Flyback Energy Storage, Magnetizing Inductance, and Peak Current — Chapter 5

Chapter 4 established the electrical, control, safety, thermal, mechanical, and manufacturing requirements. Chapter 5 now converts the confirmed power, timing, and operating-mode requirements into three tightly coupled magnetic targets: energy per switching cycle, primary-referred magnetizing inductance, and peak primary current.

By the end of this chapter, you should be able to calculate flyback energy per cycle; derive first-pass magnetizing-inductance and peak-current targets for DCM, BCM, and CCM; distinguish peak, average, and RMS current; evaluate tolerance and current-limit margin; and prepare the current-and-energy design brief needed for Chapter 6.

In This Chapter — Click to Expand

DESIGN ASSUMPTION — PRIMARY-REFERRED FIRST-PASS MODEL

Unless stated otherwise, the equations in this chapter refer magnetizing inductance, current, and stored energy to the primary winding. Initial derivations neglect leakage inductance, clamp action, switching transitions, semiconductor drops, controller propagation delay, and nonlinear core behavior unless those effects are being discussed explicitly.

The calculations produce design targets—not a released transformer. Chapter 6 will establish reflected voltage and turns ratio; later chapters will convert the targets into primary turns, core, gap, conductors, winding construction, losses, thermal performance, and validated hardware.

1. Energy, Magnetizing Inductance, and Peak Current Form One Design Problem

A flyback converter transfers a controlled amount of magnetic energy during each switching cycle. The transformer must store that energy during the primary interval and release the required increment during the secondary interval. The amount of energy, the inductance that shapes the current ramp, and the peak current reached during the cycle therefore cannot be selected independently.

QuantityWhat It ControlsWhat Constrains It
Energy per cycleThe magnetic energy that must be processed during each active switching cycle.Input power, efficiency, switching frequency, operating mode, burst behavior, and transient demand.
Magnetizing inductance, LmThe slope and ripple of primary magnetizing current for a given applied voltage and time.Operating mode, controller timing, peak-current target, core/turns/gap realization, and tolerance.
Peak primary current, IPKMaximum normal magnetizing current and a principal input to saturation, current-limit, conductor, and semiconductor checks.Power, duty cycle, inductance, frequency, control law, tolerances, and transient margin.
RMS currentConduction heating in the primary winding, switch, sense path, and related conductors.Waveform shape, duty cycle, minimum and maximum current, and temperature-dependent resistance.

A design can meet the output-power target and still be unsuitable if its peak current exceeds the controller limit, its RMS current produces excessive winding loss, or its inductance cannot be realized with a practical turns-and-gap combination.

SolidMag Engineering Insight

The Three Targets Must Be Solved Together

Energy per cycle determines how much magnetic energy must be processed. Magnetizing inductance determines how quickly current changes. Peak current determines the maximum electrical and magnetic stress.

Changing any one of the three changes the others. A successful design selects a compatible set that satisfies operating mode, current limit, flux, loss, thermal, winding, control, and manufacturing constraints.

2. Magnetic Energy Stored in the Magnetizing Inductance

For a linear inductance carrying instantaneous magnetizing current I, the stored magnetic energy is:

E=12LmI2E=\frac{1}{2}L_m I^2

At the maximum primary current:

EPK=12LmIPK2E_{PK}=\frac{1}{2}L_m I_{PK}^2

This quadratic relationship is fundamental. If magnetizing inductance remains unchanged, doubling peak current increases stored energy by a factor of four. Peak-current margin therefore has a strong effect on core, gap, switch, winding, and current-sense requirements.

