TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

CHAPTER 5
UNDERGROUND AND SUBMARINE CABLES

Section 1 - ASSOCIATED GUIDANCE

5-1. Relevant cable guidance. the United States, permission must be obtained
Maintenance work involving underground or sub- from the nearest District Engineer of the U. S.
marine cable changes requires an understanding of Army Corps of Engineers, who will specify depth
the basic design premises of such cables. requirements and any other pertinent conditions.
a. Types of installations. Underground cables When crossings are made in waterways under the
may be installed in conduit, in duct banks, or by jurisdiction of other authorities, those authorities
direct burial in the earth; submarine cables are should also be consulted.
usually submerged directly in the water and lie on b. Identification requirements. Because under-
the bed of the waterway. The terminal ends of both ground and submarine cables cannot be visually
underground and submarine cables are often above- traced between structure access points, it is impor-
ground. The burial depth of raceways or cables tant that they be marked at all points at which they
should never be less than the depths permitted by are accessible. Any such cables will be identified by
the NEC or the NESC and, in most cases, will be plastic or corrosion-resistant tags wherever they
more to conform to facility design practice. can be worked on and wherever they can possibly be
(1) Cable in conduit removal freplacement. Al- mistaken for another cable. Identification tags will
though it is easy enough to install several cables in be located at terminations and at least in every
one conduit and mechanically easy to withdraw structure. If tags become missing or illegible, they
them, the removal usually ruins the cable. Cables will be replaced as part of the maintenance pro-
become impacted in a conduit, and, when one is gram.
drawn out, the sheath may be stripped either from c. Cable impacts. The major cause of electric fail-
the withdrawn cable or from one of the other cables. ure is the breakdown of insulation. Even under nor-
Therefore, when one cable of a set in a conduit fails, mal conditions, an electrical cable experiences
all cables must be replaced. stress that will gradually weaken it, leading to fail-
(2) Direct-burial cable reinstallation. Direct- ure. Cable tests provide data which permits the
burial cables being replaced must be installed below anticipation of cable failures. An understanding of
the frost line. items which can accelerate insulation deterioration
b. Joint electric supply and communication cir- is of help in determining inspection and testing in-
cuits. Unlike aerial lines, joint structure use is not tervals.
allowed for electric supply and communication cir- (1) Cable Loading. The current-capacities or al-
cuits. Communication cables are installed to be lowable loading of underground cables is based on
completely isolated from electric power cables and the conductor size, material, and assumed ambient
require separate ducts and structures. Economy temperatures. Complex calculations are required to
may dictate contiguous structures and duct lines take all these effects into account. IEEE S-135-1
having a common trench excavation. Direct-burial and IEEE S-135-2 are used as the basis for ampaci-
power and communication lines should be separated ties given by the NEC. The factors used by the NEC
at least the minimum required distance, usually set represent a theoretical average value and may be
by the local communication agency. Control, alarm considered to be safe factors, especially if the load-
signalling, and other low-current and low-voltage ing is based on a loo-percent load factor. The load
circuits may be installed in electric manholes, de- factor for primary circuits on most facilities will
pendent upon facility requirements, but require probably range from 45 to 65 percent at the time of
special shielding or increased insulation levels. initial design. Voltage drop, especially at lower volt-
ages, may also have been a factor in determining
5-2. General construction guidance. the cable sizing. The actual temperature conditions
Rights-of-way for navigable waters and identifica- affecting the cable become an important consider-
tion must meet the following requirements. The in- ation. Added loads and variable loads affect cable
fluence of conditions which can generate cable fail- temperatures both directly and indirectly.
ures in the following discussion should be checked (a) Directly. In general, the higher the tem-
for their impacts. perature, the faster the rate of deterioration in the
a. Rights-of-way requirements. When the system physical properties of the insulation, including the
is being extended across navigable waters within formation of voids in solid-type or paper-insulated

5-l
TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

cable. The deterioration usually results in increased important at these points. Where recabling is re-
dielectric losses and decreased dielectric strength. quired do not use “T” splices in manholes, except
Large variations in daily temperatures accelerate where the facility’s engineering staff concur that _
the possibilities of cable sheaths cracking and avoiding their use is uneconomical.
bolted or clamped connections loosening. A, short- e. Lightning protection and grounding. Lightning
time large overload, and accompanying high tem- protection for aerial to underground primary cable
perature, can produce aging of insulation equiva- connections, and grounding and bonding of under-
lent to operation for a longer time at a smaller ground cables, contribute to the protection of the
overload. Since power surges contribute to cable cables and to the safety of the system.
aging, a cable serving large motors with full-voltage (1) Surge arresters. When a transition is made
starters having intermittent loads, or a cable sub- between overhead conductors and underground or
jected to a higher level of lightning or switching submarine primary cables, facility practice requires
surges, will probably have a shorter life than an that a surge arrester be installed at the termination
identical cable with a constant load and infrequent connecting insulated underground cables to aerial
low-level switching surges. bare conductors. A ground rod should be installed
(b) Indirectly. The temperature of the soil ad- and the metallic sheath or armor of the cable
jacent to a buried cable or conduit system must also bonded to that ground installation. The surge ar-
be considered as affecting cable life. If cable tem- rester then protects the primary cable from switch-
peratures become high enough, the moisture in the ing or lightning surge overvoltages which could
soil will migrate away from the cable causing a overstress the cable insulation. Secondary cables
considerable increase in the soil thermal resistivity. are usually protected from these over-voltages by
This may lead to thermal instability of the soil and primary surge arresters located at pole or ground-
further increase its thermal resistivity which, in mounted transformer installations.
turn, may cause excessive cable temperatures and, (2) Grounding and bonding. All noncurrent-
perhaps, even cable failure. carrying conductive materials in the structure and
(2) Cable insulation failure. Underground pri- any neutrals must be grounded. Most standard
mary distribution cables with solid-dielectric insu- structures are provided with a driven ground rod.
lation have experienced a high rate of electrical Bonding includes the metallic sheath or armor of all
---
failure after several years of operation as the result cables, cable shields, manhole hardware, the tanks
of carbonized paths (electrochemical tree design of all equipment and apparatus, and the secondary
markings) usually caused by the presence of water neutral of transformer installations. Where
in the conductor. nonmetallic-sheathed cable having a ground wire is
d. Termination and splicing impacts. Termina- used, the ground wire is usually brought out at the
tions and splices are usually the weakest point in a joint. These ground wires should be grounded to the
cable system, and the cable system is usually the neutral and the driven ground The resistance of
weakest link in an electrical system. Therefore, in- ground connections must meet the requirements
spection, including riser pole inspections, is doubly given in chapter 10, section III.
Section II - SAFETY PRECAUTIONS
5-3. Cable safety. 5-4. Structure safety.
The compact spacing of conductors and nearness to Subsurface structures such as manholes, hand-
any grounded sheaths is the reason that working on holes, equipment vaults, and splicing boxes are sub-
energized conductors even at low voltages is prohib- ject to accumulation of dangerous gases that may be
ited. A voltage detection tester should be used to combustible and/or explosive, toxic, or deficient in
ensure that the cable is not energized. Materials oxygen. Before entering any manhole or vault, it
such as a lead sheath, which will act as a shield, must be checked for these conditions.
must not be between the tester and the conductors a. Combustible gases. Combustible gases may be
of the circuit being tested. To prevent a de-energized detected by means of a test instrument or safety
circuit from being energized while it is being lamp. When using this equipment, the precautions
worked on, good safety practice requires that the and instructions provided by the manufacturer
disconnecting means at each end be tagged and should be followed. If it is determined that combus-
locked in the open position, and ground clamps ap- tible gases are present, it will be necessary to ven-
plied. tilate the manhole or vault before any work is done.

