Three-phase induction motors with a wide range of power ratings (ranging from a fraction of a kW up to a few MW), are the key elements in many industry applications. They are widely used in different commercial equipments. Today, it is very important to detect the fault as soon as it occurs. This is to avoid accidental and unpredictable shutdown of equipments due to the expansion of faults, this matter shows its importance particularly in heavy motors, which will lead to financial losses that cannot be recuperated. In addition, repairing such heavy motors is time consuming and costs a lot.Mechanical faults normally occur because of the damage applied to the mechanical components of the motor such as bearings and gear-box. The probability of these faults in electrical motors is more common, at about 50% to 60% . The most important mechanical faults are: bearing faults (40%–50%) and eccentricity (60%) . Eccentricity results in non-uniform air gap between the stator and rotor. Eccentricity can be of static type, dynamic type or a combination of both which is called mixed eccentricity. In static eccentricity, the rotational axis of the rotor coincides with the symmetrical axis, but it displaces from the stator symmetrical axis. In this case, the air gap distribution is not uniform around the rotor but it is time invariant. In dynamic eccentricity, the stator symmetrical axis coincides with the rotor rotational axis, but the rotor symmetrical axis displaces. Air gap distribution in this case is non-uniform around the rotor and also time variant. The static and dynamic eccentricities tend to coexist and mixed eccentricity must be considered in reality. In this case, the center of rotor, the center of stator, and the center of rotation are displaced with respect to each other. There are several methods to detect this fault .Among them is motor current signature analysis which due to Its noninvasive and easy implementation has a wide usage .In this project, the initial goal is current spectrum analysis in eccentricity condition .Therefore, first, some common eccentricity fault diagnosis methods are introduced .Fault indication indexes , characteristics and problems are presented and then the methods are evaluated .then the modified winding function approach to calculate the inductances is used. this theory proposes an improved analytical method to calculate the inductances in induction motors .Also modeling of the induction motor in faulty condition including the rotor and stator voltage and mechanical equations are developed.Then rotor slot harmonics and other harmonics related to eccentricity will be analyzed and the way these harmonics are generated in current spectrum is... Keywords: Induction Moror, Eccentricity, Fault Diagnosis, Condition Monitoring, Motor Current Signature Analysis (MCSA)