Incremental Energy in CCM

In CCM, the magnetizing current does not return to zero. The transformer begins the cycle with residual current IMIN and stored energy. The energy transferred during one cycle is the change between the maximum and minimum stored-energy states:

ΔE=12Lm(IMAX2−IMIN2)\Delta E=\frac{1}{2}L_m\left(I_{MAX}^2-I_{MIN}^2\right)

Using the average of the minimum and maximum primary currents:

IAVG=IMAX+IMIN2I_{AVG}=\frac{I_{MAX}+I_{MIN}}{2}

and the current excursion:

ΔIP=IMAX−IMIN\Delta I_P=I_{MAX}-I_{MIN}

the incremental energy can also be written as:

ΔE=LmIAVGΔIP\Delta E=L_m I_{AVG}\Delta I_P

The peak stored energy in CCM is larger than the energy transferred during the cycle because residual energy remains when current reaches IMIN. That residual energy creates magnetic bias and must be included in peak-flux and saturation checks.

Flyback transformer energy diagram showing the relationship between magnetizing inductance, peak primary current, and stored energy, with energy increasing as current squared.

Figure 5-1. Stored magnetic energy increases with the square of current; in CCM, the transferred energy is the difference between the maximum and minimum stored-energy states.

ENGINEERING CAUTION — LINEAR INDUCTANCE IS AN APPROXIMATION

The expression one-half LmIPK2 assumes that magnetizing inductance remains sufficiently linear over the evaluated current range. As a core approaches saturation, incremental inductance can fall and the actual current rise can become much steeper than the nominal calculation predicts.

Use the selected core, gap, turns, temperature, and material data to verify the actual bias-dependent inductance and flux margin.

3. Energy Per Switching Cycle and Power

The average power processed by the flyback stage is the switching frequency multiplied by the average energy transferred during each cycle:

PIN≈fsΔEP_{IN}\approx f_s\,\overline{\Delta E}

For fixed-frequency DCM or BCM operation in which current begins at approximately zero and the stored magnetizing energy is fully transferred before the next cycle:

PIN≈12LmIPK2fsP_{IN}\approx\frac{1}{2}L_m I_{PK}^2f_s

Including estimated converter efficiency:

POUT≈η(12LmIPK2fs)P_{OUT}\approx\eta\left(\frac{1}{2}L_m I_{PK}^2f_s\right)

For CCM, use the incremental energy rather than total peak energy:

PIN≈12Lm(IMAX2−IMIN2)fsP_{IN}\approx\frac{1}{2}L_m\left(I_{MAX}^2-I_{MIN}^2\right)f_s
Operating ConditionEnergy Used for Power CalculationImportant Qualification
DCMPeak energy above approximately zero current: ½LmIPK2.The transformer must fully demagnetize before the next primary pulse.
BCM / CrMPeak energy above approximately zero current, with the next cycle beginning near the zero-current boundary.Switching frequency commonly varies with line and load.
CCMIncremental energy between IMIN and IMAX.Residual current and magnetic bias remain from one cycle to the next.
Burst / skip operationAverage energy of active pulses multiplied by their effective repetition rate.Peak energy per active pulse may remain high even when average output power is low.

Variable-Frequency and Burst Operation

When switching frequency or pulse density varies, calculate power from the actual active-cycle energy and effective repetition rate. A nominal frequency alone may understate the energy per active pulse during startup, burst, overload, valley skipping, or boundary-mode operation.

SolidMag Engineering Insight

Power Is an Average of Switching-Cycle Energy

Output power does not directly specify transformer size. The magnetic component responds to energy per active cycle, current waveform, frequency, flux, losses, and temperature.

Two converters delivering the same average power can require very different transformers if their switching frequencies, operating modes, current limits, or burst strategies differ.

4. Primary Current Ramp and Magnetizing Inductance

During the primary switch ON interval, the applied primary voltage produces a magnetizing-current slope:

dIPdt=VP,ONLm\frac{dI_P}{dt}=\frac{V_{P,ON}}{L_m}

For an approximately constant primary voltage applied for ON time tON:

ΔIP=VP,ONtONLm\Delta I_P=\frac{V_{P,ON}t_{ON}}{L_m}

Using duty cycle and switching frequency:

tON=Dfst_{ON}=\frac{D}{f_s}
ΔIP=VP,ONDLmfs\Delta I_P=\frac{V_{P,ON}D}{L_m f_s}

The voltage used in the current-ramp equation should be the voltage actually applied across the primary winding. A detailed design may account for MOSFET, current-sense, bridge, filter, and wiring drops where they are significant.

