1. The Open Circuit Transformer Test: Direct Answer and Purpose
The open circuit transformer test (also called the no-load test) is performed to determine a transformer's core losses (iron losses), excitation current, and shunt branch equivalent circuit parameters. The direct answer to how it is done: you apply the rated nominal voltage to the low-voltage (LV) winding while leaving the high-voltage (HV) winding completely open-circuited.
Because the secondary is open, no load current flows through it. The primary draws only a small excitation current ($I_0$), which is typically 2% to 5% of the transformer's full-load rated current. The power consumed during this test is almost entirely due to hysteresis and eddy current losses in the magnetic core, as the $I^2R$ copper losses in the primary are negligible at such low currents.
For bench technicians and DIY builders working with control transformers (like the common Hammond 184 series), this test is the definitive way to verify core integrity before wiring the unit into a live control panel.
2. Safety First: CAT Ratings and Mains Precautions
When testing a step-down transformer (e.g., 120V to 24V), always apply the rated voltage to the low-voltage side if you are testing large distribution units to keep the test voltage safely low. However, for small fractional-kVA control transformers, it is standard practice to apply the 120V mains to the primary (HV side relative to the 24V secondary) because the primary is designed for that exact voltage, and the open-circuit secondary will safely step up to 24V. Just ensure the secondary terminals are capped and untouched.
3. Meter Setup and Probe Placement
Accurate measurement of the open circuit test requires three distinct measurements: Voltage ($V_0$), Current ($I_0$), and Real Power ($P_0$). Because the no-load power factor is extremely low (typically 0.10 to 0.20), standard plug-in power meters will yield massive errors. You need true-RMS meters and a low-PF capable wattmeter.
Meter Setup Block
- Voltmeter (e.g., Fluke 87V): Dial set to V AC (True-RMS). Black lead in COM, red lead in V/$\Omega$. Range set to Auto or 600V. CAT III 600V minimum.
- Ammeter (e.g., Fluke 376 True-RMS Clamp Meter): Dial set to A AC. Clamp jaw placed around only one primary conductor (the hot line feeding the primary). If measuring a 100VA transformer where no-load current is ~30mA, use the mA setting or a dedicated low-current clamp accessory, as standard 600A jaws lack the resolution below 100mA.
- Wattmeter (e.g., Yokogawa WT3000E or Hioki PW3360): Set to measure Real Power (W). Ensure the Low Power Factor (Low PF) compensation is enabled. Voltage probes across the primary terminals; current coil in series with the primary hot line.
Probe Placement per Test Point
- Primary Voltage ($V_0$): Place voltmeter probes directly across the primary winding terminals (e.g., H1 and H2). Do not measure at the breaker; measure at the transformer lugs to account for line drop.
- Primary Current ($I_0$): Clamp the ammeter around the hot conductor feeding H1. Ensure the conductor is centered in the jaw and the jaw is fully closed.
- Secondary Verification: Place a voltmeter across the secondary terminals (X1 and X2) to confirm it is open-circuited and stepping up/down correctly. No load should be connected here—not even a pilot light.
4. Step-by-Step Execution Procedure
- Visual Inspection: Check for cracked bobbins, melted insulation, or bulging epoxy potting. If the core laminations are visibly rusted or separated, skip the test and scrap the unit.
- Isolate the Secondary: Disconnect all wires from the secondary terminals (X1, X2, etc.). Cap them with wire nuts or heat shrink to prevent accidental contact.
- Wire the Primary: Connect your wattmeter current coil and ammeter clamp in series with the primary hot line. Connect the wattmeter voltage probes and voltmeter in parallel across H1 and H2.
- Energize: Turn on the mains disconnect. Allow the transformer to energize. (Note: Ignore the brief inrush current spike; wait 5 seconds for the current to settle to the steady-state no-load value).
- Record Data: Record $V_0$ (should be within $\pm$5% of nominal, e.g., 114V-126V for a 120V nominal system), $I_0$ in milliamps, and $P_0$ in watts.
- De-energize: Turn off the mains, verify dead with the voltmeter, and disconnect.
