FM Global

Property Loss Prevention Data Sheets 5-31

January 2020
Interim Revision January 2021
Page 1 of 23

CABLES AND BUS BARS

Table of Contents
Page

1.0 SCOPE .................................................................................................................................................... 3

1.1 Hazard .............................................................................................................................................. 3
1.2 Changes ............................................................................................................................................ 3
2.0 LOSS PREVENTION RECOMMENDATIONS ........................................................................................ 3
2.1 Construction and Location ................................................................................................................. 3
2.1.1 General ................................................................................................................................... 3
2.1.2 Cable Trays Spacing ............................................................................................................... 4
2.2 Occupancy ........................................................................................................................................ 4
2.3 Protection ........................................................................................................................................... 4
2.4 Operation and Maintenance .............................................................................................................. 5
2.4.1 Operation of Cables and Bus Bars ......................................................................................... 5
2.4.2 Maintenance of Cables and Bus Bars .................................................................................... 6
2.4.3 Subsea Cable .......................................................................................................................... 7
2.5 Contingency Planning ........................................................................................................................ 7
2.5.1 Equipment Contingency Planning .......................................................................................... 7
2.5.2 Subsea Cable ......................................................................................................................... 8
2.5.3 Sparing .................................................................................................................................... 8
2.5.4 Equipment Breakdown Spares .............................................................................................. 8
3.0 SUPPORT FOR RECOMMENDATIONS ................................................................................................. 8
3.1 Cable Fire Behavior ........................................................................................................................... 8
3.1.1 Cable Tray Spacing ................................................................................................................ 8
3.1.2 Smoke Detection, Manual Response, and Nonthermal Damage Considerations .................. 9
3.2 Electrical Field Tests .......................................................................................................................... 9
3.2.1 Insulation Resistance (IR) and Polarization Index (PI) ........................................................... 9
3.2.2 Offline Partial Discharge .......................................................................................................... 9
3.2.3 Very Low Frequency (VLF) AC Withstand Voltage Test ......................................................... 9
3.2.4 Withstand Testing .................................................................................................................. 10
3.2.5 Damped AC test (DAC) ......................................................................................................... 10
3.2.6 Insulation Dissipation Factor (DF) ........................................................................................ 10
3.3 Online Condition Monitoring ............................................................................................................ 10
3.3.1 Infrared Thermography (IR) ................................................................................................... 10
3.3.2 Time domain Reflectometry (TDR) ........................................................................................ 11
3.3.3 Distributed Temperature Measurement System (DTS) ......................................................... 11
3.3.4 Strain Monitoring .................................................................................................................... 11
3.4 Subsea Cable Failure Modes .......................................................................................................... 11
3.5 Subsea Cable Protection Against External Threats ........................................................................ 12
3.6 Repairing Subsea Cables ................................................................................................................ 13
3.7 Loss History ..................................................................................................................................... 14
3.8 Illustrative Losses ............................................................................................................................ 14
3.8.1 Lack of Adequate Cable Fire Stopping ................................................................................. 14
3.8.2 Oil-Filled Cable Fire at a Hydro-Electric Plant ...................................................................... 14
3.8.3 Ground Fault Shuts Down Paper Mill ................................................................................... 16
3.8.4 Cable Fire in a Paper Mill ..................................................................................................... 16
3.8.5 Cable Fire in Super Computer Manufacturing, Product Testing Lab .................................... 16
4.0 REFERENCES ...................................................................................................................................... 17
4.1 FM Global ....................................................................................................................................... 17

©2020-2021 Factory Mutual Insurance Company. All rights reserved. No part of this document may be reproduced,
stored in a retrieval system, or transmitted, in whole or in part, in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise, without written permission of Factory Mutual Insurance Company.
5-31 Cables and Bus Bars
Page 2 FM Global Property Loss Prevention Data Sheets

4.2 Other ................................................................................................................................................ 17

APPENDIX A GLOSSARY OF TERMS ....................................................................................................... 17
APPENDIX B DOCUMENT REVISION HISTORY ...................................................................................... 18
APPENDIX C SUPPLEMENTARY INFORMATION .................................................................................... 18
C.1 Insulation Failure ............................................................................................................................. 18
C.2 Cable Flammability Tests ............................................................................................................... 19
C.2.1 General ................................................................................................................................ 19
C.2.2 FM Specification Tested Cable ............................................................................................ 19
C.3 Subsea Cable ................................................................................................................................. 22
C.4 Online PD Measurement ................................................................................................................. 23

