A boost converter is a DC-DC switching regulator that steps up voltage while proportionally stepping down current. If your baseline query is converting a standard 5V USB supply to 12V at a 1A output load, a boost converter operating at 88% efficiency will draw 2.72A from the 5V source, require a 58.3% duty cycle, and need an inductor with a saturation current rating above 3.5A. Unlike linear regulators that burn excess voltage as heat, a boost converter transfers energy in discrete packets via an inductor, making it the definitive choice for battery-powered step-up applications.
The Core Conversion Math (Formulas & Substitutions)
To size your components, you need two foundational equations. The first dictates the switch duty cycle ($D$), and the second calculates the required input current ($I_{in}$). These formulas assume Continuous Conduction Mode (CCM), where the inductor current never drops to zero during the switching cycle.
Duty Cycle: $D = 1 - (V_{in} / V_{out})$
Input Current: $I_{in} = (V_{out} \times I_{out}) / (V_{in} \times \eta)$
Where $\eta$ is decimal efficiency (e.g., 0.88 for 88%).
Substituting our 5V to 12V, 1A reference values:
- Duty Cycle: $D = 1 - (5 / 12) = 1 - 0.4167 = 0.5833 \text{ (58.3%)}$
- Input Current: $I_{in} = (12 \times 1) / (5 \times 0.88) = 12 / 4.4 = 2.72A$
This means your 5V source (like a USB power bank) must be capable of sustaining nearly 3A without sagging, or the converter will trigger undervoltage lockout (UVLO) and shut down.
Neighboring Output Targets (±20% Range)
Designs rarely hit exact nominal voltages on the first prototype. Here is how the duty cycle and input current shift if your target output drifts within a ±20% band around 12V, assuming a constant 1A load and 88% efficiency from a 5V source.
| Target Vout | Variance | Duty Cycle (D) | Input Current (Iin) | Inductor Sat. Rating (Min) |
|---|---|---|---|---|
| 9.6V | -20% | 47.9% | 2.18A | 2.9A |
| 10.8V | -10% | 53.7% | 2.45A | 3.2A |
| 12.0V | Baseline | 58.3% | 2.72A | 3.5A |
| 13.2V | +10% | 62.1% | 3.00A | 3.9A |
| 14.4V | +20% | 65.3% | 3.27A | 4.2A |
Note: Inductor saturation ratings assume a 30% peak-to-peak ripple current added to the average input current.
What Fixes the Answer (and When It Breaks)
The math above is fixed by three core assumptions: Continuous Conduction Mode (CCM), 88% fixed efficiency, and a pure DC input. If any of these break, the standard formulas become meaningless.
When the Conversion is Meaningless
- Duty Cycle > 90%: If you try to boost 3.3V to 36V, the math demands an 90.8% duty cycle. In reality, parasitic resistances in the inductor DCR, the MOSFET $R_{DS(on)}$, and the PCB traces consume the voltage gain. The output will stall around 15V-20V. For ratios exceeding 1:6, you need a coupled-inductor (flyback) or a multi-stage topology.
- $V_{in} > V_{out}$: A standard boost converter cannot step down voltage. If you feed 12V into a boost converter set for 5V, the internal diode will simply forward-bias, passing $V_{in}$ minus a 0.3V-0.7V diode drop directly to the output. You need a buck or SEPIC converter instead.
- 120V vs 230V AC Mains: A boost converter is strictly a DC-DC topology. If your source is 120V AC or 230V AC, this math is entirely invalid. You cannot feed AC into a boost inductor. You must first rectify to DC (yielding ~170V DC or ~325V DC). If your goal is to step up rectified 120V AC to 400V DC (common in Active Power Factor Correction front-ends), the duty cycle is no longer static; it modulates continuously across the AC sine wave to shape the input current. For isolated mains step-up, an isolated flyback topology is required.
Quick FAQ: Discontinuous Conduction Mode (DCM)
Q: What happens to the math at very light loads?
A: At light loads (e.g., a 10mA standby circuit), the inductor fully discharges before the next switching cycle begins. This is Discontinuous Conduction Mode (DCM). In DCM, the duty cycle is no longer strictly tied to the voltage ratio; it becomes dependent on the load current, inductance value, and switching frequency. Modern ICs handle this automatically via Pulse Frequency Modulation (PFM) or Burst Mode to maintain high efficiency at light loads.
Decision Tree: Picking Your Boost IC
Don't waste time designing a controller from discrete op-amps and comparators. Use this decision tree to select a proven, integrated boost IC based on your input voltage and output current requirements.
| Condition / Requirement | Terminating Pick (Part Number) | Why This Part Wins |
|---|---|---|
| IF $V_{in}$ = 2.7V-5.5V, $V_{out}$ = 12V, $I_{out}$ < 2A, and budget is < $1.00 | MT3608 (Generic/Various) | Ubiquitous hobbyist module. Cheap, but requires manual trimming via onboard potentiometer and has mediocre EMI. |
| IF $V_{in}$ = 2.3V-5.5V, $V_{out}$ = 12V, $I_{out}$ < 2.5A, and you need >90% efficiency | TPS61230 (Texas Instruments) | Highly integrated synchronous boost (no external Schottky diode needed). Excellent for single-cell Li-Ion to 12V IoT sensors. |
| IF $V_{in}$ = 5V-12V, $V_{out}$ = 24V, $I_{out}$ > 3A (High Power) | TPS55340 (Texas Instruments) | External MOSFET controller. Allows you to pair it with a low-$R_{DS(on)}$ FET and massive inductor for 50W+ applications. |
| IF $V_{in}$ = 4V-38V (Automotive/Industrial), $V_{out}$ = 48V, $I_{out}$ > 5A | LTC3862 (Analog Devices) | Multi-phase current-mode controller. Interleaves phases to drastically reduce input/output ripple and thermal hotspots. |
Real-World Edge Cases & Component Sizing
Selecting the IC is only 20% of the battle. The remaining 80% is preventing the passive components from failing under real-world parasitics. According to Texas Instruments application note SLVA372, the most common point of failure in boost designs is inductor saturation.
1. Inductor Saturation Current ($I_{sat}$):
The average input current is 2.72A, but the current through the inductor is a triangle wave. You must calculate the peak current: $I_{peak} = I_{in} + (\Delta I_L / 2)$. If your ripple is 30%, $I_{peak}$ hits roughly 3.13A. If you select an inductor rated for 3.0A saturation, the core will saturate, inductance will drop to near-zero, and the internal MOSFET will instantly vaporize from uncontrolled current spikes. Always pick an inductor with an $I_{sat}$ rating at least 20% above your calculated $I_{peak}$.
2. Output Capacitor ESR and Ripple:
Unlike buck converters, a boost converter delivers energy to the output capacitor only when the switch is off. This creates massive RMS ripple currents in the output capacitor. Standard aluminum electrolytic capacitors will overheat and vent. You must use low-ESR ceramic (X7R/X5R) or specialized polymer capacitors. For a 12V/1A output, a bank of three 22µF, 25V X7R MLCCs in parallel is the minimum viable configuration to keep output voltage ripple under 50mV.
3. The Right-Rail-to-Switch Node (SW Node) Layout:
The node connecting the inductor, the switch, and the diode is a high-$dv/dt$ antenna. Keep this copper area as physically small as possible to minimize radiated EMI, but wide enough to handle the peak current without excessive trace heating. A 20-mil trace width is generally sufficient for 3A peak currents on 1oz copper, but always verify with a manufacturer's reference layout before spinning your own PCB.
