lEE E Std 81-1962

IEEE Recommended Guide for

Measuring Ground Resistance and
Potential Gradients in the Earth

N
c.o
....
C)

....
o

...
CO
"C
(J'J
W
W
W

Published by The Institute of Electrical and Electronics Engineers, Inc. 345 East 47th Street, New York, New York 10017
• 8H00273

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 IEEE Recommended Guide for
Measuring Ground Resistance and
Potential Gradients in the Earth

Published by

Institute of Electrical and Electronics Engineers

345 East 47th Street, New York, N. Y. 10017

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 ACKNOWLEDGMENT

The Institute wishes to acknowledge its indebtedness to those who have so freely given of
their time and knowledge and have conducted experimental work on which many of the A I E E
publications are based.

This publication was prepared by the Ground Resistance and Potential Gradient Measure­
ments Subcommittee of the Special Instruments and Auxiliary Apparatus Committee. The
Subcommittee is composed of:

John C. Parker, Choir man

A. L. Rickley, Secretary

E. L. Able George D. Lippert

John G. Anderson C. W. Lund
1). W . Bodel L. W. Matsch
E. J. Bonnesen Edgar H . Olsen
Myron A. Bostwick Stanley G. Paun
J. G. Cassan B. Lawrence Philip
E. B. Curdts C. S. Roadhouse
V. C. Detty Leon N. Sherban
C. A. Duke George S. Smith
John I. Holbeck G. Gordon Smith
Paul B. Jacob, Jr. Harry Sohon
D. T. Jones A. W. Spinks
Allen L. Kinyon T. B. Thompson
B. E. Lenehan

Appreciation is expressed to the following members of the subcommittee who participated

earlier in the development of this publication.
W. C. Ball C. H. Jensen
J. P. Holm V. H . Kraybill
R. W. Inman M. A. Leland

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 CONTENTS

SECTION PAGE
1.00 PURPOSE 5
2.00 SCOPE 5
3.00 OBJECTIVES OF TESTS 5
4.00 DEFINITIONS 5
S.OO CHARACTERISTICS OF A GROUNDING
CONNECTION 7
5.01 General 7
5.02 Characteristics 7
5.03 Earth Resistivity 7
5.04 Current Flow from an Electrode 7
5.05 Equipotential Lines 8
5.06 Theoretical Value of Ground Resistance 8
6.00 SAFETY PRECAUTIONS WHILE MAKING
GROUND TESTS 8
6.01 Station Ground Tests 8
6.02 Lightning Arrester Ground Tests 9
6.03 Small Isolated Ground Tests 9
7.00 GENERAL CONSIDERATIONS 9
7.01 Complexities 9
7.02 Stray Direct Currents 9
7.03 Stray Alternating Currents 9
7.04 Reactive Component of Impedance of a Large
Ground System 10
8.00 METHODS OF MEASURING EARTH
RESISTIVITY 10
8.01 Two-Point Method 10
8.02 Four-Point Method 10
8.03 Soil Samples 10
9.00 METHODS OF MEASURING GROUND
RESISTANCE 11
9.01 General 11
9.02 Two-Point Method 11
9.03 Three-Point Method 11
9.04 Fall-of-Potential Method 11
9.05 Ratio Method 12
9.06 Staged Fault Tests 12
10.00 AUXILIARY ELECTRODES 13
11.00 COUPLING BETWEEN TEST CIRCUITS 13
12.00 INSTRUMENTATION FOR GROUND
RESISTANCE MEASUREMENTS 13
12.01 Ratio Ohmmeter 13
12.02 Double-Balance Bridge 14
12.03 Single-Balance Transformer 15
12.04 Ammeter-Voltmeter 15
12.05 Integrity of Ground Grid Tests 15

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 13.00 POTENTIAL GRADIENT MEASUREMENTS 16
13.01 Potential Contour Surveys 16
13.02 Step and Touch Voltages 17
14.00 MODEL TESTS 17
14.01 Purpose 17
14.02 Material Required 17
14.03 Resistance Test Procedure 17
14.04 Potential Gradient Test Procedure 18
14.05 Principles to be Followed in Use of Model 18
15.00 BIBLIOGRAPHY 19

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 RECOMMENDED GUIDE FOR MEASURING GROUND RESISTANCE
AND POTENTIAL GRADIENTS IN THE EARTH
1.00 PURPOSE 3.00 OBJECTIVES OF TESTS
1.01 It is the purpose of this Guide to describe and dis­ 3.01 Measurements of ground resistance or impedance
cuss the present state of the technique of measur­ and potential gradients on the surface of the earth
ing ground resistance and impedance, earth resis­ due to ground currents are necessary to:
tivity, and potential gradients resulting from cur­ (1) Verify the adequacy of a new grounding sys­
rents in the earth and the prediction of the magni­ tem.
tudes of ground resistance and potential gradients
from scale model tests.1 Factors governing the (2) Detect changes in an existing grounding sys­
choice of instruments and the techniques for various tem.
types of work are covered, such as the purpose of (3) Determine hazardous step and touch voltages.
the measurement, the accuracy required, the time (4) Design protection for personnel and for power
and instruments available, and the nature of the and communication circuits.
ground or grounding system under test.
3.02 Scale-model tests are useful in developing new de­
1.02 The Guide is intended to assist the engineer or the signs for grounding systems and for comparing
technician in obtaining accurate, reliable data, but them with old ones.
not in the interpretation of such data. It points out
3.03 Earth resistivity measurements are useful for:
test procedures to promote the safety of personnel
and property, and to prevent interference with the (1) Estimating the ground resistance of a proposed
operation of neighboring facilities. station structure or grounding system.
(2) Estimating potential gradients including step
and touch voltages.
2.00 SCOPE
(3) Computing the inductive coupling between
2.01 The testing methods covered in this Guide include neighboring power and communication circuits.
those applicable to: (4) Designing cathodic protection systems.
(1) The measurement of the resistance and im­
pedance to earth of electrodes varying from 4.00 DEFINITIONS
small rods and plates to large grounding sys- 4 QJ Definitions as given herein apply specifically to the
terns of stations. application of this Guide. For additional definitions
(2) Ground potential surveys, including the meas­ see American Standard Definitions of Electrical
urement of step and touch voltages, and po­ Terms C42-1941, C42.35-1957 and C42.65-1957, or
tential contour surveys. the latest revision thereof approved by the Ameri­
can Standards Association. The three-part numbers
(3) Scale-model tests for laboratory determination in parenthesis preceding the titles of some of the
of the ground resistance and potential gradients definitions included in this Guide.indicate the num­
for an idealized design. ber of the corresponding definition in C42. The
definitions which do not carry the three-part num­
(4) The measurement of earth resistivity. ber will be submitted for consideration of inclusion
2.02 The methods covered herein are generally limited in the next revision of C42.
to those using direct current or periodically re­ 4.0210 (35.15.005) Ground. A ground is a con­
versed direct current, or commercial power fre­ ducting connection, whether intentional or acci­
quencies. This Guide does not claim to cover all dental, by which an electric circuit or equipment is
possible methods, even within these limitations. connected to the earth or to some conducting
body, of relatively large extent, which serves in
2.03 Extreme precision is seldom possible because of the place of the earth. It is used for establishing and
many variables encountered; therefore, the meas­ maintaining the potential of the earth (or of the
urements should be carefully made by the most suit­ conducting body) or approximately that potential,
able method available, with a thorough understand­ on conductors connected to it, and for conduct­
ing of the possible sources of error. ing ground current to and from the earth (or the
conducting body).
2.04 Definitions of terms pertinent to the subject matter
are given, with references to their sources where 4.0211 (35.15.010) Grounded. Grounded means
known. Those approved or standardized by other that the system, circuit, or apparatus referred to is
bodies are used wherever possible. provided with a ground.

