Integration of a new, complex technology which crosses several powertrain subsystem boundaries (and thereby involves multiple organizations), without introducing inadvertent failure modes, can be a difficult task. This paper illustrates an approach to addressing this task at a system level using an example new technology. In product development activities involving primarily reuse of known technologies, perhaps with minor improvements, organizational and subsystem boundaries are generally clear. Interfaces are well established and responsibilities for failure mode avoidance are generally known. Implementation of a new technology which involves multiple parts of the organization, however, presents unique challenges to failure mode avoidance. The example technology primarily impacts the exhaust system, the powertrain cooling system, and the powertrain control system. It can affect engine operation, emissions, and performance. Packaging of the system on-vehicle requires consideration of neighboring systems and structure. The impact of the various system functions and potential error states need to be communicated and understood by the areas affected. The numerous interfaces, intended and unintended, must be managed systematically to avoid potential failure modes. In a large, distributed product development organization, this can be a major effort. This paper is a case study of a systems engineering approach to failure mode avoidance (FMA) at a system level. The paper demonstrates how exchanges between subsystems at each interface are defined and highlights the importance of clear communication of responsibility for each interface in the failure-mode-avoidance process. An iterative process is described involving careful drawing of boundaries, definition of system-level functions, intensive interface analysis, and system-level failure mode effects analysis (FMEA).