Starting with voltage spikes, these tend to be characterized as several hundred volts for tens of microseconds, originating from lightening strikes or inductive coupling of load steps. Present solutions using a transient voltage suppressor in the LRU connector assembly, combined with a PI filter and ferrite bead arrangement are effective and space efficient.
A more challenging area is preventing propagation of voltage surges of typically less than 100V for periods of tens or hundreds of milliseconds resulting from load dump. This occurs when the disconnection of one load circuit induces a short and rapid increase in voltage across the alternator and therefore in other loads sharing the same supply. One solution is to use a passive network comprising a series-inductor and high value electrolytic bypass capacitor, combined with a transient voltage suppressor and fuse. Such solutions tend to be bulky and some transmission of higher voltages can still occur, requiring the downstream components to be tolerant of higher input voltages than would otherwise be necessary.
Around the industry, designers have independently developed active solutions based around discrete components using a MOSFET pass element but these typically require significant bench time to optimize the sensing, control loop and pass transistor circuitry. Keeping the MOSFET pass element from overheating and within its safe operating area is often cited as the most challenging part of the design. Sometimes a fuse is still required to protect the MOSFET from an output short circuit fault condition. Naturally the replacement of blown fuses could present unwelcome logistics complications for civil aviation, or could put important military equipment temporarily out of action during critical operations. One solution to the voltage surge problem is the LT4356 surge stopper IC, a device well suited to the task, whose operation will be described in more detail below.
Finally voltage ripple on the incoming LRU supply can present further design challenges, in particular the MIL-STD-1275D specification for military vehicles in generator mode is quite extreme (refer to Table 1). Various approaches are employed, including allowing the protection circuit to pass through the ripple to the voltage regulation stage or where voltage ripple is present at more modest amplitudes, smoothing it within the protection circuit itself. In the latter case, the protection circuit must be optimized to cope with the dissimilar characteristics of large voltage surges and small amplitudes of slowly varying ripple.
Pressure on costs, space, and weight, combined with the increasing requirement for multiple low-voltage, high-current supply rails to power complex FPGAs and processors has set the trend toward POL (point-of-load) power architectures.
The use of exotic, large, modular, bolt-down isolated regulators with multiple outputs supplying the final voltage rails at board level is giving way to distributed and highly efficient POL switching regulators such as LTC’s µModule® family (see Figure 1). These are typically powered by an isolated intermediate supply bus within the LRU which in turn is fed with 28 volts DC or more, from the aircraft or vehicle power system.
Figure 1 – One member of the Linear’s µModule family
One consequence of the shift to POL power architectures is the opportunity to re-distribute the surge protection from a central supply board to the individual circuit boards within an LRU. The lower load allows for a small and efficient solution using a dedicated overvoltage protection IC.
Power supply protection is a huge step towards better future with safer and greener technology and energy resources in our country and abroad.
Any research is useful until its costs don't overflow possible income and revenue.
William - http://www.carid.com/
It looks to me that this device could help in achieving compliance to the EMC requirements against Voltage Surge (EN61000-4-5) and Electrical Fast Transient Burst (EFT: EN61000-4-4) for the power-supplies.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.