Thank you for your excellent question regarding the effects a lightning strike will have on the grounding system of a utility substation and/or power plant. It is our pleasure to help.
Lighting is an atmospheric electrostatic discharge (spark) which can travel at speeds of 220,000 km/h (140,000 mph), and can reach temperatures approaching 30,000 °C (54,000 °F), hot enough to fuse silica sand into glass. There are some 50-100 lightning strikes occurring somewhere on the planet every second, and can carry as much as 200,000 amps of electricity, and generate over 100 million volts in a single strike.
There are a number of engineering issues to be concerned with when lightning strikes any object, including a power plant. For this discussion, we will assume that this is a basic power plant / substation with an overhead lightning protection system tied into the standard ground grid of the facility. We will discuss each issue here, in no particular order. Additionally, it should be understood that all of the engineering principles mentioned in this article are wholly dependent upon having excellent soil resistivity data and valid soil models. Soil resistivity data is, and always will be, the heart of grounding science.
When lightning strikes an above ground object, such as an aerial on a Lightning Protection System (LPS), the current will start to divide itself across the Lightning Protection System (LPS) proportionally to the impedance it encounters, on its way down to the earth. We can imagine the lightning striking the aerial and the current dividing in two as it moves from the aerial into the conductor. As the current flows down the conductor, at each conductor intersection the current will divide again and again until it finally reaches the earthing electrode system where it will finally travel into the soil and dissipate. Our primary concern with Distribution is whether or not the conductors can handle the current levels without burning open like a fuse.
Down conductors for a Lightning Protection System must terminate in a connection to an earthing electrode of some kind. Most typically, it is a single standard ground rod, and on occasion three (3) rods installed in a triangular pattern with conductors tying them together. The effectiveness of the connection to earth of each of these electrodes (or electrode systems) is measurable and is called “resistance-to-ground”. The resistance-to-ground of any given electrode will vary given the immediate soil conditions the electrode finds itself in (moisture content, specific soil chemistry, proximity of non-conductive buried rocks, proximity to other conductive buried objects and/or soils, etc.), and as a result, each electrode will have a specific resistance-to-ground.
As there will be multiple down conductors for any given Lightning Protection System, electrodes with a lower resistance-to-ground will see a proportionally larger percentage of the current. In other words, the electrodes with a better connection to the earth will see more current than the other electrodes, “un-balancing” the Lightning Protection System. The European Lightning Protection Standard EN 62305:2006 calls for the “balancing” of these electrodes by either supplementing each electrode until they all have the same resistance-to-ground, or installing a buried ground ring tying all the electrodes together. The United States has no such requirement. In your case, you will be tying to a grid, which more than qualifies as a buried ground ring.
When lightning enters a conductor, huge magnetic fields are formed as the energy passes through the conductor. These magnetic fields hold huge amounts of inductive energy and will propagate into any nearby metallic object, including the same conductor (wire) it is currently traveling on. When a conductor is routed in such manner as to enable the magnetic fields from one part of the conductor to induce energy into another part of the same conductor (imagine a tight bend or circle), this is called a self-induced magnetic coupling. Self-induced magnetic couplings such as this can quickly lead to a thermal-avalanche where the two magnetic fields keep cross-coupling into each other forming a perpetually increasing energy level, thereby increasing the heat in the conductor until it melts and burns open.
All known regulatory codes regarding Lightning Protection Systems have detailed instructions on how to properly route conductors so as to prevent these self-induced magnetic couplings (and thus thermal-avalanche). Not only must the straight current portion of the lightning strike be considered (impedance of the conductors, current-carrying capacity, etc.), but the magnetic fields that are formed and the subsequent current that will re-enter the system upon the collapse of the magnetic fields, must also be taken into account. Computer modeling along with good design and diligent installation techniques will prevent an over-current situation on any one given conductor at any point in the Lightning Protection System, both above and below grade.
Frequency Spectrum & Time-Domain
It is a well known phenomenon that lighting has both an Alternating-Current (AC) component and a Direct-Current (DC) component, at the same time. In fact, lighting will propagate through a structure at many (if not all) possible frequencies. A typical lighting strike will see a range of frequencies from 0-Hz to as 10-MHz or higher. However, the distribution of these frequencies is not even as certain frequencies will be prominent. This ‘spectrum’ of frequencies generated by a lightning strike is called the “Frequency Spectrum” and is primarily determined by the geometric shape of the structure struck by the lightning.
