Power Transformer Volt-Ampere Characteristic: Magnetization Curve, Inrush Current, CT Saturation & Ferroresonance
Abstract
The volt-ampere (V-I) characteristic of a power transformer describes the nonlinear relationship between applied voltage and magnetizing current through the core. This curve is foundational to understanding four of the most troublesome phenomena in power transformer protection: magnetizing inrush current, current transformer (CT) saturation, ferroresonance, and harmonic generation during energization. This article provides a production-engineer-level, first-principles explanation of the V-I characteristic — how to read it, what the knee point means, and why proper comprehension prevents maloperation of differential protection, avoids CT specification errors, and ensures ferroresonance-free design. All explanations are rooted in IEC 60076 and IEEE C57.13 practice.
1. Physical Origin of the V-I Characteristic
A power transformer core is constructed from grain-oriented silicon steel laminations. The B-H curve (flux density vs. magnetic field strength) of the core material is inherently nonlinear. Since:
- Magnetizing current *I* is proportional to magnetic field strength *H* (via Ampere's law: *H = NI / l*)
- Induced voltage *V* is proportional to the rate of change of flux density *B* (via Faraday's law: *V = N dΦ/dt*)
The V-I characteristic is, in effect, the derivative of the B-H hysteresis loop mapped to the terminal domain. In sinusoidal steady-state AC, the magnetizing current waveform is not a pure sinusoid — it contains significant third, fifth, and seventh harmonic content even under normal operation, with the third harmonic typically dominating.
1.1 The Three Regions of the V-I Curve
Region I — Linear (Below Knee): At low excitation levels, the core operates well below saturation. The relationship between V and I is approximately linear. Permeability μ is high and nearly constant. This region corresponds to normal rated-voltage operation, typically up to 105-110% of rated voltage for power transformers.
Region II — Knee Point: As voltage increases, the core begins to approach saturation. A small increase in voltage produces a disproportionately large increase in magnetizing current. The "knee point" is conventionally defined as the point where a 10% increase in voltage causes a 50% increase in exciting current (per IEC 60044-1 for protection CTs; for power transformers the definition is analogous but applied to the magnetizing characteristic). Modern grain-oriented silicon steel typically achieves a knee at 1.15-1.4 times rated flux density (1.5-1.7 Tesla).
Region III — Deep Saturation (Above Knee): Beyond the knee, the incremental permeability drops sharply toward that of free space (μ₀ = 4π × 10⁻⁷ H/m). The core becomes effectively an air-core inductor. Magnetizing current can reach 6-8 times rated current, limited primarily by winding resistance and system impedance rather than core reactance.
1.2 Why the V-I Curve Matters for Protection Engineers
The V-I characteristic is not merely academic — it directly determines:
- Whether differential protection will trip on inrush (harmonic content ratios)
- Whether a protection CT will faithfully reproduce primary current during faults
- Whether a capacitor voltage transformer (CVT) / inductive VT will oscillate ferroresonantly
- The open-circuit losses and temperature rise of the core
2. Magnetizing Inrush Current: The V-I Curve in Transient Action
2.1 Mechanism
When a transformer is energized at an unfavorable point on the voltage wave (e.g., voltage zero-crossing), the core flux must integrate from zero to twice the normal peak flux (2Φₘ) to satisfy the steady-state condition. Since power transformer cores are typically designed to operate near the knee of the magnetization curve (typically Bₘ ≈ 1.65-1.72 T for modern CRGO steel, with saturation at ≈ 2.0-2.1 T), this 2Φₘ excursion drives the core deep into Region III saturation.
The resulting magnetizing current can reach:
- 6-8 × rated current for modern grain-oriented steel cores
- 8-12 × rated current for older hot-rolled steel cores
- Duration: typically 0.1-1.0 seconds, decaying with the system time constant (L/R)
2.2 Remanence Effect
The worst-case inrush occurs when:
- The core has significant positive remanent flux (+Φᵣ, typically 20-80% of Φₘ after de-energization)
- The transformer is re-energized at voltage zero-crossing with polarity that adds to the remanence
Under these conditions, peak flux can reach:
Φ_peak = 2Φₘ + Φᵣ ≈ 2.5-2.8 Φₘ
Producing inrush currents as high as 10-15 pu.
2.3 Harmonic Signature: The Second Harmonic Restraint Criterion
The asymmetric, unipolar inrush current waveform is rich in even harmonics — particularly the second harmonic (100 Hz in 50 Hz systems). The second harmonic content of inrush current typically ranges from 15% to 65% of fundamental, whereas an internal fault generates negligible even harmonics.
