[SciPy-User] odeint : knowing time step that the solver tried to use...

Hi folks, I am dealing with a resolution of a set of ODE which periodically becomes to a stiff problem (but it isn't the trouble here). Odeint is used. There are some cases where the solver does not succeed to converge, a lot of time step are done before reaching 'tout'. Is there a way to get the unsuccessful time step? It may help us to determine whether our model has a problem... -- Fabrice Silva Laboratory of Mechanics and Acoustics - CNRS 31 chemin Joseph Aiguier, 13402 Marseille, France.