The increments in the chirp repetition interval caused by the utilization of Multiple-Input Multiple-Output (MIMO) modulation schemes degrades the maximum unambiguous measurable velocity in imaging radars. Commercial FMCW radars constantly deploy Doppler disambiguation methods to every detected target, even if their velocities are not ambiguous, as it is not possible to determine in advance whether the measured velocity is folded or not. In this paper, we propose an online adaptation mechanism to extend the maximum Doppler scale on-demand based on a simple but powerful ambiguity detector, which operates in a triple heterogeneous frame configuration. To the best of our knowledge, this is the first proposal for an online ambiguity detector, which triggers the dynamic configuration of certain phase profiles to reduce the chirp repetition interval and temporary increase the unambiguous Doppler scale. The approach in this paper is aligned with the cognitive radar principles, and does not require a constant execution of disambiguation methods in every frame, which leads to a lower processing overhead.