Study on Prediction and Monitoring Technique of Failure of Gear Teeth by Means of AE Method

研究了SCM415淬火、S45C正火两种奥贝球墨铸铁弯曲疲劳过程中的声发射信号、声发射累计事件计数、声发射事件百分率、声发射累计能量计数、声发射能量率、声发射幅度、声发射信号频谱和裂纹扩展。两种奥贝球铁[奥贝球墨铸铁，ADI1：t_&lt；：等温处理温度，T_i：等温转变温度)]和SACM645离子渗氮齿轮齿在不同的齿面载荷下。确定了弯曲疲劳过程中声发射参数与疲劳裂纹扩展的关系，以及齿轮材料和齿面载荷对声发射特性的影响。为用声发射方法预测和诊断齿轮轮齿弯曲疲劳失效提供了基础数据。AE累积事件COUN…齿轮弯曲疲劳过程中的t和Ae累积能量在裂纹萌生前显著增加，而在表面硬化和ADI1齿轮的情况下，裂纹扩展过程中的累积能量几乎保持不变。离子渗氮齿轮疲劳过程中声发射参数的变化趋势与表面淬火齿轮相同，但增幅远小于表面淬火齿轮。对于ADI2齿轮，裂纹萌生后，这些声发射参数增加。在归一化齿轮的情况下，很难观察到这些声发射参数随裂纹萌生的变化。因此，在渗层、ADI1和离子渗氮齿轮的情况下，使用这些声发射参数预测裂纹萌生是可能的，但在ADI2和正火齿轮的情况下，使用这些声发射参数预测裂纹萌生是困难的。2.硬化层、ADI1和离子渗氮齿轮弯曲疲劳过程中的声发射幅值在疲劳初期较小，裂纹萌生前较大，裂纹扩展较小，断齿前较大。对于ADI2齿轮，裂纹萌生后，声发射幅值变大。在归一化齿轮的情况下，几乎不能观察到大幅度的声发射信号。3.离子渗氮、渗层、ADI和正火齿轮弯曲疲劳过程中声发射信号的最大幅值依次减小。较少

Research was conducted on AE signals, AE cumulative event count, AE event count rate, AE cumulative energy count, AE energy rate, AE amplitude, frequency spectra of AE signals and crack propagation in the bending fatigue process of SCM415 case-hardened, S45C normalized, two kinds of ADI [astempered ductile iron, ADI1 : T_<LAMBDA> = 875ﾟC,T_I = 375ﾟC,ADI2 : T_<LAMBDA> = 875ﾟC,T_I = 235ﾟC,(T_<LAMBDA> : temperature of austemper treatment, T_I : temperature of isothermal transformation)] and SACM645 ion-nitrided gear teeth under different tooth loads. The relationship between the AE parameters and the fatigue crack propagation as well as the effects of the gear material and the tooth load on the AE characteristics in the bending fatigue process were determined. Basic data were obtained for the prediction and the diagnosis of bending fatigue failure of the gear tooth by means of AE method.The main results obtained from this investigation are summarized as follows.1. AE cumulative event coun … More t and AE cumulative energy count in the bending fatigue process of gear teeth increase markedly just before the crack initiation and become almost constant in the crack growth in the cases of the case-hardened and the ADI1 gear. In the case of the ion-nitrided gear, the change of these AE parameters in the fatigue process shows the same tendency as the case of the case-hardened gear, but the increasing rate is considerably smaller than in the case of the case-hardened gear. In the case of the ADI2 gear, these AE parameters increase after the crack initiation. In the case of the normalized gear, the changes of these AE parameters with the crack initiation can hardly be observed. Hence, the prediction of the crack initiation using these AE parameters is possible in the cases of the case-hardened, the ADI1 and the ion-nitrided gear, but difficult in the cases of the ADI2 and the normalized gear. 2.AE amplitudes in the bending fatigue process of the case-hardened, the ADI1 and the ion-nitrided gear are small in the early stage of the fatigue process, become larger just before the crack initiation, smaller in the crack growth and larger just before the tooth breakage. In the case of the ADI2 gear, AE amplitude becomes larger after the crack initiation. In the case of the normalized gear, AE signals with large amplitude can not be almost observed. 3. Maximum amplitudes of AE signals in the bending fatigue process of the ion-nitrided, the case-hardened, the ADI and the normalized gear tooth become smaller in the order mentioned. Less