magnetic field of a radially magnetized ring magnet

The graphical user interface of Elmer
Post Reply
bfischer
Posts: 9
Joined: 31 May 2020, 17:11
Antispam: Yes

magnetic field of a radially magnetized ring magnet

Post by bfischer »

Hi there,

how can an I specify the magnetic field of a radially magnetized ring magnet in 3D? I tried to do it similar to tutorial 8 (only in 3D): Model --> Material --> MgDyn --> Magnetization 1, 2, 3. When I'm right, Magnetization 1, 2, 3 corresponds to the axes x, y, z and I would need to do a coordinate transform but how do I have to define this in Elmer? Otherwise, I might set the coordinate system to "Cylindric Symmetric" but what's the meaning of the coordinate mapping 1, 2, 3 in this case?

Thanks
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: magnetic field of a radially magnetized ring magnet

Post by raback »

Hi

You can write small functions to make the coordinate transformations. You can use the fact that r=sqrt(x^2+y^2), sin(phi)=y/r, cos(phi)=x/r etc. So for example r*sin(phi) would be just y. MATC is usefull for testing, for example:

Code: Select all

MySinPhi = Variable "Coordinate"
  Real MATC "tx(1)/sqrt(tx(0)^2+tx(1)^2)"
-Peter
bfischer
Posts: 9
Joined: 31 May 2020, 17:11
Antispam: Yes

Re: magnetic field of a radially magnetized ring magnet

Post by bfischer »

Thanks for the explanation. I tried to implement it like this and I was able to perform a simulation. However, the resulting magnetic flux density doesn't seem to be rotationally symmetric (see attachment). Here is the code for the radially magnetized ring magnet, that I used:

Code: Select all

Material 1
  Name = "N42 NdFeB Magnet"
  Density = 7500
  Magnetization 2 = Variable "Coordinate"
  Real MATC "900.0e3*tx(1)/sqrt(tx(0)^2+tx(1)^2)"
  Magnetization 1 = Variable "Coordinate"
  Real MATC "900.0e3*tx(0)/sqrt(tx(0)^2+tx(1)^2)"
  Relative Permeability = 1.05
End
Can you please tell me whether this implementation is correct?
Attachments
Elmer Simulation V03.JPG
(60.34 KiB) Not downloaded yet
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: magnetic field of a radially magnetized ring magnet

Post by raback »

Hi,

Yes, the sif looks correct assuming that your case is centered origin. However, the picture does not look as if you would have rotational symmetry. Are only the magnets cylindircal but not the fully geometry.

You can also break symmetry by the mesh. Particularly for a coarse mesh it is quite impossible to have a perfectly symmetric results.

-Peter
Post Reply