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 4.0212 (35.15.015) Ground-Return Circuit. A 4.0223 Potential Profile. A potential profile is a
ground-return circuit is a circuit in which the plot of potential as a function of distance along a
earth is utilized to complete the circuit. specified path.

4.0213 (35.15.020) Ground Current. G r o u n d cur­ 4.0224 Surface Potential Gradient. Surface po­
rent is current flowing in the earth or in a ground­ tential gradient is the slope of a potential profile,
ing connection. the path of which intersects equipotential lines at
right angles.
4.0214 Grounding Electrode. A g r o u n d i n g elec­
trode is a conductor imbedded in the earth, used 4.0225 Touch Voltage. T o u c h voltage is the po­
for maintaining ground potential on conductors tential difference between a grounded metallic
connected to it, and for dissipating into the earth structure and a point on the earth's surface equal
current conducted to it. to the normal maximum horizontal reach, approx­
imately one meter.
4.0215 (35.15.105) Grounding Connection. A
grounding connection is a connection used in es­ 4.0226 Step Voltage. Step voltage is the poten­
tial difference between two points on the earth's
tablishing a ground and consists of a grounding
surface, separated by a distance of one pace, which
conductor, a grounding electrode and the earth
will be assumed to be one meter, in the direction
(soil) which surrounds the electrode.
of maximum potential gradient.
4.0216 Ground Grid. A ground grid is a system
4.0227 (05.30.005) Conductivity. T h e conductiv­
of grounding electrodes consisting of interconnected
ity of a material is the direct-current conductance
bare cables buried in the earth to provide a common
between the opposite, parallel faces of a portion of
ground for electrical devices and metallic struc­
the material having unit length and unit cross-
tures. It may be connected to auxiliary grounding
section.
electrodes to lower its resistance.
4.0228 (05.30.015) Resistivity. T h e resistivity of
4.0217 Ground Mat. A ground mat is a system
a material is the reciprocal of its conductivity.
of bare conductors, on or below the surface of the
earth, connected to a ground or a ground grid to Note: In the earlier work on earth currents and impedance,
earth conductivity ( \ ) was generally used, expressed in
provide protection from dangerous touch voltages. abmhos per centimeter cube. At present, most engineers use
Note: Plates and gratings of suitable area are common earth resistivity (p), expressed in ohms per meter cube
(commonly written "meter-ohms") or in ohms per centi­
forms of grotind mats. meter cube ("ohm-cm").

4.0218 Grounding System. A grounding system

CONVERSION T A B L E
consists of all interconnected grounding connections
Resistivity, (p) Resistivity, (p) Resistivity, (p) Conductivity, (X)
in a specific area. meter-ohms ohm-cm ohm-feet abmhos per cm"

4.0219 Ground Resistance. T h e ground resis­ 1 100 3.28 10-n

tance of a grounding electrode is the ohmic resis­ 10 1,000 32.8 IO-12
tance between it and a remote grounding electrode 50 5,000 164 2xl0-i3
of zero resistance. By "remote" is meant at a dis­ 100 10,000 328 IO-13
tance such that the mutual resistance of the two 200 20,000 656 5 x 10-"
electrodes is essentially zero. 1,000 100,000 3,280 IO-14
10,000 1,000,000 32,800 10-i5

4.0220 Mutual Resistance of Grounding Elec-

trodes. T h e mutual resistance of t w o ground­ 4.0229 (65.02.267) Coupling. Coupling is the as­
ing electrodes is equal to the voltage change in sociation of two or more circuits or systems in
one of them produced by one ampere of direct such a way that power may be transferred from
current in the other, and is expressed in ohms. one to another.

4.0230 (65.02.270) Capacitive Coupling. Capaci-

4.0221 (05.15.030) Electric Potential. T h e elec­
tive coupling is the association of two or more
tric potential of a point is the potential difference
circuits with one another by means of capacitance
between the point and some equipotential surface,
mutual to the circuits.
usually the surface of the earth, which is arbitrar­
ily chosen as having zero potential. A point which 4.0231 (65.02.273) Resistive Coupling. Resistive
has a higher potential than a zero surface is coupling is the association of two or more circuits
said to have a positive potential; one having a with one another by means of resistance mutual to
lower potential has a negative potential. the circuits.

4.0222 Equipotential Line or Contour. An equi­ 4.0232 (65.02.276) Direct Coupling. Direct coup­
potential line or contour is the locus of points hav­ ling is the association of two or more circuits by
ing the same potential at a given time. means of a self-inductance, capacitance, resistance

