Overview

The metal skin of a modern airplane can be considered a good approximation both to a Faraday cage and to the outer surface of a topological shielded system (see Section 3.1). As such, the skin provides the primary lightning protection for the aircraft. Generally, when lightning strikes a metal airplane, the lightning current remains in the skin of the plane as it flows between entrance and exit points. If the shielding by the plane's skin were perfect, there would be no danger to the interior electronics or to the fuel in the wings of the airplane. Unfortunately, there are openings (apertures) such as windows in the metal skin and antennas that project through insulated areas in the skin, both of which may serve as entry points for lightning electromagnetic fields. Additionally, the plane's aluminum skin is not always thick enough to avoid direct damage by a severe lightning charge flowing into the skin (see Section 2.3). Lightning often disables interior aircraft electronics, as we shall discuss in Section 9.2, and occasionally lightning can burn through the aircraft's skin, igniting fuel or releasing hydraulic fluids, examples of which are given in Section 9.3. In Section 9.4, we will briefly consider the standards for testing aircraft to make more certain they can withstand a lightning strike without serious consequences.

Contrary to the common view, most lightning discharges that strike airplanes in flight are initiated by the planes themselves. The lightning would not have occurred if the plane had not been present.

Overview: the cost of various types of lightning damage

In the early evening of July 13, 1977, lightning struck one or more 345 000 volt (345 kV) transmission lines located about 30 miles north of New York City. Those transmission lines supplied electrical power to the New York City metropolitan area. A chain reaction of electrical events ensued with the result that electricity was lost to all of the City by 9:34 p.m. It was the next night, the night of July 14, before all electrical power was restored. About 9 million people spent a night and a day without any electricity. By the time that the blackout ended, nearly 2000 stores and businesses had been looted or damaged, and more than 3500 people had been arrested. The cost of this lightning-caused blackout has been estimated at $350 million (Sugarman 1978). A detailed technical description of its causes is found in Wilson and Zarakas (1978) and a complete bibliography of publications concerning the blackout (including testimony before Congressional committees, books, and magazine articles) is found at http://chnm.gmu.edu/search_results.php?query=new%20york%20city%20blackout.

In Minnesota on June 25, 1998, two separate lightning flashes struck and caused the de-energization of two separate 345 kV transmission lines, initiating the cascading removal of overloaded lower voltage lines from service. In a short period of time, significant portions of Minnesota, Montana, North Dakota, South Dakota, Wisconsin, Ontario, Manitoba, and Saskatchewan were without electrical power and remained so for about 19 hours.

Personal safety

An eminent group of lightning safety experts, meeting at the American Meteorological Society's annual conference in 1998, formulated the following guidelines regarding safety from lightning injury and death (Holle et al. 1999):

Overview

In the previous chapter we discussed the effects of lightning on humans and animals, and we considered those situations and activities that are safe and those that are unsafe in a thunderstorm. In Section 7.1 we noted that in any outdoor group activity, like the positioning and movement of the spectators at a professional golf tournament, one individual should be designated as the responsible “weather person,” the primary person in charge of keeping track of the potential for lightning or other dangerous weather and of specifying when to get out of harm's way, using a plan that is already in place and tested. Only if one knows there is danger can appropriate action be taken to try to assure safety.

There are many situations in which it is not obvious that thunderstorms are approaching; for example, when the view of the storm and its lightning is obstructed or when nearby noise drowns out the sound of thunder. Nevertheless, the first line of defense in lightning warning is generally the visual observation of an approaching storm and the use of “flash to bang” thunder-ranging, that is, counting the time delay between seeing the light from the lightning and hearing its thunder. The time difference is about 5 seconds for each mile (about 3 seconds for each kilometer) of distance between the lightning and the observer, since sound travels about 1/5 mile (about 1/3 kilometer) per second while the light from a flash reaches the observer in a very small fraction of a second, that is, virtually instantaneously since light travels at 186 000 miles per second (300 000 kilometers per second).

One of the questions I've most been asked is “Do lightning rods really work?” The question can be phrased more accurately as “Does the standard lightning protection eliminate the possibility of lightning damage to structures?”; and the answer is “almost always.” Nevertheless, there are many individuals who erroneously believe that installing lightning protection on a structure significantly increases the risk of that structure's being struck and hence damaged by lightning, a relatively common view during the nineteenth century when little was known about the physics of lightning. Among those individuals who still so believe, as brought to my attention recently by an investigative newspaper reporter, are important officials in at least several high-profile fire departments, individuals who should know better. There are both international and national standards for installing lightning protection systems, and in this book I examine the assumptions underlying the common approach taken in all the standards and the resultant adequacy of those standards.

