Core Geometry and Magnetic Material Selection — Chapter 9

Chapter 9 selects the physical magnetic platform that must realize the electrical and air-gap targets developed in Chapters 5 through 8. The core geometry determines magnetic cross-sectional area, winding-window area, mean turn length, thermal surface area, bobbin options, and manufacturability, while the ferrite material determines loss, saturation behavior, permeability, and temperature performance.

In This Chapter — Click to Expand

Table of Contents

1. Core Geometry and Material Must Be Selected Together

A flyback transformer core cannot be selected from output power alone. The same 60 W converter may require very different magnetic structures depending on switching frequency, operating mode, peak current, allowable flux density, isolation requirements, winding construction, thermal environment, and mechanical envelope.

The core geometry and ferrite material also interact. A larger core can reduce flux density and winding resistance, but it adds volume and material. A low-loss ferrite grade can improve high-frequency performance, but it may not be available in the preferred geometry or gap range. The correct solution is therefore a core-set decision rather than a single-number lookup.

---

2. What the Core Must Provide

A candidate flyback core should satisfy all of the following simultaneously:

• Adequate effective core area for the selected turns and flux-density limit
• Enough winding-window area for primary, secondary, auxiliary, insulation, margins, and leads
• A practical A_L or gap option for the target magnetizing inductance
• Acceptable core loss at the actual frequency, flux waveform, bias, and temperature
• Adequate saturation margin at maximum temperature
• Reasonable mean turn length and winding resistance
• A bobbin and pin arrangement compatible with isolation and PCB requirements
• Acceptable thermal surface area and heat-transfer path
• Mechanical fit inside the available envelope
• Stable sourcing and repeatable manufacturing

Core Parameter	Design Significance
Effective area, Ae	Controls flux density for a given turns count and volt-second product
Minimum area, AMIN	Can govern local maximum flux density
Window area, Aw	Determines available space for copper and insulation
Area product, AP	Useful first-pass screening measure combining core and window capability
Effective path length, le	Used in reluctance and magnetic calculations
Effective core volume, Ve	Used to estimate total core loss
Mean length per turn	Strongly affects copper length and DCR
Available AL values	Determines whether the target magnetizing inductance can be realized practically
Bobbin winding width	Influences turns per layer and winding arrangement

---

3. Area Product as a First-Pass Screening Tool

A common first-pass core-screening parameter is the area product:

where A_P is area product, A_e is effective magnetic area, and A_w is winding-window area.

Area product is useful because it combines magnetic capability and winding-space capability in one parameter. It is particularly useful for rejecting cores that are clearly too small.

---

4. Usable Winding Window and Window Utilization

The published window area is not fully available for copper. A first-order usable winding-area relationship is:

where K_U is the window-utilization factor.

The utilization factor must account for:

• Wire insulation
• Primary-to-secondary insulation barrier
• Margin tape or molded margins
• Layer insulation
• Triple-insulated wire diameter
• Auxiliary windings
• Electrostatic shields
• Lead exits and transitions
• Winding irregularity
• Manufacturing clearance
• Bobbin walls and flanges

A transformer that theoretically fits using bare-copper area may be impossible to manufacture once insulation and practical winding clearances are included.

---

5. Common Flyback Core Geometries

Several ferrite geometries are commonly used in flyback converters. Each shape changes winding length, package profile, shielding, window shape, thermal behavior, and manufacturability.

EE and EI Cores

EE and EI families are widely available, mechanically straightforward, and compatible with many standard bobbins. They offer flexible window shapes and are often convenient for custom gapping and prototype work.

ETD, ER, and EER Cores

ETD, ER, and EER families use rounded or optimized center posts that can reduce mean turn length and improve copper utilization. They are common in power-conversion applications with moderate to higher power density.

PQ Cores

PQ cores can provide a favorable combination of core area and winding area in a compact package. Their geometry can reduce mean turn length and provide good magnetic shielding, but the winding window can become restrictive for complex isolation structures or heavy-current secondaries.