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 or a combination of these which is common to the sonnel safety, adequacy of insulation and continuity
circuits. of service.
4.0233 (65.02279) Inductive Coupling. 5.02 Characteristics
(1) In communication circuits, inductive coupling The characteristics of a grounding connection vary
is the association of two or more circuits with with the composition and physical state of the soil
one another by means of inductance mutual as well as with the extent and configuration of the
to the circuits. buried electrode. Earth in any given locality is
Note: This term, when used without modifying composed of various combinations of dry earth,
words, is commonly used for coupling by means
of mutual inductance, whereas coupling by means swampy ground, gravel, slate, sandstone or other
of self-inductance common to the circuits is natural materials of widely varying resistivity. It
called direct inductive coupling.
may be relatively homogeneous over a large area,
(2) In inductive coordination practice, inductive or it may be effectively saucered in granite, sand
coupling is the interrelation of neighboring or other matter having a high resistivity and thus
electric supply and communication circuits by be practically insulated from the surrounding area.
electric or magnetic induction or both. Consequently, the characteristics of a grounding
4.0234 (05.30.010) Effective Conductivity. The connection (ohmic resistance) vary with the sea­
effective conductivity of a material to a periodic sons which affect temperature, moisture content
current is the effective conductance between the and compactness.
opposite, parallel faces of a portion of the material Calculations and experience show that, in a given
having unit length and unit cross-section. soil, the effectiveness of a ground grid is dependent
largely upon the over-all size of the ground grid.
4.0235 (05.30.020) Effective Resistivity. The ef­
The addition of buried conductors and driven rods
fective resistivity of a material to a periodic cur­
within an enclosure nevertheless aid somewhat in
rent is the effective resistance between the oppo­
reducing the ground resistance. This reduction di­
site, parallel faces of a portion of the material
minishes with the addition of each successive con­
having unit length and unit cross-section.
ductor or rod. A good method for reducing the
ground resistance of a transmission line tower or
5.00 CHARACTERISTICS OF A mast is to install radial counterpoises.
GROUNDING CONNECTION After the installation of a substation or other
grounded structure, the settling of the earth with
5.01 General annual cyclical weather changes tends to reduce the
Connections to earth have resistance, capacitance ground resistance substantially during the first year
and inductance, all of which affect their current- or two.
carrying ability. The resistance of the connection is
of particular interest to those concerned with power 5,.03 Earth Resistivity
frequencies because it is affected by the resistivity Earth resistivity, expressed in ohm-centimeters, is
of the earth in the area of the connection. The the resistance of a cubic centimeter of earth meas­
capacitance and inductance values are of interest to ured between opposite surfaces. It can also be ex­
those concerned with higher frequencies, such as pressed in meter-ohms, (by dividing centimeter-
are associated with radio communications and light­ ohms by 100). Resistivity varies not only with the
ning. type of soil, but also with its temperature, mois­
Ground resistance measurements are made: (1) to ture content, compactness and homogeneity. Mix­
determine the actual resistance of the ground con­ tures of various materials, e.g., loam, sand, gravel,
nections; (2) as a check on calculations; (3) to rock, etc., may also be encountered. Moreover, bur­
determine: (a) the rise in ground potential, and its ied conductive objects also affect resistivity. The
variation throughout an area, that results from literature indicates that the values of resistivity
the flow of ground fault current in a power system, vary from 1.0 to 100 ohm-centimeters (0.01 to 1
(b) the suitability of a grounding connection for meter-ohm) for sea water and up to 1011 ohm-
lightning protection, and (c) the suitability of a centimeters (109 meter-ohms) for sandstone.
grounding connection for radio-frequency trans­ The resistivity of the earth increases slowly, with
mission at a transmitter; and (4) to obtain data decreasing temperatures, from 25C to 0 C. Below
necessary for the design of protection for buildings, 0 C the resistivity increases rapidly. The rate of
the equipment therein, and any personnel that may change of earth resistivity with temperature is dif­
be involved. ferent for different types of earth.
Ground connections of all power and communication
systems should be studied to determine the varia- 5 .04 Current Flow from an Electrode
tions in ground potential that can be encountered The amount of current flow from an electrode is
during ground fault conditions so as to assure per- usually determined primarily by the voltage and

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 impedance of the external circuit, rather than by which may be encountered. This gradient will be
the impedance of the electrode itself to the sur­ highest near the grounding electrode.
rounding earth. The latter is usually important
The distance between equipotential lines, meas­
primarily in determining the potential rise of the
ured along the surface of the earth radially from
electrode, and the magnitude of the gradients which
the grounding connection, will vary with a num­
will occur in its vicinity, during the current flow.
ber of factors. These include: variations in resis­
However, it should be pointed out that for low-
tivity of the earth, the presence of buried pipes,
voltage systems and also for small grounding
conduit, railroad rails, steel fences, metallic cable
systems, the impedance of the electrode to the sur­
sheaths and the presence of overhead lines carry­
rounding earth may greatly affect the current flow.
ing ground current.
The impedance of a grounding electrode is usually
measured in terms of resistance because the re­ As indicated in Section 5.04, a preponderance of
actance is generally negligible. This resistance will the ground current tends to return to the source,
not usually vary greatly from year to year after under the transmission line through which it flows.
the first year or two following the burial of the Consequently, it will be found that the ground po­
ground grid. Although the ground grid may be tential under the transmission line carrying fault
buried only a foot or two below the surface, the current will have a steeper gradient than in the
variation of the resistance for larger stations seems adjoining earth. This results in changing the pat­
to bear little relationship to the variation of the re­ tern of the equipotential lines whenever different
sistivity at the burial level. This is because the bulk transmission lines are faulted. Therefore, equipo­
of the current from a large ground grid would nor­ tential lines cannot be established simply by meas­
mally flow almost directly downward in the area uring resistance from the grounding connection to
under the station, unless the area is situated over various points around it.
a high-resistance stratum. When once established, the voltage between the
Although the above statements appear to be con­ equipotential lines for a given fault condition can
tradictory they are, nevertheless, true. Records be expected to vary directly with ground fault
which have been kept of large area ground grids current magnitude. This assumes no change in
over a period of eighteen years show little variation resistivity of the earth around the grounding con­
in the measured value of resistance, whereas re­ nection, during the flow of fault current.
sistivity measurements in the same area show wide
variations (as much as 17 to 1 at shallow depths). 5.06 Theoretical Value of Ground Resistance
It should be recognized that the resistance of a Calculated or theoretical value of the resistance of
grounding connection with a small number of driv­ an electrode to remote earth can vary considerably
en rods may vary more closely with that indicated from the measured value because of the following
by resistivity measurements. This indicates that factors:
the resistance of large area ground grids is propor­
tional to resistivity measurements made for greater (1) Condition of the soil at the time each measure­
depths where less variation is encountered. ment is made.
The current flow from a transmission line fault (2) Extent of the resistivity survey; i.e., number
to a substation ground grid tends to follow the and dispersal of tests, probe spacings, etc.
transmission line. Depth of mean current path is
(3) Area and configuration of the buried electrode.
directly proportional to the square root of the
earth resistivity and inversely proportional to the (4) Effect of adjacent buried conductors.
square root of the frequency. Thus resistance tends
In order to decrease the sources of error in estab­
to increase the cross-sectional area of the current
lishing the relationship between earth resistivity
path, whereas inductance tends to decrease it and
and ground resistance it is advisable to take re­
to tie it more closely to the transmission line. This
sistivity and resistance measurements under simi­
tendency will also affect the pattern of the current
lar weather and moisture conditions.
path away from the electrode.
Where resistivity varies with probe spacing and
5.05 Equipotential Lines consequently with depth, the measured resistance of
As a result of current flow from an electrode to an electrode buried in the vicinity will generally
earth and through its earth path, equipotential lines correlate more closely with resistivity values at
plotted at right angles to these current lines will deeper levels than those taken near the surface.
assume a shape controlled by the path of the cur­ Recent findings substantiate this relationship and
rent. The density of equipotential lines, having indicate that measurements for substation resistiv­
equal voltage differences between them, across a ity surveys should be taken at probe spacings of
path in a given direction determines the step voltage fifty feet or more.

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 6.00 SAFETY PRECAUTIONS WHILE high-potential gradient around the ground con­
MAKING GROUND T E S T S nection.