List of Figures
Fig. 1. Elevation view: uniform multilevel tray spacing ................................................................................... 4
Fig. 2. Cable vault with obstructed ceilings and adequate sprinkler spacing in aisles or walkways ............. 5
Fig. 3. Three stages of one DAC excitation; charging, switching, and discharging through inductance ..... 11
Fig. 4. Illustration of burial protection index .................................................................................................. 13
Fig. 5. Cable ships by market ....................................................................................................................... 14
Fig. 6. Fire causes by frequency .................................................................................................................. 15
Fig. 7. Electrical breakdown by frequency .................................................................................................... 15
Fig. 8. ASTM E-2058 Fire propagation apparatus. ..................................................................................... 20
Fig. 9. Typical HVAC subsea power cable: 3 phases bundled with fiber-optic elements ............................ 22
Fig. 10. Accessory armor hang-off ................................................................................................................ 23

List of Tables
Table 1. Cable Tray Spacing .......................................................................................................................... 4
Table 2. Descriptions of Electrical Tests ......................................................................................................... 7
Table 3. Comparison of FM Large-Scale Cable Flammability Test with Other Flammability Tests ............ 21

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 3

1.0 SCOPE
This data sheet covers covers cables and bus bars for all voltages. This includes, but is not limited to,
communication, power, distribution, and subsea cable.
This data sheet does not cover transmission networks.

1.1 Hazard
Cables and bus bars are primarily susceptible to fire and electrical hazards. Electrical failures can typically
be attributed to manufacturing and installation defects, as well as operational lifecycle. Insulation deterioration
can result from mechanical damage and aging, the latter often accelerated by processes such as thermal
breakdown, electrical and water treeing, and partial discharge. Cable joints and terminations tend to be the
weakest points in the system due to additional mechanical, electrical, and thermal stresses present. When
cables or bus bars fail, the failure can lead to ignition of nearby combustibles as well as cable jacketing and
insulation.
Plastic used as the cable jacket or as insulation for conductors is combustible, and many types of cable
insulations, when involved in a fire, will continue to burn beyond the area of ignition. The type of insulation
and the quantity and arrangement of the cables (horizontal or vertical) determine how rapidly the fire will
spread and how much smoke will be generated. Several factors can increase the size of a loss, including
lack of automatic fire protection, delayed discovery, combustible accumulations in cable trays, unprotected
cable penetrations, and the loss of major process equipment for extended periods of time while waiting to
replace the cables.
For additional information on hazards associated with cables and bus bars, refer to the following FM Global
publications:
• Understanding the Hazard: Grouped Cables (P0218)
• Understanding the Hazard: Fire in Electrical Rooms (P0245)

1.2 Changes
January 2021. Interim revision. Updated contingency planning and sparing guidance.

2.0 LOSS PREVENTION RECOMMENDATIONS

2.1 Construction and Location

2.1.1 General
2.1.1.1 Use FM Approved fire stops where cables, cable trays, or bus bars penetrate fire-rated floors or walls.
Provide a fire resistance rating equivalent to the rating of the wall or floor. Use FM Approved fire stopping
installation contractors.
2.1.1.2 Route control, signal, and power wiring to minimize exposure to fire and explosion from the
surrounding occupancy.
2.1.1.3 Install power cables in cable trays separate from control, signal, and instrument wiring.
2.1.1.4 Stack trays vertically in order of voltage (highest voltage on top). Locate instrument and signal wiring
in the lowest tray.
2.1.1.5 Provide covers or enclosures to keep debris and other combustible material out of cable trays where
needed. Ensure covers are made of material compatible with the tray.
2.1.1.6 Use noncombustible cable trays wherever possible.
2.1.1.7 Install cables and bus bars in accordance with internationally recognized standards and best industry
practices.
2.1.1.8 When the occupancy or equipment requires redundant power supplies, ensure they are routed
independently and not exposed to a common failure or exposure.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

