I get the message "Error: solver exceeded maximum iterations" when solving examples 6, 7 and 8 using the appropriate example input files. Is this to be expected? At least with example 6 the output seems OK. Of course all this writing to the screen drastically increases the run times. -- Peter Johnston.
JBP: there are a couple of possible issues here:
- The solver is working properly, but hasn't been given "enough time" to get to the desired tolerance
- Often, the solver will converge fairly quickly to a "good enough" answer, but may take a long time to get to a finer tolerance
- Try setting "
solvertol=1.0e-3" or even "solvertol=1.0e-2" and see if the solver converges
- Try setting "
maxiter=200" or "maxiter=500@@" and see if that helps (of course, it will result in slower run-times)
- The simulation is ill conditioned and the solver has done as much as it can do but still cannot get to the desired tolerance
- The simulation is producing errors and hence the solver cannot possibly converge -- often due to Not-A-Number (NAN) or Infinity values creeping into the solution (i.e. the simulation blew-up)