5. Expected Readings: Good vs. Bad Values
Let us look at concrete numbers using a standard Hammond 184F24 (120V Primary, 24V Secondary, 100VA, 50/60Hz) as our baseline. The rated full-load primary current is $100VA / 120V = 0.833A$ (833mA).
| Parameter | Symbol | Good Reading (Healthy Core) | Bad Reading (Failing Unit) | Physical Meaning |
|---|---|---|---|---|
| No-Load Current | $I_0$ | 25 mA to 45 mA (3-5% of FLC) | > 85 mA (>10% of FLC) | Indicates magnetizing impedance. High current means shorted turns or degraded core. |
| Core Loss Power | $P_0$ | 1.5 W to 3.5 W | > 8.0 W | Represents hysteresis and eddy currents. High wattage means shorted core laminations. |
| Voltage Ratio | $V_1 / V_2$ | ~5.0 (120V in yields ~24V out) | < 4.5 or > 5.5 | Confirms turns ratio. Deviations indicate shorted primary/secondary turns. |
If you are modeling this transformer in SPICE or MATLAB, you can extract the shunt branch parameters directly from this test:
Core Loss Resistance ($R_c$): $R_c = V_0^2 / P_0$. (e.g., $120^2 / 2.5W = 5,760 \Omega$).
Magnetizing Reactance ($X_m$): First find core loss current $I_c = P_0 / V_0$. Then magnetizing current $I_m = \sqrt{I_0^2 - I_c^2}$. Finally, $X_m = V_0 / I_m$.
6. Decision Tree: Diagnosing Transformer Health
Use this decision path to determine the exact fate of the transformer on your bench. Do not guess; follow the data to the terminal action.
| Observed Symptom | Root Cause Diagnosis | Concrete Action / Part Pick |
|---|---|---|
| $I_0$ is >10% of full-load current, but $P_0$ is normal. | Shorted turns in the primary winding. The effective turns ratio has dropped, lowering impedance. | Scrap and Replace. Do not rewind sub-1kVA potted units. Order a direct replacement (e.g., Hammond 184F24 or Magnetics Inc. equivalent). |
| $P_0$ is 2x to 3x expected, $I_0$ is slightly elevated, unit runs physically hot (no load). | Core lamination insulation failure. Eddy currents are circulating across the laminations instead of being contained within them. | Scrap and Replace. Core degradation is irreversible. Replace with a new unit featuring grade A silicon steel laminations. |
| $I_0$ and $P_0$ are normal, but Secondary Voltage is 15% low. | Shorted turns on the secondary winding, or the wrong primary voltage tap is selected (e.g., wired to 240V tap on a 120V source). | Verify Taps. Check primary wiring against the nameplate. If taps are correct, Scrap and Replace. |
| All readings ($I_0$, $P_0$, $V_{ratio}$) are within the 'Good' table ranges. | Transformer is healthy. Core and windings are intact. | Clear for Service. Proceed to load testing or panel installation. |
7. Common Mistakes That Yield Misleading Data
If your numbers look wrong, you likely fell victim to one of these three bench errors. Understanding power factor measurement nuances is critical here.
1. The Low Power Factor Wattmeter Error
At no-load, a transformer is essentially a massive inductor. The phase angle between voltage and current is close to 90 degrees, resulting in a power factor (PF) between 0.10 and 0.20. Standard digital power meters (like a Kill-A-Watt or cheap inline wattmeters) assume a PF near 1.0 and suffer from internal phase-shift errors in their current shunts. At a PF of 0.15, a standard meter might read 10W when the true core loss is only 2W. The Fix: You must use a true power analyzer with explicit low-PF compensation, or calculate power using the volt-amp method with a phase-angle meter.
2. Leaving a 'Hidden' Load on the Secondary
It is common in retrofit scenarios to leave a panel pilot light, a snubber capacitor, or a high-impedance PLC input wired to the secondary. Even a 24V LED indicator drawing 10mA will reflect back to the primary, artificially inflating your $I_0$ and $P_0$ readings, making a healthy transformer look like it has core losses. The Fix: Physically disconnect all secondary wires. An open circuit means exactly that: infinite impedance, zero secondary current.
3. Using an Average-Responding Multimeter
While the steady-state no-load current is largely sinusoidal, the magnetic core's non-linear B-H curve introduces odd harmonics (especially the 3rd harmonic) into the excitation current. An average-responding clamp meter (which simply multiplies the average by 1.111) will under-read this distorted waveform by up to 15%. The Fix: Only use True-RMS clamp meters (like the Fluke 376 or equivalent) that sample the waveform and calculate the actual root-mean-square value, capturing the harmonic content accurately.
For deeper theoretical background on the equivalent circuit models derived from this test, refer to standard electrical engineering texts and resources like the Open Circuit Test guidelines on Electrical4U. By strictly following the meter setup, respecting the CAT III safety boundaries, and applying the decision tree above, you will definitively separate healthy magnetic cores from silicon-scrap-bound failures on the very first test.