ChangePrimary Current-Ramp Effect When Other Variables Are Fixed
Increase VP,ONIncreases current slope and current excursion.
Increase duty cycle, DIncreases ON time and current excursion.
Decrease switching frequency, fsIncreases ON time and current excursion.
Decrease magnetizing inductance, LmIncreases current slope and peak-current rise.
Increase magnetizing inductance, LmReduces ripple but can create residual CCM current and a different gap/turns requirement.

The current ramp is an electrical relationship. The selected core, turns, and gap must later realize the target inductance while maintaining acceptable flux density, energy storage, winding fit, loss, and tolerance.

5. Selecting Magnetizing Inductance for DCM

In idealized DCM, primary magnetizing current begins each cycle at approximately zero. Therefore:

IPK≈ΔIPI_{PK}\approx\Delta I_P

Using the current-ramp relationship:

IPK≈VP,ONDLmfsI_{PK}\approx\frac{V_{P,ON}D}{L_m f_s}

Using the DCM energy-per-cycle power relationship:

IPK≈2PINLmfsI_{PK}\approx\sqrt{\frac{2P_{IN}}{L_m f_s}}

Solving directly for magnetizing inductance when input power, applied primary voltage, duty cycle, and frequency are known:

Lm≈VP,ON2D22PINfsL_m\approx\frac{V_{P,ON}^2D^2}{2P_{IN}f_s}

The same assumptions also produce a useful current relationship:

IPK≈2PINVP,ONDI_{PK}\approx\frac{2P_{IN}}{V_{P,ON}D}

These equations are internally consistent only when they refer to the same operating point and idealized current-starts-at-zero waveform.

Primary RMS Current in DCM

For a triangular primary current that rises from zero to IPK during duty cycle D and is zero during the remainder of the switching period:

IP,RMS=IPKD3I_{P,RMS}=I_{PK}\sqrt{\frac{D}{3}}

ENGINEERING CAUTION — CURRENT STARTING AT ZERO DOES NOT ALONE PROVE DCM

The primary-side energy and current equations can describe a current pulse beginning at zero, but true DCM also requires the secondary demagnetization interval to finish before the next cycle. Reflected voltage and turns ratio determine that timing.

Chapter 6 verifies the secondary conduction fraction and zero-current interval. Do not finalize operating mode from primary current alone.

6. Magnetizing Inductance at the Boundary of Conduction

Boundary conduction mode occurs when magnetizing current reaches approximately zero at the instant the next switching cycle begins. The primary pulse still begins near zero, so the same peak-energy relationship applies, but the secondary demagnetization interval consumes essentially all of the available OFF time.

At a specified operating point, a primary-side current-starts-at-zero inductance candidate can be calculated as:

Lm,0≈VP,ON2D22PINfsL_{m,0}\approx\frac{V_{P,ON}^2D^2}{2P_{IN}f_s}

Calling this value the true BCM inductance requires an additional timing condition:

D+DS≈1D+D_S\approx1

where DS is the secondary conduction fraction. Chapter 6 will show that DS depends on reflected voltage and the actual winding ratio.

Variable-Frequency Boundary Control

Many BCM or critical-mode controllers initiate the next cycle when demagnetization is detected. The switching frequency then varies with line voltage, load, magnetizing inductance, and reflected voltage. The design must be checked at the minimum and maximum actual switching frequencies, not only at one nominal value.

SolidMag Engineering Insight

Boundary Operation Is an Electrical and Timing Condition

A calculated inductance can produce a current pulse that starts at zero, but BCM is established only when the secondary current also reaches zero at the cycle boundary.