This property is the basis for second harmonic restraint in transformer differential relays:
- If I_2nd / I_fundamental > 15-20%, the relay classifies the event as inrush and blocks tripping
- A typical factory setting is 15%, adjustable to 20% for transformers with low-loss amorphous cores that produce less second harmonic
Critical caveat: Modern low-loss core materials (laser-scribed domain-refined steel, amorphous metal) produce significantly lower second harmonic content during inrush (sometimes as low as 7-12%). Standard 15% second harmonic restraint may fail to distinguish inrush from internal faults. For such transformers, alternative blocking methods should be considered:
- Waveform gap detection (dead-angle principle)
- Fifth harmonic restraint (for overexcitation conditions)
- Flux-based restraint algorithms (requires terminal voltage measurement)
3. CT Saturation: When the V-I Curve Attacks Measurement
3.1 The CT Excitation Characteristic
A current transformer is, electrically, a transformer operating with a current-source primary. Its equivalent circuit consists of an ideal transformer with the magnetizing branch (Xₘ) in parallel with the secondary burden. The CT's own V-I characteristic (more commonly called the excitation characteristic) describes the relationship between secondary terminal voltage (Vₛ) and exciting current (Iₑ).
The secondary excitation voltage is given by:
Vₛ = Iₛ × (R_burden + R_winding + R_lead)
If Vₛ exceeds the CT knee-point voltage (Vₖ), the CT core saturates, shunting an increasing fraction of the secondary current through the magnetizing branch. The result is ratio error and phase error that can cause differential protection to either false-trip (for through-faults with asymmetric CT saturation) or fail-to-trip (for internal faults).
3.2 CT Saturation from DC Offset
The DC component of the primary fault current (asymmetric fault, energization) is particularly dangerous because it pushes the CT flux excursion asymmetrically around the B-H loop. Even a modest DC offset (time constant 50-150 ms) can accumulate enough flux-time area to saturate a CT that would otherwise handle the symmetrical AC component comfortably.
The required CT dimensioning for avoiding saturation with DC offset follows:
K_tf = (X/R + 1) × I_fault / I_rated
Where K_tf is the transient factor and X/R is the system X/R ratio at the fault location. For a 220 kV substation with X/R = 15, the transient factor is 16 — meaning a CT must handle 16 times rated current to avoid saturation.
3.3 Knee-Point Voltage Selection (Practical Rule)
For transformer differential protection, a widely-used practical rule is:
Vₖ ≥ K × I_sn × (R_CT + R_lead + R_relay)
Where:
- K = 2 for Class P CTs (IEEE C-class equivalent)
- I_sn = rated secondary current (typically 1A or 5A)
- R_CT = CT secondary winding resistance
- R_lead = total lead loop resistance (both directions)
- R_relay = relay burden
For high-impedance differential schemes, the requirement is more stringent:
Vₖ ≥ 2 × I_f_max × (R_CT + R_lead)
Where I_f_max is the maximum through-fault current in secondary terms.
4. Ferroresonance: Self-Sustaining Oscillation on the V-I Curve
4.1 Mechanism
Ferroresonance is a nonlinear resonant phenomenon that occurs when a saturable inductance (transformer or VT winding) is connected in series or parallel with capacitance (cable capacitance, grading capacitance of circuit breakers, CVT capacitor divider). The nonlinear V-I characteristic of the iron core provides the essential ingredient.
Under certain system configurations — notably an ungrounded transformer winding connected to cable through an open breaker pole or a single-pole switching event — the series LC circuit can settle into one of several stable operating modes:
- Fundamental frequency ferroresonance: 50/60 Hz oscillation, overvoltage 1.5-3.0 pu
- Subharmonic ferroresonance: Frequency = f₀/n (typically n=3), dangerously high currents
- Quasi-periodic or chaotic ferroresonance: Non-repeating waveform, unpredictable
4.2 Conditions for Ferroresonance
Ferroresonance requires:
- A saturable inductance (transformer, VT)
- A capacitive element (cable, bus capacitance, grading capacitor)
- A low-loss circuit (low resistance damping)
- A triggering event (switching, fault clearing, fuse blowing)
The critical condition is approximated by:
X_C / X_m < 1 (where X_m is the unsaturated magnetizing reactance)
When the capacitive reactance is smaller than the unsaturated magnetizing reactance, the circuit can "jump" from the capacitive stable point to the inductive saturated stable point through the ferroresonant region.
