Diagnosis of bearings with advanced methods gained remarkable roles in the previous years. This article focuses on the manufacturing defects and methods to reveal and classify them. During manufacturing several faults could emerge because of the grinding operation, tool wear and chatter vibration. Inproper handling of the bearing parts because of the collosion to each other and the storing box that causes deformation. To reveal these problems several methods are applied in industry. For deeper surface analysis nitric acid can be used to initate the finished surface of the roller then natrium-carbonate that nautralize the elements. Vibration analysis in its standard Fourier form is not a new achivement but other mathematical tools could be applied to condition monitoring such as wavelet transform. It is an efficient tool for analyzing the vibration signal of the bearings because it can detect the sudden changes and transient impulses in the signal caused by faults on the bearing elements. In this article five different wavelets, Daubechies, Gaussian, Coiflet, Mexican hat, Meyer are compared according to the Energy to Shannon Entropy ratio criteria to reveal their efficiency for fault detection.