By John Pantzer, Skywire Design
The US National Electrical Code, published by the National Fire Protection Association (NFPA), addresses requirements for grounding (or earthing, as it’s called in certain parts of the world) in Article 250 of NFPA 70. For lightning protection systems, which are also connected to the grounding system, NFPA 780, Standard for the Installation of Lightning Protection Systems, provides additional detailed information. This standard goes into detail about what is required for driven rods that conduct lightning strike currents into the ground by contact with the soil.
For a stand-alone residence and many commercial buildings, a ground system that presents no more than 25 ohms of resistance to ground is acceptable. For telecommunications facilities, however, 10 ohms is a more common specification, and resistance below 5 ohms is better. Soil, which receives and dissipates the grounded currents, is a combination of decayed rock and organic material. The more rock in the soil, and the drier the soil, the more poorly it conducts electric current. Conversely, moist soil with clay and organic matter to retain that moisture, conducts current much better. Designing an effective ground system is essential for personnel safety, proper equipment operation, and lightning protection. Standard practices get the job done well for many grounding systems. But sometimes soil conditions and the history of war create unique conditions that warrant unique solutions. Skywire Design has designed for such a location and achieved excellent results. Let’s talk about Guam.
A Geologically Difficult Site
Geologically, Guam is challenging for grounding because the northern half is a limestone plateau. The resistivity of limestone can range from 10,000 to 1,000,000 ohm-cm. The unit ohm-cm describes a material’s resistance to the flow of electric current through a cube of material one centimeter on a side, and standardizes comparisons among various soils and substrates. For example, the maximum resistivity of loamy surface soil is about 5,000 ohm-cm.
In Guam, on land controlled by the US military, safety rules exist to protect excavators and builders from hazards remaining from World War II. During the second Battle of Guam, when the US recaptured the island from Japan, in thirteen days, 836 rounds of 16-inch, 5,422 rounds of 14-in., 3,862 rounds of 8-in., 2,430 rounds of 6-in. and 16,214 rounds of 5-in. shells pounded the island. In addition, aircraft from Task Forces 53 and 58 dropped 1,131 short tons (1,026 t) of bombs and rockets.
In December 2019, a farmer in Guam made the local news by bringing a land mine, uncovered by his cow, to his mayor’s office, and finding the mayor away, left the mine in a cardboard box on a nearby stage for the mayor to find on his return. Thankfully, no one was injured by this lax handling, but the island is still holding unexploded ordnance after all these years. Excavation rules permit construction to proceed only after electronically scanning the ground to be disturbed, and then lifting only six inches of soil before scanning again.
Installing grounding systems here becomes a slow, expensive process. For a 10-ft. long by 3-in. diameter ground rod, a hole must be bored to install the rod rather than simply driving the rod by slide hammer or pushing it in with a backhoe. The boring proceeds 6 in. at a time, in a repeated sequence of drilling, then lifting out the boring bit, lowering a scan head for another reading, pulling out the scanner, reinserting the boring bit to drill another 6 in. and repeating the sequence 20 times for every hole. A reasonable borehole is 6 in. in diameter, and it all has to be backfilled and firmly compacted to put in a single standard ground rod.
Designing Around Cost and Soil
Two conditions in Guam compelled a special design: soil conditions and economics. To reduce costs, the choice was made to minimize the number of ground rods. Since rods must be located beyond (outside) the dripline of the structure they are to protect and also need to be spaced at least 20 ft. apart for 10 ft. rods, in easier conditions this would mean 12 rods around each of these radomes. While the interconnecting conductor, buried 18 in. below the surface, would have to remain, it was decided to install only four ground rods around each radome.
The resistance to ground R of a ground rod is given by the equation:
R = (0.366ρ/l) × log(3l/d)
where ρ is the resistivity of the soil in ohms-cm, l is the length of the ground rod in centimeters, and d is the diameter of the ground rod in centimeters.
This indicates resistance will be reduced by using longer ground rods and larger diameters, although the effect of increasing diameter is blunted by the logarithm function. A 10-ft. rod length is standard. Earlier, a 6-in. diameter borehole was presented, as this was a practical size given the available boring bits. This diameter suggests a ground rod of the same diameter might be a good idea.
A solid copper rod of that diameter would be prohibitively expensive. Fortunately, it’s also completely unnecessary. By arranging four standard 10 ft. × ¾ in. rods parallel to each other in a square 4 in. × 4 in. and welding them together with copper flat bar, a compound ground bar was created that just fit into a 6 in. diameter borehole. Additional non-conductive spacers were installed to keep the rods spread apart during insertion. Conductive fill was then poured into the hole to replace all the material removed by the drill.
This large-diameter composite ground rod had eight times the surface contact of a single copper rod with the surrounding ground. Increase in surface contact is what gives larger diameter rods their reduced resistance to ground. With the reduced resistance of the larger composite ground rod, using only a third of the normal rod count was more than compensated. Having to drill only one-third as many boreholes also reduced the project cost.
Verifying the Result
Everything else about the design proceeded in the usual manner, with the result being a ground system with only 2 ohms resistance to ground, a truly stellar result. Designs, with all the simplifying assumptions involved and imperfect data on conditions, can appear on paper to meet standards, but performance testing is the judge of success. The predicted resistance for the site in Guam was 8.6 ohms, slightly below the imposed maximum of 10 ohms. One can imagine our desire to measure the new system’s resistance to confirm its suitability.
A standard test for measuring the resistance to ground of a single rod is the Fall of Potential method. A meter designed for this purpose has three test leads. One lead is connected to the ground rod. A second lead is connected to a short conductive stake pushed into the ground at least five rod lengths distant. Current from the meter flows into the ground rod, through the soil, out the stake, and back to the meter. This stake is called the current probe, since it provides a path for the current back out of the soil to the meter.
When a second stake, also connected to the meter, is pushed into the soil anywhere along the line between the ground rod and the current probe, a voltage can be measured at that point. This stake is called the potential probe. By Ohm’s Law, resistance (ohms) = voltage (volts) / current (amps), the resistance to ground can be determined by a series of measurements along that line. Each measurement is plotted against distance from the ground rod.
For a current probe sufficiently far away from the ground rod or system being measured, the resulting curve will show a flat spot at about 0.6 times the distance from the ground rod to the current probe. Nearer to the ground rod, the resistance will be progressively lower, and nearer to the current probe, the resistance will increase to a maximum. The flat part of the curve indicates the resistance to be measured — the resistance to ground of the single rod.
To measure a system of multiple rods, the current probe must be inserted at a distance at least five times the length of the diagonal system of rods. For instance, imagine four 10 ft. rods arranged in a square at a minimum 20 ft. spacing. The diagonal of that square is 28.2 ft., so the current probe would be placed 141 ft. from the nearest corner of the square.
The current probe in the actual measurement was 900 ft. away from the grounding system that surrounded the two antenna pads. In the end, some conditions were a little better than those assumed for the design, and the actual resistance was a stellar 2 ohms. This ground system will serve the site well.