SDG #219 Adding Output Capacitance to the Full Bridge Rectifier

This follow-up to the earlier active rectifier testing adds the one component most real power supplies actually have: a reservoir capacitor on the output. That single change alters the picture considerably, because a capacitor turns a smooth resistive conduction angle into short, high peak current bursts as the diodes only conduct near the top of each half cycle. The practical consequence is that the average current rating on a datasheet stops being the number you should be designing to, and repetitive peak current becomes the limiting factor.

Three rectification approaches are compared under the same loaded conditions: ordinary 1N4007 silicon diodes, the NMLU1210TWG MOSFET-based active diode, and a discrete MOSFET bridge controlled by the LT4320 ideal diode bridge controller. Voltage drop and efficiency are plotted against load current so the crossover points are visible rather than assumed. The silicon bridge loses roughly two forward drops regardless of load, which hurts badly at low output voltages, while the MOSFET-based options behave like resistances and therefore look best at moderate current, with their own quiescent and gate drive overheads showing up at light loads.

The most instructive part is the closer look at what the LT4320 is actually doing on a cycle by cycle basis, including how it decides when to turn the MOSFETs on and off with a capacitive load and what that means for conduction timing. For anyone designing a low voltage mains or transformer-fed supply, the takeaways are clear: capacitor sizing, peak current rating, and rectifier technology all interact, and picking an active rectifier only pays off when the load current and output voltage justify the extra complexity and cost.