In this guide we follow one lightning surge from the sky to the ground, with animated diagrams, and see how a small device keeps a large transformer alive.
Quick answer: A surge arrester is a device connected between a live conductor and the ground. At normal voltage it acts like a closed door, so almost no current flows through it. When a voltage spike arrives, from lightning or from switching, the door opens in microseconds. The surge current goes to ground, and the voltage at the equipment is held at a safe level. Insulation coordination is the design step that checks this held voltage stays well below what the equipment insulation can survive.
The problem: a spike taller than the insulation
Every piece of power equipment has insulation. Insulation is the material that keeps live parts apart from each other and from the ground. It can only hold back a certain voltage before it breaks down.
For short impulses, this limit is called the BIL, or Basic Insulation Level. The BIL is the peak impulse voltage the insulation is tested to survive. A transformer on a 132 kV network, for example, often has a BIL of 650 kV.
Normal grid voltage is far below the BIL. But some events create short, very tall voltage spikes. We call them overvoltages. So what happens when a spike is taller than the insulation limit?
The insulation fails. A spark punches through it. This is called a flashover if it happens through air, or a breakdown if it happens inside solid or oil insulation. Inside a transformer, this damage is often permanent.
There are two main sources of overvoltage. Lightning makes a very fast, very tall spike. Switching, which means a breaker opening or closing, makes a slower spike that is usually lower. The diagram below compares one spike with and without an arrester.
A water pipe has a pressure relief valve. Most of the time the valve stays shut. If the pressure jumps, the valve opens and dumps the extra water. The pipe does not burst. A surge arrester is the relief valve of a power line. It dumps extra voltage into the ground.
What is inside a surge arrester
A modern arrester looks like a short insulator. Inside, it is surprisingly simple. Its heart is a column of round discs made of zinc oxide (ZnO). Each disc is a metal-oxide varistor, or MOV. A varistor is a material whose resistance changes strongly with voltage.
The discs are stacked and sealed inside a housing of porcelain or polymer. The housing has sheds, the ribbed rings on the outside. They make the surface path longer, so rain and dirt cannot easily form a leak path. The top terminal connects to the line. The bottom terminal connects to the ground.
Older arresters also had spark gaps in series with the discs. Most modern arresters are gapless. The zinc-oxide discs do the whole job alone.
The core idea: millions of tiny gates
Why does zinc oxide behave like a valve? The answer is in the zoom above. The disc is made of tiny zinc-oxide grains. Between the grains are very thin grain boundaries.
Each boundary acts like a small electronic gate. Below about 3 volts, the gate blocks current. Above about 3 volts, the gate lets current through easily. A disc holds millions of these gates in series and in parallel. So the whole column has a sharp switching voltage. The designer sets it by choosing the number and height of the discs.
This makes the arrester a non-linear resistor. Its resistance is not fixed. It is huge at normal voltage and very small during a surge. The graph below shows this as a voltage–current curve. The current axis jumps by a factor of 1,000 at each tick, so we can see tiny and huge currents on one graph.
Two points on this curve matter most. The first is the normal operating voltage. Here the arrester carries only a tiny leakage current, so it does not heat up. Datasheets call the highest safe value the MCOV, or Maximum Continuous Operating Voltage. The second is the residual voltage. This is the voltage across the arrester while a standard surge current, often 10 kA, flows through it. The residual voltage is the arrester’s protective level.
A surge arrester is not a fuse or a breaker. It does not cut the circuit. Power keeps flowing to the load the whole time. The arrester only opens a short side path to ground for a few microseconds, then closes it again.
Step by step: one surge, four moments
Let us follow one lightning surge from start to finish. Each card shows one moment.
1Normal service
The grid runs at normal voltage. The gates are closed. Only a tiny leakage current flows.
2The surge arrives
Lightning hits the line. A steep voltage wave races toward the transformer at nearly the speed of light.
3The gates open
The voltage passes the knee. Within microseconds the arrester conducts. Thousands of amps flow to ground, not into the transformer.
4Back to normal
The surge is gone in a few tens of microseconds. The gates close again. The arrester is ready for the next one.
Insulation coordination: making the levels fit
An arrester only helps if its protective level sits well below the equipment’s BIL. Choosing the arrester and the equipment so these levels fit together is called insulation coordination.
Picture a building with floors. The normal voltage lives on a low floor. The arrester’s protective level is a middle floor. The equipment BIL is the roof. A surge can climb only to the arrester’s floor. Insulation coordination makes sure there is enough space between that floor and the roof.
That space is the protective margin. Engineers often check it as a ratio: BIL divided by the arrester’s protective level. IEEE guidance (C62.22) asks for a ratio of at least about 1.2 for lightning surges, which is a 20% margin. For slower switching surges it asks for at least about 1.15. The chart below uses typical values for a 132 kV transformer.
Why the arrester must sit close to the equipment
There is one more effect that surprises many engineers. A surge is a travelling wave. When it hits the end of a line, such as a transformer winding, it reflects, which means it bounces back. The arriving wave and the reflected wave add up. So the voltage at the transformer can rise above the arrester’s protective level.
The further apart they are, the bigger this extra voltage. A simple estimate is: voltage at the transformer ≈ arrester voltage + 2 × steepness × travel time between them. Use the slider to see the effect. This simple formula gives a worst-case value, so it is good for building intuition.
Assumes arrester residual voltage 330 kV, surge steepness 500 kV/µs, wave speed 300 m/µs. Real studies use detailed models.
With these values, the arrester protects well up to about 60 metres. Beyond that, the margin becomes too small. This is why arresters are mounted right at the transformer terminals. It is also why short, straight earth leads matter. Every extra metre of lead adds voltage during the surge.
Where surge arresters are used
Surge arresters appear across the whole power system. They guard power transformers, which are costly and slow to replace. They sit at substation entrances, where overhead lines enter the yard. They protect cable terminations, where an overhead line meets an underground cable. They also protect motors, generators and capacitor banks.
On distribution poles, we can see small arresters next to pole-mounted transformers. In buildings, much smaller surge protective devices (SPDs) follow the same idea at low voltage. The rule is always the same. The arrester is a cheap guard. It keeps the costly equipment behind it below its BIL.
Key takeaways
- A surge arrester connects a live line to ground. It works like a pressure relief valve for voltage.
- Its active part is a column of zinc-oxide (MOV) discs. Millions of tiny grain-boundary "gates" make it strongly non-linear.
- At normal voltage it carries only microamps. During a surge it carries kiloamps while the voltage stays nearly flat.
- The residual voltage at a standard current, often 10 kA, is the arrester's protective level.
- Insulation coordination keeps the BIL well above the protective level: a ratio of about 1.2 or more for lightning.
- Mount the arrester close to the equipment. Distance and long leads add voltage through wave reflection.
Related guides on Electrical Blog:
A Friendly Guide to Installing REF Protection in Power Transformers
Short Circuit Analysis Theory: Fault Current Calculations and X/R Ratios