Safeguarding electrical installations from lightning strikes
With spring comes the lightning season, so this is the right time to brush up on a powerful natural event and how to protect equipment from it.
Lightning protection is important in the design of communications circuits and systems. Get it wrong and nothing may happen for a while, but you will certainly know about it when the time comes.
And cleaning up the mess after a lightning event can be a very expensive, time-consuming exercise.
If your job entails responsibility for critical communications systems, read on – this could help save your installation from disaster.
Nature of lightning
Lightning is the discharge of static charge build-up in clouds, usually where there is violent movement of warm air (thunderstorms, tornados, dust storms, and even in smoke plumes associated with volcanic eruptions and bushfires).
The build-up can be of either polarity, and clouds can reach as high as 15km, so cloud-to-cloud lightning is by far the most common. When the build-up is in the lower part of a cloud, the ground takes on a charge of the opposite polarity (basic electrostatics – unlike charges attract). The cloud and ground effectively act like a parallel plate capacitor, with air as the dielectric.
The voltage gradient to ground (electric field strength) just before a cloud-to-ground lightning strike can easily exceed 100kV/metre. No wonder people say the hair on their arms starts to prickle and stand on end preceding a nearby strike.
When the intervening air finally does break down, it forms a conductive ionised channel between the cloud and ground and a lightning strike occurs. This consists of one or more short pulses of very high current, each with a very fast rise time. These pulses equalise the charge difference between the cloud and the ground.
The strike point on the ground is not always predictable, particularly as the air is often turbulent at the time. Lightning doesn’t always strike the highest point or follow the shortest path to ground through the air. It depends on where the air finally breaks down to form the conductive channel.
Apart from the blinding flash and the boom, lightning strikes are accompanied by phenomena that are less well known.
When a current flows in a conductor, it is always accompanied by an encircling magnetic field. The magnitude of this field is proportional to the magnitude of the current.
The peak current associated with a typical lightning strike is of the order of 30,000A, with about 10% of strikes reaching a peak of 300,000A. Clearly this is huge, and so is the associated magnetic field.
The following is a practical example of what this can do.
Some years ago, when I was testing the lightning protection earthing system of an installation on a mountaintop, a thunderstorm came along. So we stopped work and waited for the storm to pass.
At the height of the storm, one of the buildings was hit (the joke later was that I arranged it, just to prove my worth).
The current was conducted safely to earth by the building frame, which consisted of a metal roof and vertical steel I-beam columns. One of the rooms, full of cathode ray tube (CRT) monitors, was in a corner of the building where there was one column.
At the instant of the lightning strike, every monitor screen went green, magnetised by the intense magnetic field. It took many operations of the de-gaussing circuitry to restore each monitor to working order.
If there is lightning nearby, you can often hear a quiet click just before the boom. This click is caused by magnetostriction – a phenomenon exhibited by ferromagnetic materials, which physically change shape in the presence of a magnetic field.
The magnetic field travels much faster than the sound, so if you hear a click during a thunderstorm, you can expect a loud boom shortly after.
This is also why you hear a squeal from the ferrite transformer associated with high-frequency inverter circuits. The shape of the transformer ferrite core pulses in sympathy with the magnetic field and the core walls generate a sound wave on the outer surfaces, much like a loudspeaker cone.
These circuits often run in the kilohertz frequency range, hence the squeal.
Faraday’s law of electromagnetic induction tells us that when a conductor cuts the lines of flux associated with a magnetic field, a voltage is induced along that conductor.
It doesn’t matter whether the conductor is moving through the field or the field is moving in relation to the conductor. It is the principle used in all motors, generators and transformers.
With lightning, the current builds from nothing to a huge value in a very short time, then quickly collapses to nothing. The associated magnetic field does likewise.
Any conductor immersed in this changing field will have a voltage induced into it. What’s more, the voltage induced is proportional to the rate at which the lines of flux of the magnetic field cut the conductor.
In a lightning strike, the rate of change of current (hence magnetic field) is very fast, typically 1microsecond (one millionth of a second) from zero to peak value. So the induced voltage can be large if the conductor is anywhere near the lightning path.
It doesn’t matter whether the event is from cloud to ground or cloud to cloud – the same rule applies. So cloud to cloud events are not as harmless as they seem.
When lightning strikes the ground, very large currents flow away from the strike point, both along the surface and into the ground body.
As stated by Ohm’s law, the paths of least resistance are favoured. If there is a metallic water pipe buried near the strike point, it will tend to collect the current and conduct it a long way before it eventually dissipates into the earth.
If you are holding a tap in your house at the time and your stainless kitchen sink is touching an earthed appliance, the consequences can be quite nasty.
Similarly, if lightning strikes a railway track, under some circumstances it can be conducted a very long distance before it dissipates, particularly in dry conditions.
The strike current sets up an enormous potential gradient area in the vicinity of the strike. Remember that typical peak currents of about 30,000A can be expected, and the ground is usually not all that conductive.
If the strike point presents a resistance of, say 50Ω, a potential rise in excess of a million volts will ensue. If the ground is of uniform resistivity, this voltage diminishes exponentially with distance from the strike point. A nearby large animal with front and back legs far apart may be killed, even if it is not hit directly.
Lightning producesa flash because the air in the conductive channel is ionised and suddenly heated to a very high temperature by ohmic loss. Thunder is caused by the rapid expansion of this heated air.
In terms of the energy contained in lightning and its effect on a good conductor, not a lot happens. Close inspection of a lightning rod that has been struck by lightning will reveal a small pit mark at the strike point.
This last fact provides the clue to preventing damage. Lightning should be ‘persuaded’ to attach itself to a good conductor, then guided by a secure, direct conductive path to earth where the energy is safely dissipated in a well-designed earthing system.