This paper looks at the problem of adaptability to the calibration of a traveling wave head, when wavelet transform singular point detection technology is applied to traveling wave ranging of DC lines. It also considers the difficulty of unifying organically the arrival time of the traveling wave head its velocity. An improved scheme is proposed. For the calibration of the traveling wave head, an improved Hilbert-Huang transform with full adaptive decomposition capability for different fault signals is used, and the traveling wave head is calibrated using the peak frequency point of high-frequency abrupt change. To unify organically the arrival time of a traveling wave head and its velocity, we propose the use of a Newton interpolation algorithm to compensate for the wave velocity one by one, and substitute the wave velocity after each compensation into the ranging formula considering the change of wave velocity. When the convergence accuracy of the ranging result meets the set convergence condition, the algorithm loop is terminated, and the ranging result gradually approximates the true fault distance. We perform fault simulation in PSCAD and the calculation of fault data in Matlab. The results show that the ranging method has high accuracy, strong robustness, and a certain resistance to transition resistance, and is suitable for various common fault types. This work is supported by the National Natural Science Foundation of China (No. 51767023) and the 2017 Natural Science Foundation of the Xinjiang Uygur Autonomous Region (No. 2017D01C028).