Short T.A. Electric Power Distribution Handbook

Подождите немного. Документ загружается.

Grounding and Safety 669

Ground rods are the most common grounding electrode. The NESC

requires at least an 8-ft (2.45-m) rod; iron or steel rods must be at least 5/8

in. (15 mm) in diameter; copper-clad stainless steel, or stainless steel-clad

rods must be at least 1/2-in. (12 mm). Rods should be driven to at least 8 ft

(2.45 m). If a driven rod hits a rock bottom, the depth may be less than 8 ft

(2.45 m). Strategies are to drive the rod at an angle or to use another type

of ground such as a buried strip. NESC requires buried strips to be at least

10 ft (3 m) and have a total surface area of at least 5 ft

(0.47 m

). Guy anchors

are not allowed as grounding electrodes.

The resistivity of a ground rod is

where

r = soil resistivity, W-m

L = length of the ground rod, m

a = radius of the ground rod, m

As a good approximation for 10-ft (3.3-m) ground rods, the resistance in

ohms equals the ground resistivity in W-m divided by three. Divide earth

resistivity by 2.5 for an 8-ft rod.

Rod diameter has little impact on the electrode resistance; the diameter is

needed mainly for mechanical strength and to ensure the rod has enough

material to survive corrosion.

Twenty percent of the resistance of a ground rod occurs within 1.2 in.

(3 cm) of the rod, and half of the resistance is in the ﬁrst 6 in. (15 cm) (IEEE

Std. 142-1991). These numbers vary with the ground resistivity.

The number of rods and the length are the main factors that inﬂuence the

resistance. Rods can be driven deeper. Normally, crews drive a section and

then thread on another rod section to pound it in deeper. Deeper rods are

best when lower levels of the soil have lower resistivity, normally because

of more moisture.

Multiple ground rods in parallel effectively reduce the overall resistance.

But, if two ground rods are too close together; they act as one ground rod

with a larger diameter, reducing much of the gain of using parallel rods.

One common rule: separate the rods by a distance of at least the length of

one of the ground rods. The NESC requires ground rods to be at least 6 ft

(1.8 m) apart. Even with these separations, the resistance of n ground rods

in parallel is more than 1/n times the resistance of a single rod. Table 13.2

shows multiplying factors and ﬁnal resistances for rods placed symmetri-

cally one rod length apart.

The NESC allows other grounding electrodes. Water piping systems make

fantastic grounds if they are available; grounds less than 1 W are common.

Unfortunately, conducting water pipes are not typically installed anymore,

1ln

1791_book.fm Page 669 Monday, August 4, 2003 3:20 PM

670 Electric Power Distribution Handbook

so water system grounds are less common. Gas piping systems are not

allowed as grounding electrodes.

The NESC allows the concentric neutral of a cable to be a ground for direct-

buried cables of at least 100 ft (30 m). The cable neutral must be bare or have

a semi-conducting jacket with a radial resistivity not exceeding 10 W-m

(meaning that 30 m of cable has a resistance of 0.33 W across the jacket).

The NESC allows other “made” electrodes other than ground rods includ-

ing buried wires, strips, and plates. These are especially appropriate in areas

with a rock bottom. Buried wires must have at least a 0.162 in. (4 mm)

diameter, be at least 100 ft (30 m) long, and be at least 18 in. (0.45 m) deep

(less is allowed if a rock bottom is hit). Strips must be at least 10 ft (3 m)

long, buried at least 18 in. (0.45 m) deep, and have a total surface area on

both sides of 5 ft

(0.47 m

). Plates must be buried at least 5 ft (1.5 m) deep

and have a total surface area of at least 2 ft

(0.185 m

). Both plates and strips

must have a thickness of at least 0.06 in. (1.5 mm) wide for nonferrous metals

and at least 0.25 in. (6 mm) wide for ferrous metals. Table 13.3 shows resis-

tance calculations for several different conﬁgurations.

