Zero voltage switching is usually seen in reference to switching power supplies but the same principles would likely apply to high frequency induction oscillators as well.
To operate switching power supplies at higher frequencies, which will permit smaller power supplies, transistor switching losses at turn-off and at turn-on must be fundamentally decreased. This is achieved in 'resonant converters' by associating a resonating LC circuit with the switching transistor, to render its current sinusoidal rather than square wave in shape. It is then arranged to turn the transistor on and off at the zero crossings of the current sine wave. As a result there is little overlap of falling current and rising voltage at the switching point and hence very much reduced switching losses. Circuits which turn on and off at zero current are referred to as zero current switching (ZCS) types.
However, switching losses can occur at turn-on even though there is no overlap of rising voltage and falling current at the zero crossing of the current sine wave. Considerable energy is stored on the relatively large output capacitance of a power MOSFET. When the MOSFET is turned on once per period T, it dissipates 0.5Co((2Vdc)^2)/T watts in the MOSFET. Circuits designed to cope with this problem are called zero voltage switching (ZVS) types. When 1/T=40kHz the watts can add up quickly.
There are a very large number of resonant converter topologies available with a large range of operating modes. As yet, resonant converters do not have the flexibility of PWM converters. They do not cope well with large line and load changes. Further, component tolerances are more critical.
