On 08/23/10 12:00, Robert Cimrman wrote:
Hi Oleg,
On 08/22/10 20:32, oleg100 wrote:
Hello!
I try to apply constant volume forces to subregion (not whole domain). For example I use sfepy-release-2010.3/examples/standalone/interactive/ linear_elasticity.py
I add follow line:
ff=domain.create_region('ff','nodes in x< %.10f' % ( min_x + 0.1*(max_x - min_x)))
And replace "omega" in line which define t2 with "ff".
The result is error:
sfepy: infs or nans in the residual: [-0.78190974 0. 0. ..., 0. 0. 0. ] sfepy: False
/srv/src/sfepy-release-2010.3/examples/standalone/interactive/sfepy/solvers/nls.py(197)__call__()
196 --> 197 if self.log is not None: 198 self.log(err, it)
ipdb>
What I do wrong?
It should work that way -> it's a bug. I will check it out.
Thanks for the report! r.
It is fixed now in our git repo [1]. If you click at the 'commit' id, you will see the one line fix that was needed.
Let me know if that works for you.
r.