To convert a standard 12V nominal input to a regulated 5V, 2A output, the correct topology is a Buck (step-down) converter. Using the ideal duty cycle formula D = V_out / V_in, the baseline calculation is 5V / 12V = 0.416 (41.6%). However, assuming a realistic 88% efficiency (η) for a modern synchronous buck IC like the Texas Instruments TPS5430, the actual duty cycle shifts to D = V_out / (V_in × η), which substitutes to 5 / (12 × 0.88) ≈ 0.473 (47.3%). The converter's PWM controller dynamically adjusts this pulse width to maintain regulation as the input source fluctuates.
Because real-world power sources like lead-acid batteries or unregulated wall adapters rarely sit exactly at their nominal voltage, here is how the required duty cycle shifts across a ±20% input variance to maintain that 5V output:
| Input Voltage (V_in) | Variance from 12V Nominal | Ideal Duty Cycle (D) | Real Duty Cycle (η=88%) |
|---|---|---|---|
| 9.6V | -20% (Battery sag) | 52.1% | 59.2% |
| 10.8V | -10% | 46.3% | 52.6% |
| 12.0V | 0% (Nominal) | 41.6% | 47.3% |
| 13.2V | +10% (Alternator) | 37.9% | 43.0% |
| 14.4V | +20% (Absorption) | 34.7% | 39.4% |
Core Topologies and Voltage Conversion Ranges
Selecting from the various types of DC-DC converters depends entirely on the relationship between your input and output voltages, and whether galvanic isolation is required for safety or noise rejection. Below is a data-dense comparison of the primary topologies used in 2026 bench and embedded designs.
| Topology | Voltage Relationship | Transfer Function (Ideal) | Typical 2026 IC Example | Max I_out / V_in Range |
|---|---|---|---|---|
| Buck | V_out < V_in | V_out = D × V_in | TI TPS5430 / MP2315 | 3A / 4.5V to 36V |
| Boost | V_out > V_in | V_out = V_in / (1 - D) | MT3608 / TI TPS61040 | 2A / 2.5V to 24V |
| Buck-Boost | V_out ≷ V_in | V_out = (D / (1 - D)) × V_in | TI TPS63020 / LTC3530 | 2A / 1.8V to 5.5V |
| SEPIC | V_out ≷ V_in (Non-inverting) | V_out = (D / (1 - D)) × V_in | LT8705 / LM5118 | 5A / 4V to 80V |
| Flyback (Isolated) | V_out ≷ V_in (Isolated) | V_out = V_in × (D / (1-D)) × (Ns/Np) | UCC28740 / LinkSwitch-TNZ | Varies / 85V to 265V AC |
The Assumptions That Fix Your Conversion Math
The formulas listed above are not universal laws; they rely on strict operating assumptions. If you ignore these, your prototype will fail to regulate under load.
Continuous vs. Discontinuous Conduction Mode (CCM vs. DCM)
The standard duty cycle equations assume the converter is operating in Continuous Conduction Mode (CCM), meaning the current through the inductor never drops to zero during the switching cycle. If your load current drops below the critical threshold (often seen in IoT sensors sleeping between transmissions), the converter enters Discontinuous Conduction Mode (DCM). In DCM, the output voltage becomes heavily dependent on the load current and the inductor value, rendering the simple D = V_out / V_in formula mathematically useless. Modern ICs mitigate this by implementing Burst Mode or Pulse Frequency Modulation (PFM) at light loads.
The Dropout Voltage Limitation
A buck converter cannot output a voltage higher than its input minus its internal voltage drops. This is the dropout voltage. For a non-synchronous buck using a Schottky diode, the dropout might be 1.5V. For a synchronous buck using low-Rds(on) MOSFETs, it might be 0.2V. If you attempt to program a 5V output from a 5.1V input on a standard buck, the conversion is meaningless—the controller will simply run at 100% duty cycle, passing the input voltage minus the conduction losses straight to the output.
Scaling to Mains: 120V AC, 230V AC, and 3-Phase Shifts
DC-DC topologies do not plug directly into the wall. When your input shifts from a 12V battery to a 120V or 230V AC mains supply, the entire front-end architecture must change. You cannot use a standard 40V-rated Buck IC directly on a rectified mains bus.
- 120V AC Mains: After a bridge rectifier and bulk capacitor, 120V AC becomes approximately 160V DC. A standard non-isolated buck converter would require a massive, inefficient duty cycle (e.g., 5V / 160V = 3.1%) and a MOSFET rated for >200V. Instead, designers shift to an isolated Flyback topology (like Power Integrations' LinkSwitch family) which uses a transformer turns ratio to step down the voltage safely while providing galvanic isolation.
- 230V AC Mains: The rectified DC bus voltage jumps to roughly 320V DC. The Flyback topology still works here, but the primary switch must now be rated for at least 600V to handle the reflected voltage and leakage inductance spikes. The duty cycle shrinks further, often necessitating Quasi-Resonant (QR) switching to maintain efficiency.
- 3-Phase Industrial (400V+ AC): Rectified 3-phase yields a DC bus exceeding 600V DC. Flyback converters become impractical due to extreme voltage stress and transformer leakage inductance. At this level, the topology shifts to an LLC Resonant Converter or a Phase-Shifted Full Bridge. These topologies use zero-voltage switching (ZVS) to eliminate switching losses at high voltages and high power levels (typically >500W).
FAQ: Edge Cases and Meaningless Conversions
When is a Boost converter duty cycle calculation meaningless?
The boost formula V_out = V_in / (1 - D) implies that as D approaches 1 (100%), V_out approaches infinity. In reality, this conversion becomes meaningless around D = 0.85 to 0.90. Beyond this point, parasitic resistances in the inductor and MOSFET dominate, and efficiency collapses to zero. Furthermore, boost converters suffer from a Right-Half-Plane (RHP) zero. If you suddenly demand more output voltage by increasing the duty cycle, the output voltage will momentarily drop before rising, because the inductor is disconnected from the output for a longer period. You cannot mathematically force a boost converter past its physical parasitic limits.
Can I use a Buck-Boost for a simple 12V to 5V conversion?
Technically yes, but practically it is a poor choice. A traditional inverting buck-boost inverts the output polarity (-5V). A non-inverting 4-switch buck-boost (like the TI TPS63020) will work, but it requires two inductors or a coupled inductor, four switching MOSFETs, and suffers from higher switching losses than a simple buck. Analog Devices and TI design guides universally recommend using a buck-boost topology only when the input voltage rail will cross over the output voltage during operation (e.g., a 3.7V Li-ion battery draining down to 3.0V while powering a 3.3V logic rail).
Why does my SEPIC converter output drop under heavy load?
SEPIC converters require a coupling capacitor (C_s) in series with the second inductor. This capacitor must carry the full RMS switch current. If you use a standard electrolytic or low-grade ceramic capacitor with high Equivalent Series Resistance (ESR), the I²R losses will cause severe voltage droop and thermal failure. Always use low-ESR X7R or X5R MLCCs for the SEPIC coupling capacitor.
