November_EDFA_Digital

edfas.org 63 ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 20 NO. 4 and other possible electrical detractors from outside is a mandatory part of such failure anamnesis. In the case of broken stitchbonds, interconnect opens, or soldering problems, a PCB evaluation on bending risks in assembling, system integration, and application needs to be part of the anamnesis—frequently accompanied by further failure analysis on the PCB or system level, or even process audits within the supply chain. Most of the examples highlighted in my article about early life failures in automotive electronics in this issue of EDFA are based on failed semiconductor devices—while most root causes have been found outside those devices, within the systems. Of course, every system and every system-related failure requires a unique anamnesis approach, depend- ing on many circumstances of the observed failure. But the basic principle, the top-down approach, remains the same. A team of EMPA and RoodMicrotec colleagues summarized the most important questions for starting a successful failure anamnesis in the form of a roadmap, shown on this page. Of course, the questions for any new case will focus in different directions, but the roadmap may supply some help in finding which questions could be the most useful for determining root cause. For those of us who are accompanying semiconductor processes including packaging, the future challenge will remain tobe the technology driver for device failure analy- sis methods and tools. 3D failure localization methods within stacked devices, backside device editing by FIB, and highest resolution imaging in the lower nanoscale, even at higher working distances, will be their medium- term topics. Their findings are the indispensable input to successfully controlled processes. Our colleagues faced with field failures and customer returns need to open their minds beyond the scope of device failure analysis. This sounds like an easy claim as long as all participants in the system supply chains are openminded andwilling to cooperate, which is frequently not the case at the beginning. There must be large diplo- matic efforts in order to avoid the feeling of “passing the buck” to someone. Theseexperts needan interdisciplinary knowledge base of applications, systems, circuitry, and processes around soldering and PCBs. The real challenge is to get ridof strictly device-focused 8Ds and complement or sometimes even replace this systemwith coordinated, multi-company cooperation ina combined failure analysis and anamnesis. Download and print the full size version of this roadmap at bit.ly/2N3kxAf.