problem with time dependent phase transition model

Hi all,
I've built a model which involves a phase transition.
I want to see how long does it take for the phase transition to occur.
when I run the model with range(0,2,800) it gives logical results
with the typical graph of temperature vs. time where the temperature stays the same
as long as the phase transition occurs. But, when I run with (0,2,900), the phase
transition disappears and I get a graph with a continues temperature rise.
Is there anyone who encountered the same problem or has a solution?

thank you,
Zvika