I am also often asked about the efficacy of the two commercially available methods of “non-standard” lightning protection, non-standard in that they are not recommended in either the primary international standard or the US standard. Each method involves the use of unusual lightning rods (air terminals), in one case purportedly to attract the lightning from relatively far away so that fewer air terminals are needed and in the other case purportedly either to discharge the thundercloud (the latter possibility being originally proposed by Benjamin Franklin based on his laboratory experiments) or, at minimum, to emit enough electrical charge into the atmosphere surrounding the air terminal to repel lightning from the local vicinity.

Overview

More than 250 years ago Benjamin Franklin (1753) described the basis of a scheme for the protection of structures:

Types of lightning

Lightning is a very long electrical spark, “very long” meaning greater than about 1 kilometer. Most lightning is generated in summer thunderstorms and is characterized by a length of 5 to 10 km, at the extreme about 100 km. The longest electrical sparks that can commonly be generated in the laboratory measure 1 to 3 meters, with the maximum being 10 to 20 m. In swirling desert sand storms, sparks occur that are a few meters in length (Kamra 1972a, b). Sparks this short are generally not called lightning. Some volcanoes produce kilometer-length electrical discharges in their ejected material, so-called “volcano lightning” (e.g., Anderson et al. 1965, Brook et al. 1974a, b, McNutt and Davis 2000). Even longer discharges, called “nuclear lightning,” were produced by near-surface thermonuclear (H-bomb) explosions in the 1950s (testing that has since been discontinued), the electrical charge source for the nuclear lightning being the negative charge (electrons) blasted upward into the atmosphere by the detonation (e.g., Uman et al. 1972, Williams et al. 1988). Video records taken both by the Galileo orbiter as it circled the planet Jupiter and by earlier fly-by spacecraft document that lightning occurs in the clouds of that planet (e.g., Borucki and Magalhaes 1992, Little et al. 1999), and there is evidence that lightning also occurs on Saturn.

History

Ships, the name given to large sea-worthy vessels, are generally made of electrically conducting metal, whereas boats, smaller vessels including speedboats, powerboats, motorboats, rowboats, and sailboats, are more often constructed of electrically insulating material such as fiberglass or wood. The metal shells of modern ships can be considered to be both an approximate Faraday cage and the outer surface of a topological shielded system (see Section 3.1), with the contact region between the electrically conducting hull and the ocean providing the grounding connection, and the ocean being the “ground.” Thus, modern ships do not suffer much lightning damage. What damage does occur is generally limited to exposed communication antennas, radars, and insulating covers for equipment. On the other hand, wood and fiberglass boats seldom encounter lightning without being damaged in some way. The US lightning protection standard NFPA 780:2004 devotes Chapter 8, Protection for Watercraft, to the methods of protection for powerboats and sailboats.

Ten years or so after Benjamin Franklin proposed a method for the lightning protection of both houses and ships (see Section 4.1), those principles were applied to the protection of wooden ships, wood being the only material from which ships were constructed at that time. The history of the lightning protection of the ships of the British Royal Navy is particularly interesting and has been reviewed by Bernstein and Reynolds (1978), from which some of the following discussion is taken.

Modifying the cloud electrification process

In many circumstances it would be valuable to suppress or eliminate lightning. One potential way to do so would be to eliminate or reduce the source of lightning, the cloud charge. The primary charging mechanism in a thundercloud is thought to involve interactions between soft hail (graupel) falling under the influence of gravity and lighter ice crystals rising in the cloud's updrafts, all in the presence of liquid water drops not yet frozen but at an altitude where the temperature is colder than the freezing temperature (32 ° or 0 °). Additional information is found in Section 1.1. Since cloud electrification is related to the falling precipitation within the cloud, and since this electrification is what produces lightning, it is reasonable to suppose that the amount of lightning is related to the amount of precipitation that is falling inside and below the cloud. Observationally, this is the case. In fact, one can often estimate the amount of rainfall (melted ice forms) that will reach the ground below a thunderstorm by counting the number of lightning flashes in the storm; the more lightning, the more rainfall (e.g., MacGorman and Rust 1998, Table 7.6; Takayabu 2006; Gungle and Krider 2006). For example, for nine isolated thunderstorms in Florida, Gungle and Krider (2006) found that each cloud-to-ground lightning flash was accompanied by an average rain volume of 2.6 × 104 ± 2.1 × 104 m3 (or a mean rainwater mass of 2.6 × 107 kg) as measured by rain gauges on the ground.

Does it work?

The most important thing we know about lightning protection is that if implemented in a manner consistent with existing standards, the protection will be successful most of the time. Many individual examples illustrate the value of lightning protection systems. Some are given below.

