Documente Academic
Documente Profesional
Documente Cultură
:
St.d 575-1988 SHEATH-B(DNDING METHODS FOR SINGLE-CONDUCTOR CABLES AND THE
10
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.
ANSUIEEE
CALCULATION OF INDUCED VOLTAGES AND CURRENTS IN CABLE SHEATHS Std 575-1988
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 0
1 I * I
I 1 I I 1 1 1 1 1 1 r I 1 1 1 1 1
280 t //
f = 60 Hz
8o V 2
1 I
3 4
I
5
I 1
6
1 1 1
7 8 9 1 0
1
20
1
30
I
RATIO S/d
Fig 1
Induced Sheath Voltage Gradient for a Conductor Current of 1000 A
11
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.
ANSUIEEE
Std 575-1988 SHEATH-BONDING METHODS FOR SINGLE-CONDUCTOR CABLES AND THE
CONDUCTOR
0O0/
TREFOIL
000 b
FLAT FORMATION
0
FLAT FORMATION
(CLOSE) (WIDE)
k I
PARALLEL CONDUCTOR
I 1
v i
L/ 2
I
I I
Fig 2
Transposition of Parallel Conductor to
Reduce Induced Voltage with
Power Cables in Flat Formation or Trefoil
current can flow only by way of the ground itself. of this conductor from the cable circuit should be
Because the resistivity of the ground is very high sufficiently close to limit the voltage rise of the
compared with that of good conductors, the sheath to an acceptable level during a single-
return current is very widely diffused through phase fault. The size of this conductor must be
the ground and the mean effective depth of the adequate to carry the full expected fault current
power frequency components is many hundred for the cable system.
meters. Because the returning current, on aver- The parallel ground continuity conductor is
age, is so remote from the conductor current, the usually insulated so as to avoid any corrosion risk
voltage gradients induced along parallel conduc- and it will be subject to voltage induction from the
tors, including the cable sheaths, are very high. power cables in the same way as any other paral-
Furthermore, in the absence of a parallel ground lel conductor. To avoid circulating currents and
conductor, the occurrence of a ground fault in the losses in this conductor it is preferable, when the
immediate vicinity of a cable could cause a major power cables are not transposed, to transpose the
difference to arise between the ground potentials parallel ground continuity conductor using the
at the two ends of a cable system. Depending to methods described in Appendix D, D3.
some extent on the particular design of the volt-
age limiters (if any) employed, hazards could then 5.4.4 Circuit Arrangements. Figures 3 and 4
ensue to personnel or equipment. show the application of single-point bonding to
Accordingly, it is recommended that a single- single length and multiple length circuits respec-
point bonded cable installation be provided with tively. These diagrams do not show the discon-
a parallel ground continuity conductor that is necting boxes to permit testing of the sheath
grounded at both ends of the route. The spacing insulation.
12
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.
ANSI/IEEE
CA1,~~lJIA'~lON
OF INDlJCED VOLTAGES AND CURRENTS IN CABLE SHEATHS Std 575-1988
Fig 3
Single-PointBonding Diagrams for Circuits
Comprising One Cable Length Only
(a) End-Point Bonding (b) Midpoint Bonding
Fig 4
Single-PointBonding Diagram for Circuit
Comprising Three Cable Lengths
13
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.
ANSI/IEEE
Std 575-1988 SHEAT-BONDING METHODS FOR SINGLE-CONDUCTOR CABLES AND THE
Fig 5
Cross-BondedCables Without Transposition
MINOR
SECTIONS 7
I
f
4- MAJOR SECTION
Fig 6
Cross-Bonded Cables with Transposition
14
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.
ANWIEEE
CALCULATION OF INDUCED VOLTAGES AND CURRENTS IN CABLE SHEATHS Std 575-1988
SECTIONALIZING
JOINTS
I \ !
