An isolated DC-DC converter designed for a wide input-voltage range must manage switching losses and component stresses as operating conditions change. The hybrid T-type zero-voltage-switching pulse-width-modulated converter (HT2-ZVS-PWM) addresses this challenge with a six-switch power stage, a transformer, and selectable three- or four-level modulation. A 1.2 kW laboratory prototype was evaluated with a 240–480 V DC input, a 300 V output, and a 50 kHz switching frequency.
The design is relevant to engineers assessing isolated power-conversion architectures where input voltage may vary substantially. Its operating strategy combines a T-type switching cell with soft switching and different modulation modes, rather than relying on one fixed switching pattern across the full input range.
How the hybrid T-type isolated DC-DC converter is arranged
The converter has six switches. Four are exposed to the maximum input-voltage level, while the two switches forming the T-type cell are exposed to half that level. This division of voltage across the switching devices is a defining feature of the topology.
On the input side, two capacitors form the DC bus. The switching stage generates a waveform that is applied to a transformer, providing galvanic isolation. On the secondary side, a four-diode rectifier and capacitive output filter deliver DC voltage to the load. The transformer’s leakage inductance also plays a role in the converter’s switching transitions.
This arrangement combines features of a full-bridge converter and a multilevel switching cell. The intended result is a power stage that can use different voltage levels at its switching nodes while maintaining an isolated output.
Three-level and four-level PWM operation
The HT2-ZVS-PWM uses two modulation strategies. Three-level modulation is applied over part of the operating range, while four-level modulation is used in another region. In the reported design, the choice of modulation is associated with the converter’s duty-cycle operating region: three-level modulation covers values below 0.5, and four-level modulation covers values above 0.5.
Changing the modulation pattern lets the converter adapt how it synthesizes the voltage applied to the transformer as input conditions vary. The analysis considers multiple operating modes, including continuous and discontinuous conduction. This distinction matters because the current in the transformer’s leakage inductance does not behave the same way in each mode, affecting both the switching sequence and the conditions available for soft switching.
For engineers evaluating the design, modulation selection is not simply a matter of choosing a waveform. It is connected to duty cycle, conduction mode, and the current available during switching transitions. The work also analyzes device stresses and dynamic behavior, providing a basis for selecting component ratings and assessing control response.
What zero-voltage switching contributes
Zero-voltage switching (ZVS) means that a switch is turned on when the voltage across it has fallen to zero. Compared with hard switching, this can reduce the voltage-current overlap during turn-on. In the proposed converter, the modulation strategies are designed to provide ZVS for all six switches in continuous-conduction mode.
The switching transition depends on the interaction of the leakage-inductance current and the effective capacitances associated with the switches. During dead time, this current helps change the voltage across devices before the next switch is turned on. Consequently, the timing and available current are important design considerations; ZVS should be assessed at the intended operating points rather than assumed from the topology alone.
The thesis also examines discontinuous-conduction operation, but its stated all-switch ZVS result applies to continuous conduction. That distinction is important when translating the analysis into a practical design or operating envelope.
Prototype results across the input range
Experimental testing used a 1.2 kW prototype with a 240–480 V input range, a 300 V output, and a 50 kHz switching frequency. Results included operation at a fixed point as well as operation across the wider input range. The reported fixed-point efficiency was 97.38% at nominal power.
For wide-input-range operation, the reported efficiencies were 96.35% at a 240 V input and 60% of nominal output power, and 95.72% at a 480 V input and 70% of nominal output power. These figures correspond to different load levels and should not be read as a direct, like-for-like efficiency comparison. The results demonstrate operation at the tested conditions; they do not establish performance at every possible input, load, or thermal condition.
The experimental work also examined switch and rectifier waveforms, input and output behavior, and converter response to load changes. Together with the theoretical analysis and design methodology, these measurements provide practical evidence for the proposed operating approach.
Practical implications for power-conversion design
The hybrid T-type approach offers a structured way to address a broad DC input range: use a switching cell that divides voltage levels, select between three- and four-level PWM according to the operating region, and use leakage-inductance current to support ZVS transitions in continuous conduction.
For system designers, the central takeaway is that topology, modulation, and magnetic design need to be considered together. Input range and load conditions determine the relevant operating modes, while switch voltage exposure and soft-switching conditions influence component selection and timing. The prototype results support the converter’s feasibility at the stated test points, while the detailed operating-mode analysis helps identify what must be checked when adapting the design to a particular application.