Two-pole grounding techniques, butt plates and wire wraps (see Figure

13.14) are available but have several deﬁciencies (Clapp, 1997). A driven

ground provides more than twice the surface area of a butt plate. Addition-

ally, pole movement can reduce the contact between the butt plate or wire

wrap and the earth, and the pole can draw moisture out of the soil enough

to increase its resistivity. The NESC only considers butt plates and wire wraps

as appropriate in areas with low ground resistivity (less than 30 W-m), and

even then, two such grounds must be used, and they must not be used as

the sole ground at transformer locations.

13.4.1 Soil Characteristics

Soil resistivity is the resistance of a certain volume of soil. Normally, resis-

tivity is speciﬁed in W-m, the resistance between opposite faces of cube of

TABLE 13.2

Resistance of Multiple Rods

Number of Rods

Multiplying

Factor

Total Resistance as a

Percentage of One Rod

2 1.16 58%

3 1.29 43%

4 1.36 34%

8 1.68 21%

12 1.80 15%

16 1.92 12%

20 2.00 10%

24 2.16 9%

Source: IEEE Std. 142-1991, IEEE Recommended Practice for

Grounding of Industrial and Commercial Power Systems.

1791_book.fm Page 670 Monday, August 4, 2003 3:20 PM

Grounding and Safety 671

soil with a volume of 1 m

. Units of W-cm are also common — the resistance

between faces of a cube with a volume of a cubic centimeter: W-cm = 100 ¥

W-m.

Soil resistivities vary widely. Rich, moist organic soil may have a resistivity

of 10 W-m, while bedrock may have resistivities greater than 10

W-m. Resis-

tivity depends on the soil grain characteristics (size, variability, and density)

as well as moisture and chemical content. See Table 13.4 for typical values.

TABLE 13.3

Ground Electrode Resistance Calculations

Sources: Dwight, H. B., “Calculation of Resistance to Ground,” AIEE Transactions, vol. 55, pp.

Hemisphere

radius a

R =

2πa

One ground rod

length L,radius a

R =

2πL





 1

Twoground rods

s > L;spacing s

R =

4πL

4πs

1 

...

Twoground rods

s < L;spacing s

R =

4πL

+ln

 2+



16L

512L

...

Buried horizontal wire

length 2L,depth s/2

R =

4πL

+ln

 2+



16L

512L

...

Right-angle turn of wire

length of arm L,depth s/2

R =

4πL

+ln

 0.2373 + 0.2146

+0.1035

 0.0424

...

Three-point star

length of arm L,depth s/2

R =

6πL

+ln

+1.071  0.209

+0.238

 0.054

...

Four-point star

length of arm L,depth s/2

R =

8πL

+ln

+2.912  1.071

+0.645

 0.145

...

Six-point star

length of arm L,depth s/2

R =

12πL

+ln

+6.851  3.128

+1.758

 0.490

...

Eight-point star

length of arm L,depth s/2

R =

16πL

+ln

+10.98 5.51

+3.26

 1.17

...

Ring of wire, depth s/2

diameter of ring D,

diameter of wire d

R =

2π

+ln

Buried horizontal strip

length 2L,section a by b,

depth s/2, b < a/8

R =

4πL

 πab

2(a + b)

+ln

 1+



16L

512L

...

Buried horizontal round

plate, radius a,depth s/2

R =

4πs

1 

...

Buried vertical round

plate, radius a,depth s/2

R =

4πs

320

...





1791_book.fm Page 671 Monday, August 4, 2003 3:20 PM

672 Electric Power Distribution Handbook

FIGURE 13.14

Butt plate and wire wrap grounds. (From RUS 1728F-803, Speciﬁcations and Drawings for 24.9/

14.4 kV Line Construction, United States Department of Agriculture, Rural Utilities Service, 1998.)

TABLE 13.4

Resistivity of a Variety of Soils along with Resistances of a 10-Foot Ground Rod

Soil

Average

Resistivity, WW

-m

Resistance of a

5/8 in. (16 mm)

¥¥

10 ft (3 m) Rod, WW

Well graded gravel, gravel-sand mixtures, little or

no ﬁnes

600–1000 180–300

Poorly graded gravels, gravel-sand mixtures, little

or no ﬁnes

1000–2500 300–750

Clayey gravel, poorly graded gravel, sand-clay

mixtures

200–400 60–120

Silty sands, poorly graded sand-silts mixtures 100–500 30–150

Clayey sands, poorly graded sand-clay mixtures 50–200 15–60

Silty or clayey ﬁne sands with slight plasticity 30–80 9–24

Fine sandy or silty soils, elastic silts 80–300 24–90

Gravelly clays, sandy clays, silty clays, lean clays 25–60

17–18

Inorganic Clays of high plasticity 10–55

3–16

Soils highly inﬂuenced by moisture.

