XANSYS Message: 20681 [Go back to message list]
[bookmark on del.icio.us]
No rating yet
Subject: 2D EMAG FEA for a 2hp surface mounted magnet
Author: Mike Yaksh
Date: 2001-01-31 10:10:00Check if there are any interior gaps by selecting from the full set, the
exterior nodes
alls
nsel,,ext
If any nodes show up on the interior, this will be an edge. If no
boundary conditions was applied to these nodes, the flux lines will be
perpendicular, else if by some means these nodes were constrained, then
the edge would be flux parallel.
If for some reason some of the nodes on the side of the model did not get
the CE's (for a single pole), then those nodes on the side which have
neither a CE or a constraint, would be a flux normal condition. It only
takes a single node in iron to make a flux line to be normal to an edge
which has no conditions applied to it.
About torque:
1) define an element component of the rotor (magnet+iron) and use fmagbc,
before solving. In post processing use, torqsum
2) you can use torqc2d in post processing. If the mesh is reasonably
refined, the torque should not vary signifcantly. Try different radii near
the magnets. Select out the air elements then use the macro
Bills suggestion for using CEs to join the two models together is often
used, and has been presented in a number of ANSYS papers over the years.
original message;
i'm trying to a solve 2D Emag FEA problem of a 2hp surface mounted pm
motor. The modeling is a single pole modeling. After modeling the
geometry and meshing, it seem to work fine. I set parallel to lines
boundary condition to the opposite outside boundary and set the
PERBC2D on the symmetry boundary with odd-symmetry. The problem SOL
is fine. After viewing the 2D flux lines, it seems unrelistic.
the flux lines crosses the air gap and bend sharply to the side
(lelf) .. It should went straight through and curve to both side,
right. Can anybody tell me why it doing this ?
david