Let’s start with what matters most: in every medium voltage (MV) network, there are discreet yet reliable guardians. They don’t flash or make noise, but their reaction can save the entire system from failure. We’re talking about power system protection – intelligent setups that not only detect irregularities faster than a human but also act before we even notice something is wrong.

This article explains how these systems work, what types exist, and why their correct configuration determines the continuity of supply and the safety of both people and equipment.

If you design, modernize or operate MV networks – this read will help you structure your knowledge and avoid mistakes that could cost hundreds of thousands of euros.

What will you learn from this article?

• What exactly do protection systems in MV networks do?

• What are the most common types of protection and when are they used?

• What does selectivity mean and how does it impact reliability?

• What to consider when designing a protection system?

Reading time: 5 minutes

Power system protection – the guardians of the network

Imagine a busy highway where every vehicle travels at a set speed in its own lane. Suddenly – an accident. What does the system do? It activates hazard lights, pulls over, alerts other drivers and manages the flow. In distribution networks, something very similar happens – only instead of cars, we have electricity, and instead of drivers – measurement and protection devices.

Power protection systems are setups that monitor current, voltage and phase angle values, comparing them to predetermined thresholds. When they detect an anomaly – they respond automatically. Their main task is to disconnect the damaged section of the network before it leads to:

• transformer damage,

• line destruction,

• electric shock to a person,

• shutdown of the entire substation.

These systems allow us to talk about supply continuity, the protection of life, and the limitation of the effects of short circuits and overloads.

4 types of protection in MV networks – when and how they operate

Medium voltage networks use several types of protection systems. Each has its role, and the choice depends on network topology, the presence of distributed generation, and the required selectivity.

1. Overcurrent protection (O/C – Overcurrent)

This is the classic of all protection systems – proven, fast, and reliable guardian of the network. It reacts instantly when the current exceeds a threshold set according to the installation parameters. In the case of a fault or overload, its main task is to disconnect the supply quickly, before the conductors overheat and insulation starts to degrade.

Why does it matter? Because even a few milliseconds of overload can cause local damage, and a few seconds’ delay may result in burned-out equipment.

Technical insight: MV systems use time-current curves (e.g., IDMT type), which can be adjusted to match the load characteristics. This enables precise control over the protection’s response time – so it always reacts before devices farther from the fault location, but doesn’t trip the system unnecessarily.

2. Earth fault protection (E/F – Earth Fault)

Think of it as a sensitive radar that picks up even the slightest “escaping” current. Its job is to detect current flow to earth – a potentially dangerous phenomenon, especially in networks with an isolated neutral point or resistance grounding. Even though the current values are low, their uncontrolled presence can lead to serious consequences – from winding overheating to substation fires.

Operating range: Typical earth fault currents detected by these protections range from 0.5 A to 30 A – and effective detection depends on the type of current transformers and measurement accuracy.

When is it essential? In long cable lines or networks with a wide-spread topology, where current leakage to ground may “blur” and not cause a significant current surge. There, E/F protection acts as a detector of issues other systems might miss.

3. Differential protection (DIF – Differential)

The most “just” of all protections – it only operates when there is a real current loss between the input and output of the protected device. By comparing the currents on both sides of a transformer, switchgear, or line, it detects internal faults that classic overcurrent protections won’t see.

Practical case: In a 20/0.4 kV transformer, if the current difference between the MV and LV sides exceeds 5% of the rated value – DIF considers it an internal fault and immediately disconnects the device. Such a response can save windings from irreversible damage.

Technical requirement: Differential protections require synchronized current transformers with a precise class and logical signal processing – which makes installation more complex, but worth every minute of engineering effort.

4. Directional protection

This is a tool for modern networks – where energy no longer flows just “top-down,” but also bottom-up thanks to distributed sources. Directional protection analyzes the current vector relative to voltage, determining exactly in which direction the fault occurred. This distinction is essential in networks with RES – otherwise, we risk an incorrect system response.

Typical case: In a ring network with a 1 MW PV plant, a fault on the grid side can be “fed” both from the substation and the PV source. Without directional recognition, the protection may disconnect the wrong section – causing outages in healthy parts of the network.

The solution? Directional protections installed at coupling points or line ends, properly coordinated with the rest of the system. This allows precise identification of the fault location and current direction – and ensures action only where it’s needed.

Designing a protection system – selectivity, timing and critical points

Designing a protection system in medium voltage networks is a process much like precisely lining up dominoes – every decision impacts the next, and a single mistake can trigger a cascade of consequences. It’s not just about selecting devices; it’s primarily a strategy for protecting infrastructure that must act automatically, reliably, and in full sync with the logic of the entire power system. To achieve that level of precision, engineers ask themselves three fundamental questions – just like doctors analyzing symptoms before choosing a treatment plan.