5-2
 TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

If these tests indicate the presence of an explosive when carbon monoxide is present. By comparing the
mixture in the structure, an injection of carbon di- change in color with the color chart furnished, the
oxide (CO,) into the structure may be made, before concentration can be determined. Tests of this type
ventilating, to reduce the possibility of an explosion. can be made in less than one minute.
Ventilation is best provided by a power-driven por- c. Asbestos-cement fireproofing. Follow instruc-
table ventilating blower. Before the structure is tions for handling in chapter 15, section II.
purged with a blower, or CO, is injected into it, d. Ventilation. Even when tests indicate there are
personnel in connecting structures should be no combustible or toxic gases, it is good practice to
warned, as the gas may be blown through the ducts force-ventilate a manhole or vault whenever person-
into connecting structures. If CO, is used, the struc- nel are in it. This is especially important if cable
ture must be purged with fresh air before it is en- splicing is being performed.
tered by personnel. e. Protection of open structures. Open structures
b. Toxic gases. A calorimetric indicating gel tube should never be left unguarded. A barricade should
manufactured under specifications of the National be placed around the structure opening prior to re-
Bureau of Standards, commonly referred to as the moving the structure cover.
NBS carbon monoxide detector, is available to test f. Ladders. Portable ladders used for access to
for toxics. It is used by breaking the seals of the manholes or vaults should be checked before use to
tubes and aspirating gas to it from the atmosphere ensure that they are firmly placed and will not
to be tested. Chemicals within the tube change color wobble or tilt.
Section III - INSPECTION
5-5. Frequency of underground system in- intended. Look for signs of traction on cable termi-
spections. nations or direct-burial cable which may be a result
The frequency of inspection is largely determined by of expansion and contraction of the cable.
the importance of the equipment or facility it serves a. Cable supports. Check mountings and sup-
or contains. Inspections can vary in frequency from ports to ensure they are secure. Remove rust and
6 months to 5 years, but a 2-year cycle of inspection corrosion and clean and repaint supports with
is recommended. Records should be kept of each corrosion-resistant paint.
inspection. b. Duct entrances. End bells are usually used to
prevent cable damage at duct entrances. If they
5-6. Structure inspections. were not installed, or are damaged, strips of hard
Inspect structures and check their cleanliness and rubber or similar material should be used to protect
their physical condition, such as cracking of walls, the cable at the duct entrance.
roofs, or floor slabs, spalling of concrete, and the c. Testing. Cable insulation integrity cannot be
condition of frames and covers. Inspect for corrosion visually checked; it requires some type of insulation
of pulling eyes; driven grounds; and other miscella- testing to determine whether the cable is reaching
neous fixtures such as cable racks, arms, and insu- an insulation breakdown that will lead to a cable
lators. fault. Testing is described in section VII.
d. Cable faults. Inspection alone may reveal the
5-7. Cable inspections. location of a cable fault or it may be a more compli-
Walk the route of underground direct-burial cable cated process requiring test equipment. Visual and
circuits to inspect for changed conditions. Changes test procedures are covered in section V.
in grade caused by washouts can expose cables to
5-8. Underground equipment inspections.
damaging conditions. Adjacent new construction
should be closely monitored. Examine connections Special maintenance for such distribution equip-
to equipment terminals or cable terminations, ment in underground locations includes the follow-
whether in the structure or above-ground. Check in ing:
structures for the condition of duct entrances, fire- a. Keep items clean and protected from corrosion.
proofing, splices, cable tags, and ground connections b. Check equipment covers to be sure that their
to cable shielding and sheathes. Anchors for subma- gasketing is water-tight.
rine cables should be inspected occasionally to be c. Keep nuts and bolts free from rust by applica-
sure they are in good condition and functioning as tions of paint or heavy grease.
TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

Section IV - MAINTENANCE AND REPAIR OF DUCT SYSTEMS

5-9. Structure maintenance and repair. will bond against sound concrete. The hole can then
Maintenance and repair of structures is a continu- be patched with calcium-chloride putty or quick-
ous procedure but is seldom extensive. Pump out setting cement mortar.
structures as necessary to allow complete inspec- (3) Floor lea k age. If water is entering through
tion. Major breaks or settlement of structures caus- the floor of the structure, clean the floor and remove
ing large cracks require investigation of the struc- accumulated silt. If the floor is generally sound ex-
tural condition and rebuilding to eliminate the cept for the joints at the wall or for isolated cracks,
cause. repairs can usually be made as covered above. If the
a. Duct Line entrances. Grout up chipped concrete concrete of the floor shows evidence of general po-
at the mouth of the duct line as necessary. Heavily rosity or disintegration, it is better to pour a new
loaded cables will crawl because of expansion and floor, as follows:
contraction, which results from the alternate heat- (a) Where a reduction in headroom of 4 to 6
ing and cooling effects of changing loads. The mouth inches (100 to 150 millimeters) will not affect the
of the duct line must be kept clean and free of burrs utility of the structure, a new floor may be poured
and small patches of concrete that will damage directly over the old floor. Otherwise, the old floor
cables. should be broken out and a new floor poured.
b. Water leaks. Most structure repair require- (b) When necessary to break out the old floor,
ments consist of stopping water leaks in the floor the first step is to excavate for a temporary pump
and walls of frequently entered structures. Depend- sump about 12 inches (300 millimeters) below the
ing on the terrain, the pumping of one structure old floor level if the structure has no existing sump.
may involve the removal of water from adjacent When the floor has been removed, continued pump-
structures. In applicable locations, all vacant ducts ing may be necessary. Further excavation will be
should be plugged with standard duct plugs, and all necessary if added headroom is desirable. A perma-
occupied ducts should be sealed so as to prevent nent sump or a storm drain connection should be
water or gas from entering vaults or any users’ considered when the new floor is poured.
premises.
(1) Occupied duct sealing. Use a nonhard- 5-10. Duct line maintenance and repair. --
ening sealing material that will not harm the ca- Most damage to duct systems results from new un-
ble to seal occupied ducts. These nonhardening related construction and settling of ducts. Too often,
compounds consist of emulsified vegetable oils con- the new construction fails to locate an adjacent duct
taining fibers or asphalt compounds. Oakum is line accurately and damages the line. Ducts some-
often packed around the cables as backing for times settle where they cross older understructures,
the sealing compound. Use a ready-mixed com- whose overlay was completed without adequate
mercial sealer and follow the manufacturer’s direc- backfilling and tamping. Duct settling is often not
tions. apparent unless cable failure results or an empty
(2) Wall leakage. Water leakage through the duct is rodded in preparation for pulling in new
walls of the structure will usually occur along joints cable. In either event, the condition must be inves-
or void areas. Some leakage may be found where the tigated and repaired. A new structure at the point of
ducts enter the structure. Using a cold chisel and settlement may possibly be the quickest and cheap-
hammer, chip out the porous area so that the patch est repair.
Section V - CABLE FAULTSAND FAULT LOCATIONS
5-11. Cable faults. intended to prevent extended outages due to tran-
Whenever cable insulation breaks down, resulting sient disturbances on aerial lines. But repeated
in an underground cable fault, fuses should blow or reclosing on an underground cable fault tends to
circuit breakers should open to prevent further sys- create unusually high fault resistances. Reclosing
tem damage. Faulted circuit indicators (FCIs), serves to aggravate an underground cable fault
where provided, may also provide an indication of a which may then stress upstream circuitry.
cable fault. b. Aerial-to-underground line connections. Fuse
a. Reclosing on a fault. The practice of applying protection is required to be provided at or near riser
automatic reclosers on medium-voltage aerial dis- poles where such connections are made. When any
tribution lines presents a problem when under- aerial lines feeding underground cable systems are -
ground distribution lines are supplied from aerial provided with automatic reclosing, that feature
lines which have reclosing features. The recloser is should be designed so that any permanent fault on
5-4
 TM 5-684/NAVFAC MO-200/AJMAN 32-1082