6.01 Station Ground Tests 603 Small Isolated Ground Tests

If there is no possibility of fault currents into the
It should be impressed on all test personnel that a
ground to be tested, the only precaution concerns
lethal potential can exist between the station
possible high-potential gradients around the test
ground and a remote ground if a system fault in­
electrodes. If current is passed into a remotely lo­
volving the station ground occurs while ground
cated electrode, as in the fall-of-potential method,
tests are being made.
it is worthwhile to insure against a curious person
Since one of the objects of tests on a station being allowed near the current electrode while tests
ground is the establishment of the location of an are in progress. To a much lesser extent, in rural
effectively remote point for both current and po­ areas grazing cattle should not be allowed near the
tential electrodes, the leads to these electrodes must test current electrode.
be treated as though a possible potential could
exist between these test leads and any point on the
7.00 GENERAL CONSIDERATIONS
station ground grid. Some idea of the magni­
tude of this possible potential may be gained from 7.01 Complexities
the consideration that even in the larger stations The measurement of ground resistance and the po­
the ground grid will have an impedance in the tential gradients resulting therefrom, introduce a
order of 0.08 ohm and an "average" maximum number of complexities not encountered in other
ground fault current is in the order of 5,000 resistance and potential measurements.2
amperes; therefore, under these conditions the po­ It may be necessary to make multiple measure­
tential to a remote point will be in the order of ments and to plot trends. Stray currents and other
400 volts. For higher ground impedance or greater factors may interfere with the measurements.
fault currents, the rise of station ground voltage With development and industrial growth adjacent
will be higher than 400 volts. In very few large to power substations, it becomes increasingly dif­
stations is the maximum rise of station ground ficult to pick a suitable direction or location for a
voltage less than 400 volts, and this value can be reference probe to make a resistance test More­
taken as the minimum to be expected with maxi­ over, the connection of overhead ground wires,
mum fault current into any station. For smaller buried water pipes, cable sheaths, etc, all have the
substations the fault currents are usually less than effect of physically distorting and enlarging the
5,000 amperes; but at the same time the ground ground grid.
impedance is greater than 0.08 ohm, with the net
effect of retaining the 400-volt minimum rise of In testing substation ground grids a precaution
station ground potential. should be to let the earth settle for a year after
construction for uniform compactness to insure
All of the preceding discussion points to the neces­ greater accuracy. Nevertheless, ground resistance
sity of caution when handling the test leads, and measurements should be made immediately after
under no circumstances should the two hands or the ground grid has been installed to be certain that
other parts of the body be allowed to complete the there are no major omissions of grounded com­
circuit between points of possible high potential ponents normally connected into the ground grid.
difference. It is true that the chances are extremely
remote that a station ground fault will occur while 7.02 Stray Direct Currents
test leads are being handled, but this possibility Conduction of electricity in the soil is electrolytic
should not be discounted. and the flow of direct current results in chemical
The use of rubber gloves is advisable in many in­ action and polarization potential difference. Direct
stances. The determination of when rubber gloves potentials are produced between various types of
should be used should be in accordance with the soil and between soil and metal by galvanic action.
safety procedures and practices adopted by the par­ Galvanic potentials, polarization, and, if present,
ticular organization involved. stray direct currents may seriously interfere with
direct-current measurements. Therefore, periodic­
6.02 Lightning Arrester Ground Tests ally reversed direct current is used in making
These grounds fall in a special category because measurements. However, when using periodically
of the extremely high short-duration currents car­ reversed direct current for resistance measurements
ried by lightning arrester grounds. These currents the resulting values will be fairly close, but they
may be in excess of 50,000 amperes for surge cur­ may not be accurate for all alternating-current
rents, with a possibility of power-follow currents applications. When an alternating current of a
in the case of a defective arrester. An isolated given frequency is used for ground resistance
lightning arrester ground should never be tested measurements, the results may not be applicable
with the arrester in service because of the possible for other frequencies.

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 7.03 Stray Alternating Currents tween the two outer electrodes and the potential E
Stray alternating currents in the earth, and even between the two inner electrodes is measured with
in the grounding system under test, and also test a potentiometer or high-impedance voltmeter. Then
electrodes present an additional complication. The E/I gives the mutual resistance R in ohms which
effects of stray alternating current may be miti­ can be substituted in the equation below. The re­
gated in ground resistance measurements by util­ sistivity p in the terms of which "a" and "b" are
izing a frequency that is not present in the stray measured is:
current. Most measuring devices use frequencies 4*aR
within a range of SO to 100 cycles per second.
The effects of potentials resulting from stray al­
ternating currents may also be mitigated, if they V a2+4b2 V a2+b*
are reasonably steady and the harmonic content is It should be noted that this does not apply to
small, by balancing such potentials in the measur­ ground rods driven to depth "b"; it applies only to
ing device with an equal, but oppositely phased small electrodes buried at depth "b", with insul­
potential. In some cases, simultaneous measure­ ated connecting wires. However, in practice, four
ments of stray currents and the resultant potential rods are usually placed in a straight line at inter­
will give a useful indication of ground resistance. vals "a", driven to a depth not exceeding 0.1 a.
Then we assume b=0 and the formula becomes:
7.04 Reactive Component of Impedance of a Large P = 2TSLR
Ground System and gives approximately the average resistivity of
The impedance of a large ground system may the soil to depth "a". Since "a" is usually meas­
have a significant quadrature component.3 There­
ured in feet and the resistivity required in metric
fore, certain precautions should be taken when
units, it is convenient to use one of the forms :
measuring the 60-cycle impedance of a large
grounding system. For such measurements the test P = 1.92ai? meter-ohms, or 192 &R ohm-cm.
device should be operated at approximately sys­ A set of readings taken with various probe spac-
tem frequency of 60 cycles per second, but the ings gives a set of resistivities which, when plotted
test frequency should be slightly above or below against depth on log-log paper, indicate whether
60 cycles per second for the most accurate results. there are distinct layers of different soil or rock,
Most instruments employ a periodically reversed their respective resistivities, and the depth of the
current for these measurements. Staged ground upper layer. In inductive coordination work, spac-
fault tests can also be used for this purpose. ings up to 3,000 feet often have been used. For
these long spacings, the resistance is of the order
8.00 METHODS OF MEASURING of a few hundredths of an ohm, and a sensitive
EARTH RESISTIVITY direct-current potentiometer with a battery supply
as high as 180 volts may be required. For the
8.01 Two-Point Method shorter spacings, the four-terminal instruments
Rough measurements of the resistivity of undis­ shown in Figures 12.01, 12.02 and 12.03 are con­
turbed earth can be made in the field with the venient and adequate. For some instruments cor­
Shepard Soil Resistivity Meter and similar two- rections may be required for the potential probe
point methods. The apparatus consists of one small resistances. In such cases correction factors can
and one smaller iron electrode, both attached to an
usually be obtained from the supplier of the in­
insulating rod. The positive terminal of a battery is
strument.
connected through a milliammeter to the smaller
electrode and the negative terminal to the other 8.03 Soil Samples
electrode. The instrument can be calibrated to read The resistivity of soil samples may be measured
directly in ohm-centimeters at nominal battery volt­ with the four-point method (Paragraph 8.02) by
age. This type of apparatus is easily portable and placing the sample in a small rectangular test cell
with it a number of measurements can be made in of square cross-section having four identical, equal­
a short time on small volumes of soil either in ly spaced electrodes, in a straight line. Reasonable
holes driven in the ground or in the walls or bot­ accuracy cannot be obtained if the size of the box
tom of excavations.
is too small in relation to the electrode spacing "a".
8.02 Four-Point Method The determination of soil resistivity from the value
The most accurate method in practice, of measur­ of resistance measured between opposite faces of a
ing the average resistivity of large volumes of soil sample of known dimensions is less accurate
undisturbed earth, is the four-point method.4 Small than the four-point method, because the interfacial
electrodes are buried in four small holes in the resistances of the soil sample and the test elec­
earth, all at depth "b" and spaced in a straight trodes are included in the measured value.
line at intervals "a". A test current / is passed be­ It is difficult and in some cases impossible to obtain

