Precise detection of the misalignment can reduce maintenance cost, power loss and improve life of rotating machinery. In this paper, Hilbert Huang Transform (HHT) of the run-up response of a rotor is used to detect coupling misalignment. A finite element analysis is carried out to simulate desirable level of possible parallel and angular misalignment. All the best conventional detection techniques like Fourier spectrum analysis, Continuous Wavelet Transform are compared with results obtained by HHT. The effect of Signal to Noise Ratio and angular acceleration on all the methods to detect misalignment is discussed. The comparison of the methods is oriented towards detecting least possible misalignment level. Finally, experiments are performed on a test rotor rig to detect induced misalignment using the proposed method. It is observed that this method is sensitive to low misalignment, and consumes less testing time. Copyright © 2014 by ASME.