Development Of Automatic Detection System For Grinding Burn Of Bevel Gear Tooth Surface

With improvement of the level of design and processing,traditional geometric size of the detection has been unable to meet the actual needs of high-performance bevel gear.Added transmission error,structure borne noise,grinding burn and other testing items in Bevel Gear Tester to achieve the whole information evaluation is an important trend in the development of the bevel gear tester.As an important indicator of bevel gear performance detection.Grinding burns seriously affect the mechanical properties and service life of gears.However,all of the bevel gear tester on the market could not provide the function of grinding burn detection.Aiming at the function of grinding burn detection of the gear geometry and performance tester,the automatic detection system for grinding burn of bevel gear tooth surface is developed.This paper mainly includes:(1)Study on the mechanism of grinding burns and the principle of Barkhausen Effect.The specific design requirements of the automatic detection system for grinding burn of bevel gear tooth surface are analyzed.Considering the function and characteristics of the gear geometry and performance tester,the paper presents system general scheme and the design scheme of the MBN detection module,mechanical structure and detection software.(2)According to the system general scheme,the magnetic Barkhausen noise signal detection module and the mechanical structure are designed in detail.The error characteristic matrix is used to analyze the precision of mechanical structure.And the total deformation and equivalent stress of the mechanical mechanism under the limit working condition are analyzed by ANSYS.On the basis of this,the calculation method of motion control parameters is designed according to the structural characteristics of the gear geometry and performance tester.(3)According to the principle of Barkhausen Effect,the extracting method of magnetic Barkhausen noise signal envelope and eigenvalue is designed out.And the BP neural network model between the magnetic Barkhausen noise signal eigenvalue and the workpiece surface hardness is constructed.The accuracy of this model is verified by the detection of actual workpiece.(4)The system general scheme of the grinding burn detection software and the testing process are designed.Taking into account the characteristics of the motion control system of the gear geometry and performance tester,the communication variables between the grinding burn detection software and the motion control system are determined.On this basis,the motion control program of grinding burn detection system is designed(5)Using the detection system to detect the bevel gear pair,the correctness of the detection principle is verified by the experiment,and the detection system is stable and reliable operation.