When I take a first order approximation to my model the BK conditions are satisfied and I get vaguely sensible looking impulses. However, if I take a second order approximation it reports that the BK conditions are not satisfied ("no stable equilibrium"). My understanding was that determinacy (or otherwise) was determined entirely by the first order approximation, perhaps I'm wrong?
I am using the daily windows snapshot (dynare-2010-06-13-win.zip) so this may just be a bug with the head. If this is a bug let me know and I'll PM you the model file.
T