Power System Fundamentals · Training Guide

Resonance and Ferroresonance in Power Systems Part Four — Mitigation of Ferroresonance

Reading time ≈ 28 min

Start here

The mitigation principle

Ferroresonance can occur at any voltage level when three main conditions exist together: sufficient capacitance, nonlinear inductance, and low losses in the circuit.

The capacitance may come from cables, overhead lines, GIS, transformer bushings, grading capacitors, open circuit breakers, or stray capacitances. The nonlinear inductance usually comes from a saturable magnetic device such as a voltage transformer, power transformer, or reactor. Low losses allow the oscillation to continue because there is not enough damping to remove energy from the circuit.

Mitigation is therefore based on a simple principle: either prevent the risky circuit configuration from forming, or add enough damping so that any oscillation dies out quickly.

What this page teaches

why damping is the main mitigation principle;
how open-delta resistors suppress VT ferroresonance;
when wye-connected resistors are used;
how saturable reactors can help;
why some methods are passive and others are operational;
why power-transformer ferroresonance needs a switching and energisation strategy;
why mitigation must be validated by EMT simulation.

Three ways to mitigate

Ferroresonance mitigation works by changing the damping, detuning the circuit, or changing the switching condition — every method is a variation of one of these three.

Table 1 — Ferroresonance mitigation methods with their purpose, application and cautions.

Mitigation Method	Main Purpose	Typical Application	Main Caution
Open-delta resistor	Add damping to the VT secondary circuit	VT ferroresonance	Thermal rating and continuous duty
Wye-connected resistor	Provide a phase damping path	Unearthed systems / VT circuits	Losses and insulation duty
Saturable reactor	Limit the abnormal ferroresonant state	VT mitigation	Correct sizing and saturation behaviour
Delta tertiary loading	Add a damping path	Transformer ferroresonance	Thermal and operational implications
Soft energisation	Avoid a severe transient trigger	Black-start / weak system	Requires operational control
Switching procedure	Avoid a risky topology	Lines, transformers, VTs	Must be practical and enforceable

Valid only for what was studied

A mitigation method is only valid for the topology and switching condition studied — no method is universally suitable.

Design and layout

Passive methods for VT ferroresonance

Voltage transformer ferroresonance can be mitigated using passive or active methods. Passive methods aim to prevent dangerous configurations from occurring. Active methods allow the configuration to exist but introduce losses or switching actions to suppress the oscillation. In practice, both approaches may be needed.

A good design may reduce the probability of ferroresonance, but maintenance activities, switching errors, circuit-breaker replacement, substation extensions, or added capacitance can unintentionally increase the risk later.

Passive mitigation starts with the substation layout and switching arrangement. Risky layouts should be avoided where possible. This includes arrangements where an electromagnetic VT can be left connected to an isolated busbar or line section that is still weakly energised through grading capacitors or stray capacitance.

One practical design approach is to install VTs on the source side of disconnectors or in a position where they are not left trapped between open switching devices. This reduces the chance that a VT remains connected to a small capacitance-fed island.

Capacitive voltage transformers may be used where suitable because they normally include ferroresonance-suppression circuits. Electromagnetic VTs with higher saturation points may also reduce the likelihood of entering ferroresonance, although they do not remove the risk completely.

Interlocking and carefully defined switching sequences can also reduce the probability of dangerous configurations. For example, switching logic can be arranged so that a VT is not left energised through breaker grading capacitors after a busbar or line is isolated.

However, passive methods should not be treated as permanent guarantees. Substation modifications, circuit-breaker upgrades, changes in VT type, or additional connected equipment can change the capacitance and create a new ferroresonance risk.

Damping devices

Active methods for VT ferroresonance

Active mitigation methods introduce damping or change the circuit condition after a risky configuration occurs.

Common active methods include resistors connected in the VT secondary circuit, resistors in open-delta connections, resistors in wye-connected secondary circuits, saturable reactors in series with resistors, grounding switches, and air-core reactors on the high-voltage side.

The purpose of these devices is to remove energy from the ferroresonant circuit. If enough energy is dissipated, the oscillation cannot continue.

