Trouble Flipping Magnetization of Object

[leahkong02]

I'm trying to simulate magnetic field of six axially magnetized ring magnets, where there are 3 pairs of 2 oppositely magnetized rings in each of the three coordinate directions. However, when I'm doing both the graphical output and magnetic field calculations, the field and magnetization in the y direction does not seem correct. The y-direction keeps behaving like two same polarized magnetized rings instead of opposite polarity. I'm not sure if this is a result of my using rotation objects incorrectly, as I haven't used scipy rotation objects before this, but my rotation seems to work in the x direction, and I've tried to change the rotation object to different permutations of 90 degrees about the x axis. Here is my code:

import magpylib as magpy
from scipy.spatial.transform import Rotation as R

inner_r = 60
outer_r = 35
h = 10
mag = [0, 0, 1000]
nmag = [0, 0, -1000]
distance = 70
ndistance = -70
dim = (outer_r, inner_r, h, 0, 360)

a = magpy.magnet.CylinderSegment(mag, dim, [0, 0, distance])
b = magpy.magnet.CylinderSegment(nmag, dim, [0, 0, ndistance])
c = magpy.magnet.CylinderSegment(mag, dim, [0, distance, 0])
d = magpy.magnet.CylinderSegment(nmag, dim, [0, ndistance, 0])
e = magpy.magnet.CylinderSegment(mag, dim, [distance, 0, 0])
f = magpy.magnet.CylinderSegment(nmag, dim, [ndistance, 0, 0])

c.orientation = R.from_euler('x', 90, degrees = True)
d.orientation = R.from_euler('x', 90, degrees = True)
e.orientation = R.from_euler('y', 90, degrees = True)
f.orientation = R.from_euler('y', 90, degrees = True)

rings = magpy.Collection(a, b, c, d, e, f)
magpy.show(rings)

And here is what the graphics look like:

Any help navigating this issue would be much appreciated!

[OrtnerMichael]

hi @leahkong02

Thanks for reaching out. There are two problems

1. There is a Bug in our Matplotlib graphic output
2. There is also a Bug in your code (the first rotation should be with respect to -90°)

Both together lead to confusion that is however easily solved !
Your correct code should be like this:

import magpylib as magpy

inner_r = 60
outer_r = 35
h = 10
mag = [0, 0, 1000]
nmag = [0, 0,-1000]
distance = 70
dim = (outer_r, inner_r, h, 0, 360)

a = magpy.magnet.CylinderSegment( mag, dim, [0, 0,  distance])
b = magpy.magnet.CylinderSegment(nmag, dim, [0, 0, -distance])
c = magpy.magnet.CylinderSegment( mag, dim, [0,  distance, 0])
d = magpy.magnet.CylinderSegment(nmag, dim, [0, -distance, 0])
e = magpy.magnet.CylinderSegment( mag, dim, [ distance, 0, 0])
f = magpy.magnet.CylinderSegment(nmag, dim, [-distance, 0, 0])

c.rotate_from_angax(-90, 'x')
d.rotate_from_angax(-90, 'x')
e.rotate_from_angax(90, 'y')
f.rotate_from_angax(90, 'y')

for x in [a,b,c,d,e,f]:
    print(x.getB((0,0,0)))

magpy.show(a,b,c,d,e,f, backend='plotly')

The output should be

[ 0. 0. 10.17666237]
[ 0. 0. -10.17666237]
[ 0. 10.17666237 0. ]
[ 0. -10.17666237 0. ]
[10.17666237 0. 0. ]
[-10.17666237 0. 0. ]

which shows that the field computation is good, and the graphic

Here the graphic output is created with the optional Plotly packge. Matplotlib will still give a bad output - this will be fixed with the next update.

does this answer your question ?

[leahkong02]

this helps, thank you so much!! magpylib has been super useful in my research, and I really appreciate your prompt response :)

[OrtnerMichael]

glad to help :)

[Alexboiboi]

Hi @leahkong02, thanks for your feedback about the library ;)

• The display issue is tracked here Bad display of magnetization #571 and will be fixed in the coming release, very soon.

[Alexboiboi]

Hi @leahkong02, we have released a patch version that also fixes the wrong display of magnetization arrows like in this case.

you can upgrade your version with

pip install -U magpylib

or by giving the explicit version number

pip install magpylib==4.1.2

or if you use conda:

conda install -c conda-forge magpylib=4.1.2

:-)

[leahkong02]

Thank you so much!!