4.3 Mitigation Strategies
| Strategy | Mechanism | Application |
|---|---|---|
| Damping resistor | Dissipates resonant energy | Permanently connected resistor on VT secondary (open-delta) or tertiary winding |
| Load resistor | Lowers Q-factor of resonant circuit | Resistive burden permanently connected to auxiliary VT winding |
| Three-phase simultaneous switching | Prevents single-phase energization | Gang-operated breakers, controlled switching |
| Grounded-wye primary (transformers) | Prevents series LC path | Solidly grounded neutral eliminates floating neutral resonance mode |
| Anti-ferroresonance CVT design | Built-in damping circuit | Modern CVTs per IEC 61869-5 with resonance suppression |
Standard damping: A resistor of 25-100 Ω connected across the open-delta VT secondary (rated for continuous duty) has proven effective in preventing VT ferroresonance on isolated-neutral systems.
5. Practical Implications for Production Engineers
5.1 Factory Testing
Every power transformer should undergo a magnetizing current measurement as part of routine testing (IEC 60076-1, Clause 10.6). A V-I curve is recorded at rated frequency, typically from 5% to 110% of rated voltage in 10-15 steps. The curve shape is a sensitive diagnostic for:
- Shorted core laminations (excessive current at all voltage levels)
- Core joint gaps (localized saturation points)
- Inter-turn winding short circuits (increased exciting current)
- Residual magnetization (shifted zero-crossing)
For production QA, compare the V-I curve against a fingerprint curve from the first unit of the design — deviations exceeding 10% warrant investigation.
5.2 Field Testing: CT Excitation Curve
CT excitation (V-I) testing is one of the most valuable field diagnostic tests. Using a variable-frequency or variable-voltage test set:
- Apply voltage to CT secondary with primary open-circuited
- Record exciting current at each voltage step
- Plot V-I curve and identify knee point
- Compare with factory data and between phases
A shift in the knee point to lower voltage indicates shorted turns in the CT secondary winding — a critical finding that demands CT replacement before placing the transformer in service.
FAQ
Q: What exactly is the "knee point" on a transformer V-I curve and how is it defined?
The knee point (also called the saturation point) is the point on the excitation curve where a 10% increase in applied voltage produces a 50% increase in exciting current. This is the formal IEC 60044-1 / IEEE C57.13 definition for protection CTs. For a power transformer's own magnetizing characteristic, the knee point typically occurs at 110-120% of rated voltage, corresponding to a flux density of 1.8-2.0 T in the core. Above the knee, the core incremental permeability drops dramatically, and the magnetizing current waveform becomes sharply peaked and rich in harmonics. Knowing the knee point is essential for setting overexcitation protection (Volts/Hz relay) and for predicting inrush current severity.
Q: Why does second harmonic restraint sometimes fail for modern transformers?
Modern transformer cores use domain-refined (laser-scribed) grain-oriented silicon steel and amorphous metal alloys that produce a much narrower B-H hysteresis loop. These low-loss materials generate significantly less second harmonic content during inrush — sometimes as low as 7-12% of fundamental. A standard 15% second harmonic restraint threshold will not block tripping for these transformers, potentially causing false trips on energization. Mitigation strategies include: lowering the restraint threshold to 10% (with risk of reduced sensitivity for internal faults with CT saturation), adding waveform symmetry detection, using fifth harmonic for cross-blocking, or implementing flux-based differential algorithms. Always consult the transformer manufacturer's inrush test data when commissioning low-loss transformers.
Q: How can I tell if my CT is going to saturate during an external fault?
Calculate the required CT secondary voltage during the maximum through-fault: V_s = I_f_max × (R_CT + 2 × R_lead + R_relay), divided by the CT ratio. Compare this to the CT rated knee-point voltage V_k. If V_s / V_k > 0.5 for a Class 5P CT (or > 1.0 for a Class PX CT), the CT is at risk of saturation, particularly if the fault contains a DC offset. A practical rule: for transformer differential protection, specify Class PX (IEC) or Class X (BS) CTs with V_k ≥ 2 × V_s for high-impedance schemes. For low-impedance numerical relays, Class 5P20 CTs are generally adequate if the knee-point voltage satisfies V_k ≥ (R_CT + R_burden) × I_sn × accuracy limit factor.
Q: What conditions in my substation could trigger ferroresonance?
Watch for these ferroresonance-prone configurations: (1) an ungrounded-wye transformer winding connected through a length of underground cable to an open circuit breaker, (2) a voltage transformer on an isolated-neutral bus with grading capacitors across the breaker, (3) single-phase switching or fuse blowing on a lightly loaded transformer with significant cable capacitance, (4) a transformer with a delta tertiary that is unloaded and connected via an open breaker pole. In all cases, the key ingredient is a series LC circuit where C (cable/bus capacitance) and L (unsaturated transformer magnetizing inductance) form a resonant circuit that can be triggered by a switching transient. If your substation has any of these configurations and you observe sustained overvoltage (1.5-2.5 pu) with distorted waveforms after switching events — investigate ferroresonance.
