System reliability and dependability is a major topic in many fields and fault detection and isolation (FDI) maybe one way to achieve this. FDI is a key element for maintenance on demand, which could decrease service cost and time significantly. In railway industry, the early detection of faults could lead to significant advantages. This paper addresses the detection of faults in the suspension system. The secondary lateral and anti-yaw dampers are the most critical parts in railway suspension. In this work, a Hybrid Extended Kalman Filter (Hybird EKF) is used to capture the nonlinear characteristics of the suspension elements. In order to detect and isolate faults, a nonlinear residual generator is used. This residual generator is improved until it is possible to distinguish between different faults.