What Are Rectifier Diodes? Power Supply Topology & Design Guide

Rectifier diodes are semiconductor devices built around a PN junction that permit current flow in only one direction. In power supply design, they are the critical first active stage that converts alternating current (AC) into pulsating direct current (DC). While a basic rectifier circuit simply blocks the negative half of an AC sine wave, modern power conversion requires selecting diodes based on reverse recovery time, forward voltage drop ($V_f$), and thermal limits. Whether you are building a bench supply or an embedded IoT gateway, understanding how rectifiers interact with your chosen regulator topology dictates your efficiency, heat output, and noise floor.

Rectifier Topologies: Linear vs. Switching for a 24W Load

When designing a 12V @ 2A (24W) power supply, the choice between a linear and switching topology fundamentally changes how your rectifier diodes are utilized. A linear supply relies on a heavy 50/60Hz step-down transformer followed by a standard silicon bridge rectifier. A modern offline switching supply rectifies the 120VAC mains directly, filters it to high-voltage DC, and uses a high-frequency flyback transformer with a fast-recovery or Schottky rectifier on the secondary side.

Design Example: 120VAC Input Protection and Rectification

Before AC reaches the rectifier bridge, it must survive the real-world grid. For an offline switching supply targeting 120VAC nominal (acceptable range 108–132VAC), you need a robust input protection stage. Never connect a rectifier bridge directly to an unprotected AC outlet.

• Overcurrent Protection: Use a 2A 250V slow-blow fuse. The slow-blow characteristic prevents nuisance tripping during the initial charging of the bulk capacitor.
• Transient Suppression: A 130V RMS Metal Oxide Varistor (MOV), such as the Littelfuse TMOV14RP130E, clamps voltage spikes from inductive loads switching on the same grid branch.
• Inrush Limiting: An NTC thermistor (e.g., Ametherm SL32 2R005, 5Ω at 25°C) limits the massive current spike when the 100µF 400V bulk capacitor charges from zero volts on startup.
• Primary Rectification: The KBP210G is a 2A, 1000V bridge rectifier. While 120VAC peaks at only 170V, grid transients can easily exceed 400V. A 1000V rating provides the necessary derating margin to prevent avalanche breakdown.

The Regulator Stage: Dropout and Headroom Math

If you opt for the linear route using a 15VAC step-down transformer, the peak DC after the KBP210G bridge is roughly 19V ($15V \times 1.414 - 2.2V$ bridge drop). To drop this to 12V at 2A, you might choose an LM338 linear regulator. The LM338 requires a minimum dropout voltage (headroom) of about 3V at 2A.

Thermal Management and Diode Derating

A common bench mistake is assuming a diode's amperage rating tells the whole thermal story. The KBP210G is rated for 2A average forward current. In a bridge configuration, current flows through two diodes in series during each half-cycle. With a typical $V_f$ of 1.1V per diode, the total bridge voltage drop is 2.2V.

If you pull a continuous 2A DC load through this bridge, the power dissipated as heat is $P = V \times I = 2.2V \times 2A = 4.4W$. The standard KBP through-hole package has a junction-to-ambient thermal resistance ($R_{\theta JA}$) of roughly 50°C/W. Without a heatsink, the temperature rise will be $4.4W \times 50°C/W = 220°C$. This exceeds the 150°C maximum junction temperature ($T_J$) and will physically melt your solder joints and destroy the silicon.

The Fix: You must either bolt the bridge to a chassis heatsink using thermal paste, or practice component derating by stepping up to a 4A or 6A bridge (like the GBU610) for a 2A load. A larger silicon die and thicker lead frames drastically lower the thermal resistance, allowing the part to run cool at ambient room temperature without forced air.

Ripple, Noise, and Filtering Expectations

Rectifiers output pulsating DC, not clean DC. The filter capacitor smooths this, but the residual variation is called ripple. In a 120VAC full-wave rectified linear supply, the ripple frequency is 120Hz. The peak-to-peak ripple voltage ($\Delta V$) is calculated as:

$\Delta V = \frac{I_{load}}{f \times C}$

For a 2A load and a 4700µF capacitor: $\Delta V = \frac{2}{120 \times 0.0047} = 3.54V$. You will see a 3.54V sawtooth wave riding on your DC bus. A linear regulator easily rejects this low-frequency ripple, resulting in a clean output.

In a switching supply, the secondary-side Schottky rectifier (e.g., MBR20100CT) switches at frequencies upwards of 65kHz to 150kHz. This creates high-frequency ringing and electromagnetic interference (EMI) due to the parasitic inductance of the diode leads and PCB traces. To manage this noise expectation, you cannot rely on bulk electrolytic capacitors alone, as their Equivalent Series Inductance (ESL) makes them ineffective at high frequencies. You must place a 100nF X7R ceramic MLCC and a small ferrite bead directly across the Schottky diode's cathode and the output ground plane to shunt the high-frequency switching noise away from your sensitive load.

Frequently Asked Questions

What are rectifier diodes used for in a bridge configuration?

In a bridge configuration (using four diodes or a single integrated bridge package), rectifier diodes are used to achieve full-wave rectification. Unlike a half-wave rectifier that discards 50% of the AC sine wave, a bridge rectifier flips the negative half-cycle into a positive polarity. This doubles the ripple frequency (from 60Hz to 120Hz on a US grid), which significantly reduces the bulk capacitance required to smooth the DC output and improves overall transformer utilization.

What are the differences between Schottky and standard silicon rectifier diodes?

Standard silicon rectifiers (like the 1N4007) have a relatively high forward voltage drop ($V_f \approx 0.7V$ to $1.1V$) and a slow reverse recovery time ($t_{rr}$), making them suitable only for low-frequency 50/60Hz AC rectification. Schottky rectifiers use a metal-semiconductor junction, yielding a much lower $V_f$ (typically 0.3V to 0.5V) and virtually zero reverse recovery time. This makes Schottky diodes mandatory for the secondary side of high-frequency switching power supplies, where a slow silicon diode would overheat and fail due to switching losses and reverse current spikes.

What are rectifier diodes' maximum reverse voltage limits and how do I select one?

The maximum reverse voltage limit is specified as the Peak Repetitive Reverse Voltage ($V_{RRM}$). When selecting a diode, you must calculate the peak AC voltage ($V_{RMS} \times 1.414$) and apply a minimum 2x safety derating factor to account for grid transients and inductive kickback. For example, rectifying a 24VAC transformer yields a 34V peak; a 100V rated diode (like the 1N4002) is sufficient. However, for offline 120VAC mains rectification (170V peak), you must use a diode rated for at least 600V, though 800V or 1000V (like the 1N4007 or KBP210G) is the industry standard for reliable long-term survival.