Differential protection · Glossary

Differential protection is a protection principle in electrical power engineering. It compares the currents at the boundaries of a clearly defined protection zone. In fault-free operation, the sum of these currents is ideally zero: what flows into a transformer, a line, a generator or a busbar flows out again elsewhere. If a significant difference arises, the protection device interprets it as an indication of a fault within this zone. It then trips the associated circuit breakers.

The basis is the current balance according to the node principle, that is, Kirchhoff's current law. What is measured is not the fault itself, but its effect on the currents at the measuring points. In the event of a short circuit within the protected item of equipment, part of the current flows off into the fault location. The currents at the zone boundaries then no longer match. If the fault lies outside the zone, high short-circuit currents do flow through the equipment, but they leave again at the measuring points. In this case, the differential protection is meant to remain stable and not trip. Put illustratively: it behaves like a cash register that reconciles inflows and outflows. High turnover is not a problem as long as nothing is missing in the end.

Because of this clear delimitation, differential protection is a particularly selective method of power system protection. Selectivity means that, as far as possible, only the faulty part of the network is disconnected and all other equipment remains in operation. In transmission and distribution networks this is central, because unnecessary disconnections can quickly turn a single fault into a larger supply interruption. A busbar fault in a substation, a winding fault in a transformer or a cable short circuit in a city centre produce high currents, mechanical forces and severe thermal stress. Protection technology must limit such faults quickly without unnecessarily disconnecting adjacent parts of the network.

Protection zone, measurement and tripping

Where the protection zone begins and ends is determined by the location of the current transformers. These instrument transformers convert high operating and short-circuit currents into small measuring currents that protection devices can process. For a transformer, the current transformers are typically located on the high-voltage and low-voltage sides; for a line, at the line ends. For a busbar, the currents of all connected bays are recorded.

From these measured values, the protection device forms a differential current. Even without a fault, in practice it is not exactly zero. Transformer errors, the magnetising current of a transformer, tap positions of a tap changer or the capacitive charging currents of long cables and lines produce small so-called spurious currents. Modern devices therefore work with a stabilised characteristic. In addition to the differential current, they consider a stabilising or restraint current, which represents the magnitude of the through-flowing currents. Tripping occurs only when the ratio of differential current to stabilising current lies in the tripping region of the characteristic. The characteristic typically rises in several sections. This is necessary because current transformers become inaccurate or can saturate at very high currents. Without stabilisation, an external short circuit could appear as an internal fault if the transformers on both sides react differently. Many devices additionally have an unstabilised high-current stage. It trips immediately at very large differential currents, that is, at values that a merely through-flowing short-circuit current cannot reach.

In line differential protection over longer distances, the measured values are recorded at different locations. It therefore needs a communication link between the line ends, which must be fast and reliable. The measured values must be assigned precisely in time. For this, one uses either a propagation-delay measurement over the channel, which requires delays that are as symmetrical as possible in both directions, or an external time reference, for example via satellite time or IEEE 1588 (PTP). The transmission time feeds directly into the tripping time. In the event of communication disturbances, the device must adopt a defined state, for example block or switch over to a backup protection function. In this respect, line differential protection differs from locally operating protection functions that evaluate only measured quantities at one location.

Distinction from overcurrent protection and distance protection

Differential protection is often confused with simple overcurrent protection, because both can trip on short circuits. The difference lies in the decision rule. Overcurrent protection responds as soon as a current exceeds a limit value. Whether the fault lies inside or outside a particular item of equipment is initially unknown to it. Selectivity is achieved there through current steps, time grading or direction detection. Differential protection, by contrast, responds to a current difference within a defined zone. A high current alone does not trip it as long as the balance is correct. This is important for external short circuits, because then very high currents can flow through a transformer or a busbar without that equipment being faulty.

Differential protection must also be distinguished from distance protection. Distance protection calculates an apparent impedance up to the fault location from voltage and current and is used mainly for lines. It manages without continuous communication and grades fault zones. Differential protection, by contrast, needs measurements at all zone boundaries, but in return responds very quickly and unambiguously within the zone. Both methods are often combined, for example as main and backup protection, but fulfil different tasks.

Another common misconception is equating it with the residual current device (RCD) in low-voltage installations. It, too, evaluates a current difference. However, it serves mainly to protect persons and against fire in the case of fault currents to earth, with tripping thresholds in the milliampere range. Differential protection in networks and installations, by contrast, protects equipment and network sections against internal short circuits. This involves significantly higher currents, and the tripping conditions are different.

Particularities of transformers, machines and busbars

For a transformer, differential protection is more demanding than for a simple conductor section. The currents on the two sides differ according to the transformation ratio. In addition, the vector group shifts the phase angle of the currents. Digital protection devices adjust magnitude and phase angle computationally so that the balance is correct for the normal load current.