long underground feeder will blow the associated 5-12. Visual methods of cable fault locating.
cable riser fuses within a time period that limits to Since visual inspection can be the easiest and quick-
one reclosure, the damaging effects of automatic
est way to locate a cable fault, it should be tried
reclosing on the faulted cable feeder.
first. Visual inspection may require checking sec-
c. Check of associated equipment and lines.
ondary effects, such as leaks from the cable-
Blown fuses and open circuit breakers may be
insulating medium. If visual inspections are not ef-
caused by a cable fault or by faults on equipment or
fective, then testing devices will need to be used.
other lines connected to cables. Preliminary tests
should be made to determine that the fault is actu- a. Faults in exposed cable and splices. A quick
ally in the cable and not in associated equipment. check may be made by driving over the route of the
d. Faulted circuit indicators. As noted, reclosing cable and looking for such things as a displaced
on faulted cable circuits stresses the circuit ele- structure cover, smoke coming from a structure, or
ments and can increase potential personal hazards. indication of damage caused by digging operations.
On critical feeders, FCIs are often installed to re- A more detailed inspection may be made by exam-
duce service restoration time by providing a conve- ining the terminal equipment and the cable and
nient means of determining fault current occur- splices in the structure. Look particularly for the
rences, location, and direction on underground presence of compound on the cable sheath, smoke,
circuits. and odors of a burnout. Observe the requirements of
(1) Operation. An FCI can be a single or section II.
multiphase device which senses fault current has b. Faults in submarine cable. Oil slicks may occur
passed through the line conductors at the point on the surface of the water near the location of the
where the FCI is installed. The FCI is designed to fault, or bubbles may appear where the cable is
provide a fault current indication by a flag, a light
faulted. Applying high current from a low-voltage
emitting diode (LED) display, or other means. The
source to the faulted cable may cause bubbles to rise
current sensing is done by detecting the magnetic
to the surface near the fault, thus determining the
field strength generated by the circuit’s alternating
approximate location. For short submarine cables,
current.
establishing a line of sight between the terminal
(2) Location. Most FCI s are installed on under-
ground distribution current-carrying elements such ends and patrolling this area may aid in locating a
as cables, switch and transformer elbow termina- point of failure. Maps used in laying the cable will
tions, and separable connectors. They are also used be helpful in establishing the cable route.
on aerial lines. c. Faults in gas-pressurized cable. When a fault
(3) Application. The proper application of FCIs occurs on gas-pressurized sulfur-hexafluoride (SF,)
is crucial to their correct operation. Units must be cable, do not re-energize the cable until the follow-
correctly designed for indicator trip and reset meth- ing steps have been taken.
ods. Inrush restraint, time delay, and coordination (1) Low gas pressure. Gas pressurized cable is
may be necessary. Other considerations when se- usually pressurized at 20 to 80 pounds per square
lecting FCIs means that their initial provision re- inch gage (135 to 550 kilopascals gage). A record of
quires engineered design. the installed gas pressure should be kept for all
(4) Maintenance use. Use FCI sets to locate dis- gas-pressurized cable sections.
tribution faults. Normally where installed, the (a) Checking pressure. Check the gas pres-
number of FCI sets will be one less than the number sure at terminations and splices with a tire gage. If
of cable sets which can be sectionalized. The faulted the pressure has dropped to zero or is dropping, the
cable section will be between a “fault” and a “nor- fault damaged the conduit or jacket containing the
mal” indication. gas. Since gas is electro-negative, a gas detector
(5) Concerns. If units are damaged they must similar to that used for refrigeration gases can be
be replaced with like units having the same fea- used to locate the leak at terminations, splices, and
tures. Their operation in regard to trip and reset other points.
must be understood. If loads are changed and the (b) Detecting leaks. To detect a gas leak along
unit does not have an adaptive trip (tripping on a a buried duct or conduit line, a pipe can be driven
sudden increase above the nominal current followed into the ground above the line and the probe in the
by a loss of current) then the trip setting must be pipe may detect the gas. Another method to detect a
changed. The reset may require manual means or leak in the line is to inject gas at one end and
may be reset by other actions such as predeter- measure the pressure drop at access points. Dry
mined time, current, voltage, or other sensing meth- nitrogen can be used for this method. Then plot the
ods. pressure reading to locate the spot in accordance
5-5
TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

with the manufacturer’s directions. After a leak is (4) Open circuit. The continuity of the conduc-
located, a repair can be made in accordance with tors is determined by grounding the conductors at
section VI. the far end and then testing between each conduc-
(2) Satisfactory gas pressure. If the pressure is tor and ground. If the conductors are continuous,
satisfactory, do not re-energize the circuit until the the resistance reads low; and, if an open circuit
fault has been located and repaired, since the high exists, the tester will indicate a very high resis-
fault current reimposed on the failed cable can fur- tance.
ther damage the cable.
5-14. Cable fault locating test methods.
5-13. Determining type of cable fault. The methods generally used may be separated into
Use fault locating equipment when a check of asso- two major classifications: terminal measurement
ciated equipment and lines confirms that the fault methods and tracer methods. Except in the case of
is actually in the cable, and visual methods fail to faults on series lighting circuits (which usually re-
locate the fault. Since no single test will locate all sult in considerable carbonization because of the
types of faults, the type of fault must be determined constant-current system involved) the resistances of
in order to use the best test method to locate it. To faults are often quite high, ranging from several
determine the type of fault, any source of direct- hundred ohms to megohms when measured at a
current voltage can be used with a voltmeter or a low-voltage level.
suitable low-voltage lamp. A portable testing set, a. Terminal measurement methods. Terminal
such as a multimeter of the volt-ohm meter (VOM) measurement methods involve determining the cho-
type is most commonly used. The section of cable sen electrical value of the faulted conductor from
under test must be disconnected from feeders, one of the cable terminations to the fault, and com-
buses, and equipment. Alternating current should paring this value with the same electrical value on
not be used, because the charging current of the unfaulted cable. The proportions of the electrical
cable is sufficient to prevent accurate indications of values in regard to the length of the unfaulted cable
the condition of the cable. provides the fault distance. The effectiveness of all
a. Types of faults. Cable insulation failures result terminal measurement methods is dependent upon
in low- or high-resistance faults, because one or a the accuracy of installation records. While most of _
combination of the following conditions occur. the work is done at one terminal, access to the other
(1) One or more of the conductors may be terminal may be necessary to connect or disconnect
grounded. conductors as required. Terminal methods include
(2) Two or more conductors may be short the Murray loop, the capacitance bridge measure-
circuited. ment method, the quarter-wave or half-wave reso-
(3) One or more conductors may be open nance methods, and the pulse (time domain
circuited. reflectometer) method.
b. Checking for fault types. b. Tracer methods. These methods require test
(1) Grounded conductor. In checking for a equipment at the cable terminal but rely on checks
grounded conductor, the VOM is successively con- along the cable tracer to locate the fault. Tracer
nected between each conductor and ground with the methods include the modulated direct-current
far end of the cable open circuited. A good conductor method, the modulated alternating-current method,
will indicate a resistance commensurate with that the impulse (thumper) method, the audio frequency
of its insulation. A grounded conductor will show a (tone tracing) method, and the earth gradient
very low resistance. method.
(2) High-resistance grounded. Some installa- (1) Tracer method warning. Some of the tracer
tions are grounded through a high resistance. These methods of fault locating can ignite residual gas in
systems operate like an ungrounded system and the the vicinity of a fault and cause explosions. The
first ground fault does not trip out the system, but likelihood of such an occurrence, while extremely
only sounds a warning. The ground fault can be remote, cannot be ignored.
traced using an integral system pulser and a detec- (2) Structure testing. Normal gas tests with
tor furnished as a part of the system. combustible gas detectors should be made prior to
(3) Short ci rcuit. In checking for a short circuit, entering structures during all fault-locating opera-
the VOM is successively connected between each tions, regardless of the urgency of the situation or
possible combination of conductors. Far ends of the the type of fault-locating equipment being used. It is
cable must be open-circuited. A low reading indi- also advisable to use a carbon monoxide (CO) tester
cates a short circuit between the conductors under to check the atmosphere in structures where fault
test. repairs are to be made, particularly in cases where
5-6
 TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