1. Which network points are critical to the operation of the installation?

Not all parts of the network are equally important – it may sound obvious, but in protection design it’s crucial. Critical points are those where a failure can shut down an entire facility, cut power to large groups of users, or damage costly components. These typically include:

• MV and LV switchgear outputs, where all energy destined for distribution is concentrated,

• Transformers, as connection points between voltage levels, requiring both differential and time protection,

• Coupling points and ring connections, which may serve as backup supply paths – their protection must be integrated into the logic of the entire system.

Practical example: in a large industrial facility, a critical point might be the switchgear feeding continuous production lines. Its failure, without selective protection, could halt production for hours and generate losses amounting to hundreds of thousands of euros.

2. How fast should the protection react?

Reaction time is one of the most important features of any protection system. It must be short enough to minimize fault consequences, but long enough to allow for selectivity. It’s a tightrope walk between safety and operational stability.

Too fast a response may lead to so-called overtripping – disconnecting too large a section of the network unnecessarily. Imagine a fault in one circuit causing an entire substation to shut down. That’s costly – and avoidable.

Too slow a response means the fault lasts longer – increasing the risk of insulation damage, conductor overheating and electric arcing. For anyone near the equipment, it’s a life-threatening hazard.

In modern systems, relay response times are selected with millisecond precision, and selectivity analysis is supported by tools like ETAP, DigSILENT PowerFactory or Simaris.

3. How to ensure selectivity of operation?

Selectivity is as complex as it is essential in design. It means that in case of a fault, only the protection closest to the fault location operates – the rest of the system remains untouched. It’s a bit like an automatic hotel door lock: one door can shut, but it won’t lock down the whole floor.

To achieve selectivity, you need to:

• accurately define current (I>) and time (t>) thresholds,

• use time-current characteristics of the inverse type (e.g., IDMT – Inverse Definite Minimum Time),

• analyze network topology and account for the influence of distributed sources that can reverse fault current flows.

Example: in a 15 kV ring network with a PV plant and three MV switchgears, lack of selectivity may result in a RES-side fault cutting power to the entire area – even if only one segment was affected. Applying directional and differential protection at coupling points eliminates this risk.

So, designing a protection system is an engineering art – requiring not only knowledge of standards and devices, but also imagination, analysis, and experience. Because every fault is a test not just for the hardware – but for the decisions made at the design stage.

Not just automation, but strategy

Power system protection in medium voltage networks is far more than the sum of algorithms, tripping thresholds and contactors. It’s a comprehensive technical risk management strategy that must account for system dynamics, the presence of distributed sources, user requirements and budget constraints. There is no one-size-fits-all solution – each protection system is a response to specific network conditions, loads, topology, and potential fault scenarios.

The selection and configuration of protection systems is a task that blends engineering competence with operational responsibility. On one hand, it requires knowledge of relay characteristics, transformer ratios, limit values and response algorithms. On the other – an understanding of how each decision will impact infrastructure performance in real-world conditions. A protection system engineer acts like a strategist planning defensive maneuvers – assessing critical points, modeling system responses, and seeking the most effective action scenario.

Short circuits, overloads, overvoltages – all of these appear suddenly and demand immediate decisions. Reaction time is measured in milliseconds, but the consequences – if protection malfunctions – may last for weeks and generate costs reaching tens or even hundreds of thousands of euros. Power loss, equipment damage, danger to people, production downtime – these are real outcomes that can be prevented by thoughtful protection architecture.

That’s why a well-designed protection system is not an expense to be postponed. It’s an investment in reliability, resilience and predictability of network performance. A decision that pays off every single day – even if the system doesn’t activate for a long time.

If you're curious how a medium voltage network is structured and why it matters when designing protection – check out our latest article:

How is a MV Network built? Understand it before you design protection

Before anything reacts – make sure the system knows what to do

Power protection systems don’t operate by accident. Their effectiveness is the result of thoughtful design, a deep understanding of energy flow logic, and the ability to anticipate scenarios that haven’t happened yet. It’s a structure of responsibility – not a default settings package.

Because in a world of voltages and currents, details matter: will the overcurrent threshold trip before the differential? Will the directional relay distinguish PV feed-in from a fault? And will it be your design that keeps the lights on at the critical moment?

If you’re here, it means you’re not chasing shortcuts – you’re after clarity, consistency and real-world performance. And thanks to professionals like you, networks run more smoothly, and transformers don’t have to prove their resilience under stress.

And if now you’re looking for equipment that matches your precision in design – check out what we have for you today.

Also see which MV transformers are available immediately – complete with full documentation and engineering support.

Join the Energeks community on LinkedIn and share the knowledge that powers the future of electrical engineering.

Thank you for being with us. And remember – the best protection is the one that operates exactly where it should. Not sooner. Not later.

Sources:

IEEE Xplore – “Directional Protection Schemes for MV Networks”

ABB White Paper – “Fundamentals of Protection Relays in Distribution Systems”

elektro.info – Selektywność działania zabezpieczeń