According to Schonland (1950):

We noted in Section 10.1 that William Snow Harris compiled a record of lightning damage to the unprotected wooden ships of the British Royal Navy.

Overview

Both overhead (above-ground) and underground power and communication lines are susceptible to lightning-caused interruption and permanent damage. In an open field, the probability of lightning striking a buried cable is less than, but not that much different from, the probability of its striking an overhead line, as we shall discuss in Section 12.3. If a lightning flash would randomly strike the ground in the “near vicinity” of the location of either an underground or an overhead line in the case where the line were not present, that lightning would likely strike the line when the line was present. “Near vicinity” is typically a horizontal distance of some tens of meters. The attraction of lightning to objects projecting above ground level has been discussed in Section 1.5 and Section 3.3 and will be further considered in this section.

The number of strikes to an overhead power line per kilometer of line per year, Nkm, can be roughly calculated by using the concept of an “equivalent collective area” on the ground in which the lightning would strike if the line were not there (see Section 1.5). We assume that all flashes that would have hit ground within a horizontal distance equal to two line heights on either side of the line center (if the line were not there) will strike the line. The resulting equivalent collective area is shown shaded in Fig. 12.1. For a length of line ℓ, that area is 4hℓ, where h is the line height.

General

Trees are protected in much the same way that structures are protected, with the added provision that lightning current must not be allowed to flow in the major root system. Current in the major tree roots can result in root damage that leads to the death of the tree. Grounding of the lightning current is therefore best accomplished by locating the primary grounding electrodes, ground rods and/or buried horizontal wires, near the outer edge of the branches where there are few significant roots. NFPA 780:2004 contains an Appendix, called Annex F, discussing the protection of trees, and the American National Standard Institute document, ANSI A300 (part 4: 2002), provides similar information and some additional information regarding the protection of specific types of trees.

According to NFPA 780:2004, all trees with trunks within 3 m (10 feet) of a structure or with branches that extend to a height above the structure should be protected. The three primary reasons for the NFPA-recommended protection are to avoid damaging side flashes (see Section 4.4) from the tree to the structure (the ground electrode of the tree is to be bonded to the structure protection system), to avoid a tree fire that could presumably spread to the structure, and to avoid the explosive and potentially damaging splintering of the tree due to the superheating of the moisture in the tree (see Section 11.3).

Overview

The primary purpose of a lightning grounding system is to provide a means to direct lightning current from the down conductors into the Earth while minimizing the voltage rise on the protection system. For example, if a relatively high peak current Ip of 100 kA is injected into a grounding electrode with a resistance Rgr of 25 ohms, the peak voltage Vp on the metal components of the above-ground protection system due to the current flowing through the resistance Rgr will be Vp = Rgr × Ip, and hence will be equal to 2.5 × 106 volts (see Sections 2.3 and 4.4). This voltage level will lead to side flashes from the above-ground system to any isolated (or grounded at a distance) conducting bodies (metallic or human) within about 5 m of the protection system, since the average electric field for breakdown in air is about 500 kV m− 1 (see Section 3.3). To eliminate such side flashes and the danger to individuals near the protection system, one should bond (electrically connect) nearby metallic objects to the lightning protection system conductors. Since half of the peak lightning currents for first strokes are larger than about 30 kA and 90 percent are larger than about 10 kA, and since it is generally impractical to obtain grounding resistances below about 10 ohms except in the very best conducting soils, the lowest expected peak voltage on a down-conductor system will generally be hundreds of thousands of volts.

Overview

As noted in Section 1.4 and Section 3.2, it is common to consider separately (1) the lightning protection of a structure and (2) the lightning protection of the electrical power, electronic equipment (e.g., television, DVD, burglar alarm system, computer), and communication systems (e.g., telephone, cable television) located within that structure, although, as we have discussed in Section 3.1, it is preferable to consider the two aspects of protection in a unified way as part of an overall topologically shielded and surge-protected system. The electrical power, electronic equipment, and communication systems within a structure are generally connected to the outside world by conducting wires, primarily 50 Hz or 60 Hz utility power wiring and telephone cables. Outside metallic wires are exposed to direct lightning strikes and suffer significant induced voltages and currents due to lightning occurring with a few hundred meters of those wires. In this chapter we consider primarily the protection of low-voltage systems. By “low voltage” we mean any voltage equal to or below the normal household power level, 480 volts generally being the highest “low voltage” satisfying this definition. In Section 12.2 we will examine the use of surge arresters on distribution and transmission power lines where the voltages may vary from some thousands of volts on distribution lines to about a million volts on transmission lines.

Surge protective devices (SPDs) are known by a number of different names including “arresters,” “surge arresters,” and “lightning arresters.”