Defense programs are demanding more from power supplies: higher power density, tighter EMI margins, smaller physical envelopes, and better efficiency, all at once.
ACT engineers address these competing pressures through expert implementation of soft switching topologies, a capability that competitors relying on older hard-switched architectures cannot match. The difference shows up in every dimension of performance that matters to a defense program: size, weight, efficiency, thermal management, and EMI compliance. Understanding why requires a look at the physics of how a power supply switches.
In a hard-switched converter, transistors turn on and off while significant voltage and current are simultaneously present at the switching device. Every transition dissipates energy as heat. At tens or hundreds of thousands of switching events per second, those losses accumulate into a substantial system loss. Increase the switching frequency to shrink the magnetics, and the thermal penalty grows in proportion. Hard switching also drives fast dv/dt and di/dt transitions that couple into parasitic paths in the PCB and heatsink structure, generating the electromagnetic interference that MIL-STD-461 limits.
The result is a set of tradeoffs that become increasingly difficult to resolve as SWaP requirements tighten. Push the frequency up to reduce size, and the switching losses and EMI signature both grow. Add filtering to control the EMI, and the size and weight savings are partially erased. Hard switching creates a performance ceiling that legacy suppliers cannot design around.
Soft switching solves the hard switching problem by ensuring that each switching transition occurs at or near zero voltage, zero current, or both. When a transistor turns on with zero volts across its drain-to-source junction, the crossover energy that would otherwise become heat is eliminated. When it turns off with zero current in the power path, the trailing-edge loss disappears as well. The dominant loss mechanism shifts from switching loss to conduction loss, a condition that is predictable.
The mechanism behind this is the resonance between the parasitic output capacitance of the switching device and the parasitic or intentional inductance in the circuit. These form an LC tank that oscillates during the dead time between switching events. The inductance discharges the device capacitance, driving drain-to-source voltage to zero before the gate fires. ACT engineers utilize this behavior through control schemes, various topologies, and designing the LC tank circuit to maintain soft switching conditions across the operational range of the power supply.
Zero Voltage Switching (ZVS) and Zero Current Switching (ZCS) are the two primary conditions ACT designs toward, and the topologies used to achieve them include the LLC resonant converter, the Phase-Shifted Full Bridge, the Dual Active Bridge, and the Quasi-Resonant Flyback. Each exploits resonant behavior in a different way, and ACT engineers select and implement the appropriate topology based on the specific requirements of the program.
Because soft switching holds switching loss near zero, ACT engineers can push switching frequencies significantly higher than a hard-switched design would allow at equivalent power levels. Higher frequency directly reduces the size of magnetic components, which in power supply design are among the largest contributors to physical volume and weight. The ability to shrink the magnetics without a thermal penalty is what allows ACT to deliver compact, high-power-density supplies that meet demanding SWaP requirements.
This also enables ACT designs to scale with modern wide-bandgap semiconductor devices. GaN and SiC switching devices have extremely small parasitic capacitances and are capable of very high slew rates, but those slew rates produce severe EMI in a hard-switched circuit. Soft switching is what allows ACT to capture the performance potential of these devices, operating them at high frequency and high power density while keeping the EMI signature within specification.
The efficiency gains from soft switching translate directly into lower junction temperatures across the switching devices. A converter that dissipates less switching loss runs cooler at a given power level, which has compounding effects on long-term reliability. Lower operating temperature keeps devices well within their operating range and reduces electromagnetic interference and thermal cycling stress on solder joints and component leads, leading to an improvement in mean time between failure (MTBF).
Because soft-switched devices carry more thermal headroom, ACT engineers can, in some cases, achieve higher power throughput from a given device footprint rather than stepping up to a larger or more expensive component. Lower junction temperatures also reduce the external cooling burden: smaller heatsinks, potential elimination of forced-air cooling, and feasibility of conduction-cooled solutions where fan-cooled designs would otherwise be required.
The dv/dt and di/dt transitions of a hard-switched converter are the primary source of conducted and radiated EMI in most power supply designs. Fast voltage and current transitions couple into parasitic paths within the PCB stackup, heatsink structure, and interconnects, producing differential-mode and common-mode noise that must be suppressed to meet MIL-STD-461 limits.
Soft switching controls these transitions at the point of generation. Because the device transitions when voltage or current is at or near zero, the slew rate of the event is inherently reduced. Less high-frequency energy is injected into the system to begin with, which means less filtering is required to meet compliance. For ACT customers, this means a lower risk of late-stage EMI qualification failures and a filter design that does not grow back the size and weight that was engineered out of the rest of the supply.
Soft switching is not required in every application. At lower power levels and more relaxed SWaP and EMI requirements, hard-switched designs remain viable. However, as performance demands tighten across any of the key parameters, the tradeoffs of hard switching become increasingly difficult to manage. Programs that need higher switching frequencies for transient response, tighter physical envelopes, compliance with MIL-STD-461, or improved efficiency at any power level are programs where soft switching is not a preference but a requirement.
ACT engineers implement soft switching as standard practice for high-performance designs across vetronics, avionics, maritime electronics, sensor systems, and counter-UAS applications. The ability to scale power upward while holding physical size and thermal load constant is what separates ACT’s designs from suppliers still working within the constraints of hard-switched architectures.
If your program is facing SWaP constraints, efficiency requirements, or EMI qualification challenges, contact ACT to discuss how our design capabilities can be matched to your high-performance requirements.