Just as the length of an antenna determines the best frequency to broadcast/receive radio signals, the same is true for lighting. The lightning will adjust its frequency based on the structure (antenna) it strikes and will resonate due to impedance imbalances between the structure and the earth. In the case of lightning, all structures are antennas. Most structures make up very complex antennas, with the buried portion adding a further complexity as it will affect the lightning strike following many of the same principals found in the half-wave length theory for antennas. All the various variables and nuances of the calculations can become very complex, given all the different types of materials to be found on a typical structure, the erratic path-ways of the lightning protection system, and the variations found in the soil, only computer modeling can adequately calculate the expected frequencies for a lightning strike on a given structure.
To calculate the frequency spectrum of a lightning strike, one must first develop an accurate model of the structure (including material types), with the lighting protection system, and sitting in the actual soil conditions found at the site. Numerous frequencies must be run individually through the model until an accurate profile can be developed. Mathematical algorithms have been developed to assist in the proper selection of the test frequencies to reduce run times and improve the statistical accuracy of the profiles. But needless to say, many hundreds if not thousands of individual frequencies must be run through the computer simulation to accurately determine the frequency spectrum. The final result is a graph showing the entire range of expected frequencies on the x-axis, with magnitude on the y-axis. Typically, the simulation will demonstrate that two or three frequencies will resonate through the structure during a lighting strike.
Time Domain is the length of time it takes a lightning strike to clear out of your structure. The actual lighting stroke itself will start and stop in a very short time frame, typically only a few micro-seconds. However, as described earlier, magnetic fields will from in not only the lighting protection system, but in the structure itself. The time it takes to generate the magnetic fields to full strength, the resonation of energy through the structure, and the time it takes to collapse the fields (steel is at least 250X more magnetic than copper and therefore holds magnetic fields for far longer) is the Time Domain of the lightning strike. While the lighting stroke may start and stop in microseconds, the time your system will be impacted by the electrical energy from the stroke, will almost certainly be many thousands of times longer.
Conducting studies related to Frequency Spectrum & Time Domain can have substantial benefits. The primary benefits include:
- The ability to demonstrate compliance with European Lightning Protection codes (EN 62305:2006) without over-engineering. These codes have strict rules that in many cases will simply result in over designed lightning protection systems for many buildings. A good Frequency Spectrum & Time Domain study can prove the effectiveness of a lightning protection system without adding huge installation costs due to over engineering. Or it will at least prove the necessity of those costs.
- The Frequency Spectrum and Time Domain can be used in an Electrical Coordination and Short-Circuit Fault Study, to improve the settings of the over-current protection devices and reduce down-time due to unintentional power outages caused by lightning strikes. For many facilities, such as military facilities, hospitals, data centers, and power plants, a single power outage can have huge consequences and often is measured in millions of dollars in lost revenues.
- The Frequency Spectrum and Time Domain can be used for improving surge-protection systems. While all sites will want a broad-spectrum frequency protection for unwanted surges and transients, the addition of specially-tuned surge protectors designed to stop the resonant frequencies determined during the Frequency Spectrum Study can prove especially useful for protecting vital equipment from the impact of lightning strikes.
- Lighting strikes are different from standard electrical utility faults in not only dramatic amounts of current, but in the near limitless voltage potential and the high frequencies that will be generated. The Frequency Spectrum and Time Domain are critical for accurately calculating the effects of a lightning strike in regards to Human Safety Studies involving Step & Touch Voltage Hazards, Ground Potential Rise (GPR) studies, and Electromagnetic Interference studies.
Additionally, once the resonant frequencies of lighting strike have been determined, the Distribution can be better analyzed as impedances can now be properly calculated.
Hazardous Step & Touch Voltages for Human Safety during a Lightning Strike
Human Safety is always a paramount concern, and when dealing with lightning strikes, the critical issues can compound quickly. Personnel touching a conductive object or even stepping near a lighting strike can suffer serious injury and even death. Calculating Step & Touch Voltage Hazards during a lightning strike is in principle the same as when one calculates the hazard during a line-to-ground fault. However, we must take into account the new frequencies and clearing time that will be generated during the strike, as determined by the Frequency Spectrum and Time Domain analysis.
