Whitney AV Solver: non-isotropic permeability possible?

[crobar]

Hello,

I am interested in the 3D magnetic capabilities of Elmer, mainly for the simulation of electrical machines. It is common for electrical machines to be made of a laminated material, i.e. thin sheets of steel stacked on top of one another with a non-conductive varnish between them. The purpose of this is to reduce losses in the machine by preventing circulating currents in a certain direction (usually the axial direction in a rotating machine). The result of this is a material which has a different magnetic permeability in different directions, i.e. non-isotropic permeability. Similarly it has a non-isotropic electrical conductivity, since this is the main point of constructing it in this way.

My question is, is it possible to model this in Elmer? In the GUI at least, it doesn't seem like this is possible, as there is no option for multiple values of these parameters.

It might be even nicer if you could assign values of permeability, magnetisation etc. as a function based on, say, the node locations in the mesh or similar. The 2D simulation program FEMM for instance lets you specify the magnetisation as a formula where you may use certain variables, such as the x and y location in the mesh to determine the magnetisation when solving. However, since I haven't looked in the manual yet to see if this kind of thing is possible yet, I won't waste people's time asking about it here just now.

[raback]

Hi

Electric conductivity can be orthotropic and usually all material parameters can be function of anything using UDFs and MATC expressions.

-Peter

[crobar]

Hi

Electric conductivity can be orthotropic and usually all material parameters can be function of anything using UDFs and MATC expressions.

-Peter

Ok thanks, I'll have a look into MATC and UDFs then.