Mangostatic problem no DOF error

Analysis issues
1

Setup up a magnostatic problem but I get this error

STEP 1

Newton-Raphson iterative procedure is active

Decascading the MPC’s

Determining the structure of the matrix:

*WARNING: no degrees of freedom in the model

Using up to 0 cpu(s) for setting up the structure of the matrix.
number of equations
0
number of nonzero lower triangular matrix elements
0

Using up to 0 cpu(s) for the heat flux calculation.

The pertainant input cards are

*NODE, NSET=NALL
*ELEMENT, TYPE=C3D4, ELSET=C3D4
*ELSET, ELSET=Domain1
*ELSET, ELSET=Domain2
*ELSET, ELSET=Domain3
*MATERIAL, NAME=air
*CONDUCTIVITY, TYPE=ISO
1.0
*ELECTRICAL CONDUCTIVITY
1.0
*MAGNETIC PERMEABILITY
1.0, 1
*MATERIAL, NAME=magnet
*ELECTRICAL CONDUCTIVITY
667
*MAGNETIC PERMEABILITY
987500, 2
*MATERIAL, NAME=domain3
*ELECTRICAL CONDUCTIVITY
1.0
*MAGNETIC PERMEABILITY
1.0, 3
*SOLID SECTION, ELSET=Domain1, MATERIAL=air
*SOLID SECTION, ELSET=Domain2, MATERIAL=magnet
*SOLID SECTION, ELSET=Domain3, MATERIAL=domain3
*STEP
*ELECTROMAGNETICS, MAGNETOSTATICS, NO HEAT TRANSFER
0.25, 1, 1.e-5, 1.e+30
*EL FILE
EMFB, EMFE
*END STEP

Does anyone know the problem?

2

I reported this issue here: Joule heating analysis - no degrees of freedom in the model but came to the conclusion it was because I had no air domain (this procedure seems to be limited to quite specific subset of cases).

You have 3 domains (including air) and this kind of setup is mentioned in the documentation. You also have the required material properties (maybe even redundant for some materials because only the air needs magnetic permeability).

Can you share the whole input deck (via some hosting website) ? Is your model based on some example from the documentation ?

3

https://github.com/mrkearden/magnostatic

The problem is a copy of a FEMM, and Elmer, that have good agreement, after some work

I also included the FEMM and Elmer files

4

I also include the FEMM Files anf the Elmer Files, they give good results.

5

As I’ve mentioned, this procedure in CalculiX is limited to a specific class of problems and you may have to fulfill some other assumptions described under 6.9.22 Electromagnetism.

Also see this official example file: induction.inp

For instance, you may have to use shell elements for the solid part.

The problem is that CalculiX needs to generate some MPCs and it may need a specific set of domains, element types and materials to do this.

6

OK thanks, that gives me a direction to work with