A very effective method of obtaining a low-resistance earth connection is to bury a conductor in the form of a closed loop in the soil at the bottom of the excavation for building foundations.
The resistance R of such an electrode (in homogeneous soil) is given (approximately) in ohms by: 
where
L = length of the buried conductor in metres
The quality of an earth electrode (resistance as low as possible) depends essentially on two factors:
Installation method
Type of soil
Installation methods
Three common types of installation will be discussed:
Buried ring (see Fig. E20)
This solution is strongly recommended, particularly in the case of a new building. The electrode should be buried around the perimeter of the excavation made for the foundations. It is important that the bare conductor be in intimate contact with the soil (and not placed in the gravel or aggregate hard-core, often forming a base for concrete). At least four (widely-spaced) vertically arranged conductors from the electrode should be provided for the installation connections and, where possible, any reinforcing rods in concrete work should be connected to the electrode.
The conductor forming the earth electrode, particularly when it is laid in an excavation for foundations, must be in the earth, at least 50 cm below the hard-core or aggregate base for the concrete foundation. Neither the electrode nor the vertical rising conductors to the ground floor, should ever be in contact with the foundation concrete.
For existing buildings, the electrode conductor should be buried around the outside wall of the premises to a depth of at least 1 metre. As a general rule, all vertical connections from an electrode to above-ground level should be insulated for the nominal LV voltage (600-1,000 V).
The conductors may be:
Copper: Bare cable ( ≥ 25 mm2) or multiple-strip (≥ 25 mm2) and (≥ 2 mm thick)
Aluminium with lead jacket: Cable ( ≥ 35 mm2)
Galvanised-steel cable: Bare cable ( ≥ 95 mm2) or multiple-strip ( ≥ 100 mm2 and ≥ 3 mm thick)
The approximate resistance R of the electrode in ohms: 
where L = length of conductor in metres
ρ = resistivity of the soil in ohm-metres (see “Influence of the type of soil” )
Fig. E20: Conductor buried below the level of the foundations, i.e. not in the concrete
Earthing rods (see Fig. E21)
For n rods: 
Vertically driven earthing rods are often used for existing buildings, and for improving (i.e. reducing the resistance of) existing earth electrodes.
The rods may be:
Copper or (more commonly) copper-clad steel. The latter are generally 1 or 2 metres long and provided with screwed ends and sockets in order to reach considerable depths, if necessary (for instance, the water-table level in areas of high soil resistivity)
Galvanised (see note (1)) steel pipe ≥ 25 mm diameter or rod ≥ 15 mm diameter, ≥ 2 metres long in each case.
Fig. E21: Earthing rods, connected in parallel
It is often necessary to use more than one rod, in which case the spacing between them should exceed the depth to which they are driven, by a factor of 2 to 3.
The total resistance (in homogeneous soil) is then equal to the resistance of one rod, divided by the number of rods in question.
The approximate resistance R obtained is: 
if the distance separating the rods > 4L
if the distance separating the rods > 4L
where
L = the length of the rod in metres
ρ = resistivity of the soil in ohm-metres (see “Influence of the type of soil” below)
n = the number of rods
Vertical plates (see Fig. E22)
For a vertical plate electrode: 
Rectangular plates, each side of which must be ≥ 0.5 metres, are commonly used as earth electrodes, being buried in a vertical plane such that the centre of the plate is at least 1 metre below the surface of the soil.
The plates may be:
Copper of 2 mm thickness
Galvanised(1) steel of 3 mm thickness
The resistance R in ohms is given (approximately), by: 
where L = the perimeter of the plate in metres
ρ = resistivity of the soil in ohm-metres (see “Influence of the type of soil” below)
Fig. E22: Vertical plate - 2 mm thickness (Cu)
(1) Where galvanised conducting materials are used for earth electrodes, sacrificial cathodic protection anodes may be necessary to avoid rapid corrosion of the electrodes where the soil is aggressive. Specially prepared magnesium anodes (in a porous sack filled with a suitable “soil”) are available for direct connection to the electrodes. In such circumstances, a specialist should be consulted
Influence of the type of soil
Measurements on earth electrodes in similar soils are useful to determine the resistivity value to be applied for the design of an earth-electrode system
| Type of soil | Mean value of resistivity in Ωm |
|---|---|
| Swampy soil, bogs | 1 - 30 |
| Silt alluvium | 20 - 100 |
| Humus, leaf mould | 10 - 150 |
| Peat, turf | 5 - 100 |
| Soft clay | 50 |
| Marl and compacted clay | 100 - 200 |
| Jurassic marl | 30 - 40 |
| Clayey sand | 50 - 500 |
| Siliceous sand | 200 - 300 |
| Stoney ground | 1,500 - 3,000 |
| Grass-covered-stoney sub-soil | 300 - 500 |
| Chalky soil | 100 - 300 |
| Limestone | 1,000 - 5,000 |
| Fissured limestone | 500 - 1,000 |
| Schist, shale | 50 - 300 |
| Mica schist | 800 |
| Granite and sandstone | 1,500 - 10,000 |
| Modified granite and sandstone | 100 - 600 |
Fig. E23: Resistivity (Ωm) for different types of soil
| Type of soil | Average value of resistivity in Ωm |
|---|---|
| Fertile soil, compacted damp fill | 50 |
| Arid soil, gravel, uncompacted non-uniform fill | 500 |
| Stoney soil, bare, dry sand, fissured rocks | 3,000 |
Fig. E24: Average resistivity (Ωm) values for approximate earth-elect
Measurement and constancy of the resistance between an earth electrode and the earth
The resistance of the electrode/earth interface rarely remains constant
Among the principal factors affecting this resistance are the following:
*Humidity of the soil
The seasonal changes in the moisture content of the soil can be significant at depths of up to 2 meters.
