Abstract: Disclosed techniques achieve soft-switching conditions by determining relative timing of switch states and events, and making recurring timing adjustments in successive cycles (i.e., cycle-by-cycle) to reduce timing errors that introduce switching losses. Timing adjustments provide a prediction of when an optimal soft-switching condition will exist during a subsequent cycle so that switch-actuation signals are provided, irrespective of inherent signaling and feedback delays, in advance of actually observing the condition, thereby subsequently changing a switching state within a desired threshold of the targeted soft-switching condition. Error in the prediction is observed and compensating corrections applied during the next cycle.

Abstract: Disclosed techniques achieve soft-switching conditions by determining relative timing of switch states and events, and making recurring timing adjustments in successive cycles (i.e., cycle-by-cycle) to reduce timing errors that introduce switching losses. Timing adjustments provide a prediction of when an optimal soft-switching condition will exist during a subsequent cycle so that switch-actuation signals are provided, irrespective of inherent signaling and feedback delays, in advance of actually observing the condition, thereby subsequently changing a switching state within a desired threshold of the targeted soft-switching condition. Error in the prediction is observed and compensating corrections applied during the next cycle.

Abstract: Disclosed techniques achieve soft-switching conditions by determining relative timing of switch states and events, and making recurring timing adjustments in successive cycles (i.e., cycle-by-cycle) to reduce timing errors that introduce switching losses. Timing adjustments provide a prediction of when an optimal soft-switching condition will exist during a subsequent cycle so that switch-actuation signals are provided, irrespective of inherent signaling and feedback delays, in advance of actually observing the condition, thereby subsequently changing a switching state within a desired threshold of the targeted soft-switching condition. Error in the prediction is observed and compensating corrections applied during the next cycle.