Tiza Group

Home / Guides / Standards and grid connection

Overvoltages Transferred Through a Transformer

How overvoltages pass from the high-voltage to the low-voltage winding (capacitive, inductive and temporary transfer) and when an arrester is needed.

By Shadai García, CEO and protection engineer at Tiza Group.

winding Acapacitiveinductivewinding BLVarresterloada transferred overvoltage usually has a capacitive and an inductivecomponent, superimposed on the power-frequency voltageprimary (winding A)secondary (winding B)capacitive: first, in the MHz rangeinductive: later, with changing shape and amplitudecapacitive division: from 0 to at least 0.4ChainWave at ATransferWave at BLV arrester© tizagrp.comwinding Acapacitiveinductivewinding BLVarresterloada transferred overvoltage usually has a capacitive and an inductivecomponent, superimposed on the power-frequency voltageprimary (winding A)secondary (winding B)capacitive: first, in the MHz rangeinductive: later, with changing shape and amplitudecapacitive division: from 0 to at least 0.4ChainWave at ATransferWave at BLV arrester© tizagrp.com
Overvoltage transferred through a transformer: capacitive and inductive paths from winding A to B. Location in the system and operating principle. Illustrative: real values come from the nameplate and the manufacturer's manual.
Short answer

Equipment connected to the low-voltage side of a transformer fed by an overhead line is not free from lightning and switching overvoltages: they pass from the high-voltage to the low-voltage winding by capacitive and inductive transfer, and temporary ones by voltage unbalance. The effect depends on the transformer construction, its connection and the load on the receiving side.

Short answer

Equipment connected to the low-voltage side of a transformer is not free from the overvoltages of the overhead line on the high-voltage side. Lightning and switching surges pass from one winding to the other, partly through the capacitances between windings and partly through magnetic coupling. Earth faults on the primary side can also raise the secondary voltage. IEC 60071-2 treats the subject in its annex E and recalls, in clause 7.2.1.1, that these effects must be considered in insulation co-ordination, with low-voltage-side arresters if needed.

Three transfer paths

Annex E describes three transfer modes:

  • Electrostatic or capacitive. It passes through the capacitances between windings.
  • Oscillatory. It occurs through the natural oscillations of the primary or secondary circuits. It is normally small and of secondary importance, unless amplified by resonance. The standard does not develop it.
  • Normal electromagnetic. It depends mainly on the transformation ratio, the leakage inductance and the load impedance.

A transferred overvoltage usually has a capacitive and an inductive component, superimposed on the power-frequency voltage. The capacitive one is in the megahertz range and is seen first in the transferred wave. The inductive one comes later, and its shape and amplitude change over time because the voltage distribution in the primary winding also changes.

There is one special case: the capacitive rise of the neutral potential during earth faults and other asymmetrical events. It appears in transformers with an exceptionally large turns ratio between high and low voltage, such as generator transformers or those with a tertiary, and when the capacitance on the low-voltage side is low.

What the transferred value depends on

Annex E lists the factors that matter:

  • the construction of the transformer, above all that of its windings (disc, interleaved) and their order around the core legs;
  • the leakage inductances and the damping of the windings;
  • the transformer capacitances;
  • the winding connection (vector group) and the connection to the network;
  • the shape of the incoming wave.

Some of these factors are hard to calculate. For that reason the standard points out that the most practical way to obtain a quantitative estimate is to measure, for example with a recurrent surge test. The equations it presents serve only as a rough estimate, and for specific values the data must be requested from the manufacturer. They are not reproduced in this guide.

Transferred temporary overvoltages

An earth fault on the primary produces unbalance in the phase-to-earth voltages. This can cause phase-to-earth overvoltages on the secondary when it has an isolated neutral and a markedly low rated voltage compared with the primary.

Their magnitude depends on the primary voltage during the fault, on the ratio between the transformer capacitances and the additional capacitors connected to the secondary. The capacitances come from the routine test records of the transformer.

They can be excessive if the phase-to-earth capacitance of the secondary is too low. The standard gives an example: in a 110 kV transformer, the standard power-frequency withstand voltages can be exceeded if the rated voltage of the secondary is 10 kV or less. They are also excessive if the secondary with isolated neutral is left completely disconnected from the network during an earth fault on the primary side.

The corrective measure is to connect additional capacitors between phase and earth on the three phases of the secondary.

Overvoltages transferred by the capacitive path

They are almost always critical in the high-to-low-voltage direction. They come from the rise in potential of the primary winding due to an incoming fast-front or slow-front overvoltage. They pass to the secondary through the winding capacitances.