At a depth of 1 metre the resistivity and therefore the resistance can vary by a ratio of 1 to 3 between a wet winter and a dry summer in temperate regions
*Frost
Frozen earth can increase the resistivity of the soil by several orders of magnitude. This is one reason for recommending the installation of deep electrodes, in particular in cold climates
*Ageing
The materials used for electrodes will generally deteriorate to some extent for various reasons, for example:
*Chemical reactions (in acidic or alkaline soils)
*Galvanic: due to stray DC currents in the earth, for example from electric railways, etc. or due to dissimilar metals forming primary cells. Different soils acting on sections of the same conductor can also form cathodic and anodic areas with consequent loss of surface metal from the latter areas. Unfortunately, the most favourable conditions for low earth-electrode resistance (i.e. low soil resistivity) are also those in which galvanic currents can most easily flow.
*Oxidation
Brazed and welded joints and connections are the points most sensitive to oxidation. Thorough cleaning of a newly made joint or connection and wrapping with a suitable greased-tape binding is a commonly used preventive measure.
Measurement of the earth-electrode resistance
There must always be one or more removable links to isolate an earth electrode so that it can be tested.
There must always be removable links which allow the earth electrode to be isolated from the installation, so that periodic tests of the earthing resistance can be carried out. To make such tests, two auxiliary electrodes are required, each consisting of a vertically driven rod.
Ammeter method (see Fig. E25)
Fig. E25: Measurement of the resistance to earth of the earth electrode of an installation by means of an ammeter
When the source voltage U is constant (adjusted to be the same value for each test)
then: 
In order to avoid errors due to stray earth currents (galvanic -DC- or leakage currents from power and communication networks and so on) the test current should be AC, but at a different frequency to that of the power system or any of its harmonics. Instruments using hand-driven generators to make these measurements usually produce an AC voltage at a frequency of between 85 Hz and 135 Hz.
The distances between the electrodes are not critical and may be in different directions from the electrode being tested, according to site conditions. A number of tests at different spacings and directions are generally made to cross-check the test results.
Use of a direct-reading earthing-resistance ohmmeter
These instruments use a hand-driven or electronic-type AC generator, together with two auxiliary electrodes, the spacing of which must be such that the zone of influence of the electrode being tested should not overlap that of the test electrode (C). The test electrode (C) furthest from the electrode (X) under test, passes a current through the earth and the electrode under test, while the second test electrode (P) picks up a voltage. This voltage, measured between (X) and (P), is due to the test current and is a measure of the contact resistance (of the electrode under test) with earth. It is clear that the distance (X) to (P) must be carefully chosen to give accurate results. If the distance (X) to (C) is increased, however, the zones of resistance of electrodes (X) and (C) become more remote, one from the other, and the curve of potential (voltage) becomes more nearly horizontal about the point (O).
In practical tests, therefore, the distance (X) to (C) is increased until readings taken with electrode (P) at three different points, i.e. at (P) and at approximately 5 metres on either side of (P), give similar values. The distance (X) to (P) is generally about 0.68 of the distance (X) to (C).
[a] the principle of measurement is based on assumed homogeneous soil conditions. Where the zones of influence of electrodes C and X overlap, the location of test electrode P is difficult to determine for satisfactory results.
[b] showing the effect on the potential gradient when (X) and (C) are widely spaced. The location of test electrode P is not critical and can be easily determined.
Fig. E26: Measurement of the resistance to the mass of earth of electrode (X) using an earth-electrode-testing ohmmeter