Couenne returns infeasibility for feasible instances

[Luca-Mencarelli]

Dear all,

I am using Couenne to solve several non-convex MINLPs. The problem is that when I set "min 0" as objective function Couenne finds a feasible solution, nut when I change the objective function Couenne returns infeasibility. I have already tried to deactivate all the bound bound tightening options, but Couenne returns always that the instance is infeasible. I can provide the .nl files if necessary.

Thanks for all your help. Best regards,
Luca