The transformer turns ratio gives you the output of an ideal transformer, and no real transformer is ideal. Winding resistance, leakage inductance and magnetizing current all take their share, so the loaded output is lower than the ratio predicts while the unloaded output is higher than the nameplate. Measuring one condition and assuming the other is a reliable way to be surprised.
What the ideal model leaves out
The familiar relationship, that the voltage ratio equals the turns ratio and the current ratio is its inverse, assumes perfect coupling, no winding resistance, no core loss and no magnetizing current. Every one of those assumptions is violated by a physical part. Each winding has real resistance, so current through it drops voltage that never reaches the load. Not all flux produced by the primary links the secondary, and the portion that does not behaves as a leakage inductance in series with the winding, dropping additional voltage that grows with current and frequency. The core requires a magnetizing current to establish flux at all, drawn even when the secondary is open. The ideal equation remains an excellent starting point for design, but it is a first approximation and the corrections are not small.
Why the output sags as you load it
Regulation is the name given to the difference between no-load and full-load output voltage, expressed as a percentage of the rated output. It is not a defect; it is an unavoidable consequence of the resistances and leakage inductances in the windings, and it is larger for small transformers than for large ones because small windings use finer wire with proportionally more resistance. Manufacturers account for this by winding the secondary with extra turns so that the output reaches its nameplate value at rated load rather than at no load. The direct consequence is the measurement that puzzles people: connect a meter to an unloaded secondary and you read noticeably more than the nameplate says. The transformer is not mislabeled and is not faulty. It is simply being measured in a condition its label does not describe.
Rectification moves the target again
Feed that secondary into a rectifier and a reservoir capacitor and the DC you get is related to the peak of the AC waveform rather than its RMS value, less the forward drops of the rectifying devices. A lightly loaded supply therefore sits at a DC voltage well above the secondary’s RMS rating, which surprises anyone who sized downstream components against the AC number. As load increases the capacitor discharges further between peaks, ripple grows, and the average DC falls, so the same supply that measured comfortably high at idle can drop below what a regulator needs at full load. Both ends of that range have to be checked: the high end against component voltage ratings and regulator input limits, the low end against dropout and ripple requirements.
The input is not fixed either
Nameplate output assumes nameplate input, and supply voltage varies with time of day, distance from the transformer, and load elsewhere in the building. A transformer that produces the right output at nominal input will produce proportionally less at the low end of the supply tolerance and proportionally more at the high end. Design margin has to absorb the combination of low input plus full load at one extreme and high input plus no load at the other, because both can occur. Frequency matters too where equipment might be used in a region with a different mains frequency, since running a transformer designed for a higher frequency at a lower one increases the flux swing and can drive the core toward saturation. Never assume a transformer intended for one supply system is suitable for another.
Measuring it properly, and safely
To characterize a transformer, measure the secondary open circuit, then at a known resistive load near its rating, and record both. The difference gives you the regulation you will actually experience. Measure the input voltage at the same time, because the output figures mean little without it. Keep in mind that a transformer under test is mains-referenced equipment and the primary side is dangerous; do not probe a live primary casually, keep one hand away from the circuit, work on an insulated surface and follow the appropriate safe work practices. Treat a step-up secondary with particular respect, since it can produce voltages far above the supply. If the purpose of the transformer is isolation, verify that the isolation is intact rather than assuming it, because a failed interwinding insulation is not visible from outside.
What to check before you act
- Measure secondary voltage at no load and at rated load, and record both.
- Note the input voltage during every measurement so the results mean something.
- Check rectified DC at both light load and full load against component ratings.
- Treat the primary side as live and follow safe work practices throughout.
Common questions
My transformer reads higher than its rating with nothing connected. Is it faulty?
Almost certainly not. Secondaries are wound to reach the nameplate voltage at rated load, so an unloaded reading is expected to be higher.
Does a transformer with better regulation always run cooler?
Better regulation usually means lower winding resistance, which reduces copper loss under load. Core loss is a separate matter and may be unchanged or even higher.
The practical takeaway
Use the turns ratio to start a design, then apply the transformer’s stated regulation and check the result at both load extremes and both ends of the input tolerance. Measure loaded and unloaded, record the input voltage alongside, and treat the primary side as a live mains circuit throughout.
References and further reading
- OSHA: wiring design and protection requirements for electrical installations
- CDC NIOSH: electrical safety fundamentals for anyone working on powered equipment
- NFPA: what the National Electrical Code covers and how it is organized
Numerical scenarios are illustrative unless identified otherwise. Follow the exact product instructions; component ratings and local installation requirements can differ.
