To drive a 12V, 3A load from a 24V DC source using a constant current (CC) buck converter (assuming 90% DC-DC efficiency), the required input current is 1.67A and the switching duty cycle is 50%. If your 24V source is fed by a 120V AC-DC power supply (assuming 85% AC-DC efficiency and a 0.9 Power Factor), the AC wall draw shifts to 0.39A. These baseline numbers anchor your wire sizing and upstream breaker selection.
The Core Conversion: DC Input Current and Duty Cycle
A constant current buck converter steps down voltage while strictly regulating output current, making it the standard topology for driving high-power LEDs, battery charging stages, and laser diodes. Unlike a constant voltage (CV) buck where output current varies with the load, a CC buck adjusts its output voltage dynamically to maintain a fixed amperage.
The two critical conversions for DC-DC stage design are Input Current ($I_{in}$) and Duty Cycle ($D$). Here are the formulas with our baseline values (24V in, 12V out, 3A constant current, 90% efficiency) substituted:
Duty Cycle ($D$):
$D = \frac{V_{out}}{V_{in}} = \frac{12V}{24V} = 0.50$ (50%)
Input Current ($I_{in}$):
$I_{in} = \frac{V_{out} \times I_{out}}{V_{in} \times \eta} = \frac{12V \times 3A}{24V \times 0.90} = \frac{36W}{21.6V} = 1.67A$
What fixes this answer? The assumption that locks these numbers in place is the converter efficiency ($\eta$) and the fixed input voltage. If your input voltage sags under load (common with unregulated AC-DC bricks), the duty cycle must increase, and the input current will spike to maintain the 36W output.
Neighboring Values: ±20% Output Current Range
In practical bench testing, LED forward voltages drift with temperature, and you may need to dim or boost the string. Here is how the input current scales if you adjust the constant current setpoint by ±20% while maintaining a 12V string voltage and 24V input.
| Output Current ($I_{out}$) | Output Power ($P_{out}$) | Input Current @ 24V ($I_{in}$) | Duty Cycle ($D$) |
|---|---|---|---|
| 2.4A (-20%) | 28.8W | 1.33A | 50% |
| 2.7A (-10%) | 32.4W | 1.50A | 50% |
| 3.0A (Base) | 36.0W | 1.67A | 50% |
| 3.3A (+10%) | 39.6W | 1.83A | 50% |
| 3.6A (+20%) | 43.2W | 2.00A | 50% |
Real-World Component Selection: CC Buck ICs and Modules
Theoretical math only gets you to the schematic. On the bench, you need a silicon switcher that can handle the inductor saturation current without thermal throttling. Below is a data-dense comparison of common constant current buck controllers and hobbyist modules available in 2026.
| IC / Module | Topology | $V_{in}$ Range | Max $I_{out}$ | Dimming Support | Best Application |
|---|---|---|---|---|---|
| XL4015 (Generic Module) | Async Buck | 5V - 36V | 5A | Manual Potentiometer | Hobbyist LED strips, bench prototyping |
| LM3409 (TI) | Hysteretic (PFM) | 6V - 42V | 1.5A (Ext FET) | PWM / Analog | Precision architectural lighting |
| TPS92515 (TI) | Sync Buck | 4.2V - 65V | 2.5A | PWM / Analog | High-efficiency automotive / PoE |
| PT4121 (Generic Module) | Async Buck | 6V - 30V | 1.2A | PWM only | Low-cost MR16 bulb retrofits |
Bench Note: When using cheap XL4015 modules for constant current, the onboard current-sense shunt resistor is often rated for only 2W. At 5A, it dissipates over 3W and will desolder itself or drift wildly in resistance. Always swap the shunt for a 5W Kelvin-sense resistor if pushing past 3A continuously.
AC Mains Shift: 120V vs 230V and the Power Factor Trap
A buck converter is strictly a DC-DC stage. However, if your 24V DC source is an AC-DC power supply (like a Mean Well LRS-35-24) plugged into the wall, you must calculate the AC input current to size your branch circuit wiring and breakers. This is where regional voltages and Power Factor (PF) drastically shift the numbers.
Assuming our 36W DC output, an 85% efficient AC-DC front end, and a 0.9 PF, the AC input power required is $\frac{36W}{0.85} = 42.35W$. The apparent power (VA) is $\frac{42.35W}{0.9} = 47.05VA$.
- 120V AC (North America): $I_{ac} = \frac{47.05VA}{120V} = \mathbf{0.39A}$
- 230V AC (EU/UK/AU): $I_{ac} = \frac{47.05VA}{230V} = \mathbf{0.20A}$
- 3-Phase (Industrial Scaling): For a single 3A LED driver, 3-phase is irrelevant. But if you scale this to a 15kW lighting array (400+ drivers), you distribute the load across three phases. The per-phase current drops to roughly $\frac{1}{3}$ of the single-phase equivalent, balancing the neutral and allowing smaller feeder conductors.
When is this conversion meaningless?
If your AC-DC power supply lacks Active Power Factor Correction (APFC) and the Power Factor is unknown, converting output DC watts to AC input amps using standard formulas is meaningless for breaker sizing. Cheap non-PFC supplies draw current in sharp, high-amplitude spikes near the peak of the AC sine wave. This high harmonic distortion means the RMS current is much higher than the real power suggests. In this scenario, you must ignore the math and measure the apparent power (VA) directly with a true-RMS clamp meter, sizing your upstream breaker and wire gauge on the VA rating, not the Wattage.
FAQ: Edge Cases and Thermal Limits
What happens if the LED string voltage exceeds the input voltage?
A buck converter can only step voltage down. If your constant current setpoint pushes the LED forward voltage ($V_f$) to 25V, but your input is only 24V, the converter will drop out of regulation. It will pass the input voltage directly to the load (minus the dropout voltage of the switch and inductor, usually ~1.5V), and the current will fall below your 3A target. Always ensure $V_{in} > V_{out(max)} + 2V$.
How do I calculate the inductor value for a CC buck?
Unlike CV bucks where you target a specific ripple current percentage (usually 30% of $I_{out}$), CC bucks often use hysteretic control or fixed on-times. If using a standard PWM controller like the TPS92515, you calculate the inductor based on the maximum allowable ripple current to prevent LED flicker and acoustic noise. A 22µH to 47µH shielded ferrite inductor is standard for 1A-3A LED applications at 500kHz switching frequencies.
Can I put a capacitor across the output of a CC buck converter?
Generally, no. Placing a large electrolytic capacitor directly across the output of a constant current buck converter can cause severe startup overshoot and control loop instability. The converter will dump maximum current into the empty capacitor, spiking the voltage until the feedback loop catches up, which can instantly blow your LED string. If you need to filter high-frequency PWM dimming noise, use a small ceramic capacitor (100nF - 1µF) or an LC pi-filter specifically designed for the dimming frequency.
