Unknown Transformer Identification Tool: Turns, Power and Rewinding
Introduction
An unmarked transformer in a drawer is a common situation for anyone who salvages electronic equipment. The questions are always the same: what voltage does it provide, how much current can it handle, and can I wind a different secondary on it? The other pages on this site go from design to hardware; this one takes the opposite route, from the actual component to the numbers, using only a caliper, a multimeter, a scale, and a length of thin wire.
The key to the whole method is the test winding. If you wind ten turns of any wire through the window and measure the voltage that appears across them while one winding is energized, you obtain the volts per turn for that core. Dividing the applied voltage by that value gives the number of turns on the unknown winding: it is a direct estimate derived from the voltage ratio, and its accuracy depends on the measurements.
What makes the method convenient and safe is that this ratio does not depend on the excitation level. Apply twelve volts and you obtain the same turns estimate you would obtain at two hundred and thirty volts. The assumed rated voltage is used later, in a separate card, only to determine the flux density at which the manufacturer intended the transformer to operate — the value that tells you whether your assumptions are plausible.
How to energize it safely
Do not connect an unknown transformer directly to the mains. The reason is simple: you do not know whether the winding you think is the primary really is the primary, whether it is rated for 230 V or 110 V, or whether it has shorted turns — and all three mistakes can reveal themselves in the worst possible way.
The correct approach is to energize it from a low, isolated AC voltage, preferably current-limited, supplied by a second transformer whose characteristics are known. Twelve or twenty-four volts are usually sufficient; if in doubt, start even lower. The core will operate at a flux density far below its rated value while the voltage ratio — the only quantity needed at this stage — remains unchanged. If the winding being energized has shorted turns, current can rise quickly: stop the test if the current is abnormal, because even a low-voltage source can overheat or damage the winding if it can supply substantial current.
But there is one condition that makes everything else safe: you must energize the correct winding. A transformer works in both directions, and if voltage is applied to the low-voltage side it acts as a step-up transformer. Applying 12 V to the nominal 24 V secondary of a 230/24 V transformer produces about 115 V on the primary: the core is not overexcited, but a dangerous voltage still appears on the other winding. If instead 12 V is applied to a winding rated for 6 V, the 230 V side can rise to about 460 V and the energized winding is being driven at twice its rated voltage. In that case the core can saturate and current can increase rapidly.
For ordinary mains transformers, a useful practical starting point is the winding with the higher resistance: the high-voltage winding usually has many turns of thin wire, while the low-voltage winding has fewer turns of thicker wire. However, this is not a universal rule: multi-winding transformers, taps, auxiliary windings, and special constructions can be ambiguous. For this reason, start with only a few volts AC and current limiting, measure all induced voltages, and verify the ratio before increasing the excitation.
If in doubt, start even lower: three or four volts are often enough to obtain a useful reading from the test winding with a decent multimeter. Measure before touching any terminal. If another winding shows a voltage higher than the applied voltage, the calculator flags it: that winding has more turns than the energized winding, and the connection must be checked before proceeding.
If for some reason you must test at full voltage, a series incandescent lamp is the classic current limiter and it works well: with a healthy transformer the lamp remains off or only faintly lit; with a short circuit it lights and limits the current. For the measurements required on this page, however, there is normally no need to exceed a few tens of volts, and the best practice is to remain at the minimum voltage that gives reliable readings.
Reference parameters
These are not quantities measured on the transformer: frequency, current density, and fill factors are reference values used by the following cards. Change them only if you know that the transformer operates under conditions different from the defaults.
Card 1 — The measured core
For EI laminations, the center-leg width and stack thickness are normally sufficient. The calculator compares the measurement with a table of EI sizes and, when the match is unambiguous, uses the tabulated dimensions to derive window area and volume; otherwise it retains a proportional geometric estimate. The overall lamination width is optional, but it helps distinguish very similar sizes such as EI38.4 and EI41.
Weight is a useful cross-check. The iron volume is obtained from the dimensions and the density of silicon steel is known, so the measured weight should agree once the stacking factor is considered. If it does not, one of the measurements is wrong — finding that out here takes thirty seconds; finding it after winding can cost an afternoon.
On EI laminations, measure the center leg on the assembled stack; it is the column that carries the bobbin. For the tabulated EI sizes, the overall width is normally about three times the center-leg width. Some very similar sizes cannot be identified reliably from a single dimension, so the tool reports the ambiguity instead of forcing a match.
Card 2 — The test winding
Ten turns are a good compromise: enough to produce an easy-to-read voltage, but few enough to pass quickly through the window. Almost any thin insulated wire will do; it carries essentially no load current and is used only for measurement.