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 a useful approximation of soil resistivity from pair of ground electrodes in series and substituting
resistivity measurements on samples. This is due the resistance values in the equation, the value of
to the difficulty of obtaining representative, homo­ rr may be established. If the two auxiliary grounds
geneous soil samples, and in duplicating the origin­ are of materially higher resistance than the ground
al soil compaction and moisture content in the under test, the errors in the individual measure­
test cell. ments will be greatly magnified in the final result.
For this measurement the electrodes must be at
9.00 M E T H O D S O F MEASURING some distance from each other; otherwise absurd­
ities may arise in the calculations, such as zero or
GROUND RESISTANCE
even negative resistances. In measuring a single-
9.01 General driven ground the distance between the three sep­
In this section, only general methods are covered. arate ground electrodes should be at least 15 feet,
5 6 7
. , - For the instrumentation available refer to with a preferable spacing of 25 feet or more. For
Section 12.00. While in this section the ohmic larger area grounds, which are presumably of
value is called "resistance", it should be remem­ lower resistances, spacing in the order of the di­
bered that there is a reactive component that should mensions of the ground field is required as a min­
be taken into account when the ohmic value of imum. This method becomes awkward for large
the ground under test is less than one-half ohm. substation grounds, and some form of the fall-of-
This reactive component has little effect in grounds potential method is preferred.
with an impedance higher than one ohm. The resis- g Fall-of-Potential Method
tance of a ground electrode usually is determined
This method has several variations and is applic­
with alternating or periodically reversed current to
able to all types of ground resistance measure­
avoid possible polarization effects when using direct
ments. As mentioned in Section 9.01, the "resis­
current. The frequency of this alternating current
tance" of a large ground has an appreciable com­
should be at or near the power frequency.
ponent of reactance ohms when the resistance is
9.02 Two-Point Method less than 0.5 o h m ; therefore the measured value is
In this method the total resistance of the unknown in "impedance" ohms and should be so considered,
although the generally used terminology is "resis­
and an auxiliary ground is measured. The resis­
tance of the ground".
tance of the auxiliary ground is presumed to be
negligible in comparison with the resistance of the The method involves passing a current into the
unknown ground, and the measured value in ohms ground to be measured and noting the influence of
is called the resistance of the unknown ground. this current in terms of voltage between the ground
The usual application of this method is to deter­ under test and an auxiliary "potential" electrode.
mine the resistance of a single-rod driven ground An auxiliary "current" electrode is used to permit
near a residence that also has a common municipal passing a current into the ground to be tested. This
water supply system that uses metal pipe without current electrode should be effectively outside the
insulating joints. The water pipe ground is as­ "influence" of the ground to be tested. This influ­
ence is sometimes called "extent" of station ground
sumed to be in the order of one ohm and must be
and may be considered as the distance beyond
low in relation to the permissible driven ground
which there is a negligible effect on the measured
maximum, which is usually in the order of 25 ohms.
rise of ground voltage caused by ground current.
Obviously, this method is subject to large errors
Theoretically the influence extends to infinity; but
for low-valued driven grounds but is very useful
practically there is a limit, because the influence
and adequate where a "go, no-go" type of test is
varies inversely as some power of the distance from
all that is required.
the ground to be tested. This influence is deter­
9.03 Three-Point Method mined and allowed for during the test on ground
This method involves the use of two auxiliary grids or deep-driven ground rods of one ohm or
grounds with the resistances of the auxiliary less. In the case of small-area or single-rod driven
grounds designated "r 2 " and "r 3 " and with the grounds the influence is assumed to be negligible at
ground to be measured designated " n " . The resis­ 100 feet, and the potential electrode is placed mid­
tance between each pair of grounds is measured way between the current electrode and the ground
and designated " n 2 " , "ri 3 ", and "r 2 s'\ where under test.
r\i = t\ -j- r 2 etc. Solving the simultaneous equa­ In a test of large-area, low-impedance grounds,
tions, it follows that the potential electrode is moved away from the
( n 2 ) —■ (r 2 3 ) + (r«) ground under test in steps. A value of impedance
n = is obtained at each step. This impedance is plotted
2
as a function of distance, and the value in ohms at
Therefore, by measuring the resistance of each which this plotted curve appears to level out is

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 taken as the impedance value of the ground under If it is found by test that the current electrode is
test within the influence of the station ground, the cur­
rent electrode must be moved farther from the
A representative curve for a low impedance ground
ground under test; otherwise the indicated value of
is shown as Curve A in Figure 9.04-1. Curve B
impedance of the ground under test will be low
illustrates the effect of coupling as described in
in Curve A of Figure 9.04-1.
Section 11.00. The data for this figure were taken
from a test made on a station that had a ground Curves for a higher impedance ground are shown in
grid approximately 400 by 480 feet (192,000 square Figure 9.04-2. There is no preferred placement of
feet). Distances were measured from the station electrodes in this case, as may be seen from the
fence; hence the impedance is not zero at zero indicated values. Since the coupling effects are a
distance on the curve. very small part of the total value in ohms, the
most practical placement of the electrodes should
In Curve A, Figure 9.04-1, the distance at which
be chosen, usually with the test electrodes in the
the plotted impedance appears to level out is about
same physical direction.
600 feet; this 600-foot distance may be termed
the influence of this particular ground; this distance
9.05 Ratio Method
may be used to locate the "remote" ground for
neutralizing transformers used on communication In this method the resistance of the ground under
circuits. test is compared with a known resistance, usually
by using the same electrode configuration, as in the
fall-of-potential method. Since this is a comparison
TO 1.2 OHMS AT
method, the ohm readings are independent of the
IOOO FUT test current magnitude if the test current is high
enough to give adequate sensitivity.
amurr tucnooe noo
— monFtme
9.06 Staged Fault Tests
Staged high-current tests may be required for
. EliCTROU m
DHXOTONAS those cases where specific information is desired on
— mcrnoot
a particular grounding installation. Also, a ground
POTWMTtAL £L£C7B00£ impedance determination can be obtained as auxil­
OFPOSTTEONtKTION
CUMfNTBUCTROOe iary information at the time of actual ground faults
by utilizing an oscillograph or one element of the
automatic station oscillograph.
In either case the instrumentation is the same. The
MPtMMCE HfASUBCP fXCM
cam* of STATION ONOONO object is to record the voltage between selected
OHO IN BOTH TUTS
points on one or more oscillograph elements. The
0 200 400 600 800 IOOO 1200 FEET voltages to be recorded will probably be of such
DISTANCE BETWEEN POTENTIAL ELECTRODE AND STATION FENCE
magnitude that potential transformers will be re­
Figure 9.04-1 quired. The ratio of the potential transformers and,
therefore, the maximum voltage that can be ap­
plied to the transformers may be determined in
advance of the staged tests by using the fall-of-
potential method at practical values of test current.
The only problem involved is in the calibration of
the oscillograph circuit, which is composed of a
potential transformer with a possible high resis­
tance in the primary. This resistance is composed
of the remote potential ground in series with a long
lead. A satisfactory method of calibration of the
deflection of the oscillograph element may be made
by inserting a measured voltage in the primary
circuit in series with the lead and the remote po­
tential ground as used during the test.
The location of the actual points to be measured
is, of course, dependent on the information desired ;
0 200 4C0 COO 600 1000 1200 FEET but in all cases due allowance must be made for
DISTANCE BETWEEN POTENTIAL ELECTRODE AND STATION FENCE coupling between test circuits, as given in Section
Figure 9.04-2 11.00.
12

