There is no reason why an IRF cannot have a sharp turning point. It very much depends on the parameterization and the endogenous propagation mechanisms you put into the model. If you want to find out where it comes from, you need to simplify your by taking away features. At some point, the kink must go away.