With fast variations of the primary voltage, only the parts of the winding near the terminals take part, so the distributed capacitances of the winding are decisive. Only the manufacturer can give the value of the series capacitance of the winding.

The result is a capacitive divider. The division ratio can range from zero to at least 0.4. A delta-connected low-voltage winding with a wye-connected high-voltage winding reduces that value further. In addition, a factor that depends on the overvoltage class and on the winding connection modifies the result:

Primary overvoltage classPower-frequency voltage factor
Slow frontEqual to 1, whatever the connection
Fast frontGreater than 1: about 1.15 for wye-delta or delta-wye, about 1.07 for wye-wye or delta-delta

For the primary value of a fast-front overvoltage, the arrester protection level on the primary side can be taken. For a slow-front one, the peak value of the phase-to-earth voltage, assuming the arresters do not operate.

Winding losses and the connected load damp the capacitive peaks. They are critical above all in transformers with large step-down ratios and when little capacitance is connected to the secondary. Arresters connected to the secondary limit these voltages effectively, and additional capacitors improve the protection, above all for equipment that does not tolerate fast rise times, such as generators and motors.

Overvoltages transferred by the inductive path

This is normally the most important transfer mode and occurs already with moderate variations of the voltage. In this mode the transformer works essentially in its normal mode, so power-frequency methods apply. The difficulty is obtaining the transformer parameters, which is why direct measurement gives more reliable information.

The magnitude depends on:

  • the primary voltage, including arrester operation;
  • the duration of the incoming overvoltage;
  • the transformer characteristics (number of windings, ratio, short-circuit impedances, vector group);
  • the surge impedances of the lines connected to the secondary;
  • the load characteristics.

The response of the secondary circuit is the hardest part to predict. For disc-winding transformers connected to an overhead line without appreciable load, the standard gives typical values of the secondary response factor ranging from 0.3 to 1.3 for fast fronts, depending on the secondary voltage. With three-winding transformers, values even above 1.7 have been recorded. When the secondary is a cable, the usual values do not exceed 1.0. For specific values the manufacturer must be consulted.

Inductive high-to-low overvoltages can be critical if:

  • the secondary winding is not connected to the network;
  • the secondary has a low rated voltage and high power, as in generator step-up transformers;
  • the winding is the tertiary of a three-winding transformer.

They can also be dangerous for the phase-to-phase insulation of delta secondaries, even if all terminals have phase-to-earth arresters. In that case phase-to-phase arresters may be needed. Adding capacitors normally does not reduce inductively transferred overvoltages. High voltages can also occur when an overvoltage passes from the low-voltage to the high-voltage winding, above all if it causes resonant voltage rises.

What this means for the design

Phase-to-earth and phase-to-phase protection are studied case by case, with information from the transformer manufacturer. In general, arresters between all phases and earth, and between phases when needed (for example on wye-delta transformers), give adequate protection. The overvoltage picture of an industrial installation is in transient overvoltages in the plant.

To go further, see Temporary overvoltages: causes and typical values, Very fast front overvoltages (VFTO) in GIS, Lightning overvoltages in a substation, Power transformer: parts, functions and tests and Slow-front overvoltages: switching and fault.

When to contact us

If a transformer feeds a motor, a generator or a medium-voltage switchboard from an overhead line, or insulation failures have already occurred on the low-voltage side, the transferred overvoltage is defined with the manufacturer’s data and a study. We carry out your insulation co-ordination study and quote the scope, remotely and worldwide.

References

  • IEC 60071-2 (1996 edition), Insulation co-ordination — Part 2: Application guide, annex E and clause 7.2.1.1.

Frequently asked questions

Why does a high-voltage overvoltage reach the low-voltage side of the transformer?

By three paths: capacitive transfer between windings, which appears first and in the megahertz range; inductive transfer, which follows it and is the most important; and, in earth faults, the transfer of the temporary voltage through unbalance of the primary voltages.

Does adding capacitors on the low-voltage side reduce all transferred overvoltages?

Not all of them. Capacitors reduce those transferred by the capacitive path and the temporary ones. According to the standard, they normally do not reduce those transferred by the inductive path.

When is an arrester needed on the low-voltage side?

When the low-voltage winding is not connected to the network, when it has a low rated voltage with high power (for example generator step-up transformers) or when it is the tertiary of a three-winding transformer. Each case must be studied with manufacturer data.

WhatsApp