This thread is a small gem because it tackles one of the most counterintuitive ideas in electronics: how do you get more voltage out of a circuit than you put in, without violating conservation of energy? The asker compares a boost converter to a capacitor, which is a reasonable first guess but ultimately wrong — and the comment section walks through why with admirable patience.

The core trick of a boost converter is the inductor, not the capacitor. Here's the intuition the thread builds up:

• A switch (usually a MOSFET) rapidly connects an inductor to ground, allowing current to ramp up and store energy in the inductor's magnetic field.
• When the switch opens, the inductor refuses to let its current change instantly (that's its defining property — V = L·di/dt). To maintain that current, it generates whatever voltage is necessary, often well above the input.
• That voltage spike pushes current through a diode into an output capacitor, which smooths it into a steady DC level higher than the input.
• Repeat thousands of times per second, and you have a stable boosted output.

What's educational here isn't just the mechanism — it's the conservation-of-energy framing several commenters use. You don't get "free" voltage. Power in (roughly) equals power out, minus losses. If you boost 5 V to 20 V, you're trading current: the input draws about 4× the current the output delivers. That single insight unlocks why boost converters can't magically run a 100 W device off a coin cell.

The thread also touches on the practical cousins — buck (step-down), buck-boost, and SEPIC topologies — and why switching converters dominate modern electronics over linear regulators (efficiency, often 85–95% versus a linear regulator's wasted heat).

For a beginner, the OP's instinct to reach for the capacitor analogy is telling. Capacitors store energy in an electric field and resist voltage changes; inductors store energy in a magnetic field and resist current changes. Boost converters exploit that current-inertia property in a way capacitors simply can't replicate alone. Understanding this duality is one of the bigger conceptual leaps in learning analog electronics.