What does the term "ground" mean"?
Exact definitions are available from the National Electric Code (NEC) and IEEE. It is important to remember that the term tends to be user-specific; it means what the speaker intends. When a dialogue is in progress, make sure that the participants are referring to the same thing. A ground is an intentional electrical connection with the earth by means of an electrode, that references the connected electrical system to earth (i.e., zero) potential. The term ground is also used to refer to the electrode itself, as well as the conductors that extend throughout the system to tie the equipment to ground. By extension, ground may also mean the third wire in a typical "hot-neutral" line cord. The term ground may also be used to indicate an unintentional connection to earth, where a piece of equipment has faulted and shorted out. So before any discussion proceeds, always be sure to determine the manner in which ground is being used.

How does a ground tester perform its test?
A dedicated ground resistance test instrument has a unique design, specifically fashioned to deal with the features that set ground testing apart from more familiar types of measurement. Unlike familiar test items (motors, transformers, wire and cable), the earth is obviously not a designed circuit, does not confine current flow to a specific path, has no beginning or end. Furthermore, a ground resistance test is a volumetric measurement that must be capable of measuring the resistance of a volume of soil surrounding the ground electrode full-circle, not in a straight-line path as in a familiar circuit. To accomplish this, the tester has a current circuit and a potential circuit. Using metal probes driven into the soil, a test current is established between the ground electrode and the current probe. Similarly, a voltage probe contacts the soil at a distance commensurate with the volumetric resistance under test, and a measurement made of the associated voltage drop. With these two parameters, voltage and current, the instrument calculates and displays, via Ohm's law, the resistance exhibited by the soil to a flow of current from the ground electrode. For purposes of visualization, think of the item under test as not only the electrode, but also the ball or block of soil surrounding it to a distance beyond which the rest of the earth, because of its vastness, contributes no more appreciable resistance.

Why should I bother to test the ground?
At time of installation of a new facility, electrical system, or piece of equipment, there is generally a ground-resistance specification that the ground rod or grid must meet. An appropriate test provides the necessary verification. Subsequently, however, this ground must continue to be periodically tested. Lightning damage, corrosion, and the rigors of freezing and thawing can disrupt or deteriorate a ground electrode. Seasonal variations in temperature and moisture can have profound effects upon the electrical properties of the soil. The water table may fall, so that an electrode that was installed in a moist, conductive soil may end up dry and isolated. If the system ground has been tied to the water pipes, repairs that are made with nonconductive materials (plastic pipe and couplers) can render the system highly resistant to current flow, hence an inadequate ground. Additions and improvements to the electrical system and equipment in a facility can result in the ground no longer being sufficient. Sophisticated electronic equipment can suffer immeasurably from noise and power disturbances that would have passed unnoticed by cruder electrical equipment, thereby requiring a substantially improved ground. All of these contingencies and more make ground maintenance an integral part of electrical upkeep.

What do I do if my ground is bad?
Frequently, a ground test will reveal an inadequate ground. Don't blame the tester, improve the ground by going deeper, adding rods, or treating the soil. Driving rods deeper generally improves their performance. This is not simply because there's more rod, going deeper brings the rod into moister, more stable conditions, and is especially effective if the water table is reached. Adding rods, or expanding the size of a grid, affords more parallel current paths to decrease resistance. To prevent added rods from acting as one, and thereby defeating their purpose, they should be separated by a distance at least as great as their length. Finally, soil can be treated in a number of ways to improve conductivity. These include the addition of salts to increase current-carrying ions, or backfills with water-retaining materials like bentonite. But don't ignore the EPA. First be certain that no ordinances are violated before adding chemicals to the soil. And don't forget that these may leach away, and so these solutions must be periodically maintained just like the electrode itself.

A variation of the treatment theme is the availability of several types of specialized ground rods that may be filled with ionic solutions, or may extract atmospheric moisture and leach it into the soil. These products are available from manufacturers of grounding materials.

What is the difference between "resistance" and "resistivity"?
Although these terms sound almost alike, don't confuse them. They have very different meainings. The efficiency of a buried electrode (rod, grid, plate, parallel ground field, etc.) is evaluated in terms of resistance. This is a measure of how well the electrode can disperse current into the surrounding soil. The independent electrical properties of the soil itself, however, are described in terms of resistivity. If you are making a resistance measurement, you are testing a particular installed ground. If making a resistivity measurement, you are testing the soil itself.

Resistance is measured in Ohms; resistivity is commonly given in units of Ohm-centimeters, although for convenience other units, such as Ohm-feet, can sometimes be used. The resistivity of the local soil combined with the configuration of the electrode make up the resistance which that particular electrode experiences. In field practice, resistivity measurement is frequently done first, in order to locate a good site for a ground and to theoretically calculate its optimum design. Then resistance testing is done second, to verify that the design has in fact met the requirement.

What is "Fall of Potential"?
This is the name given to the accepted testing procedure that is described in IEEE Standard No. 81. It is also referred to as the three-point method because it entails three points of contact with the soil: the test ground plus voltage and current probes. A fairly large number of resistance readings are taken by moving the voltage probe at regular intervals (e.g., five or ten feet) between the ground under test and the current probe. These are plotted on a graph of resistance versus distance and should produce a characteristic curve. Resistance should rise steadily in the electrical field of the ground under test and also within the field of the current probe. But in between, there should be a "flat" area in which resistance holds steady (it may vary by a few digits since soil is not uniform, but the overall graph should be noticeably horizontal through mid range). This is the value which is to be taken as the sought-after resistance measurement with the point at 61.8% of the total distance representing the value that is closest to the theoretical true.

Fall of Potential has been devised as a means of verifying that the true volumetric measurement has been attained. Otherwise, readings may vary depending only on where the voltage probe is placed. To understand the relationship that underlies this, it can be visualized that a fault current being carried to ground by the protective system will at first encounter a resistance from the surrounding soil. But the current in this case is not restricted to a single path, as in a wire. It can, and does, spread out 360 º around the ground electrode. But at a certain distance, depending on local soil conditions, the "freedom to wander" has become essentially infinite; that is to say, once away from the immediate environment of the electrode, the earth becomes so vast as to offer only a negligible added resistance. So it is the effect of the immediate environment, up to its maximum, that must be measured. Readings made too close to the tested ground will be artificially low, and impart a false sense of the ground's adequacy. By observing where it maxes out on the graph, one can be certain that the correct value has been determined. [In addition, the current path becomes constricted again in the vicinity of the current probe producing the second rising curve on the typical graph. In a real life situation, of course, this second rise does not exist. A lightning strike, for instance, can disperse throughout the earth. The second rise observed on a Fall of Potential graph is an artificially imposed condition of the test procedure, deriving from the necessity of introducing a test current.]

Fall of Potential is a thoroughly reliable procedure, as the correct reading can be observed from the graph. Its limitation is that it takes time to record all the required values and construct the graph. To speed testing without losing reliability, any number of specialized procedures have been developed. But if they are recognized by respected organizations like IEEE, they are derived from the basic concept of Fall of Potential. In these alternate methods, mathematics are substituted in place of graphic observation as the means of distinguishing between acceptable and bad tests. In Fall of Potential, if insufficient probe spacing has been used it will be directly observable from the graph: there will be no flat area from which to take the reading. Simplified methods use mathematical tests instead. Brief calculations are required and these throw out bad tests by not yielding coherent results. Mathematical tests are simplifications of calculus taking advantage of the rate of change of slope. The necessity for using one or another of these procedures is to have an objective way of knowing that the resistance measured as the test result is the maximum resistance that a fault current will encounter, not some lesser value taken off some other point on the Fall of Potential curve.