Step & Touch Voltage Hazards are calculated using the strike amperage at the point of contact, the frequency, the X/R Ratio, the clearing time (i.e. Time Domain), and the specific soil resistivity conditions. Calculating the X/R Ratio (Reactance (or X) divided by Resistance) of a lighting strike can ultimately be a very complex equation; even though it is nothing more than an expanded version of Ohm’s law. The reactance of the strike will vary based on the frequency. As the Frequency Spectrum analysis will tell us which frequencies will have the highest magnitude (usually only 2 or 3 peaks) it is possible to calculate the most probable X/R Ratio for a given strike. However, if the computer simulation is already taking into account the frequency of the strike, a high default X/R Ratio (such as 1) may provide more accurate and safer results.
Once all of the engineering factors have been determined, a computer simulation can be run to determine the Step & Touch Voltage Hazards that will be present at the site, for the site-specific frequencies that will be seen during a lightning strike. It is of course best to run all of the possible frequencies during the lightning strike, however limited computer modeling may only enable a certain number of possible frequencies to be calculated. The frequencies with the highest magnitude are obvious choices when computing power is limited. Once the safety parameters (crushed rock, shoes, etc.) have been applied to the computer model, the over-voltage hazards will appear and standard mitigation techniques for reducing these hazardous voltages can be applied, thus making the site safe.
Electromagnetic Interference and Ground Potential Rise during a Lighting Strike
Another key factor during a lightning strike is electromagnetic interference. As mentioned earlier in this article, large magnetic fields will form in both the Lightning Protection System and in the buried Grounding/Earthing System. Additionally, the Ground Potential Rise (or Earth Rise Potential) will cause scalar voltages to form across the surface of the earth with the potential decreasing with distance. These large magnetic fields and ground potential rise effects will transfer voltages and currents (by inductive, capacitive and through-earth coupling) into nearby pipelines, railways, communication lines, homes, industrial facilities, farms, and other such utilities, whether buried or aboveground.
The current and voltages that are induced into the nearby utilities can cause great damage. One of the primary concerns is on the induced voltages/currents that can form on the data and/or shield lines of Telco/computer wires causing damage as these transient currents will flow through sensitive electronic equipment on their way to earth. The neutral wires from the utility power company can carry stray currents formed from the electromagnetic fields up and into homes and industrial parks. Nearby gas pipelines can have their protective coatings compromised by the stress-voltages caused by the difference in potential between the surrounding earth and the pipe.
Again using the highest magnitude frequencies of the lightning strike, as determined by the Frequency Spectrum Analysis, the Electromagnetic Fields and the Ground Potential Rise can be accurately calculated and the impact on the surrounding infrastructure can be properly analyzed and potential problems mitigated.
When calculating the effects of a standard electrical utility fault on a grounding system, we know that the utility fault will have clamped voltages at a very specific frequency (60 Hz). With a lightning strike, the potential voltages are virtually unlimited and will have multiple frequencies based upon the geometric shape of the structure that is struck. The hazards presented by an electrical fault at 60 Hz are very different from the hazards presented by an electrical fault at 60 kHz or even 60 MHz. These order-of-magnitude increases in frequency present unique engineering challenges and is why lightning strikes are so dangerous. The bottom line is that Ground Potential Rise and Step & Touch Voltage Hazard studies conducted using power company (utility) fault data is simply not applicable for a lightning strike.
So in summation, all of the factors presented above need to be taken into account when building an effective lightning protection system; be it a Utility substation, telecommunication site, or simply just a personal home.
Remember that Distribution, Frequency Spectrum, Time-Domain, Hazardous Step & Touch Voltages for Human Safety, Electromagnetic Interference, and Ground Potential Rise, are all important parts of the puzzle that is called “Lightning Protection Systems”.
We hope you have found this information useful. If you should have any questions, please feel free to call one of our engineers and they will gladly speak with you free of charge.
The Engineering Team at E&S Grounding Solutions