Magnetizing inductance, duty cycle, reflected voltage, load, and switching frequency must all agree. Chapter 6 closes that timing loop.

7. Selecting Magnetizing Inductance for CCM

In CCM, primary magnetizing current begins the ON interval at nonzero current IMIN and rises to IMAX. The ON-interval average primary current required for idealized input power is:

IAVG,ON≈PINVP,ONDI_{AVG,ON}\approx\frac{P_{IN}}{V_{P,ON}D}

The current excursion remains:

ΔIP=VP,ONDLmfs\Delta I_P=\frac{V_{P,ON}D}{L_m f_s}

For a linear triangular ramp:

IMAX=IAVG,ON+ΔIP2I_{MAX}=I_{AVG,ON}+\frac{\Delta I_P}{2}
IMIN=IAVG,ON−ΔIP2I_{MIN}=I_{AVG,ON}-\frac{\Delta I_P}{2}

CCM requires:

IMIN>0I_{MIN}>0

The primary RMS current over the complete switching period is:

IP,RMS=D3(IMIN2+IMINIMAX+IMAX2)I_{P,RMS}=\sqrt{\frac{D}{3}\left(I_{MIN}^2+I_{MIN}I_{MAX}+I_{MAX}^2\right)}

The transferred energy increment is:

ΔE=12Lm(IMAX2−IMIN2)\Delta E=\frac{1}{2}L_m\left(I_{MAX}^2-I_{MIN}^2\right)

while the residual energy at minimum current is:

EMIN=12LmIMIN2E_{MIN}=\frac{1}{2}L_m I_{MIN}^2
Comparison of DCM and CCM flyback operation showing primary current, stored magnetic energy, current reset behavior, peak current, ripple, and residual energy between switching cycles.

Figure 5-2. DCM transfers energy from approximately zero current, while CCM transfers the incremental energy between nonzero minimum and maximum current levels.

ENGINEERING CAUTION — DO NOT APPLY THE DCM POWER EQUATION TO CCM PEAK CURRENT

In CCM, one-half LmIMAX2 is the total peak stored energy, not the energy transferred during that cycle. Using the DCM expression with IMAX can overstate transferred power because residual energy remains at IMIN.

Use the incremental energy or the actual current waveform and control law.

8. How Magnetizing Inductance Changes the Design

Lower Magnetizing Inductance Tends TowardHigher Magnetizing Inductance Tends Toward
Steeper primary current rampShallower primary current ramp
Higher peak current for the same ON timeLower ripple and potentially lower peak current
Higher primary and switch conduction stressGreater residual current and magnetic bias in CCM
Potentially smaller inductance targetMore demanding turns-and-gap realization for the selected core
Greater sensitivity to propagation delay and current-limit overshootGreater sensitivity to mode transition and demagnetization timing
Potential DCM operationPotential CCM operation
Potentially greater copper and clamp stressPotentially greater peak stored energy and bias flux

These are design tendencies, not universal outcomes. The controller can change duty cycle, frequency, and current limit as line and load vary. The final inductance target should be selected through an iterative process.

Practical Selection Sequence

  1. Choose the operating mode or allowable mode range from Chapter 3.
  2. Use Chapter 4 requirements to identify input power, voltage, duty-cycle, frequency, and current-limit bounds.
  3. Calculate a first-pass Lm and current waveform at each important operating corner.
  4. Check peak and RMS current against controller, MOSFET, winding, and current-sense limits.
  5. Check whether primary turns, core, and gap can realize the inductance with acceptable flux and energy storage.
  6. Iterate inductance, duty cycle, frequency, reflected voltage, core, turns, and current limit until all constraints agree.

SolidMag Engineering Insight

The Best Magnetizing Inductance Is Not the Largest or Smallest Value

Magnetizing inductance is a system optimization variable. A lower value may increase peak and RMS current; a higher value may increase residual current, magnetic bias, and the difficulty of realizing the target with practical turns and gap.