U STATION
ROUND
Fig 7
Sectionalized Cross-Bonded Cable
with Three Major Sections
ible exactly by three, the circuit can be arranged three. Balanced voltage conditions within a given
to consist of one or more major sections in series. major section consisting of four minor sections
At the junction of two major sections and at the can be achieved by subdividing one minor section
ends of t h e circuit, t h e sheaths a r e bonded into two subsections, as follows:
together and grounded, although the grounds at (1) One short length (or subsection) followed
the junctions of major sections will generally be by two equal lengths (or minor sections) with
only local ground rods (See Fig 7 in which each another short length (or subsection) completing
separate major section is connected as in Fig 6). the major section; the combined length of the two
5.6.4 Modified Sectionalized Cross Bonding. subsections should be equal to the length of one
In this moditied version of the sectionalized cross- minor section as shown on Fig.. 8 and 9.
bonding system, it is not necessary to have the (2) One short length (or subsection) followed
number of minor sections exactly divisible by by one longer length (or minor section) then
15
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.
ANSI/IEEE
Std 575-1988 SHEATH-BONDING METHODS FOR SINGLE-CONDUCTOR CABLES AND THE
L L2
Fig 8
Modified Sectionalized Cross-Bonding
Type 1 Without Transpositions
L1 L L L2
t 2-
I I
I I
I
Fig 9
Modified Sectionalized Cross-Bonding
Type 1 with Transpositions
L1 L L2 L
- I -
- REVERSED
CROSS BONDING
L1 and L2 = LENGTH OF SUBSECTIONS
L = LENGTH OF MINOR SECTIONS
L1 .t L2 = L
Fig 10
Modified Sectionalized Cross-Bonding
Type 2 Without Transpositions
16
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.
ANSI/IEEE
CALCIJIATION OF INDLJCED VOLTAGES AND CURRENTS IN CABLE SHEATHS Std 575-1988
L1 L L2 L
t
REVERSED
I-
TRANSPOSITION
AND CROSS BONDING
Fig 11
Modified Sectionalized Cross-Bonding
Type 2 with Transpositions
J/ GROUND-1 1
Fig 12
Continuous Cross Bonding
another short length (or subsection) followed by 5.6.5 Continuous Cross Bonding. In this sys-
one longer length (or minor section) to complete tem the sheaths are cross bonded at the end of
the major section; the two longer lengths (or each minor section throughout the whole cable
minor sections) should be equal and the com- route. The three sheaths are bonded and grounded
bined length of the two subsections should be at the two ends of the route only, as shown in
equal to the length of one minor section as shown Fig 12. It is again generally desirable that the
on Figs 10 and 11. In this case, the first cross cables are transposed so that each conductor
bonding must be reuersed. occupies each of the three positions for one third
17
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.
ANSI/IEEE
Std 575-1988 SHEATH-BONDING METHODS FOR SINGLE-CONDUCTOR CABLES AND THE
I c
SINGLE-POINT
BONDED LENGTH
4
CROSS-BONDED SYSTEM ---
Fig 13
Termination of Cross-Bonded System with
Single-PointBonded Length
of the total length. The number of matched minor fore generally some imbalance, and it may be
sections should preferably be exactly divisible by necessary to calculate the circulating sheath cur-
three, but this becomes less important as the total rents that are present so as to assess their effect
number of minor sections increases (see 5.6.7). on the cable rating. See [ 131 and [ 151 for methods
5.6.6. Mixed Systems. When the number of of calculation.
minor sections is not exactly divisible by three, the
system may consist of a mixture of Kirke-Searing 5.7 Sheath Sectionalizing Joints. When t h e
(regular and modified) and single-point bonded sheath losses of single-conductor cables must be
lengths. When necessary, on account of a large reduced or eliminated, sheath sectionalizingjoints
number of minor sections having unequal lengths, are required for interrupting the electrical conti-
the cross bonding may be of the continuous type. nuity of the sheath circuit. To perform their func-
Figure 13shows the arrangement of a final single- tion satisfactorily there are several major factors
point bonded length at the end of a cross-bonded involved in the design of these joints. Mechanically
system. they must be rugged, impervious to moisture, and
5.6.7 Imbalanced Systems. It is not generally fluid tight under all operating conditions. Electri-
possible to divide the route length into exactly cally, they must be designed to withstand the vol-
matched minor section lengths, nor is it always tage stresses occurring under fault, and lightning
possible to maintain a constant spacing of the and switching surge conditions.
cables throughout the route. In continuous cross- One of the quantities that must be evaluated
bonded systems, it may also be desirable to have a before a sheath sectionalizing joint can be used in
total number of minor sections not exactly divisi- a bonding scheme is the insulation strength re-
ble by three. In practical systems, there is there- quired at the joint. This can be determined by
18
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.
