Key takeaways
- A transformer coil is sized by volts per turn and the window, then checked for loss and temperature rise.
- An inductor coil is sized by the energy it must store, which usually means a gapped core and a saturation check.
- Insulation between windings is a safety function, not a winding detail, and it sets creepage and clearance.
- Copper loss and core loss trade against each other, so the cheapest coil is rarely the coolest one.
Two coils, two jobs
A transformer coil transfers energy from one winding to another through a shared magnetic field. The primary sets the flux, the secondary picks it up, and the turns ratio sets the voltage. The coil is judged by regulation, efficiency, isolation and temperature rise.
An inductor coil stores energy in its own magnetic field. It is judged by inductance, saturation current, loss and, at higher frequency, by the behaviour of the winding against the core. The two products use similar wire and similar bobbins, but the design logic is different.
A supplier that winds both will ask different questions. For a transformer, they concern the turns ratio, the isolation and the inter-winding capacitance. For an inductor, the gap, saturation current and inductance at the operating point.
Turns ratio and volts per turn
The turns ratio sets the secondary voltage in a transformer. The secondary voltage is set by the ratio of secondary to primary turns, less the drops. The ratio is a design input, not a winding guess.
The core sets the volts per turn. A larger core carries more flux per turn, so fewer turns are needed for the same voltage. Fewer turns mean less copper and lower loss, but a larger and more expensive core. The design balances the two.
Flux density has to stay below saturation at the highest supply voltage and lowest frequency. A transformer designed at 50 Hz and run at 60 Hz has margin; the reverse can saturate the core. The rating should state the frequency range.
For an inductor, the turns set the inductance for a given core and gap. Inductance rises with the square of the turns, so a small change has a large effect, which is why the drawing needs the core, the gap and the inductance tolerance together.
The winding window and current density
The window is the space available for copper. It has to hold the primary, the secondary, the insulation, the bobbin wall and the creepage path. A window that is full on the drawing is usually too full in production.
Current density is the current per unit area of copper. A higher density means a smaller coil and more loss. A lower density means more copper and a larger coil. The choice is a thermal one, and the acceptable value depends on the cooling path and the duty.
The fill factor decides how much of the window is copper. A layer wind gives a higher fill than a random wind, and foil fills more than round wire. The bobbin, tape and layer insulation take the rest.
The winding order matters too. Splitting the primary around the secondary reduces leakage inductance, at the cost of more layers and insulation. That is a common trade in a transformer feeding a fast-changing load.
Copper loss and core loss
Copper loss is the current squared times the resistance of the winding. It rises with temperature because copper resistance rises, and it depends on the mean turn length, which grows with each layer.
Core loss has two parts. Hysteresis loss depends on the flux swing and the material. Eddy current loss depends on flux, frequency and lamination thickness or ferrite particle size, and can dominate at high frequency.
The two losses trade against each other. More turns reduce the flux per turn and the core loss, but add copper and raise the copper loss; fewer turns do the opposite. The optimum is usually where the two are similar, adjusted for material cost.
At high frequency, skin and proximity effect push current to the outside of the conductor and between adjacent turns. Stranded or litz wire reduces the effect, at a cost and a fill penalty.
Loss is a temperature, not a number
A loss figure in watts only becomes useful when it is turned into a temperature rise. The same loss can be comfortable in an open frame and unacceptable in a sealed potted unit.
Insulation between windings
The insulation between primary and secondary is a safety function. It has to withstand the working voltage, the transient overvoltage and the test voltage, and maintain creepage and clearance over life.
The requirements depend on the application and the standard. Reinforced insulation, basic insulation and functional insulation have different thickness, spacing and test requirements. The drawing should state which one applies, not leave it to the winding house.
Layer insulation manages the voltage between turns and layers. Bank or split winding can reduce the stress. The insulation material has to match the thermal class and the varnish, or the system rating falls.
A coil that passes a high potential test may still have a creepage path that is too short for the pollution degree of the application. The test proves the insulation at that moment; it does not prove the spacing.