The correct value is the one that produces an acceptable current waveform, energy transfer, flux, loss, control behavior, winding construction, and tolerance margin across the complete operating envelope.

9. Peak Current, Current Sense, and Control Margin

The calculated normal operating peak current must remain below the minimum effective controller current limit after tolerances, propagation delay, blanking, slope compensation, and current-sense errors are considered.

Current-Sense Threshold

For a simple primary current-sense resistor:

ILIM≈VCS,LIMRCSI_{LIM}\approx\frac{V_{CS,LIM}}{R_{CS}}

A worst-case analysis should use the minimum current-limit threshold and maximum effective sense resistance when determining the minimum available limit:

ILIM,MIN≈VCS,LIM,MINRCS,MAXI_{LIM,MIN}\approx\frac{V_{CS,LIM,MIN}}{R_{CS,MAX}}

Propagation-Delay Overshoot

Current continues rising during the time between threshold crossing and actual MOSFET turn-off. A first-order overshoot estimate is:

ΔIDELAY≈VP,ONtDELAYLm\Delta I_{DELAY}\approx\frac{V_{P,ON}t_{DELAY}}{L_m}

The estimated maximum switched current is then:

ISW,PK≈ISENSE,TRIP+ΔIDELAYI_{SW,PK}\approx I_{SENSE,TRIP}+\Delta I_{DELAY}

Current-Limit Margin

A useful margin definition is:

MI=ILIM,MIN−IPK,MAXIPK,MAX×100%M_I=\frac{I_{LIM,MIN}-I_{PK,MAX}}{I_{PK,MAX}}\times100\%

The required margin depends on controller behavior, fault strategy, core saturation characteristic, temperature, transients, startup, and product reliability targets. There is no universal percentage suitable for every design.

ENGINEERING CAUTION — CURRENT LIMIT IS NOT A NORMAL OPERATING TARGET

A transformer should not operate continuously at the controller’s minimum current-limit threshold. Current-limit operation can include propagation-delay overshoot, reduced effective inductance, input transients, and thermal stress.

Normal peak current, transient peak current, protection current, and hard saturation should be treated as separate limits.

10. Peak, Average, and RMS Current Serve Different Checks

Current QuantityPrimary Engineering UseDo Not Substitute It For
Peak currentCurrent limit, switch peak rating, peak ampere-turns, flux, saturation, and gap-energy checks.RMS heating or average input current.
Average ON currentInput-power and control calculations during the primary conduction interval.Peak stress or full-period RMS current.
Full-period RMS currentPrimary winding, MOSFET conduction, sense resistor, lead, and PCB heating.Peak flux or current-limit checks.
Minimum current in CCMResidual stored energy, magnetic bias, mode identification, and current ripple.Zero-current assumption used in DCM.

A common error is to size primary wire from average current or to check saturation from RMS current. Each current metric answers a different engineering question.

Current Density Is Only a First Pass

After primary RMS current is known, a preliminary copper area may be estimated from an allowable current density:

ACU≈IP,RMSJALLOWA_{CU}\approx\frac{I_{P,RMS}}{J_{ALLOW}}

The final conductor must also be checked for skin effect, proximity effect, number of layers, termination resistance, insulation, window fill, and temperature. Chapter 10 of the guide will develop conductor selection in detail.

SolidMag Engineering Insight

Peak Current Protects the Magnetic Design; RMS Current Protects the Thermal Design

Peak current establishes the maximum magnetic and semiconductor stress. RMS current establishes resistive heating. Average current helps describe power flow and control.

A complete current specification should report all three rather than presenting one current rating as though it serves every design check.

11. Tolerance and Operating-Corner Analysis

The nominal current waveform is not the worst-case waveform. Magnetizing inductance, switching frequency, duty cycle, primary voltage, efficiency, current-limit threshold, and controller delay all vary.

