More than only one K1 about 3 HOT 7 CLOSED

That's not possible unless you hack engine/anisotropy.go. What exactly are
you trying to accomplish?
On Apr 30, 2014 5:03 PM, "Gabriel Rodriguez-Rodriguez" <
[email protected]> wrote:

Is it possible to define more than only one K1/anisU pair?
In OOMMF it's possible to add several anisotropies just by naming them
separately, so you can define several uniaxial anisotropy axis with
different values (ej. K1(001)=1E5J/m3 AND K1(100)=1E3J/m3).
Is there any way to add more than only one anisotropy direction/magnitude?

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/16
.

from 3.

We are simulating thin films with mixed anisotropy: Mainly out of plane (001), but with a weaker (about 100 times weaker) in-plane component (100).
Anyway, this (weak) secondary in-plane anisotropy is big enough to break system magnetic degeneration (in-plane), something we find useful in order to study geometry-tuneable magnetic systems.

from 3.

If Ka = Ku1(001), Kb = Ku1(100), then I think you can just set AnisU =
(Kb/sqrt(Ka²+Kb²), 0, Ka/sqrt(Ka²+Kb²), Ku1 = sqrt(Ka²+Kb²). Unless I'm
missing something, that should be equivalent to what you asked for.

On Wednesday, April 30, 2014, Gabriel Rodriguez-Rodriguez <
[email protected]> wrote:

We are simulating thin films with mixed anisotropy: Mainly out of plane
(001), but with a weaker (about 100 times weaker) in-plane component (100).
Anyway, this (weak) secondary in-plane anisotropy is big enough to break
system magnetic degeneration (in-plane), something we find useful in order
to study geometry-tuneable magnetic systems.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/16#issuecomment-41818722
.

from 3.

As far as I know, you can't add directly two anisotropy directions like vectors because the resulting energy landscape is completely different from the resulting of the addition of the two separated energy terms (one for the in-plane anisotropy, and the other one for the out-of-plane). If we define an AnisU with the two components (x,0,z) compensated with the strength of the anisotropy energy densities, the results will differ from the system where we have the two separated terms. If I'm not wrong, your approach is to understand a problem with two anisotropies like a problem with an effective anisotropy which direction is parallel to the vector resulting from the addition of the two original.
Anyway, we are trying to introduce the in-plane anisotropy effect in the same way that you pointed.

from 3.

If Ka = Ku1(001), Kb = Ku1(100), then I think you can just set AnisU =(Kb/sqrt(Ka²+Kb²), 0, Ka/sqrt(Ka²+Kb²), Ku1 = sqrt(Ka²+Kb²). Unless I'm missing something, that should be equivalent to what you asked for.

The 2nd order anisotropy isn't linear so it wouldn't work: (u.m)^2 + (v.m)^2 ≠ [(u+v).m]^2.

This feature would be useful especially in cases where you have hard and easy axis terms simultaneously ---and since they will have different signs, a new term in mumax is needed.

from 3.

I'm wondering if there has been any change on this issue? I'd like to have the same functionality although in my case I would like to be able to simulate the application of a strain being applied at an axis not co-linear to the uniaxial anisotropy. As far as I can see, looking at the latest code, it is still not possible. Is this correct and if so, is there any scope for adding it in? Thanks!

from 3.

In the latest version of mumax it's possible to add custom energy terms, e.g. to define any combination of anisotropies you want. Please have a look at https://github.com/mumax/3/blob/master/test/custom_anisotropy.mx3

from 3.