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 10.00 A U X I L I A R Y ELECTRODES troduced with a potential electrode resistance of
1,000 ohms, and some instruments will permit
In the two-point method it is inferred that exist­ 10,000 ohms in the potential electrode. This means
ing electrodes will be used; that is, a very low that little attention needs to be paid to obtaining
resistance reference electrode must be available; a low resistance; therefore the potential electrode
otherwise another method should be used. may be moved easily from point to point, and in
In the three-point method two additional test elec­ many cases a simple rod can be pressed into the
trodes are required that should have a resistance soil to act as an adequate potential electrode.
value in the same general order as the ground
being tested. A s a general rule, the additional
11.00 C O U P L I N G B E T W E E N T E S T C I R C U I T S
electrode should not have a resistance of more
than five times that of the ground being tested; T h e effect of coupling is of importance only when
otherwise the precision of measuring may give measuring ground impedances of one ohm or less.
absurd results, as in the case of a low test ground When the ground to be tested is a very large area
resistance being used with two additional high- and may have an impedance of less than one-tenth
resistance electrodes. This simply means that, if ohm, the coupling may introduce an error equal
the ground being tested is a cluster of driven to the value of the ground under test. The defi­
rods, each of the two additional electrodes should nitions of various types of coupling are given in
be a cluster of at least three driven rods. The Sections 4.0229 to 4.0233.
most practical arrangement of an auxiliary ground The inductive coupling in ground testing causes
in this case is three rods in a triangle, with about an error when the current in the lead to the test
10 feet to each side. The use of auger- or screw - current electrode induces a voltage into the poten­
type ground rods is advisable where the earth is tial lead. Also, the return test current flowing in
relatively soft, while the use of driven rods is the earth from the current electrode causes a
advisable where the soil is extremely hard or voltage drop in the earth because of the resistive
rocky. coupling and the direct inductive coupling of the
The requirements of the electrodes in the fall-of- return earth path. For the " average" earth resis­
potential method are not so critical. The current tivity this is in the order of 0.03 ohm per 100
electrode should have a resistance low enough to feet, but this value may vary widely. If the effects
allow sufficient test current to flow with the of coupling have an appreciable influence on the
available test supply voltage. The resistance of measured results, these effects may be held to a
the current electrode does not affect the accuracy minimum by placing the potential electrode in a
of the results when this electrode is properly physical direction opposite from that of the cur­
placed outside the influence of the ground to be rent electrode to avoid inductive coupling between
tested. In the ammeter-voltmeter method a mini­ the current and potential test leads. Where oppo­
mum of two amperes of test current is desirable. site directional placement of the current and po­
For safety reasons the maximum test voltage tential test leads is not practical, a satisfactory re­
should not exceed 440 volts. This means the cur­ sult usually can be obtained if a physical angle of
rent electrode resistance should not exceed ap­ 90 degrees between current and potential is real­
proximately 200 ohms. In most locations a cluster ized. Separate current and potential test leads
of three driven rods, four feet deep, with triangu­ should be available in order that no part of the
lar spacing of 10 feet, will have a resistance of current test lead circuit will be included in the
less than 200 ohms, although in a very dry loca­ measurement of a low impedance ground.
tion it may be necessary to pour water around the
rods to lower the current electrode resistance to 12.00 I N S T R U M E N T A T I O N FOR
the lowest practical value, which may be over 200 GROUND RESISTANCE MEASUREMENTS
ohms. The addition of salt to the water poured
around the test electrode is of very little value; 12.01 Ratio Ohmmeter
the moisture is the main requirement. The poten­ A commonly used instrument for measuring
tial electrode in the fall-of-potential method must ground resistance is shown in Figure 12.01.
meet only the requirements of the voltage-indi­ Current from the hand-cranked direct-current
cating element of the instrument being used. (See generator is reversed periodically by the current
Section 12.04). reverser and flows in the earth between ground
With ratio-type instruments (Figure 12.01, for " X " under test and auxiliary electrode "C". The
example) the sensitivity, but not the accuracy, is fall-of-potential between " X " and the other auxil­
affected by the current electrode resistance, which iary electrode " P " is rectified by the potential
may be up to 200 ohms without causing a sensi­ reverser, which is on the same shaft, and there­
ble loss of sensitivity. With good commercial in­ fore, operates in synchronism with the current re­
struments, a negligible measurement error is in­ verser. The coils operate in a field provided by a

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 permanent magnet The current coil tends to turn
the pointer toward zero, while the potential coil
tends to turn the pointer toward a higher ohm
reading. The operating current through these coils
is furnished respectively by the current through
and the voltage drop across the ground under
test; therefore the scale of the instrument can be
marked in ohms. A suitable range switch pro­
vides a divider to the scale values.

By connecting terminals "Pi" and "Ci" (also,

"P2" and "C2") together, the instrument becomes
a two-terminal ohmmeter anl may be used in any 6B0UND UNOER
TEST
MUX. POTENTIAL
WOUND
AUX. CURRENT
OffOUND
of the methods; but the connections to separate
Figure 12.02 Double-Balance Bridge
auxiliary grounds, as shown in Figure 12.01, are
preferred. For grounds over one ohm the "Pi"
and "Ci" terminals may be connected together to 12.02 Double-Balance Bridge
use a common lead to the ground under test. This bridge method for measuring ground resis­
tance is shown in Figure 12.02.

The synchronous reversing switch (combination

In this method current from the alternating-cur­
current and potential reverser) used in this in­
rent source flows in two parallel circuits. The
strument makes it relatively insensitive to stray lower circuit includes fixed resistance "A", ground
voltages in the potential circuit. In most cases a "X" under test, and auxiliary current ground "C".
cranking speed, which eliminates the effect of rela­ The upper circuit includes fixed resistance "B"
tively large stray voltages, can be used. Some dif­ and an adjustable slide rheostat on which two
sliders, "Sa" and "Sb", make contact. With the
ficulty may be experienced in obtaining a reading
detector switch closed to the left, slider "Sa" is
in an extreme case of a ground of less than 0.5 adjusted until the detector shows a balance. The
ohm with stray voltages of more than 10 volts. currents in the two branch circuits are then in­
versely proportional to resistances "A" and "B".
The switch then is closed to the right, and slider
"Sb" is adjusted until the detector again shows a
balance. The potential drop between "X" and "P"
is then equal to the drop in portion "Rb" of the
CL C. GENERATOR slide rheostat, and the resistance of the ground
under test then is given by Rx = RbA/B. The
scale over which "Sb" moves can be calibrated to
read Rx directly.