When a transformer is energised, an inrush current (rush) can occur that reaches a multiple of the rated current. It flows only on the side being switched on and therefore appears entirely as differential current in the protection device. Typical of this current is a high proportion of the second harmonic, that is, twice the system frequency, which is largely absent in a genuine short circuit. Transformer differential protection devices evaluate these harmonics and block or stabilise tripping if a set proportion is exceeded. A similar effect can occur when a parallel transformer is switched in (sympathetic inrush). In the case of overexcitation, that is, overvoltage or underfrequency, the fifth harmonic is often used as an additional criterion. Transformer design and neutral earthing also influence the parameterisation. For earth faults in earthed windings, there is a separate, more sensitive earth-fault differential protection (restricted earth fault).

Besides transformers, generators and large motors are also equipped with differential protection. For these applications, the international standard IEC 60255-187-1 (2021, with corrigendum 2023) sets minimum requirements for function and testing. Further parts of the series of standards deal with busbar differential protection and line differential protection.

For a busbar, the protection task is particularly critical, because many lines, transformers or generators converge there. A fault can therefore mobilise very high short-circuit power. Busbar differential protection must include all connected bays and cope with changing switching states. If a bay is switched to another busbar, a coupler is opened or a busbar is split, the protection zone changes. The protection must know this topology, for example via disconnector positions. Otherwise it either disconnects too much or fails to detect a fault.

Why the term matters in the power system

Differential protection shows that security of supply does not depend on sufficient generation or capable lines alone. A power grid remains manageable only if faults are detected quickly and spatially contained. Differential protection typically decides within a few tens of milliseconds. Together with the breaking time of the circuit breaker, this gives the total fault clearing time. These decisions determine whether a short circuit damages only a single item of equipment or destabilises a larger network area.

Economically, avoided damage, shorter outage times and lower consequential costs count. A transformer fire, a destroyed busbar or a long-persisting cable short circuit causes more than repair costs. It ties up reserves, changes switching states, burdens alternative paths and can affect industrial consumers or critical infrastructure. Fast, selective protection reduces the fault energy and limits the area that must be disconnected.

Organisationally, it involves more than choosing a protection device. Network operators, plant operators, planners, manufacturers and testing service providers must agree on a common protection concept. The current transformers must suit the protection task. Settings must be documented, tested and updated after network modifications. A wrongly set differential protection can be just as problematic as a missing one: it can trip unnecessarily for external faults or fail to respond, or respond too late, to internal faults.

Typical misinterpretations and limits

“Differential protection is simply a particularly fast short-circuit protection”: Speed is a consequence of the principle, not the principle itself. It is fast because it monitors a clearly defined zone and needs no time grading for selectivity. Its decision is based on the current balance, not on the absolute current magnitude.
“It detects every fault in the equipment”: It detects only faults that produce a measurable current difference above the pick-up threshold at the zone boundaries. Very high-resistance faults, inter-turn faults involving only a few turns, gradual insulation ageing, thermal overload or mechanical damage may lie below it. Separate devices exist for these, for example Buchholz relays for oil-filled transformers, temperature monitoring, earth-fault and overload protection, and condition diagnostics.
“Protection is an add-on component fitted afterwards”: Differential protection is closely tied to plant layout, transformer design, communication paths, switchgear logic and network operation. New infeeds, changed short-circuit power or different operating modes can make a review of the protection concept necessary.

The energy transition also changes the boundary conditions. Generators connected via power electronics supply limited and controlled short-circuit currents in the event of a fault, usually only slightly above rated current. Large synchronous generators, by contrast, supply a multiple of their rated current. Battery storage, wind farms and photovoltaic inverters can limit, control or switch off their currents very quickly. The balance logic of differential protection remains valid. However, the available fault current contributions, the burden on the transformers and the margin between fault currents and pick-up thresholds change. Protection concepts must suit the actual mode of operation of the network, not a simplified picture of high short-circuit currents from a few large power plants.

Differential protection does not describe a market rule, a network capacity or general grid stability. It designates a specific protection function that monitors a defined electrical area via current balances. Its strength is the fast, selective detection of internal faults. Its limits lie where measurement, zone definition, communication or the nature of the fault do not produce a clear difference.

Sources

IEC 60255-187-1 Edition 1.0 2021-07: Measuring relays and protection equipment – Part 187-1: Functional requirements for differential protection – Restrained and unrestrained differential protection of motors, generators and transformers (Leseprobe) — IEC (Leseprobe über iTeh Standards); 2021-07. Vorwort und Abschnitt 1 Scope
IEC 60255 1xx: Protection relay functional standards for all — Megger; 2021-10. Auflistung der Normteile 187-1, 187-2, 187-3
SIPROTEC Differentialschutz 7UT612 V4.0 Handbuch — Siemens AG; unbekannt. Kapitel 1 Funktionsübersicht, Kapitel 2.2 Differentialschutz
SIPROTEC Differentialschutz 7SD610 V4.2 Handbuch — Siemens AG; unbekannt. Abschnitte Grundprinzip/Stromwandlerkennlinie und Einschaltstabilisierung
Implementing IP/MPLS network-based synchronization for line differential protection and control — CIGRE Science & Engineering; unbekannt. Abschnitt 2 Line differential protection relay communication

Related glossary entries

Edited by eCMS on 11.11.2026 18:58