ANSI/IEEE
CALCULATION OF INDUCED VOLTAGES AND CURRENTS IN CABLE SHEATHS Std 575-1988
calculating the maximum voltage appearing across ground faults than a parallel ground continuity
the joint due to faults and lightning and switching conductor. Hence, the voltages induced in parallel
surges. This subject is discussed in Appendix E. cables are less during ground faults in a cross-
bonded system than for a similar single-point
5.8 Choice of Bonding System. Impedance-bond- bonded system.
ing methods are less satisfactory than the other 5.8.3 Choice of Cross-Bonded System. For
methods described. For this reason these methods long cable circuits, there is a choice between sec-
are not recommended for general use. tionalized cross bonding (see 5.6.3 and 5.6.4) and
Bonding transformers may be economical in continuous cross bonding (see 5.6.4).The relative
some isolated cases such as advantages are as follows:
(1) Suitable balancing for cross bonding is im- 5.8.4 Advantages of Sectionalized Cross
possible and single-point bonding is unacceptable Bonding
(that is, no empty duct is available for a ground (1) Since each major section forms a separate
continuity conductor). electrical mesh, it is relatively straightforward to
(2) A spare cable ( a fourth cable for a single calculate the sheath currents when the lengths or
circuit or a seventh cable for a double circuit) is spacings of the minor sections are not uniform.
installed; in this case, reconnecting the cross In a nonuniform section having an equilateral
bonding whenever the spare cable is needed is a cable configuration the ratio of sheath loss with
lengthy and complex operation, whereas recon- cross bonding to that with solid bonding is given
necting of bonding transformers is simple and by
straightforward.
X
Users of this method may refer to 5.5 and [9], - = [1-3(AlA, + A,A, + A,A,)]
P O I , P O I , [211, and Wl. Y
Further discussion will therefore be limited to where
consideration of the other bonding methods.
5.8.1 Use of Single-Point Bonding. A min-
x = cross-bonded loss
= solidly bonded loss
imum of three minor sections are needed to form
a cross-bonded system, and it is normal practice
A A,, A, = per unit lengths of the three minor
sections, that is A , + A, + A, = 1
to use sheath sectionalizing insulators only at
joint positions. Hence, cross bonding is not nor- EXAMPLE: When
mally applicable to cable circuits comprising only
A, = 0.4
one or two lengths, and, for such circuits, single-
point bonding is widely used.
A, = 0.2
5.8.2 Advantages of Cross Bonding. Although
A, = 0.4
the cable sheaths of a single-point bonded system the loss with cross-bonded sheaths is 4% of the
are generally of a cross-sectional area and con- loss with solidly bonded sheaths
ductivity that makes them quite capable of carry- (2) The sheath bond at the junction of each
ing short-circuit currents due to through faults in major section allows fault current due to a cable
the power system, they are unable to do so failure to be distributed between the three sheaths
because they are grounded at one point only. A except within the major section containing the
parallel ground continuity conductor is therefore fault.
recommended (see 5.4.3), and this adds appreci- (3) The sheath bonds and grounds at the junc-
ably to the cost of the cable system. tions of major sections tend to reduce transient
The principal advantage of cross bonding is sheath voltages.
that, while induced sheath currents are inhibited (4) The number of sheath voltage limiters
during normal balanced load operation, the sheaths required is reduced.
do form a continuous path from end to end of the (5) The sheath bonds at the junction of major
cable circuit and are grounded at both ends. sections ensure that there will be no charging
Sheath currents can therefore flow during ground current flow beyond the neutral points of the
faults, and the necessity for the parallel ground bonds irrespective of any inequality in the lengths
continuity conductor is removed. In addition to of the minor sections.
the economy achieved by the elimination of the 5.8.5 Advantages of Continuous Cross
ground conductor, the cable sheaths function Bonding
more effectively as screening conductors during (1) The effects of nonuniform minor sections
19
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.