In testing high resistance grounds the alternating-

current source may be a buzzer operating from
dry cells, and the detector may be a telephone
receiver. The tone of the buzzer usually can be
recognized and balanced out even in the presence
of considerable background noise caused by stray
alternating currents. Resistance at "P" merely
reduces the sensitivity of the detector. Excessive
resistance at "C" may limit the range of resistance
y\n*QBOUND UNDER TEST ^AUXILIARY OR REFERENCE GROUNDS^
that can be measured. The locations of electrodes
" P " and "C" are determined by the same con­
POTENTIAL CURRENT
siderations as in the fall-of-potential method,
Figure 12.01 Ratio Ohmmeter given in Section 9.04.
14

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 12.03 Single-Balance Transformer voltmeter requirements, when there is no stray
voltage, are simply that the impedance of the volt­
An instrument that uses a single balance to give
meter be high in relation to the reference elec­
a bridge type of measurement is shown schematic­
trodes.
ally in Figure 12.03.
When accuracy is required, the impedance of the
In this instrument a battery is used to drive a reference electrode must be considered when
vibrator that has two sets of contacts. The first measuring the voltage caused by current flow into
set of contacts reverses the direction of flow of the measured ground. It is obvious that less error
primary current to a transformer that provides is introduced when using a high-impedance volt­
test current between the auxiliary current elec­ meter, and this error will become negligible when
trode and the ground under test. The second set a vacuum tube type of voltmeter is used.
of contacts gives "sense direction" to the balancing
galvanometer, which then can indicate whether When there is a stray voltage in the ground to be
the dial setting is low or high. measured and this voltage is large compared with
the voltage caused by the test current, this stray
When the slider of the potentiometer is adjusted voltage must be balanced out, both in magnitude
until there is no potential between the slider and and phase, before test current is applied. The volt­
auxiliary electrode "P", as shown by a galvan­ meter in this case should be frequency selective,
ometer null, the portion of rheostat "Ri" bears a because only one frequency can be balanced out.
definite relationship to the resistance of the Usually the only case where such a selective-
ground under test. Therefore the potentiometer frequency voltmeter is required is in the measure­
can be calibrated in ohms with appropriate multi­ ment of large grounds with a value of less than
pliers provided by taps on the ratio transformer 0.2 ohm.
as selected by the range switch. Since a negligible
current flows in the potential electrode circuit at A simplified schematic diagram of the test con­
balance, the resistance of the potential electrode nections for a selective-frequency voltmeter-am­
does not affect accuracy but does have an effect meter circuit is given as Figure 12.04. The test
on the sensitivity of the galvanometer. current is measured by taking the voltage drop
across a 0.1-ohm shunt and is monitored by an
ammeter while the voltmeter is being used to
This instrument is relatively insensitive to stray measure the voltage between the potential elec­
voltages; and only in an extreme case will diffi­ trode and the ground under test. This arrange­
culty be experienced, as noted in the last para­ ment provides a form of ratio measurement and
graph of Section 12.01. thus limits the errors to scale errors of the instru­
ment and ratio errors of' the shunts and multipliers.

12.05 Integrity of Ground Grid Tests

In this test the object is to determine whether the
various parts of the ground grid are connected with
low-resistance copper. This copper is, of course,
shunted by the surrounding earth, which usually
has a very low impedance.

The best method for making integrity-of-ground-

grid tests uses a large but practical direct current
and some means of detecting the voltage drop
caused by this current. Direct-reading ohm-
meters can be used if the sensitivity is adequate.
TEST VfiorttfTlAL "cVtUHMT
The ammeter-voltmeter method, using alternating
Figure 12.03 Single-Balance Transformer current, cannot be used satisfactorily for this
test, because the reactance ohms of a large cop­
per wire is greater than the resistance ohms; also,
the buried wire in this case is shunted by the sur­
rounding earth, which path may have slightly
12.04 Ammeter-Voltmeter
less reactance ohms than the wire. Therefore a
There are no particular requirements for the am­ continuity test for buried wire would give inde­
meter in any of the measurement methods. The terminate results if alternating current is used.
15

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 sible hazardous potential gradients in the vicinity
of grounded electrical structures during fault
conditions. 8» 9 The voltage drop to points sur­
rounding the structure is measured from a known
reference point and plotted on a map of the loca­
tion. A potential contour map may then be drawn
by connecting points of equal potential with con­
tinuous lines. If the contour lines have equal volt­
'•HOUND M I N T Or LAROEST
age differences between them, the closer the lines,
POWCR TRANSFORMERS
the greater the hazard. Actual gradients due to
I ^-STATION FENCE ground fault current are obtained by multiplying
test current gradients by the ratio of the fault cur­
Figure 12.04 Selective-Frequency Voltmeter- rent to test current.
Ammeter Circuit

By extension of this reasoning, it is practically The most accurate measurements of potential

impossible to sensibly lower the impedance be­ gradients are made with the voltmeter-ammeter
tween two ground grids of any distance apart, method. A known current, in the order of 100
each of which has an impedance in the order of amperes, held constant during test, is passed
0.1 ohm at 60 cycles per second. The addition of through the ground grid to a remote ground
copper connectors, however large, will not lower test electrode and returned through an insulated
the reactance ohms between the two ground grids. conductor. A remotely located ground test elec­
The resistance component can be lowered by addi­ trode is necessary to prevent gradient distortion,
tional connectors, and this component is used to caused by the mutual impedance of inadequately
determine the integrity of the ground grid. spaced ground electrodes. This distance may
vary from 1000 feet for a small ground grid to a
mile or more for larger installations. Measure­
One practical "integrity" test consists of passing ments should be made with a very-high-impedance
about five amperes into the ground grid at the two voltmeter from a point on the ground grid to
points to be checked. The voltage drop across these points on the surface of the earth along profile
points is measured with a millivoltmeter or port­ lines radial to the point of connection to the
able potentiometer, and the effective resistance is ground grid. Unless suitable means are employed
calculated from the current and voltage readings. to mask out residual ground currents, the test
From these readings and calculated ohms of cop­ current must be of sufficient magnitude to do so.
per it can be determined whether there is an ade­ At the same time care must be taken to prevent
quate connection. For those ground systems that heating and drying of the soil in contact with the
have a direct voltage between points, the change ground grid or test electrode to avoid variations
of voltage caused by the test current is used to in voltage gradients during a series of measure­
calculate the resistaance. ments. Economics and the necessary detail re­
quired will determine the number of measure­
ments to be made.
For the majority of large-ground systems in ser­
vice there will be a relatively large alternating
voltage between the points to be measured, com­
pared with the direct millivolts to be detected. When more than ore overhead line or under­
The effects of the alternating component on the ground cable are connected to a substation, po­
detector can be mitigated by shunting the moving tential gradients in and around the substation
coil in the millivoltmeter, or the galvanometer in may be quite different for faults on different lines
the potentiometer, with a capacitor of 20 micro­ or cables. Likewise, faults aft different locations
farads or more. This capacitor should preferably in large substations may result in differences in
have a liquid impregnated paper dielectric, but potential gradients in and around the substation.
some modern electrolytic condensers have so little It may, therefore, be advantageous to determine
leakage that they may be used in this application. potential gradients in and around a large sub­
station for two or more fault conditions.