For a DCM current ramp beginning near zero, a conservative first-pass operating peak can be expressed as:

IPK,MAX≈VP,ON,MAXDMAXLm,MINfs,MINI_{PK,MAX}\approx\frac{V_{P,ON,MAX}D_{MAX}}{L_{m,MIN}f_{s,MIN}}
VariationTypical Effect on Current or EnergyRequired Check
Lower LmSteeper current ramp and higher peak current.Use minimum production inductance including gap and assembly tolerance.
Lower fsLonger ON time at the same duty cycle.Use actual controller minimum frequency at the evaluated condition.
Higher applied primary voltageSteeper current slope.Use the actual voltage/duty combination, not independent maxima that cannot occur together.
Higher duty cycleLonger current-ramp interval.Check controller limits, startup, transient, and fault modes.
Lower efficiencyHigher required input power for the same output.Use the minimum required efficiency at the relevant operating corner.
Current-limit tolerance and delayRaises maximum switched current above the nominal trip point.Use minimum threshold, maximum sense resistance, and maximum delay.
Temperature and core nonlinearityCan reduce effective inductance or saturation margin.Verify current and flux using temperature-appropriate material data.
Flyback transformer tolerance-stack diagram showing how input voltage, magnetizing inductance, current-sense, timing, and temperature tolerances increase worst-case peak primary current toward the current-limit threshold.

Figure 5-3. Minimum inductance, minimum frequency, applied voltage, duty cycle, sensing tolerance, and propagation delay combine to determine worst-case peak current and current-limit margin.

ENGINEERING CAUTION — DO NOT COMBINE IMPOSSIBLE CORNERS

A worst-case calculation should be conservative, but it should also represent a physically possible controller and converter state. Maximum input voltage and maximum duty cycle may not occur together in closed-loop operation.

Create a corner matrix based on the actual control law, startup sequence, fault behavior, and component tolerances rather than blindly combining every independent maximum.

12. Worked Example — From Requirements to Energy, Inductance, and Current

WORKED EXAMPLE — 85–265 VAC TO 12 V / 5 A WITH AUXILIARY OUTPUT

This example continues the educational requirements case from Chapter 4. It uses a provisional minimum rectified bus of 100 V, 100 kHz fixed-frequency operation, 45% primary duty cycle, 88% estimated efficiency, and a current-starts-at-zero primary waveform.

The actual controller, bulk-capacitor ripple, reflected voltage, turns ratio, safety construction, core, gap, and winding are not yet selected. The example therefore establishes current-and-energy targets, not a production transformer.

InputValue
Main output12 V / 5 A = 60 W
Auxiliary output5 V / 0.1 A = 0.5 W
Total output power60.5 W
Estimated efficiency88%
Provisional minimum primary bus100 VDC
Switching frequency100 kHz
Primary duty cycle at the example point0.45
Initial current conditionCurrent begins at approximately zero

Step 1 — Calculate Input Power

PIN=POUTηP_{IN}=\frac{P_{OUT}}{\eta}
PIN=60.5 W0.88=68.75 WP_{IN}=\frac{60.5\ \mathrm{W}}{0.88}=68.75\ \mathrm{W}

Step 2 — Calculate Energy Per Cycle

ECYCLE=PINfsE_{CYCLE}=\frac{P_{IN}}{f_s}
ECYCLE=68.75100000=687.5 μJE_{CYCLE}=\frac{68.75}{100000}=687.5\ \mu\mathrm{J}

Step 3 — Calculate a Current-Starts-at-Zero Inductance Candidate

Lm=VP,ON2D22PINfsL_m=\frac{V_{P,ON}^2D^2}{2P_{IN}f_s}
Lm=1002(0.45)22(68.75)(100000)L_m=\frac{100^2\left(0.45\right)^2}{2\left(68.75\right)\left(100000\right)}
Lm≈147.3 μHL_m\approx147.3\ \mu\mathrm{H}