ANSI/IEEE
Std 575-1988 SHEATH-BONDING METHODS FOR SINGLE-CONDUCTORCABLES AND THE
may be reduced when they form part of a total Because of the infinite variety of geometrical
sheath circuit containing a number of sections. It arrangements coupled with differences in individ-
may also be possible to use a total number of sec- ual cable loading and phase rotation, a universal
tions not exactly divisible by three. solution to sheath standing voltages on multiple
(2) It is possible to monitor sheath currents circuits cannot be given here. Some of the more
throughout the whole circuit, irrespective of the common double-circuit geometries are described
number of minor sections, a t one point along the in [15] and [19].
length. A general solution requires the use of a digital
(3) At least for low-resistance faults, the moni- computer and linear algebra. However, when dis-
toring of the sheath insulation and sheath voltage cretion is used in the selection of phase rotation
limiters becomes easier because there are only and position, the effect of adjacent circuits does
two sheath bonds and ground links to be removed, not significantly increase standing voltages pro-
even on a long circuit, to enable tests to be applied vided these circuits have equal or lower balanced
from the ends of the cable circuit. phase currents.
A solution to a simple parallel double circuit is
5.9 Sheath Standing Voltage given in Appendix D, D2.
5.9.1 Single-PointBonding. Figure 1 shows the
sheath voltages per kilometer due to balanced
loads in the cable conductors. 6. Sheath Voltage Limiters
5.9.2 Sectionalized Cross Bonding. In any mi-
nor section, the sheath standing voltage per kilo- 6.1 Introduction. Sheath sectionalking insula-
meter will be as stated in Fig 1 and the longest tors in cross-bonded cable systems and the insu-
minor section shall be taken for calculating the lators in a single-point bonded-cable system may
maximum standing voltage. With the modified flashover due to overvoltages generated by light-
bonding method described in 5.6.4( 1) the maxi- ning, switching surges, or faults on the power
mum standing voltage thus calculated is reduced system. It is necessary to provide some form of
as much as 13%(see Appendix D, D5). This max- protection for these insulators under system tran-
imum reduction applies when the two short lengths sient conditions. At present, sheath voltage
(or subsections) are equal (that is, L 1 = L2 = 0.5 L). limiters are used for this purpose. The three main
See Figs 8 and 9. types are
When the major section is nonuniform, the (1) Nonlinear resistances
sheath standing voltage can be taken as that cal- (2) Nonlinear resistances in series with spark
culated for the longer of the two grounded minor gaps
section lengths. When the nonuniformity causes ( 3 ) Spark gaps
appreciable sheath current, there will be some
reduction of the shkath standing voltage. 6.2 Nonlinear Resistances. Nonlinear resistances
5.9.3 Continuous Cross Bonding. When the can provide good protection for transient volt-
whole system between sheath bonds consists of a ages. They do, however, have a limited capacity to
number of uniform minor sections exactly divisi- absorb energy and are not designed to carry a
ble by three and the cables are transposed so that 60 Hz fault current. They must be sized to with-
each conductor occupies each of the three posi- stand 60 Hz fault-current overvoltage due to sys-
tions for one third of the total length, then no tem faults external to the cable circuit, although
sheath current flows, and the maximum sheath they are not normally expected to survive over-
standing voltages per meter are as stated in Fig 1. voltages resulting from faults internal to the cable
In a practical system having variable lengths of circuit. The surge energy and 60 Hz voltages, to
minor sections, the sheath standing voltage can which t h e resistor is subjected, dictates the
be taken as that calculated for the longest minor characteristics of the resistor. Distribution class
section length. When appreciable sheath current arresters are often adequate for the surge energy
flows, the sheath standing voltages is reduced requirements when selected to withstand the
somewhat. power-frequency fault voltage without discharg-
5.9.4 Double-Circuit Systems. Where two ing. High humidity tends to reduce the effective-
closely spaced circuits are present, the sheath ness of nonlinear resistances, and they must,
standing voltages are modified by the presence of therefore, be protected from moisture by a suit-
the second circuit. able case or encapsulation.
20
Authorized licensed use limited to: MI2S - Universite Joseph Fourier. Downloaded on November 25,2010 at 09:37:32 UTC from IEEE Xplore. Restrictions apply.