substantial quantities of cable insulation have been the closed loop to be used for fault-locating mea-
destroyed by the fault. Gas concentrations in struc- surements and comparing this measurement to
tures can be dispersed by a thorough purging with a known circuit constants. Conductor continuity gen-
positive-pressure blower. Gas testers and their ap- erally will have no effect on the operation of tracer-
plication are discussed in section II. type fault-locating equipment. Faults exhibiting
both high series resistance (open conductor) and
5-15. Simplifying cable fault locating. high parallel resistance (ungrounded conductor) can
Locating faults in cables can be a complicated pro- be located by using a capacitance-type terminal
cess. The following paragraphs provide some helpful measurement device.
hints which can simplify the process and increase
fault-locating accuracy. 5-16. Cable fault locating equipment.
a. Fault reduction. In cases where the parallel Cable fault locating equipment is available from
resistance of a fault is too high to allow effective test equipment rental companies. Member compa-
application of either tracer or terminal measure- nies of the InterNational Electrical Testing Associa-
ment devices, the fault resistance must be reduced, tion (NETA) can be hired to test and to provide the
that is it should be carbonized or “burned down”. test equipment. As with all techniques used infre-
Direct-current high-potential test sets, as described quently, the skill of trained outside personnel may
in section VII can be used for this purpose. The fault well be worth the additional cost,. “Electrical Equip-
reduction is accomplished by applying a continuous ment Testing and Maintenance” covers terminal
potential between the faulted conductor and and tracer cable-fault locating methods in more de-
ground. The voltage level is adjusted to give the tail for those who wish an explanation of testing
maximum current allowed by the rating of the test technique principles. Three of the methods using
set. As the fault carbonizes, a continually decreas- less complex methods of measuring some electrical
ing voltage will be required to sustain this current. characteristics of faulted cable are shown in figure
The fault reduction has been accomplished when 5-1. Another method uses a time domain reflec-
virtually no voltage results in a steady flow of cur- tometer tester.
rent and fault-locating operations can then proceed. a. Murray loop resistance bridge method. To use
b. Conductors grounded. If one conductor of the this method, the grounded conductor must be con-
faulted cable remains ungrounded, terminal mea- tinuous at the fault and a continuous ungrounded
surement devices can be used. If the fault grounds conductor in the faulted cable must be available.
all conductors and low parallel resistance results, The accuracy of this method is directly related to
only tracer methods can be effectively applied. The the accuracy of the plans showing cable routing.
ohmic value of the fault may be used in some cases The fault is located in terms of its distance from its
to anticipate the effectiveness of the various tracer cable terminal by measuring and comparing electri-
methods that could be applied. cal characteristics of the cable’s faulted and
c. Conductor-to-conductor resistance. When a unfaulted conductors. It is essentially a Wheatstone
fault results in a low-resistance short circuit be- bridge of the slide-wire type. When the bridge is
tween two conductors and the resistance to ground balanced, the fault distance is found as indicated in
is high, reflection methods may be made effective if figure 5-1. A number of slide-wire bridges designed
the single phase fault can be reduced to a ground for fault location are available commercially. They
fault before attempting to locate the fault. If the range from inexpensive units with limited accuracy
single-phase fault cannot be reduced to a ground to more expensive units which can locate a fault
fault, one of the conductors involved may be within one foot per mile (0.2 meters per kilometer)
grounded at a termination. Depending upon the of cable length. Instructions for use, including appli-
relative location of the grounded termination, the cable mathematical formulas, should be supplied
signal pattern and its level and direction may be with the instrument.
quite different from that obtained when locating an b. Capacitance bridge measurement method. The
ordinary grounded fault. capacitance bridge measurement method is effec-
d. Conductor continuity. Use of a bridge-type ter- tive where both the parallel and series fault resis-
minal measurement device depends upon the avail- tances are high enough to treat an unfaulted and
ability of a continuous ungrounded conductor in the the faulted conductor as capacitances to a metallic
faulted cable, which can be looped to the faulted shield or sheath. This technique is simply the mea-
conductor at the far end of the circuit. The faulted surement of capacitance from one end of the faulted
conductor, though grounded, must also be continu- cable to ground and comparing it in terms of dis-
ous. The required continuity is best checked by tance with the capacitance of an unfaulted conduc-
making a bridge measurement of the resistance of tor in the same cable. Almost any alternating-
5-7
TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

‘Jumper

Point of
fault

Murry loop bridge method

120 v
3
60 Hz

Capacitance bridge measurement method

Charging current method for fault location

Figure 5-l. Terminal equipment and cable connection diagrams

current capacitance bridge is suitable, provided it cable propagation velocity “v” multiplied by “t” and
measures capacitance to ground. divided by two which results in equation 5-l.
c. Charging current method. In the absence of an d = vt/2 (eq. 5-1)
alternating-current bridge, the charging current on
the faulted conductor and on a good conductor may (1) Distance determination. The TDR/analyzer
be compared, using several hundred volts or even measures the reflection time and the fault distance
several thousand volts at 60 hertz as the voltage is automatically calculated based on the entered
supply. This circuit with its fault distance formula is velocity of the pulse travel which is usually the ratio
shown in figure 5-1. of the cable’s propagation factor to the speed of light
d. Time domain reflectometer (TDR) method. This or a value of less than one. The analyzer can deter-
method is based upon the measurement of the time mine whether the fault is open-circuited or short-
"t" it takes a generated pulse to reach a fault and be circuited based on waveform reflections as shown in
reflected back. The fault distance “d” equals the figure 5-2.

5-8
 TM 5-684/NAVFAC MO-200/AFJMAN
N 32-1082

da d,,,Ac l??;Eted

Open In the
cable

Transmitted pulse

Short in
the cable
\
\IA
1
\ /
\ 1
\ / T
”
\ Reflected
pulse

Figure 5-2. Cable fault waveform reflections

(2) TDR test settings. Tests require selecting a 5-17. Tracing the cable fault signal.
pulse duration and a propagation factor. A halving procedure can be used to trace faults
(a) Pulse duration. The pulse duration must where the signal cannot be traced along the entire
be wide enough to be interpreted by the TDR ana- length of the cable in any other way. It is time
lyzer and at least about one percent of the transit consuming and costly, and more modern methods
time for the entire length of the tested cable. TDRs utilizing the sophisticated signalling instruments
should have provisions for changing the pulse width available should always be applied if possible.
depending on cable length. a. Procedure. The procedure consists of localizing
(b) Propagation factor. This is the velocity of the fault by progressively limiting it to one half of
the pulse in an insulated cable and will vary in- the previously considered length of cable. The
versely as the square root of the product of the points along the cable route chosen for signal mea-
cable’s line constants, that is, its inductance and surement must be selected so that maximum fault
capacitance. Therefore it will vary dependent upon localization results from each and every measure-
cable insulation. A propagation factor of 500 feet ment. Thus, the first measurement should be made
(152 meters) per microsecond for medium-voltage as close to the midpoint of the circuit as possible;
cables or 600 feet (183 meters) per microsecond for the second at the one-quarter or three-quarter
low-voltage cables is sufficiently accurate when point, depending on the fault location given by the
fault distance measurements are made by two-end first measurement; and so on.
fault pin-pointing. b. Drawbacks. This procedure requires cutting
(3) Two-end fault pin-pointing. A propagation the cable. For cable in duct an access point is
factor set to any value (which must remain un- needed, or the duct line must be broken and a new
changed for both measurements) can be used to structure provided for resplicing. For direct-burial
provide a TDR-measured fault distance from each systems good cable will need to be respliced in a
end of the tested cable. These two distances will new splice box. Resplicing can introduce other pos-
either come short of meeting or overlap each other. sible trouble points.
The true distance to the fault “dt” can be calculated
by equating the determined fault distance from end 5-18. Selecting cable fault locating methods.
one “di” to the fault distance from end two “dz” and
The fault locating method differs dependent upon
the distance between these two points “dp” using
the way the cable is installed. Most installations
equations 5-2 and 5-3, respectively, based on
will be in duct lines, but direct-burial and subma-
whether the determined fault distances fall short or
rine cable installations must also be considered.
overlap.
a. Duct line. The fault-locating equipment used is
df = d, + (d2 + d,)/ (d, + d2) generally a tracer type. Pinpointing of the fault be-
df = d, - (d2 + d,)/ (d, + d,) tween structures is unnecessary. The entire length
5-9
TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

