Failure Analysis And Research On Condition Assessment Method Of IGBT Power Module Bonding Wire

With the rapid development of new energy technology,power converter has been widely used.At the same time,more severe working conditions and more optimized product performance put forward higher requirements for the reliability of power converter.As the core device of power converter,the reliability of insulated gate bipolar transistor(IGBT)has been paid more and more attention.According to the latest failure statistics on the distribution of failure rate of each component in the converter system,21% failure rate of power semiconductor devices becomes the most vulnerable component of the converter.As one of the most widely used power semiconductor devices,how to effectively improve the reliability level of IGBT power module has become the focus of research.In the study of IGBT module failure mechanism,solder layer fatigue and bonding lead failure are considered as two main failure modes,and corresponding failure evaluation criteria are established.Due to the continuous improvement of solder layer technology,IGBT module bonding lead failure has been widely concerned.For a long time,bonding lead fault has been one of the important factors that restrict the reliable operation of IGBT module,and scholars at home and abroad have done a lot of research work on it.Firstly,this paper introduces the research background of IGBT power module reliability,expounds the current research status of IGBT power module reliability at home and abroad,and analyzes the importance of failure analysis and state evaluation of IGBT bonding line.Then,taking the SKM50GB12T4 model of SEMKRON series as an example,the dynamic model of the module is established in Simplorer,and the half bridge test circuit and double pulse test circuit is built to verify the effectiveness of the dynamic model.The voltage and current waveforms of the model are extracted,and the power loss of the module is obtained by discrete integral calculation using the voltage and current data extracted in MATLAB.Secondly,the three-dimensional finite element model of IGBT module is built in ANSYS,and the loss obtained is taken as the initial condition to analyze the temperature and stress of IGBT module,and the temperature and stress at the bonding line is the largest,and when the bonding line fails,the temperature and stress change of the module is greatly affected.Finally,a full bridge inverter physical platform is built.In order to simulate the breakout of bonding line caused by bad working condition or normal aging of IGBT module,the artificial breaking test of bonding line was carried out on the unplasticized IGBT module.The high and low temperature test chamber was used to simulate the change of environment temperature and change the load so as to change the forward conduction current of the module,record the saturation voltage drop of the emitter of the IGBT module,and establish the environment temperature,forward conduction current and saturation And pressure drop.The experimental results show that the healthy IGBT and the bond line broken IGBT have good discrimination on the surface of saturation pressure drop.In order to monitor the fault of IGBT power module bonding wire quickly and effectively,the bonding wire fault is divided into three grades according to the increment of saturation voltage drop.The least square support vector machine optimized by genetic algorithm is used to model and predict the three-dimensional data.The results show that the method can accurately predict the fault grade of the bonding wire,thus realizing the condition assessment of the bonding wire of IGBT power module.