The last field is useful when the transformer already has other windings. Measure the voltage that appears on one of them while the first winding is energized, and the calculator also estimates its turn count and the ratio. This is how you determine whether an unknown secondary will provide six volts or sixty when the transformer is operated at its rated primary voltage.
If the voltage measured on the test winding is suspiciously low, or the source appears heavily loaded, stop and check before proceeding: the transformer may have shorted turns, in which case the following calculations are not meaningful.
If the voltage on the other winding is higher than the applied voltage, that winding has more turns than the energized winding. Stop the test before touching the terminals, check the winding resistances, and restart with only a few volts AC and current limiting on the most plausible winding.
Card 3 — Design flux density
Sinusoidal excitation; use RMS values when calculating flux density.
This is primarily a diagnostic card: it calculates the flux density corresponding to the entered assumptions and helps determine whether those assumptions are plausible. The number of turns has been estimated directly from the voltage ratio and the core cross-section from the measurements; the main remaining assumption is the rated voltage of the reference winding. If 230 V produces a flux density compatible with an ordinary mains transformer, the assumption is consistent, although that alone does not prove the identification.
If the result is very high, one of the assumptions may be wrong: the winding may have been misidentified, its rated voltage may be incorrect, the core cross-section may be underestimated, or the transformer may have been designed for a frequency higher than the one entered. A transformer designed for 400 Hz, for example, can use far fewer turns than an equivalent 50 Hz transformer and would appear to have excessive flux density if evaluated at the same voltage and 50 Hz. If the calculated density is very low, recheck the assumed rated voltage, turn count, and core cross-section, or consider that the transformer may simply have been designed conservatively.
The flux-density calculation uses the sinusoidal relation with RMS voltage. The automatic badge thresholds are intended mainly for 50/60 Hz mains-frequency silicon-steel cores; at other frequencies the numerical calculation of B remains valid, but core losses and suitable design flux density require a separate assessment. This tool is not intended for ferrite or powder cores used in switching converters.
Card 4 — Estimated VA rating
A transformer’s power rating is not written into the hardware itself, but it can be estimated in two independent ways. The first starts from the core: cross-section times window area gives the area product which, together with reasonable flux density, current density, and fill factor, indicates how many VA the core can support. The second starts from the copper actually present: measure the secondary wire diameter to estimate the current the manufacturer intended that winding to carry.
Comparing the two estimates is particularly useful when the measured winding is the main secondary or the only secondary. If they agree, the design appears balanced. If the wire-based estimate is much lower, that winding was designed for a modest current relative to what the core might support. On transformers with multiple secondaries, the VA inferred from a single wire size describes only that winding, not the transformer’s total rating.
To measure the wire diameter you need physical access to it; exposing the outermost secondary layer is usually enough. With enamelled wire, the caliper also measures the insulation. The difference from the actual copper diameter depends on wire size and insulation grade, so a fixed percentage correction can only be approximate. Enter the best estimate of the copper diameter.
Card 5 — Resistances and regulation
The two DC resistances are the last quantities needed for an initial estimate of load-voltage drop. In small mains transformers the primary often has a much higher resistance than the low-voltage secondary, but the absolute values depend strongly on power, voltage, conductor size, and design. Use them as clues, not as a rigid identification criterion.
Referring the primary resistance to the secondary through the square of the turns ratio gives the equivalent copper resistance. From this, the calculator estimates the resistive load-voltage drop and regulation using the ΔV/Vload convention. Actual regulation can differ because leakage reactance and the load power factor are not included here.
The displayed efficiency accounts for copper losses only. Core losses are still present and must be measured separately from the no-load input power. On a salvaged transformer this is also a useful health check: unusually high no-load consumption can indicate a core or winding problem.
Card 6 — Adding a secondary winding
This is why many people perform all the preceding measurements: the transformer has a winding they do not need and they want a different one. Once the number of turns on the reference winding has been estimated and its rated voltage assumed, the calculator derives the nominal turns per volt and from that the turns required for a new secondary, including a small indicative allowance to compensate approximately for load-voltage drop.
Available space is the uncertain part, and it cannot be known exactly without disassembling the transformer. The entered percentage is an estimate of the remaining geometric window; the calculator then compares the required bare-copper area with the usable amount based on the selected fill factor. This check is approximate: enamel, interlayer insulation, margins, and the actual winding arrangement can require more space. If the estimate is exceeded only slightly, a higher current density may be considered only after accounting for the temperature rise allowed by the application.
The turns-per-volt value used here is the nominal value, obtained as Np/Vp,nom from the turn estimate in card 2 and the assumed rated voltage in card 3. It is not the low-voltage test value in volts per turn: that test is used to determine the ratio and turn count without applying rated voltage to the unknown transformer.