Plan View

Grounding Plate

Staple

Butt Plate Wire Wrap

1791_book.fm Page 672 Monday, August 4, 2003 3:20 PM

Grounding and Safety 673

Soil resistivity often varies signiﬁcantly with depth. In most areas, resis-

tivity improves signiﬁcantly with depth as moisture increases, and normally,

grounds driven down to the water table provide a very good ground. In

other areas, an underlying bedrock signiﬁcantly increases resistivity at lower

levels (and makes driving ground rods difﬁcult). Several different layers of

soil with different resistivities may be present. A two-layer soil model is

often used for engineering analysis.

Soil resistivity is primarily affected by

• Moisture — Moisture is one of the main factors determining soil

resistivity. The dryer the soil, the higher the resistivity. Changes in

moisture level throughout the year are the biggest reason for the

change in ground electrode resistance. Once the moisture exceeds

22%, resistivity changes very little (IEEE Std. 142-1991).

• Temperature — Above the freezing point of water, temperature does

not impact resistivity signiﬁcantly. Below freezing, resistivity rises

signiﬁcantly.

• Salts — The presence of soluble salts signiﬁcantly impacts the

resistivity.

Figure 13.15 shows these three main effects for different example soils.

These effects are somewhat interrelated and depend on the soil’s speciﬁc

characteristics.

One option to reduce ground electrode resistivity is to use chemical

treatments or backﬁlls to reduce the soil resistivity near the electrode.

Bentonite backﬁlls or salt treatments such as sodium chloride or calcium

chloride signiﬁcantly reduce the soil resistivity. If soil treatments are used

to enhance ground rods, make sure to consider the effect on corrosion.

Some chemical treatments increase the metal corrosion rate. Also, consider

the environmental impacts and regulations. Most normal distribution

grounding needs outside of the substation do not warrant chemical

enhancement.

If a grounding electrode sustains signiﬁcant current, the current may dry

out the surrounding soil and increase the resistance of the electrode. With a

multigrounded system, several grounds share current, but on systems with

isolated grounds, sustained current through a grounding electrode may

degrade the electrode’s resistance (another good reason to use multiple

grounding electrodes in parallel whenever possible).

The resistivity of soil is normally very linear, but for very high voltage

stresses, the soil breaks down; and arcing in the soil reduces the effective

resistivity. For a ground rod, the soil breaks down only when high current

ﬂows into the rod, normally from lightning (a localized effect) or from faults

with very little sharing between ground rods. The breakdowns improve the

resistance of a grounding electrode.

1791_book.fm Page 673 Monday, August 4, 2003 3:20 PM

674 Electric Power Distribution Handbook

13.4.2 Corrosion and Grounding Electrodes

Corrosion is a major consideration for grounding electrodes. Ground rods

are normally copper-clad steel or galvanized steel. Most grounding wires

are copper. Steel and aluminum corrode too quickly for grounding use. One

problem with copper is its attractiveness to thieves. In problem areas, copper-

clad steel has reduced thefts since it has less commercial value.

Mixing metals signiﬁcantly increases corrosion. The problem is galvanic

corrosion. Different metals create an electrochemical battery in the soil. A

voltage develops between grounding electrodes made of different metals (or

between a grounding electrode and another metallic object in the ground).

When these electrodes are connected, current ﬂows in the earth between the

metal objects, corroding one of the metals at a much higher-than-normal rate.