between structures of a faulty cable must be re- the point indicated by the terminal measurement.
placed. If a structure does not contain a cable splice When the portion of the cable suspected to be at
it may be augmented by a resonance or radar termi- fault is raised, it is generally necessary to apply a __
nal method. tracer method to verify the fault location. The im-
b. Direct burial. Fault-locating for direct-burial pulse method is ideally suited to this application, as
installations must pinpoint the fault, so that the the noise of the discharge is usually quite evident
repairs can be made at the point of failure. Such when the faulted section leaves the water.
faults generally can be best located with impulse
(2) Cable sheath leaks. Emergency mainte-
equipment such as the TDR method. Faults can also
nance of submarine cables frequently involves re-
be located by patrolling the cable and listening for
pair of a leak in the cable sheath. When a leak is
the noise of an impulse discharge at the fault. On
longer cables it may be preferable to use some other evident, it should be located as soon as possible; the
means, such as a terminal measurement, to obtain cable raised; and the damaged section removed to
an approximate fault location. In the absence of prevent migration of moisture into the cable. Re-
audible noise, test holes must be dug so that detec- pairs should be made in accordance with standard
tor tests can be made using a tracer method. splice procedures for submarine cables covered in
c. Submarine cable. The approximate location of section VI.
a submarine cable fault must be determined by ter- d. Gas-pressurized cable. Use a high-resistance
minal measurements. method such as any terminal method, except the
(1) Locating the fault. Verification of location Murray bridge loop, or use the impulse tracer
can be made by pulling the cable out of the water at method.
Section VI - CABLE REPAIR
5-19. Underground cable repairs. open up the duct line at the fault point. If the cable
Underground cable is usually either direct burial or can be repaired without a splice, the duct line can
installed in ducts. While repair methods described be reclosed. It may be necessary to build one or
below are basically the same for any underground more new structures at the point to house any new
cable, there are some differences depending on the cable which needs to be spliced into the existing -
installation condition. cable to replace the faulted section. The method to
be used is largely a matter of judgment based on all
a. Direct burial. While there may be splice boxes,
the factors known at the time.
normally there is no structure to consider, and a
hole will have to be dug to make the repair. This
5-20. Submarine cable repairs.
may be a test hole used to pinpoint the fault loca-
tion. Such access may have to be enlarged, if the When the approximate location of the fault has been
repair involves an appreciable length of cable. The determined, the cable should be lifted to the surface
major problem in repairing direct-burial cable may for examination and repairs. The cable may be lo-
be to provide a dry environment while making the cated with a grapple hook and lifted with a barge-
repair. A temporary shelter may be required. type crane. If the cable is difficult to locate, the
b. Duct line. If the faulted cable length is in a services of a diver may be required to determine the
duct line between structures, there are several re- location and to attach a line for lifting the cable.
pair methods. Usually, only one circuit is installed After the cable has been hoisted to the surface, it
in a duct line in order to avoid cable capacity derat- should be cut at the location of the fault to deter-
ing. mine if (and how much) water has entered the
(1) Spare duct. If there is a spare duct avail- cable. The most expedient method of determining
able, the simplest solution may be to pull a new the distance the water has entered the cable is to
length of cable into this duct and connect it at both cut the cable 25 to 50 feet (8 to 15 meters) on either
ends to the good ends of the faulted cable. Then pull side of the fault. This distance depends somewhat
out the faulted cable, if possible, to provide a spare on the time that the cable has remained in the
duct. At the very least, tag at both ends to indicate water after the fault occurred and the type of cable
that this cable has been abandoned in place and is insulation. At the point where the cable has been
faulty. cut, a sample of insulation should be tested for
(2) No spare duct. The cable should be pulled moisture as covered in section VII. If there is evi-
and a new cable installed. If this is impossible be- dence of moisture at the first cut, then a second -
cause of duct damage, a new duct must be installed. section should be removed and another test made.
Alternately, it may be faster and more economical to This operation should be continued until the point
5-10
 TM 5-684/NAVFAC MO-2OO/AFJMAN 32-1082

is reached where moisture is no longer present in that the cable can be picked up at any desired point
the insulation. for repair or inspection. Maps should indicate the
a. Replacement of cable section. The cable section exact location of the cable at various points along its
which has been removed should be replaced by length, as established by measurements with sur-
splicing in a new cable of sufficient length. The veying instruments. Maps should also indicate the
cable manufacturer’s splicing kit instructions exact length of cable installed between any two ref-
should be followed for making splices which will erence points, so that any movement or drifting of
probably have to be made on a boat or barge. Care the cable on the bottom can be estimated.
should be taken not be bend the cable sharply
where it enters the water or where it rests on the 5-21. Cable repair safety.
bottom. When the cable is again laid in the water, Cable repair will involve: working with other cables
the top must be carried out and laid down in such which may or may not be energized; ensuring that
manner that there will be no sharp kinks or bends the grounds and bonds essential for safe operation
in the cable. are in good condition; and, in some cases, dealing
b. Protection of submarine cables. Because sub- with hazardous substances.
marine cable is relatively unprotected as it lies on a. Energized cables. Repair work on electric
the bottom of a body of water, special precautions cables should be done unenergized. However, other
must be taken to prevent damage from swift cur- cables in the manhole may be energized and inspec-
rents, boat anchors, or other causes. Normally, tions will usually be done with cables energized.
these precautions are taken during the original Moving cables while energized should be restricted
cable installation and are not the concern of repair to low-voltage cables in good condition and with
and maintenance personnel. The following para- adequate bending radii. When the condition of any
graphs describe steps to be taken if original precau- insulation is questionable, or cables are installed to
tions were omitted, or if original precautions were their permissible bending radius limits, a small
disturbed in the process of cable repair. change in radius to an energized cable could cause a
(1) Anchors. When a cable crossing is subjected fault.
to flow or tidal currents, cable anchors are generally b. Grounding and bonding. Inspection and main-
required to prevent excessive drifting of shifting of tenance of cable grounds are as important as the
the cable along the bottom. These anchors are usu- inspection and maintenance of cables, both for
ally made fast by a series of U-bolts which pass safety and for corrosion mitigation. Ground rods
through a common base plate, thus affording a mul- should also be inspected.
tiple grip of the cable. Other U-bolts, eyebolts, or c. Hazardous substances. Hazardous substances,
alternate means are usually provided for attach- such as lead or asbestos, should be replaced when
ment of the anchor cable or chain. Ordinarily, the the cable repair work requires their removal; such
anchors are masses of concrete sufficiently large to as pulling new cable in ducts sealed with lead or
resist the draw of the current. When the water is asbestos or splicing cables having asbestos fire-
shallow, anchors may be placed on the cable as it is proofing. Encapsulating materials are available
being reeled out. When a deep-water crossing is which will prevent asbestos fibers from becoming
encountered, attachment of the anchors to their airborne. See chapter 15, section II.
chains must be done by a diver.
(2) Warning signs. Suitable warning signs indi- 5-22. Making cable repairs.
cating the location of the shore ends of a submarine In many cases, the fault will be in either an existing
cable, and stating that anchoring of vessels is pro- splice or a termination, and the repair is compara-
hibited in the immediate vicinity of the cable any- tively simple. In other cases, the fault will be in the
where along its length, are required for every sub- cable itself, and the repair involves removing a de-
marine cable crossing. fective cable length and splicing in a good length.
(3) Pile clusters. Clusters of piles are fre- The replacement must be the same as the original
quently driven at the shore lines of important cables cable or a type of cable comparable to and compat-
where they enter and leave the water. These aid in ible with the original cable. Splice kits and termina-
locating the points where the cable is anchored. Pile tion kits should be used as much as possible. The
clusters also provide a certain amount of mechani- following paragraphs contain general instructions
cal protection for the cables and furnish platforms for the various types of cables. More detailed in-
on which to mount warning signs. structions are given by the manufacturers of the
(4) Maps. The development of accurate maps is cable and splice kits used for any specific job. After
one of the most important tools in maintenance and the repair is completed (and before backfilling for
repair of a submarine cable installation. It ensures direct-burial cable) insulation resistance and poten-
5-11
TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