EFD and Low-Profile Cores

EFD and other low-profile core families are useful where component height is constrained. The tradeoff is often reduced winding volume and tighter thermal, insulation, and conductor-placement constraints.

RM and Pot-Core Families

RM and pot-style cores provide compact construction and good field containment. They can be useful for lower-power applications, although winding access, termination geometry, and creepage must be checked carefully.

Planar Core Families

Planar magnetics use PCB windings, foil, lead frames, or flat conductors with low-profile ferrite structures. They can provide repeatable geometry and excellent thermal contact, but also introduce strong proximity effects, interwinding capacitance, current crowding, and limited turn resolution.

Core Family	Typical Strength	Common Constraint
EE / EI	Availability, flexibility, straightforward gapping	Mean turn length and package efficiency may be less optimized
ETD / ER / EER	Good copper utilization and power density	Height and bobbin constraints vary by family
PQ	Compact magnetic volume and short turn length	Narrow winding structure can complicate isolation or high-current windings
EFD / low profile	Low component height	Reduced winding volume and thermal margin
RM / pot	Good field containment	Winding access and termination complexity
Planar	Low profile, repeatable geometry, thermal integration	Capacitance, proximity loss, current crowding, turn resolution

---

6. Mean Turn Length, Copper Length, and Winding Resistance

Core geometry strongly influences the average conductor length required for each turn. A simple winding-resistance estimate begins with:

The total copper length can be approximated from the mean length per turn and the number of turns, then refined using the actual winding layers and lead geometry.

where MLT is the mean length per turn.

A physically larger core may reduce flux density but can increase mean turn length. A compact core may shorten the conductor path but force additional layers, tighter proximity fields, or more difficult insulation. The complete winding loss must therefore be evaluated rather than assuming smaller or larger is inherently better.

---

7. Selecting the Ferrite Material

Ferrite is the dominant core material for most high-frequency flyback transformers because its high electrical resistivity limits eddy-current loss. However, power ferrites differ significantly in frequency range, loss behavior, saturation characteristics, permeability, and temperature dependence.

Important material-selection variables include:

• Switching-frequency range
• Flux-density excursion
• Waveform shape and duty cycle
• Magnetic bias
• Minimum and maximum operating temperature
• Core-loss target
• Temperature-adjusted saturation margin
• Available core geometry and gapped variants
• Supply stability and cost

---

8. Core Loss and the Actual Flyback Flux Waveform

A first-order Steinmetz-type core-loss density expression is often written as:

Total core loss is then:

The definition of the flux term B_X and the applicable coefficients must match the manufacturer data or fitted loss model.

Flyback transformers do not normally experience a simple sinusoidal flux waveform. Their flux may contain unequal positive and negative slopes, zero-voltage intervals, magnetic bias, minor loops, and variable-frequency operation.

For final design, use manufacturer loss data, manufacturer design tools, generalized Steinmetz methods, waveform-aware loss models, or measured prototype behavior appropriate to the actual material and waveform.

---

9. Saturation Flux Density and Temperature

Ferrite saturation flux density generally decreases as temperature increases. Therefore the magnetic design should compare the worst-case peak flux against a temperature-appropriate design limit:

The selected operating limit should include margin for:

• Core material variation
• Effective area tolerance
• Primary-turn count
• Gap and magnetizing-inductance tolerance
• Current-limit tolerance
• Input-voltage variation
• Switching-frequency variation
• Maximum core temperature
• Startup and transient conditions

Core loss may establish a lower practical flux-density limit than hard saturation, particularly at high switching frequency. Both checks are required.

Magnetic Check	Question Answered
Peak flux vs. saturation limit	Will the material retain adequate incremental permeability?
Flux excursion vs. core-loss data	Will the ferrite generate acceptable heat?
Temperature-dependent loss	Does the worst thermal point remain acceptable?
Bias / operating mode	Does residual current move the core toward an unfavorable operating region?

