When asking how an AC to DC converter works, the direct converted answer for a standard 120V AC RMS mains input depends entirely on the topology. For a raw full-bridge rectifier with a filter capacitor, the unregulated DC output is 168.3V DC, calculated using the formula V_DC(peak) = (V_RMS × √2) - V_diode_drop, substituting values as (120 × 1.414) - 1.4V = 168.28V. However, for a practical, regulated step-down switching converter (like a 150W enclosed supply), the direct answer is a stable 12.0V DC at up to 12.5A. This is achieved by first rectifying the mains to high-voltage DC, then chopping it via a high-frequency MOSFET (often >100kHz) and stepping it down through a ferrite transformer before final LC filtering.
The Assumptions That Fix Your Output (Voltage, Phase, and PF)
The baseline assumption fixing our 168.3V raw math is a single-phase, purely resistive load with a unity power factor (PF = 1.0). If you change the input parameters, the conversion shifts dramatically:
- 120V vs 230V: On a 230V nominal European or UK mains, the raw peak shifts to
(230 × 1.414) - 1.4 = 323.8V DC. Modern switch-mode power supplies (SMPS) handle this automatically via a wide-range Active Power Factor Correction (PFC) front-end that boosts the DC bus to a fixed ~390V regardless of whether the input is 120V or 230V. - 3-Phase Shifts: For 3-phase AC, a 6-diode bridge is used. The ripple frequency jumps from 120Hz (single-phase) to 360Hz, drastically reducing the required bulk capacitance. The DC bus voltage shifts to approximately
1.35 × V_L-L(line-to-line RMS). Thus, a 480V 3-phase input yields roughly 648V DC, which is the standard high-voltage bus for industrial variable frequency drives (VFDs).
Conversion Table: ±20% Input Variance and Output Expectations
Mains voltage is rarely exactly 120V. According to ANSI C84.1 standards, utility voltage can vary by ±5% to ±10%, and brownouts can push this further. Here is how the conversion math holds up across a ±20% variance band (96V to 144V) for both unregulated raw rectification and a regulated 12V SMPS.
| Input AC RMS | Raw Peak DC (Unregulated) | Regulated 12V SMPS Output | Internal PWM Duty Cycle Shift |
|---|---|---|---|
| 96V (-20%) | 134.3V | 12.0V (Stable) | Increases to maintain 12V output |
| 108V (-10%) | 151.3V | 12.0V (Stable) | Slight increase |
| 120V (Nominal) | 168.3V | 12.0V (Stable) | Baseline (approx. 45% for buck) |
| 132V (+10%) | 185.2V | 12.0V (Stable) | Slight decrease |
| 144V (+20%) | 202.2V | 12.0V (Stable) | Decreases to prevent overvoltage |
Note: The regulated SMPS output assumes a high-quality switching regulator topology with a feedback loop (like a TL431 shunt regulator driving an optocoupler) that adjusts the primary-side MOSFET duty cycle to compensate for input fluctuations.
Decision Tree: Picking the Right AC-DC Converter Module
Do not build a mains-to-low-voltage DC supply from discrete components on a breadboard unless you are specifically studying rectifier theory. The risk of lethal shock and fire from improperly spec'd bulk capacitors is too high. Use this decision tree to select the correct pre-built, safety-certified module for your workbench.
| If Your Requirement Is... | Then Choose This Topology... | Concrete Part Pick (2026) | Approx. Cost |
|---|---|---|---|
| Mains isolated, 12V DC, up to 12.5A (CNC routers, LED arrays, ham radio) | Enclosed Switch-Mode Power Supply (SMPS) with Active PFC | Mean Well LRS-150-12 | $22 - $28 |
| Mains isolated, 5V DC, up to 2A (Raspberry Pi clusters, ESP32 gateways) | Encapsulated PCB-mount AC-DC module | Mean Well IRM-10-5 | $14 - $18 |
| Non-isolated DC-DC step-down from an existing 12V/24V battery bus to 5V logic | Synchronous Buck Converter Module | TI LM2596-based module (with Schottky diode) | $3 - $6 |
| Ultra-low quiescent current, non-isolated 3.3V for deep-sleep IoT sensors | Low-IQ Hysteretic Buck | TPS62840 Evaluation Module | $45 (Dev board) |
FAQ: Real-World Converter Failures and Edge Cases
Why does my AC-DC converter blow the mains fuse the moment I flip the switch?
This is caused by inrush current. When an AC-DC converter is first energized, the bulk DC filter capacitors (often 220µF to 470µF rated at 400V) act as a dead short until they charge to the peak AC voltage. This can pull 30A to 50A for a few milliseconds, tripping fast-acting fuses or nuisance-tripping AFCI breakers. The Fix: Install an NTC inrush current limiter (like the Ametherm MS35 10018) in series with the AC line, or use a power supply with a built-in active relay-based inrush bypass.
Can I wire two 12V AC-DC converters in series to get 24V?
Generally, no. Unless the manufacturer explicitly states the outputs are floating and series-compatible, wiring them in series can cause reverse voltage stress on the output capacitors of the lower supply during startup or short-circuit events. The internal feedback loops will also fight each other during transient loads. If you need 24V at high current, buy a native 24V supply (e.g., Mean Well LRS-150-24) rather than stacking two 12V units.
My multimeter reads 18V DC on a 12V unregulated wall wart. Is it broken?
No, it is behaving exactly as the math dictates. Unregulated wall warts consist only of a step-down transformer, a diode bridge, and a small filter capacitor. With no load attached, the capacitor charges to the absolute peak of the AC waveform (V_RMS × 1.414). A '12V' wall wart actually outputs about 12.6V AC RMS, which peaks at roughly 17.8V DC. Once you connect your actual 12V load, the voltage will sag down to the nominal 12V range due to transformer regulation and ripple. Never use an unregulated wall wart to power sensitive 12V logic without an inline linear or buck regulator.