Active methods must be selected carefully because they can affect VT burden, metering accuracy, protection performance, insulation requirements, and thermal loading.

Common damping method

Open-delta resistor

An open-delta resistor is one of the most common methods used to damp VT ferroresonance.

The open-delta connection is sensitive to zero-sequence voltage. Under balanced normal operation, the open-delta voltage is low, so the resistor does not significantly affect normal metering or protection. Under unbalanced or ferroresonant conditions, zero-sequence voltage appears and the resistor provides a damping path.

The resistor value must be selected carefully. If the resistance is too low, damping is strong, but the VT thermal burden may become excessive. If the resistance is too high, the damping may not be sufficient to suppress ferroresonance quickly.

This method is effective for many VT ferroresonance cases, but it is not universal. It may be less effective for some sub-harmonic ferroresonance modes. Therefore, the damping requirement should be checked against the specific ferroresonance mode being studied.

Phase damping

Wye-connected resistor and saturable reactor

Wye-connected secondary resistor

Another mitigation option is to connect resistors across each phase of the VT secondary winding.

This method introduces damping directly into each phase. However, it may introduce continuous losses during normal operation. It may also affect the accuracy of metering or protection circuits if the burden becomes too high.

For this reason, wye-connected resistors may be switched in only during specific conditions, such as when the VT is de-energised or when a ferroresonance risk is detected.

This approach can be useful, but it requires careful coordination with VT accuracy class, relay inputs, metering burden, thermal capability, and switching logic.

Wye resistor with saturable reactor

A wye-connected resistor may be combined with a saturable reactor to create a nonlinear damping arrangement.

Under normal operation, the saturable reactor presents high impedance. This prevents the resistor from interfering with metering or protection accuracy.

Under ferroresonant overvoltage conditions, the reactor saturates. Its impedance falls, and the resistor is effectively inserted into the circuit. This provides damping when it is needed.

This method can be accurate and efficient because it is selective. However, it is case-specific. The saturable reactor and resistor must be designed for the expected voltage, frequency content, thermal duty, and ferroresonance mode.

Wider toolkit

Other VT mitigation options

Another option is to specify circuit breakers without grading capacitors where this is technically and commercially practical. Removing grading capacitors removes one possible weak energisation path across an open breaker.

Optical or electronic voltage transformers can also remove the ferroresonance mechanism associated with the VT's own magnetic core. Since they do not rely on the same saturable iron-core measurement transformer, they avoid the conventional electromagnetic VT ferroresonance mechanism.

However, this does not mean that the bay, busbar, or surrounding substation arrangement is immune to ferroresonance. Other magnetic devices, such as power transformers, reactors, CVTs, or adjacent conventional VTs, may still participate in ferroresonance if they are connected to capacitance and low damping.

Capacitive voltage transformers can reduce some external network-driven VT ferroresonance risks, but they should be used with understanding. CVTs have their own internal ferroresonance mechanism and rely on suppression circuits. They may also have limitations in some line-reclosing applications.

Interlocking, switching procedures, and operator guidance are also important. Many ferroresonance events occur not because the equipment is poorly designed, but because the switching sequence temporarily creates a weakly energised nonlinear LC circuit.

VTs with higher saturation points are harder to drive into ferroresonance. However, if ferroresonance does occur, the resulting overvoltage may be higher before the core saturates and limits the voltage. Therefore, a high saturation point is not a complete mitigation method by itself.

Floating neutral

VT ferroresonance in unearthed or arc-suppressed systems

Unearthed and arc-suppressed systems can be especially sensitive to VT ferroresonance because the neutral point is not strongly fixed to earth.

One mitigation approach is to use VTs designed for low induction. This helps avoid core saturation during transients. However, the VT must still be able to withstand high phase-to-earth voltage, which may reach several times the normal value under neutral displacement conditions.

Resistive damping can also be applied in the open-delta circuit. The resistor may be permanently connected or switched in when required. A permanent resistor is simple, but it must be thermally rated and must not interfere with normal operation. A switched resistor may reduce continuous burden, but it adds mechanical and control complexity.

