Inductor Ripple and RMS Current Calculator (Buck, Boost, Buck-Boost)
Everything about a power inductor hangs on three currents: the average it carries, the peak the core must survive without saturating, and the RMS that heats the winding. All three follow from volt-seconds once the topology fixes the duty cycle. This page computes them exactly for the three hard-switched non-isolated converters, and tells you the inductance below which the CCM picture stops applying at all.
Choosing something to solve for makes Ripple current, peak to peak an editable target and computes the chosen input from it.
Formula
| D | duty cycle of the main switch |
| dI | peak-to-peak inductor ripple current (A) |
| Iavg | average (DC) inductor current (A) |
| f, L | switching frequency (Hz) and inductance (H) |
What this model assumes, and where it stops
Assumptions
- Continuous conduction: the inductor current never reaches zero. The page refuses and reports the critical inductance otherwise.
- Ideal lossless conversion sets the duty cycle - no diode drop, no switch or winding resistance. Real duty runs a few percent higher.
- Constant inductance over the ripple excursion. A core near saturation has less inductance at the peak than at the valley, so real ripple is worse than calculated.
- Steady state at a fixed operating point - no load steps, no current-limit events.
Limitations
- DCM is refused, not modelled. Below the critical inductance the duty cycle, ripple and RMS all follow different formulas, and light-load operation of any real converter passes through that regime.
- The RMS here is the inductor current. Switch and diode each carry only a slice of it; their RMS values are different and lower, and this page does not split them.
- Losses feed back into duty cycle. At low output voltage the diode drop alone can move D by several percent, taking the ripple with it.
- Coupled inductors, forced PWM at light load, and current-mode slope compensation all modify the picture in ways a single-inductor volt-second balance cannot capture.
When you need a 3D field solution instead
Closed-form models like the one above hold on idealised geometry. These are the cases where they stop being good enough and a full 3D electromagnetic and thermal solution is the only way to get a trustworthy answer:
- Core saturation margin at the current peak - flux crowding and local saturation are geometry effects a lumped L cannot see.
- Winding loss at the ripple frequency: the AC portion of this waveform sees skin and proximity effect, and fringing flux near a gap drives local losses far above the DC estimate.
- Thermal coupling between core and winding losses in a compact magnetic, where each heats the other and both properties are temperature dependent.
- Inductance roll-off with current in powder cores, where L at the peak differs enough from L at the valley to reshape the waveform.
Common questions
What ripple current should I design for?
The conventional target is 20 to 40% of the average inductor current. Less ripple means a physically larger inductor and slower transient response; more ripple raises the peak toward saturation, the core loss, and the output capacitor burden. The default case here runs 45% - workable, but at the point where the peak (23% above average) starts driving the core size.
Does ripple actually matter for winding loss?
Far less than intuition suggests. The RMS of a triangular waveform is sqrt(Iavg^2 + dI^2/12), so even a 40% ripple raises the RMS - and the I2R heating - only 0.7% above the pure DC value. What the ripple does drive is core loss, which sees the full flux swing, and the AC portion of the winding loss through skin and proximity effect at the switching frequency. The peak, not the RMS, sizes the core.
What happens below the critical inductance?
The valley current reaches zero before the switch turns back on, the converter enters discontinuous conduction, and every CCM formula on this page stops applying: the duty cycle becomes load-dependent, the peak current rises faster, and the control transfer function changes character. This page refuses rather than quietly extrapolating - light-load operation of every real converter passes through DCM, and it needs its own analysis.
References
- Erickson & Maksimovic - Fundamentals of Power Electronics, 3rd ed. - steady-state converter analysis
- TI SLVA477 - Basic calculation of a buck converter power stage
- TI SLVA372 - Basic calculation of a boost converter power stage