Leakage, coupling and capacitance
Not all the flux from the primary links the secondary. The flux that does not link is leakage flux, and it appears as a leakage inductance in series with the winding. Leakage causes a voltage drop under load and a slower response in a switching converter.
Interleaving the windings reduces leakage because the primary and secondary are closer together. It also raises the inter-winding capacitance, which couples high frequency noise across the isolation barrier. The design has to balance the two.
An electrostatic shield between the windings can reduce common mode noise at the cost of another layer and more insulation. It is common in power supplies that must meet an emission limit.
For an inductor, the self-capacitance sets the self-resonant frequency. Above it the part behaves like a capacitor, not an inductor. A random wind has more variable capacitance than a layer wind, which is one reason the pattern matters at high frequency.
Inrush, duty and thermal design
A transformer switched on at the wrong point in the mains cycle can draw a large inrush current. The core saturates for a few cycles and the current is limited only by the winding resistance and the supply impedance. The coil and the protection have to survive it.
The thermal design uses the total loss, the surface area and the cooling path. A coil in free air, in a sealed housing or on a heatsink has a different thermal resistance, and the ambient at the coil is the starting point.
The duty cycle matters for both products. A transformer that runs continuously reaches a steady temperature. An inductor in a switching converter may see a ripple current that adds to the loss without changing the average. The loss calculation should use the actual waveform, not just the RMS value.
The hot spot is usually inside the winding, not on the surface. A resistance measurement gives the average, and the difference can be significant in a thick winding. The thermal class should be chosen against the hot spot.
What to test and record
A transformer is checked for turns ratio, resistance, inductance or magnetising current, insulation resistance and high potential withstand. A no-load test gives core loss, and a short circuit test gives copper loss and leakage impedance.
An inductor is checked for inductance at the stated frequency and bias, resistance, saturation and high potential. Zero bias inductance is not enough if the part operates near saturation.
The records should include the core material and batch, the wire grade, the insulation materials, the winding scheme and the test conditions. A coil copied without those records is a different part, even if it measures the same on the bench.
For a safety related transformer, the file should also include the insulation system, the creepage and clearance calculation and the test report, so the buyer can show that the part meets the standard.
| Design point | Transformer | Inductor |
|---|---|---|
| Sizing basis | Volts per turn | Energy storage and inductance |
| Core gap | Usually ungapped | Often gapped to control saturation |
| Main loss concern | Copper and core balance | Copper loss and saturation |
| Insulation focus | Between windings and to core | Winding to core and layer stress |
| Key test | Turns ratio and withstand | Inductance at bias and saturation |
Some parts combine both functions, such as flyback transformers. Treat the coil as both an energy store and an isolation barrier.
Frequently asked questions
What sets the number of turns in a transformer?
The required secondary voltage and the volts per turn of the core. The core area and the flux density set the volts per turn, and the turns ratio then sets the secondary. The design is checked against saturation at the highest voltage and lowest frequency.
Why does an inductor need a gap?
A gap increases the reluctance of the magnetic circuit, which lowers the inductance for a given number of turns but raises the saturation current. It also makes the inductance less dependent on the core permeability, which improves consistency.
Can I use the same coil for a transformer and an inductor?
Sometimes physically, but not by design. A transformer winding is sized for volts per turn and isolation, and an inductor winding is sized for energy storage and saturation. The tests and the acceptance criteria differ.
How is the insulation between windings specified?
By the insulation class required by the application standard, such as basic, supplementary or reinforced, together with the working voltage, the test voltage and the creepage and clearance. The drawing should state all of them.
Sources and further reading
- IEC 61558 series, Safety of transformers, reactors, power supply units and similar
- IEC 60085, Thermal evaluation and classification of electrical insulation
- IEC 60317 series, Specifications for particular types of winding wires
Qumek Engineering Desk
Engineering team
The engineering team behind Qumek drawings, DFM notes, tolerances and inspection standards. Every guide is reviewed against the drawings and processes used on live programmes.
Reviewing a transformer or inductor winding?
Send the turns, core, frequency, isolation requirement and duty. The engineering desk will reply with a note on the window, the loss balance and the test plan.
Send the winding data