Another option is to add parallel capacitance. This can increase the zero-sequence capacitance and shift the capacitance-to-inductance relationship away from the resonant band for the fundamental-frequency mode.

However, this must not be treated as a general rule that more capacitance is always safer. Adding capacitance may suppress one ferroresonance mode while moving the system closer to another mode, including a sub-harmonic mode. For this reason, added capacitance should only be used after study verification, not as a rule-of-thumb solution.

Special bus VT configurations can also be used. One arrangement uses two line-to-line VTs and one phase-to-earth VT with a higher voltage rating. A resistive burden may then be applied across the open-delta circuit to provide damping.

Modern digital relays can sometimes increase the risk because they impose a very low burden compared with older electromechanical devices. Lower burden means less damping. In floating-neutral systems, it may also be difficult to add damping unless suitable secondary windings are available.

Configuration control

Avoiding conditions that promote power transformer ferroresonance

Power transformer ferroresonance can occur in transformer-terminated lines, weak-source energisation, single-phase switching, multi-circuit EHV corridors, and series-compensated networks.

Mitigation can be grouped into three main strategies: avoid the conditions that promote ferroresonance, minimise the energy transfer that sustains the oscillation, and control the duration of the abnormal condition through switching or protection.

The most direct mitigation method is to avoid configurations that are known to create ferroresonance risk.

Long de-energised lines should not be left connected to transformer terminals if a nearby energised circuit can supply energy through capacitive coupling. If the critical line length is exceeded, circuit breakers on both sides of the transformer may be required so that the transformer can be fully isolated.

Low flux-density transformer design can make saturation less likely. However, this is often impractical because it increases transformer size, cost, and material requirement.

In multi-circuit corridors, another option is to disconnect the parallel live circuit before isolating the transformer. This removes the capacitive energy source. However, it may require a double-circuit outage and therefore must be checked against system security and stability requirements.

Reducing sustaining energy

Minimising energy transfer in power transformer circuits

If the risky configuration cannot be completely avoided, the next strategy is to reduce the energy available to sustain the oscillation.

Artificial losses can be introduced during switching using resistors. These resistors absorb energy and help the oscillation decay.

A loaded delta tertiary winding can also provide a damping path. Connecting a suitable resistive load to the tertiary winding can absorb zero-sequence or harmonic energy and reduce the risk of sustained ferroresonance.

Phase transpositions can also influence energy transfer. Adjusting phase transpositions may reduce inter-circuit coupling and therefore reduce the amount of energy transferred from an energised circuit to a de-energised transformer-terminated circuit.

These methods do not necessarily eliminate all possible ferroresonance conditions, but they reduce the chance that the oscillation will be sustained.

Operational control

Controlling duration through operational switching

Another mitigation strategy is to allow a temporary abnormal condition to exist only briefly, and then quench it through switching action.

Disconnector operation may be used to isolate the transformer. However, this must be treated carefully because ferroresonant current may cause arcing, contact damage, or unacceptable stress on the disconnector.

Earth switches can be used to collapse the voltage and remove the ferroresonant condition. In some schemes, earth switches may be closed automatically when resonance is detected. This requires appropriate equipment ratings and careful protection coordination.

Disconnecting the parallel circuit can remove the energy source that sustains the oscillation. However, this may cause a double-circuit outage and must be assessed against system operation and stability requirements.

Single-phase switching should be avoided where it creates ferroresonance risk. Replacing fuses or single-pole switches with three-phase circuit breakers can reduce the probability of partial energisation.

Cable length should also be limited where possible. Installing the circuit-breaker cubicle close to the transformer terminals can reduce cable capacitance and therefore reduce ferroresonance risk.

Adding resistive load is another practical method. The load provides damping and helps prevent the oscillation from continuing.

Restoration risk

Black-start and weak-network energisation

Black-start and weak-network restoration conditions require special attention because the source is weak and damping may be low.

Soft energisation, also called controlled voltage build-up, can be used to reduce ferroresonance risk. In this method, the generator is brought up to speed without excitation. The circuit breaker and excitation are then closed, and the voltage is gradually increased using the automatic voltage regulator.

