Search

Of great significance is knowledge of failure modes of IGBT modules under power cycling test (PCT) in advance. It can not only precisely determine accurate utilization of physics-of-failure lifetime prediction methods but also help optimize IGBT designs. However, establishing an accurate and generic offline failure mode classification method for different IGBT modules remains a challenging problem due to the complex degradation process of IGBT modules. In this article, a data-driven convolutional neural network (CNN) based method is proposed for quad-classification of failure modes for different IGBT modules under different PCT conditions. First, in order to accurately characterize the mapping between the failure modes and the precursor parameters, a framework of precursor parameters is meticulously established. Then, mainly using collected training data from existing publications, the CNN classification model is developed by a newly dynamic tuning-multilevel particle swarm-back propagation optimization algorithm. Finally, experimental PCTs of various IGBT modules are performed. The obtained PCT data and additional testing data from existing publications are used to verify the generalizability and robustness of the proposed classification method. The superiority of the proposed method is well demonstrated through comparison with random CNN, the state-of-the-art particle swarm optimization CNN, and other intelligent algorithms.

Lift-off of aluminum bond wires is one of the main failure mechanisms of insulated gate bipolar transistor (IGBT) modules. Its root cause is the crack propagation, which causes reduction in the contact area of bond wires at the bond-metallization interface. Thus, lifetime prediction based on the classic Paris law is very appealing due to its capability to account for such a failure mechanism. However, Paris law is currently restricted due to its dependency on inaccurate but expensive means to observe the crack rate. In this article, an analytical calculation method for the crack length of bond wires is proposed by the degradation curves of VCE,on measured during normal power cycling tests. The dye-and-pry method has been used to measure crack lengths to further validate the proposed analytical calculation method. The proposed analytical calculation method can avoid time-consuming measurement of crack length, and can economically calibrate the key constants c and m of Paris law, which has been employed to predict the lifetime of bond wires of IGBT modules by finite element method. Finally, the calibrated values of c and m make Paris law have better applicability to predict the lifetime of bond wires comparing c and m values reported in existing papers.