Step 4 — Calculate Nominal Peak Current

IPK=VP,ONDLmfsI_{PK}=\frac{V_{P,ON}D}{L_mf_s}
IPK=100(0.45)(147.3 μH)(100000)I_{PK}=\frac{100\left(0.45\right)}{\left(147.3\ \mu\mathrm{H}\right)\left(100000\right)}
IPK≈3.06 AI_{PK}\approx3.06\ \mathrm{A}

Step 5 — Verify Stored Energy

EPK=12LmIPK2E_{PK}=\frac{1}{2}L_mI_{PK}^2
EPK≈687.5 μJE_{PK}\approx687.5\ \mu\mathrm{J}

Step 6 — Calculate Primary RMS Current

IP,RMS=IPKD3I_{P,RMS}=I_{PK}\sqrt{\frac{D}{3}}
IP,RMS=3.060.453≈1.18 AI_{P,RMS}=3.06\sqrt{\frac{0.45}{3}}\approx1.18\ \mathrm{A}

Step 7 — Evaluate Inductance and Frequency Tolerance

Assume magnetizing inductance can be 10% below nominal and switching frequency can be 5% below nominal at the evaluated condition:

Lm,MIN=0.90(147.3 μH)=132.5 μHL_{m,MIN}=0.90\left(147.3\ \mu\mathrm{H}\right)=132.5\ \mu\mathrm{H}
fs,MIN=0.95(100 kHz)=95 kHzf_{s,MIN}=0.95\left(100\ \mathrm{kHz}\right)=95\ \mathrm{kHz}
IPK,TOL=100(0.45)(132.5 μH)(95000)≈3.57 AI_{PK,TOL}=\frac{100\left(0.45\right)}{\left(132.5\ \mu\mathrm{H}\right)\left(95000\right)}\approx3.57\ \mathrm{A}

Step 8 — Add Propagation-Delay Current

For an illustrative 100 ns effective turn-off delay:

ΔIDELAY=100(100 ns)132.5 μH≈0.075 A\Delta I_{DELAY}=\frac{100\left(100\ \mathrm{ns}\right)}{132.5\ \mu\mathrm{H}}\approx0.075\ \mathrm{A}
ISW,PK≈3.57+0.075=3.65 AI_{SW,PK}\approx3.57+0.075=3.65\ \mathrm{A}

If the minimum effective controller current limit were 4.00 A, the resulting margin would be:

MI=4.00−3.653.65×100%≈9.6%M_I=\frac{4.00-3.65}{3.65}\times100\%\approx9.6\%

Whether that margin is acceptable depends on the controller, core saturation behavior, startup and fault conditions, current-sense tolerance, and reliability target. The values are illustrative, not recommended limits.

Step 9 — Compare a Higher-Inductance CCM Current Candidate

For comparison, assume a 300 µH magnetizing inductance at the same primary voltage, duty cycle, frequency, and input power. The primary current excursion is:

ΔIP=100(0.45)(300 μH)(100000)=1.50 A\Delta I_P=\frac{100\left(0.45\right)}{\left(300\ \mu\mathrm{H}\right)\left(100000\right)}=1.50\ \mathrm{A}
IAVG,ON=68.75100(0.45)≈1.528 AI_{AVG,ON}=\frac{68.75}{100\left(0.45\right)}\approx1.528\ \mathrm{A}
IMIN=1.528−1.502≈0.778 AI_{MIN}=1.528-\frac{1.50}{2}\approx0.778\ \mathrm{A}
IMAX=1.528+1.502≈2.278 AI_{MAX}=1.528+\frac{1.50}{2}\approx2.278\ \mathrm{A}
IP,RMS≈1.065 AI_{P,RMS}\approx1.065\ \mathrm{A}
EMIN=12(300 μH)(0.778)2≈90.7 μJE_{MIN}=\frac{1}{2}\left(300\ \mu\mathrm{H}\right)\left(0.778\right)^2\approx90.7\ \mu\mathrm{J}
EMAX=12(300 μH)(2.278)2≈778.2 μJE_{MAX}=\frac{1}{2}\left(300\ \mu\mathrm{H}\right)\left(2.278\right)^2\approx778.2\ \mu\mathrm{J}
ΔE=EMAX−EMIN=687.5 μJ\Delta E=E_{MAX}-E_{MIN}=687.5\ \mu\mathrm{J}
QuantityCurrent-Starts-at-Zero Candidate300 µH CCM Current Candidate
Magnetizing inductance147.3 µH300 µH
Minimum currentApproximately 0 A0.778 A
Peak current3.06 A nominal2.278 A
Primary RMS current1.18 A1.065 A
Peak stored energy687.5 µJ778.2 µJ
Residual stored energyApproximately 0 µJ90.7 µJ
Incremental energy per cycle687.5 µJ687.5 µJ
Primary-side mode indicationCurrent starts at zeroNonzero minimum current