---

10. Build a Core-Candidate Comparison Matrix

The fastest way to avoid premature commitment is to evaluate several candidate core sets in parallel. A candidate matrix can include:

Candidate Metric	Why It Matters
Core family and size	Mechanical fit and packaging
Material grade	Frequency, loss, and temperature behavior
Ae / AMIN	Flux-density margin
Aw	Winding and insulation capacity
AP	First-pass magnetic/winding screening
Ve	Total core-loss estimate
Mean turn length	Copper resistance and loss
Available AL values	Gap and inductance realization
Bobbin and pin layout	Isolation and PCB integration
Height / footprint	Mechanical constraints
Thermal surface area	Temperature rise
Availability / cost	Production viability

The best candidate is the one that produces the best complete transformer, not the one with the largest area product or smallest core volume.

---

11. Worked Example — Comparing Three Candidate Core Sets

Parameter	Candidate A — EE	Candidate B — ETD	Candidate C — PQ
Effective area Ae	75 mm²	90 mm²	85 mm²
Window area Aw	110 mm²	125 mm²	100 mm²
Effective volume Ve	5.5 cm³	6.3 cm³	6.0 cm³
Mean turn length	58 mm	52 mm	47 mm
Available AL near target	Yes	Yes	Limited
Relative profile	Moderate	Moderate / taller	Compact
Isolation winding space	Good	Very good	Moderate

Step 1 — Compare Area Product

Candidate B has the largest area product, but this alone does not establish it as the best design.

Step 2 — Recalculate Flux Density with the Existing 32-Turn Primary

Using 100 V, 0.45 duty cycle, 100 kHz, and 32 primary turns:

Candidate B provides the largest flux margin under these assumptions.

Step 3 — Compare Primary Copper Length

Candidate C offers the shortest primary conductor length, which may reduce DCR if the same copper area and winding arrangement can be used.

Step 4 — Compare the Complete Tradeoff

Design Consideration	Candidate A — EE	Candidate B — ETD	Candidate C — PQ
Flux margin	Lowest	Best	Good
Window area	Good	Best	Moderate
Primary copper length	Longest	Intermediate	Shortest
AL availability	Good	Good	Limited
Isolation flexibility	Good	Best	Moderate
Compactness	Moderate	Moderate	Best

Candidate B would likely receive the highest initial ranking for this isolated flyback because it combines the largest window, strongest flux margin, practical A_L availability, and reasonable mean turn length. Candidate C remains attractive where compactness and copper length dominate, but its more limited winding space and gap options could become decisive constraints.

---

12. Verify the Selected Core Set Before Proceeding

Before accepting the selected core and material, verify:

• Actual manufacturer A_e, A_MIN, A_w, l_e, V_e, and mean-turn-length data
• Material loss data over the intended frequency, flux, bias, and temperature range
• Temperature-adjusted saturation limit
• Available factory gap or A_L options
• Bobbin winding width and usable window area
• Creepage and clearance capability
• Pin and termination arrangement
• Core clip or mounting method
• Mechanical height and footprint
• Supply availability and approved alternatives
• Prototype core loss and temperature

---

13. Design Handoff to Conductor Selection

Once the core geometry, material, primary turns, secondary turns, and preliminary gap are coherent, the next major task is to design the actual conductors.

Chapter 10 will use:

• Primary RMS current
• Secondary peak and RMS current
• Auxiliary winding current
• Switching frequency and harmonic content
• Core-window geometry
• Mean turn length
• Available winding width
• Insulation requirements
• Temperature-rise target

to select primary, secondary, and auxiliary conductors and evaluate DCR, skin effect, proximity effect, current density, foil, parallel strands, litz wire, triple-insulated wire, terminations, and winding fit.

---

Technical References for This Chapter

Use the selected ferrite manufacturer’s datasheets and design tools for exact geometry, A_L, material loss, permeability, and saturation data. Core geometry and material values should be taken from the production part rather than generic family assumptions.

Related SolidMagnetics Resources