Galvanic corrosion between dissimilar metals can eat up grounding elec-

trodes. Galvanic corrosion operates on the same principal as a battery (bat-

teries release their energy with controlled corrosion). Pure metals are

unstable; they are found in nature as ores and must be reﬁned to a pure

state. Metals corrode to return to their natural state. Iron dissolves in a

corrosive electrolyte and gives up electrons, forming a positive ion. Some of

the iron ions attach to free oxygen and become iron oxide (rust). At low

FIGURE 13.15

Effect of salt, moisture, and temperature on soil resistivity. (From IEEE Std. 80-2000. Copyright

1791_book.fm Page 674 Monday, August 4, 2003 3:20 PM

Grounding and Safety 675

corrosion rates, the metal reaches equilibrium with the surrounding electro-

lyte that prevents additional iron from releasing electrons. Two different

metals separated by an electrolyte (a conducting medium) have a voltage

difference between them as they give up electrons at different rates. If these

different metals are electrically connected together, current ﬂows between

the metals and increases corrosion.

Consider a galvanized-steel ground rod next to a copper ground rod. The

galvanized rod is coated with zinc. The zinc releases more electrons into the

electrolyte, so it becomes positively charged relative to the copper rod. At

the zinc surface (the positive terminal, the anode), the ﬂow of this dc current

dissolves the zinc at an increasing rate. At the copper rod (the negative

terminal), the galvanic circuit decreases the rate of corrosion. Figure 13.16

shows the galvanic series for several different metals. When dissimilar metals

form a galvanic circuit, the most anodic of the metal corrodes (corrosion

slows down, stops, or even reverses on the cathodic metal).

In fact, most corrosion occurs by the same action. An anode and a cathode

develop between different portions of a metal, and galvanic currents ﬂow

FIGURE 13.16

Relative potentials of various metals in an electrolyte.

Pure magnesium

Magnesium Alloy

Zinc

Aluminum Alloy

Pure Aluminum

Mild Steel (Clean & Shiny)

Mild Steel (Rusted)

Cast Iron, Lead

Mild Steel in concrete,

Copper, Brass, Bronze,

Mill Scale on steel,

High silicon cast iron

Carbon, Graphite, Coke

-2.0

-1.0

0.0

Potential, volts

relative to Copper

Sulfate

(Cu-CuSO

)

Most Anodic

Most Cathodic

(Easiest to corrode)

1791_book.fm Page 675 Monday, August 4, 2003 3:20 PM

676 Electric Power Distribution Handbook

between them, corroding the anode. One part of a metal may become anodic

to another in several ways based on differences in impurities in the metal,

mechanical stresses, and oxygen concentration.

Copper has naturally good properties that limit corrosion. When copper

corrodes, it becomes copper oxide; the copper-oxide layer greatly reduces

the corrosion rate and protects the underlying copper. Copper is normally

the most cathodic element, so it rarely corrodes because of dissimilar metals,

but it may help corrode other grounding electrodes.

A zinc coating on a steel ground rod (which is a galvanized steel rod)

provides corrosion protection for the rod. Zinc by itself is resistant to corro-

sion because it develops a self-protecting zinc-carbonate ﬁlm that retards

further corrosion. In addition, the zinc provides galvanic protection to the

steel core. If part of the rod has exposed steel (due to cuts, scratches, or

corrosion of the zinc), the zinc acts as a very large anode to the small area

of exposed steel. Galvanic action reduces the corrosion on the exposed steel

(the cathode).

With lower soil resistivity, galvanic currents are higher, so grounding elec-

trodes corrode faster. Resistivity is the main soil parameter that determines

corrosion. Other factors that increase corrosion are very high or very low pH

levels, high water content, and the presence of chlorides (salts) and sulfates.

To limit excessive corrosion caused by different metals, avoid using dif-

ferent grounding rods (either stick with galvanized steel or with copper-clad

steel). Provide as much separation as possible between ground electrodes of

different metals. Consider the presence of guy anchors and other possible

buried wires including bare concentric neutral cables and water pipes. With

galvanized steel rods, use steel or aluminum pole ground leads. With copper-

clad steel rods, use copper down leads.

13.4.3 Resistance Measurements

The NESC does not specify how a ground measurement should be taken.

IEEE Std. 81-1983 documents several of the methods of measuring grounds.

Normally, on systems with a multigrounded neutral, the actual impedance

is not critical, so neither is the measurement. (Just make sure ground rods

are connected.) On systems without a multigrounded neutral, individual

grounds (and ground resistance checks) are more important.