ENGINEERING CAUTION — CHAPTER 6 MUST CONFIRM THE ACTUAL MODE

The primary current calculations above do not yet establish secondary demagnetization time. Reflected voltage and turns ratio determine whether the current-starts-at-zero candidate operates in DCM, reaches BCM, or fails to demagnetize before the next cycle.

The 300 µH candidate shows a nonzero primary minimum-current solution under the stated idealized power and timing assumptions, but Chapter 6 must still close the volt-second and secondary-conduction relationships.

SolidMag Engineering Insight

The Same Power Can Be Processed with Different Current and Energy States

The two candidates transfer the same 687.5 µJ per cycle, but they do not store or conduct current in the same way. The current-starts-at-zero candidate has higher peak and RMS current; the CCM candidate has lower peak current but retains magnetic energy and bias.

Neither is automatically superior. The final choice depends on controller behavior, turns ratio, demagnetization, core and gap realization, winding loss, semiconductor stress, transient response, EMI, and optimization priorities.

13. Design Handoff to Turns Ratio and Reflected Voltage

At the end of this chapter, the electrical design should provide a traceable set of energy, inductance, and current targets for every important operating corner.

Required Outputs from Chapter 5

  • Input power and energy per active switching cycle.
  • Target magnetizing inductance and allowable minimum/maximum tolerance.
  • Primary minimum, average, peak, and RMS current.
  • Controller current-limit threshold, tolerance, and propagation-delay margin.
  • Expected DCM, BCM, CCM, or multimode current behavior.
  • Startup, overload, burst, and transient current limits.
  • Current and energy corners that will control core, gap, winding, and semiconductor design.
  • Explicit assumptions that must be confirmed or replaced with controller and hardware data.

Chapter 6 uses these targets with output voltage, rectifier drop, duty-cycle range, and semiconductor voltage constraints to select reflected voltage and primary-to-secondary turns ratio. That step determines secondary conduction time and confirms whether the intended operating mode is physically achievable.

SolidMag Engineering Insight

Energy and Current Targets Are the Inputs to the Magnetic Structure

Magnetizing inductance and peak current are not isolated transformer specifications. They are the electrical targets that the primary turns, core, material, and air gap must realize.

Once the current-and-energy design is traceable, Chapter 6 can select reflected voltage and turns ratio without losing sight of current limit, mode, loss, and saturation constraints.

Technical References for This Chapter

Related SolidMagnetics Resources

Ready to Design a Flyback Transformer?

Move from design theory to an automated magnetic design. Enter the converter requirements and evaluate a coordinated transformer candidate, winding geometry, loss estimates, thermal performance, and CAD output.

ABOUT THE AUTHOR

Stan Gibson

Electrical Engineer & Founder, SolidMagnetics

Stan Gibson is an electrical engineer and founder of SolidMagnetics, an engineering platform focused on magnetic-component design automation. His work includes power electronics, inductor and transformer design, magnetic-core selection, winding design, thermal analysis, and manufacturable CAD development. Through SolidMagnetics, he develops technical guides, calculators, and automated design tools intended to help engineers move from electrical requirements to practical magnetic-component designs.

View Full Author Profile