Zero Voltage Switching Conditions

The Core Requirements for ZVS

At its heart, achieving ZVS is about ensuring two things happen correctly during the dead time—the brief interval when the switching element is off:

1. The voltage across the switch must be driven to zero.
2. The switch must be turned on precisely at that moment.

For this to happen, the resonant tank needs to have both enough energy to do the work and enough time to complete it. Let's break down these requirements into three primary conditions.

Condition 1: Sufficient Resonant Energy

In the last lesson, we learned that the energy stored in the resonant inductor ($L_r$) is used to charge and discharge the resonant capacitance ($C_r$), which is dominated by the switch's output capacitance, $C_{oss}$.

The fundamental energy condition for ZVS is straightforward: the initial energy stored in the inductor must be greater than or equal to the energy needed to swing the capacitor's voltage across the full range (e.g., from 0 to $V_{in}$).

$W_L\ge W_C$
$\frac{1}{2}{L}_r{I}_{initial}^2\ge \frac{1}{2}{C}_r{V}_{swing}^2$

This single relationship leads to the most critical practical constraint of ZVS: there must be a minimum amount of current flowing in the resonant inductor to store enough energy to achieve a full voltage swing.

This means ZVS is inherently dependent on the load current. At very light loads, the current may be too low, the inductor's stored energy will be insufficient, the voltage will not swing all the way to zero, and ZVS will be lost. This results in a "hard" turn-on with associated losses.

The following reading selection from a Texas Instruments application note explains this concept and provides the key equations.

From the reading, the key takeaway is the equation for the minimum primary current required for ZVS:

$I_{PRI(MIN)}=V_{IN}\sqrt{\frac{C_R}{L_R}}$

This formula is a cornerstone of ZVS design. It tells you that for a given input voltage and resonant tank, there is a hard limit on the minimum current required to achieve a lossless transition.

Condition 2: Sufficient Dead Time

The resonant voltage swing is not instantaneous. It takes a specific amount of time, dictated by the resonant frequency of the tank ($f_r\approx 1/(2\pi \sqrt{L_r{C}_r})$). The transition itself takes roughly a quarter of the resonant period.

Therefore, the dead time ($t_d$)—the duration programmed into the controller where the switch is held off—must be long enough for this resonant transition to complete.

A highly practical paper from Vishay explains this condition and its implications for design.

As the paper states, ZVS turn-on is only achieved if, within the available dead time:

• The gate capacitance of the outgoing MOSFET is discharged.
• The output capacitance of the incoming MOSFET is fully discharged close to zero.

This introduces an important trade-off.

• If dead time is too short: The resonant swing won't complete, and the switch will turn on with significant voltage across it (hard switching).
• If dead time is too long: After the voltage hits zero, the current will continue to circulate through the MOSFET's body diode for an extended period. Body diodes are typically less efficient and have higher voltage drops than the main MOSFET channel, leading to unnecessary conduction losses.

Optimizing dead time is a key part of maximizing the efficiency of a ZVS converter.

Condition 3: The Correct Turn-On Sequence

Meeting the energy and time conditions sets the stage for ZVS, but the final action is crucial. The switch must be turned on at the correct moment.

So, what happens when the resonant swing brings the switch voltage ($V_{DS}$) down to zero? The oscillation doesn't just stop. The voltage continues to swing slightly negative, to approximately -0.7V. This negative voltage forward-biases the MOSFET's intrinsic body diode, and the resonant current begins to flow through it.

This is the golden moment for ZVS. With the body diode conducting, the voltage across the switch is clamped at or near zero. The controller must then turn on the MOSFET while its body diode is conducting. This allows the current to smoothly transfer from the body diode to the main, low-resistance MOSFET channel, completing the zero-voltage turn-on.

To see this sequence in action, let's revisit a video from the previous lesson, focusing specifically on the turn-on event.

This sequence—resonant discharge followed by body diode conduction, then MOSFET turn-on—is the defining characteristic of a successful ZVS transition.

Conclusion

In this lesson, we have moved from the "how" of ZVS to the "what's required." You can now identify the essential criteria that a power converter must meet to achieve the benefits of zero voltage switching.

Key Takeaways:

• Sufficient Energy: The resonant inductor must carry a minimum current to store enough energy ($\frac{1}{2}L{I}^2$) to fully discharge the switch capacitance ($\frac{1}{2}C{V}^2$). This makes ZVS performance load-dependent.
• Sufficient Time: The controller's dead time must be long enough for the resonant voltage swing to complete but short enough to minimize losses from body diode conduction.
• Correct Sequence: The MOSFET must be turned on after its voltage has resonated to zero and while its intrinsic body diode is conducting current.

You now have a practical checklist to begin evaluating ZVS designs. When you look at a schematic or a PCB, you can start asking the right questions: What is the expected minimum load? What is the dead-time setting? How are the resonant components chosen?

In our next lesson, we will explore the trade-offs involved in meeting these conditions. Specifically, we will explain the trade-off between the ZVS operational range and circulating current losses, looking at what designers do to maintain ZVS at light loads and the efficiency price they pay for it.