Ferroresonance in Power and Instrument Transformers

Background

The term “ferroresonance” appeared in the technical paper by Boucherot (1920), to explain two possible operating conditions of the transformer core. Since then extensive experiments and research works have been devoted to describe the phenomenon, Iravani, et al. (2000). However, due to its highly nonlinear nature, the ferroresonance phenomenon has neither been well nor widely understood and still there exists misconceptions and unclear aspects of ferroresonance in the engineering community.

Ferroresonance is a special case of the resonance phenomenon, and can occur when a non-linear inductance is connected in series or parallel with a capacitance. In a linear circuit, the resonance occurs when the capacitive reactance equals the inductive reactance at the circuit source frequency and can result in excessive currents and voltages. However, due to the inherent nonlinearity of the ferroresonance phenomenon, several steady state solutions may exist for a particular excitation condition and the range of circuit parameters. It is also possible that a system disturbance causes the circuit normal, steady-state operating condition to migrate to another stable operation point with very high current and/or voltage magnitudes, i.e., ferroresonance operating point.

In the majority of ferroresonance cases, a series path including a saturable inductance and a capacitance is formed, and constitutes a series ferroresonance circuit. Another type of ferroresonance can occur during temporary power frequency overvoltage conditions. Under a normal three-phase operation, the magnetizing inductance of the transformer is in parallel with the system capacitance and if the transformer voltage is held below the saturation point, ferroresonance does not occur. However, during temporary power frequency overvoltage conditions, if the system voltage is not maintained below the core saturation point, the core is saturated and an exchange of energy between the system capacitance and the highly nonlinear magnetizing inductance of the transformer can occur. The rapid changes in core flux during this period can produce high overvoltages. Since in this case the reactance and capacitance are in parallel, this second type of ferroresonance is considered as parallel ferroresonance. An example of this type is the ferroresonance phenomenon of the inductive voltage transformer (VT) in an isolated neutral system.

Both types of ferroresonance can cause abnormal voltage (either low or high), across the transformer terminals and from terminals to ground. The high abnormal voltage due to ferroresonance is accompanied by abnormal transformer sound and, if sufficiently high, by equipment damage. The occurrence of both types of ferroresonance is often unpredictable as both depend on various parameters such as the cable length, the amount of system capacitance, the connection type and saturation characteristics of the transformers, and the amount of load or burden, Ferracci (1998). However, the occurrence of the phenomenon requires:

• 1.

  A non-linear inductance; which is the saturable iron core of the transformer or the reactor.

• 2.

  A capacitance; This can appear in the form of voltage grading capacitors in high voltage circuit breakers, conductor inter-phase capacitances, capacitance to ground of cables and long lines, series and shunt capacitor banks, CVTs, or even the stray capacitance of the busbar and the transformer windings.

• 3.

  Low power loss condition; for example very lightly loaded transformer with a modern low loss magnetic core.

• 4.

  A voltage source; which is the system supply voltage and required to provide energy for a sustained ferroresonance oscillations.