Currently, the main methods of fault location are based on traveling waves. In the presence of noise, the performance of such methods tends to decrease. Therefore, this paper presents a performance evaluation of four noise reduction morphological filters applied to the fault location methods for HVDC lines. In the analyses, different morphological filter configurations are adopted, as well as various fault conditions, sampling rates and noise levels. The tests were performed based on the one-terminal method using the Multi-resolution Morphological Gradient to identify the wavefronts. The faults were simulated for a 500 km monopole HVDC transmission line with a ground return. The line was modeled according to the CIGRE-HVDC system.