tial tests should be made to determine that the are not in potheads, the terminal ends should be
cable, including the new repair, is suitable for use. checked periodically for leakage of compound and
Refer to section VII. breaks in insulation where moisture can enter and
cause deterioration. Where the cable is not termi-
5-23. Solid dielectric cable repairs. nated with potheads, leakage of compound may oc-
The vast majority of cable used in most installations cur. In such instances, an electrical adhesive tape,
will be solid dielectric cable. Solid insulation for designed for medium-voltage splices, should be ap-
power cables is divided into two main cate- plied over the varnished cambric and should be
gories-thermosetting and thermoplastic. A thermo- painted with a sealer paint to stop the flow of com-
setting material is one that requires heat to vulca- pound. If it is suspected that the varnished-cambric
nize or crosslink it. This process causes a chemical insulation has absorbed moisture, a moisture test
reaction and the insulation will have little tendency (as described in section VII) should be made. Should
to soften if reheated. Thermoplastic insulation will the insulation become damaged, it may be replaced
soften repeatedly when heated. Ethylene-propylene with varnished-cambric and covered with friction or
rubber and cross-linked polyethylene solid dielectric adhesive tape, or the ends of varnished-cambric
cable, which are both thermosetting, should be con- tape can be secured with cotton tape and painted
sidered when replacing any old medium-voltage with an insulating paint.
cable. b. Paper-insulated lead-covered cable. A paper-
a. Use. Most facility cable will be single- insulated lead-covered (PILC) cable always has a
conductor type installed in duct. Direct-burial lead sheath and therefore requires little mainte-
concentric-neutral cable, usually installed in hous- nance. Testing at regular intervals will indicate the
ing areas, may be single-phase or three-phase. condition of the insulation. A break in the lead
Direct-burial or submarine cable, armored or gas sheath will expose the paper to moisture and a
pressurized, is usually three-conductor cable. moisture test should be made. If several layers of
b. Repair. The repair of solid dielectric medium- paper have been removed, they should be replaced
voltage cable may be accomplished by the use of with varnished-cambric tape. In order to do this, it
preassembled splice kits, which are available for the may be necessary to remove a longer section of lead
various types of cables and their protective cover- sheath. The section of lead sheath removed should
ings, if any. Particular attention should be given to be replaced with a new lead sheath and wiped in
ensure that the material used for the repair is com- place in the same manner as a splice. The repairs
patible with the cable insulation. Low-voltage described above can only be made in a structure. If
cables which do not have cable shields and are not it appears there has been a break in the lead sheath
provided with armors or metallic sheaths can be within a duct, it will be necessary to replace a
repaired with jacket repair sleeve kits which seal length of cable between structures or between ter-
and repair insulation damage. These sleeves may be minals.
appropriate on medium-voltage cables when only (1) Should a PILC cable fail again shortly after
the jacket is damaged. a repair or lengthy de-energization, the cable may
be completely unusable. However, prior to replac-
5-24. Other cable insulation and covering re- ing, make an attempt to “dry out” the cable by
pairs. forcing a low-voltage, high current through the
Cable insulations other than solid dielectric compo- cable (via an arc welder) for not less than 12 hours.
sitions are now being installed only where special Current should be as high as possible but must not
circumstances justify their use. Varnished cambric exceed 80 percent of rated cable ampacity.
and paper insulated cables, however, may still be in (2) Do not support PILC cables by using metal
service at this time. Gas-pressurized cable with straps or supports or by laying them on metal trays
solid dielectric insulation is used for underwater because of the deteriorating galvanic action.
installations to provide mechanical protection, pre- c. Lead-sheathed cable. Lead sheaths may crack
vent the entrance of water, and minimize electrical or suffer other damage as a result of fatigue due to
losses which can arise from armor protection. Pro- cable movement or bending. If a section of lead
tective cable coverings require appropriate repair. sheath is seriously damaged or badly cracked, the
a. Varnished-cambric insulated cable. Varnished- section of sheath should be removed and the area
cambric insulation requires little or no maintenance covered with a lead sleeve. The sleeve should be
where cables terminate in potheads. Some wiped in place in the same manner as a splicing .-
varnished-cambric insulated cables are provided sleeve. Where the damage is not too serious, repairs
with lead sheaths. However, where terminations may be made as follows:
5-12
 TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

(1) Scrape the lead in the vicinity of the dam- toxic gas will dissipate fast. Follow instructions for
age. SF, gas handling in chapter 15, section II.
_- (2) Preheat area using an acetylene or gas blow (4) Cable insulation. If the cable insulation
torch. Care should be taken not to melt lead. fails, the repair should be made as given for solid
(3) Apply a good flux such as stearine. dielectric insulated cables.
(4) Apply solder and heat to a point where it is f. Armor. There are various types of armor to pro-
pliable. vide mechanical protection for the cable insulation
(5) Work sold er into sheath with a paddle or and some armor is provided with a protective
stick and smooth. jacket.
(6) Wipe with wiping pad. (1) Flat metal armor. When flat metal armor is
d. Concentric-neutral cable construction. The broken, it can be repaired in place by soldering the
neutral conductor of this type of cable consists of broken part or by overlapping a short piece of armor
equally spaced strands of wire or flat strap wrapped and soldering. The armor should be thoroughly
spirally around the outside of the cable insulation. cleaned and tinned before soldering. If the armor
Concentric neutral cables may be of a single or becomes loose at the ends of the cable, it may be
multiple conductor configuration and may have an wired to the lead sheath or a lead sleeve installed at
outer protective jacket over the neutral conductors. each end.
Care must be used in handling the cable to prevent (2) Inter-Locked armor sheath. The interlocked
the concentric-neutral conductors from loosening or armor sheath may sometimes be separated during
bunching together in one place around the cable installation. Short sections of this armor can usu-
circumference. If this occurs, it may cause tracking ally be worked into place (one section at a time) by
of the insulation or affect the voltage stress distri- using a hammer and screwdriver or other blunt
bution within the insulation. The number and size instrument. The armor may sometimes become
of the concentric conductors are determined by the dented. When not too seriously dented, no repair
manufacturer in accordance with ICEA standards. should be attempted. However, if the dent is serious
Repairs should ensure that the operating character- enough to cause possible injury to the cable, a sec-
istics of the concentric neutral are not adversely tion of armor should be removed and replaced with
affected. a sleeve.
e. Gas-pressurized cable. Repairs may be re- (3) Wire armor The wire armor over submarine
quired for this cable because of a loss of pressure cable may sometimes become bent and separated.
from a leak in the outer covering or conduit or a In such cases, the armor should be replaced, bound
puncture in the solid dielectric cable. with wire, and soldered. The binding wires should
(1) Gas pressure. The high density of the gas be of a metal similar to that of the armor. If the wire
may result in a zero gage reading even though suf- armor is broken, it can be repaired in a similar
ficient gas is still present to keep out air and mois- manner. After the link has cooled it should be
ture. If the cable is under a water-pressure static- painted with a heavy coat of insulating compound to
head, then a pressure of 0.5 pounds per square inch reduce any possible dissimilar metal corrosion.
gage (3.5 kilopascals gage) is needed for each foot (4). Protective jacket over armor. Sometimes a
(0.3 meters) of water head to continue to keep water protective jacket such as rubber, thermoplastic, or
out. braid is placed over the armor for protection against
(2) Repair of gas Leaks. Follow the manufactur- corrosion. Should these jackets become damaged,
er’s recommendations. A beep detector should be the damaged area can be repaired by wrapping with
used to ensure all leaks have been repaired. Sur- thermoplastic tape or self-vulcanizing tape.
faces should be clean and smooth, and the pressure
must generally be dropped back to zero. 5-25. Other cable component repairs.
(3) Treatment of gas. The SF, gas used is non- Potheads and terminations should be checked along
toxic and odorless. However any arcing produced by with any cable fireproofmg.
the fault will result in the gas producing toxic ma- a. Terminations. The insulators should be kept
terials that smell like sulfur and rotten eggs. Do not clean, and bodies of the compound-filled termina-
breath this gas. Let it dissipate to the air and clear tions, such as potheads, should be checked for leaks.
it from a structure. Since the gas will be outside the A leak will usually be indicated by oozing out of the
shield with a solid dielectric cable, the cable can fail compound. In such instances, the leak should be
without arcing the gas. If a cable were to fail and repaired and the pothead refilled. A power factor
burn a hole through the insulation and then be test (as covered in section VII) of the termination
re-energized, a very small amount of gas in the hole and cable will give some indication of the condition
could arc and may be noticeable. The smell and of the termination so far as electrical leakage and
5-13
TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