Q: How is the transformer V-I curve used during factory acceptance testing (FAT)?
During FAT, the V-I (magnetizing) curve is recorded at rated frequency across a range of applied voltages, typically 5% to 110% of rated in 10-15 steps. The curve is compared to the design baseline. A shift of more than ±10% in exciting current at a given voltage, or a "kink" in the otherwise smooth curve, indicates a manufacturing or transport defect: shorted core laminations increase current at all voltage levels, core joint gaps create localized saturation knees, inter-laminar insulation damage increases eddy-current losses, and transport damage to core clamping can produce irregular curve shapes. For three-phase transformers, all three phases should produce similar V-I curves; a deviation >10% between phases indicates an anomaly in the deviating phase.
Q: What's the relationship between the V-I characteristic and the B-H hysteresis loop?
The V-I characteristic measured at the terminals is the time-averaged, RMS representation of the magnetizing behavior described by the B-H loop. In sinusoidal steady-state excitation, voltage V_RMS is proportional to B_peak × frequency × turns × core area (V = 4.44 × f × N × B_peak × A), and current I_RMS is proportional to H × magnetic path length / N. However, the V-I curve is an RMS representation — it averages the non-sinusoidal magnetizing current waveform — while the B-H loop shows the instantaneous relationship including hysteresis. The V-I curve is what a field engineer can measure with standard instruments; the B-H loop requires an oscilloscope or specialized fluxmeter. Both curves share the same knee-point characteristic, but the B-H loop additionally reveals coercivity and remanence, which are critical for predicting inrush severity.
Glossary
| Term | Definition |
|---|---|
| V-I Characteristic | Plot of applied voltage (RMS) vs. magnetizing current (RMS) for a transformer or CT |
| Knee Point | Point at which 10% voltage increase causes 50% current increase (IEC definition) |
| Inrush Current | Transient magnetizing current during transformer energization, 6-12 × rated |
| Second Harmonic Restraint | Differential relay logic that blocks tripping when 2nd harmonic exceeds 15-20% of fundamental |
| Ferroresonance | Nonlinear resonance between saturable inductance and capacitance |
| CT Saturation | Core saturation of current transformer causing ratio/phase error |
| Remanence | Residual flux retained in core after de-energization (20-80% of Φₘ) |
| Permeability (μ) | Ratio of B to H in the core material; drops sharply above knee point |
References / Standards
| Reference | Title |
|---|---|
| IEC 60076-1:2011 | Power transformers — Part 1: General |
| IEC 60076-5:2006 | Power transformers — Part 5: Ability to withstand short circuit |
| IEC 60044-1:2003 | Instrument transformers — Part 1: Current transformers |
| IEC 61869-2:2012 | Instrument transformers — Part 2: Additional requirements for current transformers |
| IEC 61869-5:2011 | Instrument transformers — Part 5: Additional requirements for capacitor voltage transformers |
| IEEE C57.13-2016 | IEEE Standard for Requirements for Instrument Transformers |
| IEEE C57.105-2019 | IEEE Guide for Application of Transformer Connections in Three-Phase Electrical Systems |
| IEEE C37.91-2021 | IEEE Guide for Protecting Power Transformers |
| IEC TR 61869-102:2014 | Instrument transformers — Ferroresonance in inductive voltage transformers |
*Authored by Du Fu, Production Engineer at ZY POWER. All technical assertions verified against IEC/IEEE standards and field commissioning experience. For questions or corrections, contact the ZY POWER Engineering Team.*
Download This Guide as PDF
Save this technical guide for offline reference. Includes all tables, specifications, and contact information.
Related Articles
Buchholz Relay Guide: Gas Accumulation (Alarm) vs. Oil Surge (Trip), Installation Slope 1-1.5%, DIN 42566 & Fault Gas Interpretation
The Buchholz relay — named after its inventor Max Buchholz (1921) — is the simplest, most reliable, and most widely used internal-fault detector for oil-immersed conservator-type transformers. Installed in the pipe connecting the main tank
Power Quality Fundamentals: Harmonics, Voltage Sags, Flicker, and Filtering Solutions
Power quality is the silent profit-killer of industrial plants. A single voltage sag lasting 100 milliseconds can drop a continuous process line, causing hours of restart time and tens of thousands of dollars in scrap product. Harmonic dist
Transformer Differential Protection: Percentage Restraint Characteristic, CT Matching & Yd Vector Group Compensation, 2nd/5th Harmonic Blocking, and REF Coordination
Transformer differential protection (ANSI 87T) is the primary fast protection for internal phase-to-phase and phase-to-ground faults in power transformers. Unlike a generator or busbar differential scheme — where the protected zone is elect