Lightning is one of the main paths for electrons to return from the atmosphere to the earth's surface. As in all electron flow, lightning is a form of current flow. Enormously high voltages (Table 1) between clouds and the ground cause breakdown of almost all insulator systems, leading to loud and brilliant arcing.
(Click to enlarge image)
Providing a flow path for the lightning current is central to effective lightning protection. NexTek engineers are frequently asked why certain practices are important in grounding and bonding coaxial protectors. In this article we discuss some key parameters, and illustrate their role and importance. While the issues here are oriented towards grounding coaxial lightning protectors, the same concerns apply to general lightning protection practices.
Examining the frequency domain parameters is just as important as looking at the time domain profiles. Taking the 8x20 µs impulse as a very rough representative pulse, the harmonic contents show energy starting at 20 kHz to 100 kHz Figure 1); while the energy continues to diminish above 100 kHz to about 1 MHz, there is very little energy above 1 MHz.
(Click to enlarge image)
Note that many waveforms (such as 10x350 and 6.7x70) are of longer duration, so that the harmonic content will be at lower frequencies.
While most of the parameters are of a scale that technically oriented people can understand, perhaps the most difficult parameters to grasp are the huge currents of tens of kiloamperes flowing for only a few tenths of microseconds. To give a sense of these currents and times, consider that a bare 18 AWG (1 mm diameter) copper wire, in air, normally will conduct at least 10 amperes safely, with very low self-heating temperature rise. If the current slowly rises, the temperature will increase until the melting temperature of 1065° C (1950° F) is achieved at about 83 A. This same temperature could be reached "instantly" by an 8x20 µs pulse at a current of 61 kA. A 61 kA impulse is as much of an overload to a 10-amp extension cord as a continuous, 83-amp load. The effect might be just the same: molten copper, easily hot enough to start a fire.
A. Temperature rise of the grounding conductors
When high-current pulses flow in a wire, the temperature rises due to the heat created by the current flow through the resistance of the wire material. The integral of i2dt is referred to as the action integral. The action integral, when multiplied by the resistance of the current flow path, is approximately the energy dissipated for that current flow (since P = I2R and J = I2Rt.
During a current impulse, wire temperature increases due to several factors. For a given current impulse, the two most critical factors are the wire's resistivity and diameter. Temperature-rise parameters of secondary importance are the wire's density, specific heat, temperature coefficient of resistivity and the wire's initial temperature. Figure 2 shows the impulse capability of wire with 8x20 µs current pulses.
(Click to enlarge image)
The wire's current capability is surprisingly high, even with a low temperature rise limit of 50° C. It appears that 4-mm diameter wire (~6 AWG) will withstand over 175 kA! So why are lightning grounding wires usually required to be larger?
B. Forces on a current-carrying conductor
Whenever current flows on a conductor in a magnetic field, Lorentz forces are created at right angles to the conductor. The magnetic field can originate from the earth or from the current flow in the wire itself. If the wire is straight, the force will tend to push the conductor sideways, or at right angles to the conductor direction. If the conductor is curved, then the force tends to put internal "hoop" tensile stress on the conductor.
Figure 3 shows samples of 12 AWG (0.080" or about 2 mm diameter) annealed copper wires after exposure to transient current pulses.
The top image shows the shape before the current pulse, with an inside bend radius of about 7/32" (~5mm). The middle wire, after exposure to a 55-kA, 8x20 µsec current pulse, shows that the internal hoop stresses have somewhat straightened the wire by smoothing out the relatively sharp bend. Repetitive current pulses of this magnitude would probably fracture this wire. The bottom wire has actually 'necked down' and fractured at the bend location, after a single 90-kA pulse.
According to the temperature-rise information above, impulse temperature rise was only slightly over 50° C, so wire softening or melting had no significant role in this fracture, even if follow-on or post-fracture current arcing occurred on the wire ends. The strength of the wire was exceeded by the internal forces.
Lorentz forces increase with sharper bends (lower radius) and are resisted by the size and strength of the wire material. As a rule, bends in lightning grounding wire should have at least a 10" or 30-cm radius. To eliminate bends, lugs should be positioned in line with the wire. Lugs which are in line with the ground wire are more effective than cosmetically satisfying lugs in line with an architectural component which force a directional change in the wire.
C. Terminations: It's not just the wire size; it's also the termination
Wire lug attachment (Figure 4) usually works better for continuous current than for transients.
Consider a 4-mm diameter (~6 AWG) copper grounding wire and lug. The normal voltage drop of a crimp lug is about the same as the voltage drop of 50 mm (2 inches) of wire. However, the lug junction with the wire is usually less than 12 mm (1/2") long, so the lug's heat dissipation is about four times more concentrated than it is in the wire. Normal heat dissipation can use the full surface of the lug and wire to assist in cooling. With lightning transients, there is insufficient time for the same level of heat-spreading and cooling effects.
It is not uncommon for lug transient-temperature rises to be over twice that of the attached cable. Substantial de-rating is usually appropriate for crimp or compression connections. To get more of the wire's transient capability, it is helps to use redundant connections or long-barrel terminals and proper crimp tools. Avoid the vise-grip or hammer approach for field connections.
Welded connections can get all of the wire capability and allow the use of smaller diameter wire. These may be less expensive in total cost, especially when compared to longer, upsized copper conductors. Of course, don't use excellent or redundant lugs so you can use under-sized wire!
D. Aging effects
The above test results and calculations are based on using new materials and do not account for aging, the oxidation and corrosion effect on wires and lugs. In addition, galvanic-corrosion action, salt, or other chemical exposure and soil conditions can be even more aggressive. Vibration and other mechanical loading, including loading which occurs during maintenance, have to be considered. This is another reason why larger ground cables are specified for large primary-bonding wires (those wires that may exclusively take the lightning current).
In environmentally aggressive environments, use tin-plated wire and fittings, and place the ground wire in plastic conduit or coat exposed bare metal. Keep terminals and junctions away from, or above, standing water. Use copper wiring outdoors, and do not mix copper with aluminum, even with approved CuAl (dual use) fittings.
E. Mechanical Damage
While 10, 12 and even 14 AWG (2.6 to 1.6 mm diameter) copper wires have a significant transient capacity, they are not strong enough to survive accidental snagging, pulling, or other forces. Consequently, smaller ground cables have to be physically protected or be bronze- or copper-clad steel. However, high-strength bronze or steel does not approach the conductivity of copper. If significant risk of exposure to lightning is possible, it is better to use larger, copper ground wires, rather than smaller and stronger wires.
(Note: Part 2 of this article discusses impedance, bonding, grounding resistance, and physical connections. It is available at http://www.planetanalog.com/features/showArticle.jhtml;jsessionid=H5SSGMYMRLH1WQSNDLRSKHSCJUNN2JVN?articleID=201807830.)
Lightning-stroke parameters are varied due to variation in both instrumentation and location. Some representative profiles are available at:
About the author
George M. Kauffman is Vice President of Engineering at Nextek, Inc., Littleton, MA (www.nexteklightning.com). For the past five years, he has has overseen NexTek's engineering team, developing new lightning-arrestor and DC power-conditioning products. George is a recognized leader in the lightning-protection industry and holds multiple patents in the EMC field. His technical expertise assures that each product line exhibits optimal technical performance characteristics.
Prior to NexTek, George spent eighteen years at Digital Equipment Corporation in numerous roles including manufacturing, product development, product management and software development. George holds both a BS in Mechanical Engineering and an MS in Engineering Management from the University of Massachusetts/Amherst. He is registered in Massachusetts as a Professional Engineer.