Accurate ground electrode measurements of a concentrated electrode

(ground rods, anchors, or short strips of wire) are normally performed using

a three-point method (AEMC Instruments, 1998; Biddle Instruments, 1982;

IEEE Std. 81-1983). Two auxiliary ground probes are driven into the earth

forming a straight line with the test electrode. The tester injects current I

through the electrode under test and the outer auxiliary electrode. The tester

measures voltage V between the test electrode and the inner auxiliary elec-

trode. If the middle electrode is out of the inﬂuence of the outer two elec-

trodes, then the resistance of the test electrode is V/I.

1791_book.fm Page 676 Monday, August 4, 2003 3:20 PM

Grounding and Safety 677

The fall of potential method is the most common three-point technique (see

Figure 13.17). It requires several measurements. Take several measurements

with the middle electrode at different locations between the test electrode

and the outer electrode. Plot the resistance vs. the distance, and pick the

resistance from the portion of the curve where resistance ﬂattens out. The

point where the fall-of-potential curve ﬂattens out is most likely near 62%

of the distance from the test electrode to the outer electrode. A simpliﬁed

version is just to measure near the 62% mark.

The three-point methods must be done with the ground electrode discon-

nected from the multigrounded neutral and connections to other grounds

that may interfere (including cable TV and telephone grounding connec-

tions). When disconnecting a grounding conductor from the multigrounded

neutral, crews should know that a signiﬁcant potential can develop between

the neutral and the disconnected ground if a line-to-ground fault occurs

somewhere on the line.

FIGURE 13.17

Three-point ground-electrode measurement.

ZYX

Best estimate

62%0% 100%

Position of electrode Y

Resistance

measured

Current injected

Voltage measured

Area of influence of the

tested ground rod

Test

ground

rod

1791_book.fm Page 677 Monday, August 4, 2003 3:20 PM

678 Electric Power Distribution Handbook

To have adequate spacing with the thee-point method, normally the outer

electrodes must be spaced at least 75 to 100 ft apart (23 to 30 m). Deeper

grounds or wider grounds (strips or multiple ground rods) require more

distance between the electrodes.

If an excellent ground is available nearby as a reference, the two-point

measurement technique is suitable. Just like an ohmmeter, the tester injects

current through the grounding electrode under test and the reference

ground. The resistance reading is the sum of both grounds, so if the reference

ground is very good (like a water piping system), the reading is predomi-

nantly the resistance of the test electrode. A multigrounded neutral can be

used as the ground reference but only if the frequency of the injected signal

is different from the standard power frequency (normal power-frequency

current ﬂows can foul the ground reading). As with the three-point method,

the ground electrode must be disconnected from the multigrounded neutral.

A convenient ground electrode tester is a clamp-on device that tests a

grounding electrode by clamping around the grounding lead. The clamp-on

unit only works on electrodes connected to a very good ground like the

multigrounded neutral (it relies on the good ground). It works just like the

two-point test. The clamp-on is a current transformer that couples a voltage

to the grounding lead; the voltage is in the lower kilohertz range. The

resistance of the grounding electrode is the voltage divided by the current

(if all of the parallel grounds provide a low-impedance path, the resulting

impedance is only the grounding electrode).

The clamp-on ground-resistance tester more than adequately tests ground-

rod connections on systems with a multigrounded neutral. It is fast and

convenient, crews do not have to disconnect the ground rod, and it tests the

connector continuity. Since values of individual grounds are not as critical

on multigrounded-neutral systems, the clamp-on’s accuracy is good enough

for most grounding checks on distribution circuits.

Ground resistivity is normally measured using the four-point method.

Four electrodes of the same size are driven to the same depth in a straight

line with equal spacing between them. A tester injects current through the

outer two electrodes and measures the voltage between the inner two elec-

trodes. The distance between two electrodes determines the depth at which

the average resistivity is measured. If the electrode spacing is at least 20

times the electrode depth, the resistivity is

r = 2 p A R

where

r = earth resistivity, W-m

A = electrode spacing, m

R = resistance reading from the ground tester, W

Since the rod spacing determines the depth at which the resistivity is

measured, testing with multiple electrode separations shows how the soil

1791_book.fm Page 678 Monday, August 4, 2003 3:20 PM