Electronic systems such as electronic controls, onboard computers, communications, navigation, and radar perform many critical functions onboard military and commercial aircraft. All of these systems depend on electrical power supplies for direct current (DC) power at a constant (regulated) voltage to drive solid-state electronics. With these power supplies playing an important role in the operation of aircraft systems and subsystems, flight and ground crews need health state awareness and prediction tools that diagnose faults accurately, predict failures, and project life remaining of these components.
An integrated approach to switching mode power supply health management was developed that implements techniques from engineering disciplines including statistical reliability modeling, damage accumulation models, physics of failure modeling, and sensor-based condition monitoring using automated reasoning algorithms. Using model-based assessments in the absence of fault indications, and updating the model-based assessments with sensed information when it becomes available, provides health state awareness at any point in time. The diagnostic techniques, and prognostic models, have been demonstrated through accelerated failure testing of switching mode power supplies.
Switch-mode power supplies (SMPSs) are commonly used aboard aircraft where their weight, size, and efficiency make them preferable to conventional transformer-based power supplies. In addition to regulating the voltage of DC power, these novel circuits can also serve as DC-to-DC converters that can step down (“bucking” design) voltage like conventional supplies or step up (“boost” or “flyback” design) voltage. However, early SMPS designs suffered from sudden and catastrophic failures or generated excessive electromagnetic interference (EMI). More recent SMPS designs employ protective circuits to isolate sensitive components from damaging events.
The DC-DC converter at the heart of SMPSs uses a switching element, along with capacitors and inductors, to step up or step down voltage and current accordingly. High-speed switching enables the transfer of energy packets from the input filter capacitor to the output filter capacitor. The last stage filters out any high-frequency components from the DC output. Finally, the output is feedback into a control circuit that stabilizes the DC/DC converter.
Reliability studies of switching mode power supplies have shown that the majority of failures may be attributed to a small number of components. The relative frequency of specific component failures may vary based on SMPS topology, type of component used, derating factors, and location in the system. The components that commonly fail are classified into three different categories: switching transistors, filtering capacitors, and rectifying diodes.
Three types of switching transistors are commonly used in SMPS applications: bipolar junction transistors (BJTs), metal-oxide semiconductor field-effect transistors (MOSFETs), and insulatedgate biplolar transistors (IGBTs). Each type of transistor exhibits unique failure modes and rates, but switching transistors are generally the leading cause of SMPS failures. Bipolar junction transistors are widely used in SMPS designs where low priorities for weight and efficiency allow lower switching frequencies. MOSFETs and IGBTs are more common in applications where weight and efficiency requirements mandate higher switching frequencies.
Physics of failure models are used as the basis for incipient fault detection, fault to failure progression, and remaining useful life predictions. Critical transistor failure modes include thermal runaway, gate-oxide breakdown, contact migration, and thermal fatigue. Thermal runaway and thermal fatigue can affect all types of transistors, while gate-oxide breakdown only affects MOSFETs and contact migration only affects BJTs. Thermal runaway, contact migration, and thermal fatigue also affect diodes. Critical capacitor failure modes include dielectric breakdown and thermal fatigue.
Gate-oxide breakdown is one of the major concerns regarding MOSFET devices. Damage to the gate oxide (see figure) can result in excessive leakage current, increased standby power, and a decrease in response time. Eventually, the damage will cause a MOSFET to short-circuit.
The SMPS fault to failure progression models and diagnostic features for incipient fault detection were verified through accelerated failure tests of commercially available computer power supplies. To generate failures quickly, a load emulator was designed and built to subject the test specimens to extreme electrical and thermal stress.
The load emulator successfully generated multiple switching transistor and diode failures over the course of three months of testing. While tests designed to generate capacitor failures were performed, no catastrophic capacitor failures occurred.
This work was done by Thomas Dabney and Andrew Hess of the Joint Strike Fighter Program Office; and Rolf Orsagh, Douglas Brown, and Michael Roemer of Impact Technologies, LLC. JSF-0001
This Brief includes a Technical Support Package (TSP).
Prognostic Health Management for Avionics System Power Supplies
(reference JSF-0001) is currently available for download from the TSP library.
Don't have an account? Sign up here.