refinement study down to very small mesh elements for you magnet surface, it is easier (and less time and memory intensive) to use a different method of calculating the magnetic force. There is another, much less mesh-sensitive method of computing magnetic force, but one that only gives a result when performed on a current-carrying domain. This method can therefore be used on the coil (but not the permanent magnet). However, because the forces balance, you can use it to give you a total force on the cantilever too.
This method involves performing a volume integral of each of the “Lorentz force contribution” components (this appears in a similar results section of a slice plot as the other force variables), which are calculated by default as a function of position in current-carrying conductors. You can then simply apply the opposite of the resultant force to the permanent magnet.
As with all finite element modelling, you will always of course have to perform a mesh-refinement study to check your mesh is fine enough to give you mesh-independent results. However, the method described above should not require as fine a mesh as the method you implemented.
Finally, the moving mesh was not set up correctly in your model. You had the 'prescribed mesh displacement' (boundary condition) on the boundaries of your cantilever set to zero. These boundaries will in fact be displaced by the deformation variables u,v,w.
Best regards,
Oliver Lipscombe
COMSOL Support
The main problem is likely to be one of mesh refinement. There is also an issue in the setup of the moving mesh boundary conditions.
The 'force calculation' domain condition is useful in that it is a very general way of calculating magnetic force. However, this particular method can be very sensitive to the mesh on the boundaries of the chosen domain if the field at the domain boundaries is high. Because you are computing the force on a permanent magnet (which has a considerable field) this method will likely require an extremely fine mesh on the surfaces of your permanent magnet to yield accurate results.
Rather than performing a mesh refinement study down to very small mesh elements for you magnet surface, it is easier (and less time and memory intensive) to use a different method of calculating the magnetic force. There is another, much less mesh-sensitive method of computing magnetic force, but one that only gives a result when performed on a current-carrying domain. This method can therefore be used on the coil (but not the permanent magnet). However, because the forces balance, you can use it to give you a total force on the cantilever too.
This method involves performing a volume integral of each of the “Lorentz force contribution” components (this appears in a similar results section of a slice plot as the other force variables), which are calculated by default as a function of position in current-carrying conductors. You can then simply apply the opposite of the resultant force to the permanent magnet.
As with all finite element modelling, you will always of course have to perform a mesh-refinement study to check your mesh is fine enough to give you mesh-independent results. However, the method described above should not require as fine a mesh as the method you implemented.
Finally, the moving mesh was not set up correctly in your model. You had the 'prescribed mesh displacement' (boundary condition) on the boundaries of your cantilever set to zero. These boundaries will in fact be displaced by the deformation variables u,v,w.
Best regards,
Oliver Lipscombe
COMSOL Support