losses are concerned. Installation of other termina- b. Cable fireproofing. Where cables of more than
tions should be in accordance with the instructions one circuit pass through structures, any fireproofing
of the termination manufacturer for the type of material around cable sheaths should be main-
cable involved. tained to prevent damage from adjacent cables.
Section VII - CABLE TESTING
5-26. Cable tests. sulation resistance test will measure the insulation
Tests are made on installed cable for two resistance of all items connected together, the cable
reasons-to check the condition of the cable and for to be tested must be completely disconnected from
a cable requiring maintenance to ensure that the all other cable and equipment.
repair was properly made. a. Measuring equipment. All measurements re-
a. Type of voltage tests. Insulation resistance quire a direct-current source which can be a hand-
measurements and direct-current, over- or high- cranked generator, a motor-driven generator, a
potential test (direct-current hi-pot test) are the battery-supplied power pack, rectified alternating
usual direct-current voltage tests for cables. A di- current, or its own internal power source. The mea-
electric absorption test which takes longer than a suring devices can be as follows.
standard insulation resistance test may aiso be ap- (1) A megohmeter which is a contained instru-
propriate. A fourth test is the power factor test. ment (commonly called “megger,” although “Meg-
b. Testing frequency. The periodic testing of in- ger” is the trade name of a tester of this type made
stalled medium-voltage cables is known as proof by James C. Biddle Co.) consisting of an indicating
testing, since its purpose is essentially a means of ohmmeter and an internal source of direct-current
proving that weak spots in existing cables have voltage.
been recognized before failure occurs. Cables nor- (2) A resistance bridge.
mally have a higher failure rate in the first 2 years (3) A voltmeter.
of service, which is the period when manufacturing (4) A voltmeter and micro-ammeter.
defects will show up. An alternating-current test is b. Megohmeter. The most convenient and com-
used for factory testing of new cable. Acceptance monly used way to measure resistance is to use a
and proof testing utilizes direct-current testing self-contained instrument giving a direct readings -_
which, while not as effective as alternating-current, in ohms, kilohms, or megohms. Measurement is ob-
is less liable to damage cable. Direct-current testing tained by connecting one instrument terminal to the
provides extremely valuable historical data and al- cable conductor or the equipment terminal and the
lows comparison of the acceptance testing value to other instrument terminal to the metal sheath,
the periodic proof testing values. A yearly overvolt- frame, container, or support of the insulation under
age test for the first 2 or 3 years, and then testing test. Instruments are available in voltage ratings of
every 5 or 6 years, is the optimum and can reduce 500 to 2,000 volts or more. Care must be taken to
in-service failures by a factor of about nine to one as use a voltage which does not exceed the insulation
opposed to not having a proof testing test program. rating of the item being tested. Follow specific in-
Insulation resistance tests should be based on the structions provided with the instrument being used.
importance of the circuit; once a year is usually This type of instrument is not very accurate or sen-
adequate. sitive in very low ranges and should not be used to
c. Other tests. Varnished-cambric and paper- measure a few ohms or fractions of ohms, such as
insulated cables may require moisture tests. Gas- resistances of conducting paths.
pressurized cable may require leak tests. Refer to c. Resistance bridge method. A self-contained in-
section V. strument, called a Wheatstone Bridge, containing a
battery, a galvanometer, and known resistances, is
5-27. Cable insulation resistance tests. used to compare an unknown resistance with a
Insulation resistance is the resistance which the known resistance. Each instrument contains de-
insulation presents to a flow of current, from an tailed instructions for its use. While very accurate
impressed direct-current voltage. An insulation re- results can be obtained, a bridge is basically a labo-
sistance test is a short-time test made to indicate ratory instrument and requires a fairly skilled op-
the suitability of the insulation for the purpose in- erator. It is not recommended for field work.
tended, or to indicate whether an overpotential test d. Other methods. Both the voltmeter method
can be made without damaging the insulation. It is and the voltmeter/micro-ammeter method require
not a dielectric strength test, but will give an indi- laboratory type instruments and a separate source -
cation of the insulation’s condition with respect to of direct-current voltage. They are inconvenient to
moisture and other contamination. Because any in- use in the field and are not recommended. However,
5-14
 TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

if desired, some details can be found in the “Ameri- cable insulation can still withstand over-voltages
can Electricians’ Handbook. caused by normal system surges. As noted,
e. Factors affecting insulation resistance. Person- alternating-current tests are reserved for factory
nel making and interpreting the results of insula- tests to determine whether the insulation had any
tion resistance tests should consider the following discontinuities, voids, or air pockets. Less destruc-
factors which affect the test readings: tive, direct-current tests are used for installation
(1) Temperature. Insulation resistance varies proof testing. Also, most equipment for direct-
with the temperature, and the effect of temperature voltage testing is smaller and more readily portable.
depends on many other things, such as type of insu- A 115-volt alternating current power supply is rec-
lation, amount of moisture in and on the surface, tified to provide direct-current for testing. Several
and the condition of the surface. All spot-test read- commercial types are available. Each type comes
ings should be corrected to a base temperature such complete with transformers, rectifiers, instruments,
as 40 degrees C. and controls.
(2) Moisture. The amount of moisture in the a. Voltage. Normally, the maintenance proof tests
insulation has a large influence on its resistance. performed on cables are at a test voltage of 60 per-
For meaningful results, tests of insulation resis- cent of the final factory test voltage for new
tance should be made under as near similar condi- cable/equipment. Determination of voltages for ac-
tions as practical. A long cable can be exposed to ceptance and proof tests should be made by quali-
different conditions along its length so a comparison fied electrical engineers, and such tests should be
of readings not made at the same point may be made only when specifically directed by an engi-
misleading. neering activity having jurisdiction over the instal-
f. Interpretations. Usually, because of the stored lation. Tests should be performed in accordance
capacitance of the cable there will be an initial am- with ANSI/IEEE 400. It is always appropriate to
pere dip toward zero followed by a steady rise. The conduct the insulation resistance measurement test
spot-test reading should be taken after a 60-second first; and, if the data obtained is within acceptable
voltage application. A cable 1,000 feet (300 meters) limits, to proceed with the direct-current
long will have an insulation resistance of one-tenth overpotential test.
of that for a l00-foot (30-meter) cable, provided all b. Procedure. For each cable, the test should be
other conditions of both tests were identical. A made between each conductor and every other con-
gradual decline in resistance with age is normal; ductor and between each conductor and ground. For
however, a sudden decline means insulation failure the test to ground, all conductors may be connected
is imminent and a continued downward trend indi- together. There is no need to disconnect other equip-
cates insulation deterioration, even though mea- ment from the cable, but caution must be observed
sured resistance values are above the minimum ac- to ensure the test voltage is not greater than recom-
ceptable limits. mended for any of the equipment. Some preas-
g. Dielectric absorption test. This test is usually sembled or premolded cable accessories may have a
conducted at higher voltages for extended periods of basic insulation level lower than the cable tested,
from 5 to 15 minutes. Since the current is inversely and the lower voltage should be taken from IEEE 48
related to time, insulation resistance will rise test limits or the manufacturer’s test limits, which-
gradually for a good cable but will flatten rapidly ever is smaller. If a test shows poor results, items
otherwise. Periodic readings should be taken and and/or conductors should be retested separately un-
plotted against time. The ratio of the lo-minute to til the defective portion is identified. Specific in-
the l-minute resistance is known as the polariza- structions furnished with the tester being used
tion index. A polarization index of two or higher should be carefully followed. Additional information
indicates good insulation, while a polarization index is contained in chapter 7, section II.
of less than one indicates cable deterioration c. Safety. When making high-voltage tests, all ap-
the need for immediate maintenance. plicable precautions regarding live electrical con-
ductors should be observed to avoid dangerous elec-
5-28. Cable overvoltage tests. trical shock. After testing, the terminals should be
A hi-pot or overpotential test is an overvoltage test short circuited before disconnecting the tester from
used to check a cable for its relative condition after the equipment. The short circuit should be main-
it has been repaired or otherwise worked on. Nei- tained for at least as long as the time the proof
ther the insulation resistance test nor the dielectric voltage was applied.
absorption test can determine the dielectric d. Test data. There are three types of direct-
strength of cable insulation under normal use. A current hi-pot tests commonly performed. In all
hi-pot test is the only way to gain proof that the cases, leakage (conduction) current is measured and
5-15
TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

