Bootstrap Circuit

A bootstrap circuit solves the specific problem of turning on a high-side N-channel MOSFET in a half-bridge, whose gate needs to sit several volts above its own source — but that source is the switching node, which itself swings up near the supply rail when the device turns on, so no fixed voltage on the board is ever high enough — by charging a small capacitor to the supply voltage while the switching node sits low, then letting that charged capacitor float upward along with the switching node once the high-side device turns on, carrying its stored charge above the rail exactly when the gate driver needs it there.

The trick is entirely about a floating reference frame, not about generating a new voltage from nothing. A bootstrap capacitor never produces more than VDD of voltage across itself — exactly the same as the fixed supply rail used to charge it — but because one of its plates is tied to the switching node rather than to ground, the capacitor's absolute voltage (measured against true ground) rises right along with the switching node, so a gate driver referenced to that capacitor sees a voltage that's always VDD above the node no matter how high the node itself has climbed; the whole mechanism works by redefining what "VDD above" actually means in absolute terms as the switching node moves, not by boosting the voltage itself.

This is fundamentally different from a charge pump's mechanism, and the difference matters for how each circuit is actually used. A charge pump (the technique covered in this site's charge-pump article) uses its own internal switching clock, independent of any load, to repeatedly pump charge through a diode-capacitor ladder and build up a fixed elevated voltage rail that stays available continuously; a bootstrap circuit has no clock of its own at all — it depends entirely on the half-bridge's own switching activity to recharge the capacitor every time the low-side device turns on, which means a bootstrap supply is inherently tied to and limited by the switching pattern of the circuit it's powering, not an independent, continuously-available rail.

Duty cycle limitations are the direct, unavoidable consequence of depending on the half-bridge's own switching for recharge. If the high-side device is commanded on for too long without the low-side device getting a turn-on window to recharge the bootstrap capacitor, charge leaks away from gate-drive current and the capacitor's own leakage paths, and the floating supply voltage droops — eventually dropping below what's needed to keep the high-side gate fully enhanced, which can partially turn off the device and cause excessive conduction losses or outright failure; this is exactly why bootstrap-driven half-bridges have a maximum high-side on-time specification and why some designs insert a brief forced low-side pulse purely to refresh the bootstrap capacitor even when the application doesn't otherwise need one.

Sizing the bootstrap capacitor and diode correctly is a balance between recharge speed and leakage tolerance, not a simple pick-a-big-capacitor exercise. A larger bootstrap capacitor droops less per high-side cycle and tolerates a longer maximum on-time, but it also takes longer to fully recharge during the low-side on-time and demands more peak charging current through the bootstrap diode each cycle; the diode itself needs a fast reverse-recovery characteristic since it must stop conducting cleanly the instant the switching node starts rising, or charge will leak backward out of the capacitor into the rail at exactly the wrong moment — gate driver ICs with integrated bootstrap diodes (like many commercial half-bridge driver chips) exist specifically to get this timing-critical component right without relying on a discrete diode's variable characteristics.

Read a bootstrap circuit through a *riding-the-reference-frame* lens rather than a *voltage-boosting* lens: the capacitor never stores more than one supply's worth of voltage, but because its own reference point rises with the switching node it's attached to, the same fixed charge delivers exactly the above-rail voltage a high-side gate needs at exactly the moment it needs it — a solution that works only because the circuit it's powering is itself switching, which is the opposite dependency a standalone charge pump has.