This gradual voltage build-up means the transformer core is fluxed progressively rather than being exposed to a sudden voltage step. This reduces the risk of driving the core into saturation.

Soft energisation also reduces transformer inrush current and temporary harmonic overvoltage. It can therefore support safer energisation of weak networks, auxiliary transformers, and long lightly loaded circuits during restoration.

Why black-start is the worst case

Black-start combines the most dangerous ingredients at once: a weak source, residual core flux, significant line and cable capacitance, and a transformer easily driven into saturation. With little load and low damping to absorb energy, this is when ferroresonance is most likely and most severe — controlled energisation is not optional.

Summary

Final learning points

Key takeaways

Ferroresonance mitigation is based on two principles — avoid dangerous circuit configurations, or add damping so that the oscillation cannot continue.
VT ferroresonance can be mitigated passively — by avoiding risky layouts, using suitable VT types, applying interlocking, and controlling switching sequences to prevent a VT from being left trapped on a capacitance-fed island.
Open-delta resistors are commonly used — because they provide damping mainly under zero-sequence conditions and normally have little effect on balanced normal operation.
Wye-connected resistors provide phase damping — but may introduce continuous burden and affect metering or relay accuracy. They may need to be switched in selectively.
Saturable reactor and resistor combinations provide selective damping — inserting resistance only when ferroresonant overvoltage is present. Design must be case-specific.
Optical or electronic VTs remove the VT's own ferroresonance mechanism — but do not make the surrounding bay or substation immune to ferroresonance from power transformers, reactors, or adjacent conventional VTs.
In unearthed or arc-suppressed systems, low-burden modern relays can reduce damping — and increase ferroresonance susceptibility. Adding parallel capacitance may detune one mode but move the system toward another, and must be verified by study.
Power transformer ferroresonance can be mitigated by avoiding risky configurations — limiting transformer-terminated line lengths, avoiding long unearthed de-energised lines adjacent to energised circuits, and using three-phase breakers instead of single-phase switching.
Artificial losses, loaded delta tertiaries, and phase transpositions can reduce the energy available to sustain oscillation — even when the risky configuration cannot be completely avoided.
Earth switches, reactor switching, and special protection systems can quench ferroresonance — but must be checked against equipment ratings and system-security requirements before relying on them.
Black-start and weak-network energisation are high-risk conditions — soft energisation, or controlled voltage build-up, reduces the risk of core saturation, inrush, and temporary harmonic overvoltage during restoration.
For junior engineers, the key message is simple — ferroresonance mitigation is not only about adding a resistor. It is about understanding the full circuit path, the source of energy, the nonlinear magnetic device, the capacitance, and the available damping.

Mitigation validation checklist

Before a mitigation measure is accepted, confirm that:

the topology and switching condition it was designed for are clearly defined;
the damping or detuning effect has been demonstrated by EMT simulation, not assumed;
the worst-case trigger (open phase, single-pole operation, trapped charge, black-start) has been included;
the device thermal and continuous duty rating is adequate for sustained operation;
normal metering, protection and balanced operation are not degraded;
any operational measure (switching order, interlock) is practical and enforceable in the field;
the measure has been re-checked if the network or equipment later changes.

In conclusion: effective ferroresonance mitigation always traces back to the same three levers — damping, detuning, and switching control — applied to a specific, studied topology and proven by simulation. The final part of this series draws these threads together into a set of practical lessons and a closing assessment checklist.

Part Four Reading now

Mitigation of Ferroresonance

Practical mitigation methods for VTs and power transformers — passive and active damping approaches, open-delta resistors, wye resistors, saturable reactors, and black-start energisation strategies.

Series progress 4 of 5

Part Five

Conclusions and Practical Lessons

Series synthesis — assessment methodology, EMT modelling practices, mitigation decision frameworks, and practical field lessons drawn across all five parts.

Previous Ferroresonance in VTs, Transformers and Network Models

VT ferroresonance scenarios, transformer core saturation mechanisms, grading capacitor effects, busbar and line configurations, and network topologies that sustain or trigger ferroresonant states.