the values compared either on a voltage or a time age is left on for 5 minutes and the leakage current
basis for initial to steady-state values or for a con- is read after 30 seconds, 1 minute, and then at 1
stant rate of leakage current. minute intervals thereafter up to 5 minutes. The
(1) Initial leakage current. The initial leakage leakage current is plotted against time as the initial
current upon a test voltage application will include high value reduces to a steady-state value. A con-
transient capacitive charging and dielectric absorp- tinuous decrease indicates a good cable. There
tion currents. Two other currents, corona current should be no increase in current during this period.
and surface leakage current, can be bypassed by
f. Test for cable withstand strength. A go/no-go
installing correct guarding circuits.
test is usually performed after repair if only cable
(2) Steady-state leakage current. The initial withstand strength requirements need be verified.
value will decrease to a steady-state value consis-
The test provides a rising voltage up to the specified
tent with the system’s charging current. If correctly
value applied to maintain a constant leakage cur-
done, only the volumetric leakage current will be
rent. A period of 1 to 1.5 minutes for reaching the
left. This current is of primary interest in the evalu-
final test voltage is usually adequate. The final test
ation of an insulation’s condition. The decay of tran- voltage is held for 5 minutes. If the current has not
sient current time is known as the stabilization
increased sufficiently in that time to trip protective
time. devices, the cable withstand voltage is adequate.
(3) Constant leak age current. In some cases, a
constant leakage current is measured. This is main- 5-29. Cable power factor tests.
tained by increasing the test voltage in a manner
which maintains the same current. Power factor testing is a nondestructive ac test
e. Tests for relative cable condition. Two tests are which has been utilized for many years to measure
used to determine the relative cable condition as an or test the integrity of substation insulation sys-
identification of its dielectric strength under tems including cables.
medium-voltage tests. a. Test theory. An insulation to which voltage is
(1) Leakage current versus voltage test. In this applied will act like a resistor and capacitor in par-
test, equal voltage steps are applied until the maxi- allel as shown in figure 5-3. The capacitive current
mum test voltage is reached or an indication of a I, will be much larger than I, so the angle 8 will be
breakdown voltage is indicated. It is usually recom- close to 90 degrees and the power factor (cosine of 0)
mended that no less than five and, if possible, eight will be very small.
equal steps be made with no less than 1 and up to 4 b. Cable power factor test limitations. Cable insu-
minutes stabilization time allowed. The steady- lation can be considered to consist of a simple ele-
state leakage current is plotted against the applied ment of capacitance in parallel with resistance as
voltage. As long as the slope of the plot is the same, shown in figure 5-3. The measured power factor is
the insulation is in good condition. If the leakage the average of the entire length of the cable. If a
current increases noticeably, so will the slope of the section of cable increases in power factor the high
curve. Any change in the slope indicates that any value obtained for that section will be averaged
voltage increase may cause insulation breakdown with the normal value obtained for the remainder of
and the test should be stopped. the cable. The influence that the defective section of
(2) Leakage current versus time test. This test the cable has on the overall cable power factor de-
is made after the maximum test voltage of the pre- pends on the relationship of the defective section
vious test has been determined. The maximum volt- length to the overall cable length. Thus, the ability

‘T ‘C

‘R

WATTS . ”R
POWER FACTOR= = C O S I N E ++
V O L T A M P E R E S = VI,
Figure 5-3. Insulation power factor equivalent circuit and vector diagram

5-16
 TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

to
o detect a localized fault diminishes as the length Table 5-l. Typical power factor ranges for various cable
insulating materials
of the cable under test increases.
c. Effectiveness of power factor tests. Power factor Cable insulation Power factor ranges1
tests can be effective in detecting defective insula- Polyethylene 0.001 to 0.002
Cross-linked polyethylene 0.001 to 0.002
tion for short cable runs which are common in elec- Oil and paper Less than 0.005
trical substations and industrial complexes, and for Rubber 0.005 to 0.04
indicating general deterioration and/or contamina- Varnished cambric 0.04 to 0.08
tion of longer lengths of cable insulation runs. This
’ At 20 degrees centigrade
test can be performed at any voltage that does not
exceed the line-to-ground voltage rating of the
f. Temperature correction. Temperature has an
cable. In addition to checking cable insulation, this
influence on the power factor values. However, at
test can be used to find: the operating temperature normally encountered in
(1) Any defects in the shield circuit which can the field, this influence is minimal for modern insu-
lead to localized problems. The measurements lation systems. Older forms of insulation may re-
should be performed on each end of the cable. The quire a temperature-correction factor. It is difficult
shield should be grounded for the test only at the to obtain accurate field cable temperature measure-
end where the test connections are made. An in- ments; hence, most utilities evaluate the condition
crease in power factor can indicate discontinuities of the insulation of their cables based on test data,
or breaks in the shield. uncorrected for temperature. If it appears that high
(2) Any defects in cable terminations, particu- cable temperature may have influenced the results
larly compound-filled potheads, can be determined it is recommended that a cable having a high power
by using collar tests as covered in chapter 3, section factor be retested at a time when a lower cable
VII. temperature will occur.
d. Power factor tip-up testing. Normally, power 5-30. Cable moisture tests.
factor should be independent of voltage as shown by Tests for moisture may be made on paper and
the formula of figure 5-3. An increase in power varnished-cambric insulation by removing one or
factor at an increased test voltage is usually an two layers of the insulation and dipping it into oil
indication of insulation voids. These voids are ion- heated to a temperature of 260 to 285 degrees F
ized at the higher voltage and act as resistors re- (125 to 140 degrees C). If the insulation contains
sulting in a greater resistive current and therefore a moisture, a concentration of bubbles will be emitted
greater power factor. This increase in voltage is from the paper. If there is no moisture present,
called power factor tip-up test. there will be little or no bubbling. Generally, the
(1) Two to five measurements should be made outer layers of insulation are tested first, as they
at voltages with an overall 5 to 1 ratio in order to are the most accessible for testing and the most apt
determine whether there are any significant differ- to show moisture. Where moisture is indicated in
ences in the measured power factors. the insulation, successive layers should be removed
and tested until there is no evidence of absorbed
(2) Generally differences are not a concern un-
moisture. All moisture-damaged cable should be re-
less the higher value exceeds 25 percent of the lower
placed with new cable.
value. This much change indicates that further in-
vestigation of the cable insulating quality is re- 5-31. Cable test records.
quired. It is very important that cable records be made for
e. Cable test data. Power factor data obtained any inspection or test on any circuit. Such records
from the field tests should be compared with any should flag when the next inspection or mainte-
available previous test data in order to detect any nance outage is to be made. Since these tests re-
changes. Lacking any initial test data, evaluation of quire taking the cable out of service, advantage can
the condition of the cable insulation may be made then be taken of the maintenance outage, rather
by a comparison of the field data with tabulated than taking a cable out of service for tests only. The
power factors obtained for similar insulated cables trend of the reading obtained will determine
known to be in good condition, or with the manufac- whether the cable is stable, slightly aging, or rap-
turer’s published specifications. Table 5-1 indicates idly deteriorating. Slight decreases in the insulation
typical acceptable power factors for various cable resistance each year are to be expected as the cable
insulations which may be used if no other data is ages. Tests should be made more frequently if more
available. Ranges given should not be used to justify than the usual decrease indicates that deterioration
variations in tip-up test values. is approaching a critical state.
5-17
TM 5-684/NAVFAC MO-200/AFJMAN 32-1082

Section IX - UNDERGROUND CORROSION CONTROL

5-32. Importance of corrosion control. conductors, in which moisture serves as the electro-
-
As corrosion often results in deteriorated equipment lyte.
leading to electrical outages, its control is necessary.
5-34. Prevention of corrosion.
While corrosion can occur because of many reac-
tions, underground corrosion of metallic cable Maintenance personnel must be alert to minimize
sheaths and grounds is the most common and costly the effects of corrosion inherent in exposure to the
type of corrosion found in electrical distribution sys- elements or resulting from installation methods
tems. that did not properly address galvanic effects.
a. Chemical corrosion. Prevention of chemical
5-33. Types of corrosion. corrosion is relatively easy to accomplish by proper
There are two basic types of corrosion. One is purely painting or other surface protection.
chemical in nature and is the reaction between ele- b. Galvanic corrosion. In electrical systems, gal-
ments, such as water and iron to cause rust. The vanic corrosion is caused primarily by protective
other is galvanic corrosion and is an electro- metallic cable coverings, such as lead or steel; by
chemical reaction between dissimilar metals in an the grounding system, which is usually copper or
electrolyte. An example of the first type is rusting of copper coated; and by metal conduit, either galva-
a steel nail in a glass of water with no other metal nized or ungalvanized. TM 5-811-7, MIL-HDBK-
involved in the reaction. An example of the second 1004/10 and AFI 32-1054 cover galvanic corrosion
type is corrosion between copper and aluminum in detail.

5-18