13.00 POTENTIAL GRADIENT

MEASUREMENTS Underground metallic structures, e.g., neutral con­
ductors, metallic cable sheaths, metallic water and
13.01 Potential Contour Surveys gas lines, etc., metallic structures on the surface
A potential contour survey is made to locate pos­ of the ground such as railroad rails and fences

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 and overhead ground wires in the vicinity of a sub- complex ground electrode or to predict the prob­
station, whether connected to the ground grid or able voltage gradient of a complex ground sys­
not, may have a significant effect on potential tem. 10 . »
gradients and should be considered when making
potential gradient measurements. 14.02 Material Required
When a potential gradient survey cannot be justi­ The materials required for model tests are: a tank
fied economically, potential gradients may be cal­ of nonconducting material filled with water, a
culated from ground resistance or soil resistivity small quantity of common salt or copper sulfate
measurements. The accuracy of such calculations to make the water an electrolyte, a scale model
will be dependent upon the accuracy of the meas­ of the electrode or grounding system in question,
urements, and the unknown abnormalities of the an alternating-current source of power with some
earth around and below the ground grid. The ade­ means of varying the voltage so the proper cur­
quacy of such calculations may be verified with a rent flow may be obtained, a voltmeter with a
relatively few potential gradient measurements. minimum sensitivity of 5000 ohms per volt, a
portable ammeter, a "return path" plate, a known
13.02 Step and Touch Voltages electrode shape such as a hemisphere, and a small
wire probe.
The magnitude of step and touch voltages 0 Figure
13.02 may be scaled off of a potential contour map
of the site or actually measured by the voltmeter- 14.03 Resistance Test Procedure
ammeter method. 1. Suspend the hemisphere electrode in the elec­
trolyte with the diameter at water surface.
With a small current flowing (less than 0.5
ampere) make a series of voltage readings
between hemisphere electrode and a small
wire probe, at equal distances apart on surface
E
STEP \^zl
of electrolyte between model and "return
path" plate which is immersed in the electro­
lyte as far away from model as possible. The
POTENTIAL RISE ABOVE readings should rise rapidly from zero, level
REMOTE EARTH DURING
SHORT CIRCUIT off, and then rise rapidly again to full test
voltage as probe approaches "return path"
plate. Midway on the level-off point, the
voltage may be considered as the voltage from
the electrode to "true earth" and the voltage
STEP VOLTAGE AT A GROUNDED STRUCTURE drop to that point is the "applied voltage".
This "applied voltage" and the measured cur­
rent yields the resistance, ( R = B/I).
2. From formula R = P in which R is the
rD
resistance in ohms, P is the resistivity of the
electrolyte in ohms per centimeter-cube, and
D is the diameter of the hemisphere in
centimeters, calculate P after measuring R.
3. Now suspend the model of the electrode
(which may be a simple model of a ground
rod arrangement or a complex irregular elec­
trode such as a tower footing) in the electro­
TOUCH VOLTAGE ATA GROUNDED STRUCTURE
lyte as it would be in the soil, measure R and
solve for an equivalent hemisphere. The resis­
FIGURE 13.02 tance of the actual complex electrode can then
be predicted for any other value of resistivity
such as the resistivity of the soil in which it
14.00 MODEL TESTS will be placed.

14.01 Purpose
The main purpose of a model test is to help pre­ 14.04 Potential Gradient Test Procedure
dict the probable resistance to true earth of a Measure the voltage on the surface of the elec-
17

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 trolyte at regular intervals between a selected 3. The probe should be about Y% inch round rod
point on the model being tested and the level- cut off square and should not be immersed
off point mentioned before. Plot this as a percent­ more than % inch deep.
age of the "applied voltage" against the distance 4. The model should be to scale and large
from the selected point. This shows the voltage enough to simplify its manufacture and assure
gradient. a reasonable accuracy, but should be small
enough to be convenient. A 20/1 scale is often
satisfactory.
14.05 Principlet to be Followed in Use of Model
5. The tank should not be smaller than five
1. Alternating current should be used to prevent times the model's maximum dimension. This
polarization of electrode which would cause will give error of less than 10 percent from
errors at low currents. results obtained from an infinite tank. If the
2. Current densities should be kept low—less tank size were increased to seven times
than 0.1 ampere per square centimeter of model's dimension, error will be less than 5
electrode. percent.

18

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.
 15.00 BIBLIOGRAPHY

1. Bibliography on Ground Resistance and Potential Gradients in the Earth, Ground Resis­
tance and Potential Gradient Measurements Subcommittee of the AIEE Special Instru­
ments and Auxiliary Apparatus Committee, AIEE Transactions Paper 60-1.

2. Master Test Code for Resistance Measurements, AIEE No. 550, May, 1949.

3. The Effect of Reactive Components in the Measurement of Grounding Circuits, L. H.

Harrison, AIEE Transactions, Vol. 72, Part II, 1953, pp. 340-345.

4. A Method of Measuring Resistivity, F. Wenncr, National Bureau of Standards, Scientific

Paper, 12, No. S-258, 1916, p. 469.

5. Some of the Fundamental Aspects of Ground Resistance Measurements, E. B. Curdts,

AIEE Transactions, Vol. 77, Part I, 1958, p. 760.

6. The Technique and Instrumentation of Low-Impedance Ground Measurements, C. A. Duke,

L. E. Smith, AIEE Transactions, Vol. 77, Part I, 1958, p. 767.

7. A Manual on Ground Resistance Testing, James G. Biddle Co., Publication 25-J-1952.

8. Voltage Gradients Through the Ground Under Fault Conditions, AIEE Working Group
56.1, AIEE Transactions, Vol. 77, Part III, 1958, pp. 669-692.

9. A Guide to Safety in Alternating-Current Substation Grounding, No. 80, AIEE,

10. Grounding Electrode Characteristics from Model Tests, H. R. Armstrong, AIEE Trans­
actions, Vol. 72, Part III, 1953, pp. 1301-1305.

11. Grounding Electrode Potential Gradients from Model Tests, H. R. Armstrong, L. J.

Simkin, AIEE Conference Paper 59-1201.

19

Authorized licensed use limited to: Technological Inst of the Philippines. Downloaded on January 21,2021 at 06:27:08 UTC from IEEE Xplore. Restrictions apply.