5-31 Cables and Bus Bars
Page 4 FM Global Property Loss Prevention Data Sheets

2.1.2 Cable Trays Spacing

2.1.2.1 Space cable trays in accordance with Table 1 and Figure 1.

Fig. 1. Elevation view: uniform multilevel tray spacing

Table 1. Cable Tray Spacing

Tray Width (W)1 ft (m) Vertical Separation Distance (V)2 Horizontal Separation Distance (H)3
0.5 ft (0.15 m) 4½W 1½W
1.0 ft (0.30 m) 3½W 1½W
1.5 ft (0.46 m) 3W 1W
2.0 ft (0.61 m) 3W 1W
2.5 ft (0.76 m) 2½W 1W
3.0 ft (0.91 m) 2½W 1W
1
W is the width of the cable tray.
2
Same guidelines apply for vertical separation distance from a ceiling.
3
Same guidelines apply for the horizontal separation distance from a wall.

2.1.2.2 Space fiberglass reinforced (FRP) cable trays in accordance with the NFPA 70, National Electric Code,
and the manufacturer’s guidelines.
2.1.3 Route underground cables to ensure accessibility for future maintenance or replacement.
2.1.3.1 Ensure underground cable routes are known and drawings of their locations are maintained.

2.2 Occupancy
2.2.1 Inspect and maintain cable trays free of combustible material, such as rags, paper dust, wood chips,
and oily residue, by cleaning them on a regular basis.
2.2.2 Remove discontinued cable and wire when renovating existing spaces.

2.3 Protection
2.3.1 Provide one of the following active methods of fire protection for areas containing combustible grouped
cable and/or combustible cable trays.
A. An automatic sprinkler or water spray protection installed in accordance with Data Sheet 2-0 or Data
Sheet 4-1N. Design protection for 0.2 gpm ft2 over 3,000 ft2 (8 mm/min over 279 m2) at ceiling level. If cable
is the only combustible material in the area, design protection with a temperature rating of 165° (74°C)
on 130 ft2 (12 m2) spacing. If cable is not the only combustible in the area, refer to the appropriate
occupancy- or equipment-specific data sheet for guidance on sprinkler types, ratings, and spacings.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 5

B. A total flooding, FM Approved, clean agent fire extinguishing system if cables are in rooms or enclosures
only, and designed in accordance with the manufacturer’s installation instructions and Data Sheet 4-9,
Halocarbon and Inert Gas (Clean Agent) Fire Extinguishing Systems. Design using the appropriate Class
C concentrations when the electrical cables/equipment remain energized, and Class A concentrations if
all cables within a room/enclosure are deenergized as recommended in Data Sheet 4-9.
2.3.2 Fire protection is not needed if one of the following passive means of protection is provided:
1. The cable is FM Approvals Class 3972 Specification Tested GP-1 as listed in the specification tested
section of the Approval Guide.
2. The cable or cable tray is wrapped with FM Approved wrap for grouped electrical cables.
2.3.3 When ceiling level obstructions exist, such as high-level cable trays (see Figure 2), install sprinklers
in every aisle or walkway, and in accordance with FM Global Data Sheet 2-0.

Fig. 2. Cable vault with obstructed ceilings and adequate sprinkler spacing in aisles or walkways

2.3.4 For cable tunnels or outside cable trays, design sprinkler protection using a density of 0.2 gpm/fm2
(8.1 mm/min) over the most remote 100 linear ft (30.5 m) up to 3000 ft2 (279 m2).
2.3.5 Position cable trays and runs to facilitate sprinkler placement and reduce obstructions.
2.3.6 Provide FM Approved smoke detection for areas containing combustible grouped cable and/or
combustible cable trays, arranged to alarm at a constantly attended location. For business-critical cable, (e.g.,
cable controlling safety instrumented systems (SIS), cable supplying DC power to emergency lubrication
pumps, cable with a large associated business interruption in the event of a fire), consider the use of Very
Early Warning Fire Detection (VEWFD). VEWFD will enable detection of an incipient fire and an opportunity
to initiate manual response and possible emergency response plans.
2.3.7 Provide FM Approved line-type heat detection within outdoor cable trays tied to a constantly attended
location.

2.4 Operation and Maintenance

2.4.1 Operation of Cables and Bus Bars

2.4.1.1 Operate cables and bus bars within their designed specifications and ratings, including operating
voltage, and temperature limits, ultra violet ratings and weather resistance, direct burial and other exposures.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

5-31 Cables and Bus Bars
Page 6 FM Global Property Loss Prevention Data Sheets

2.4.1.2 Include cables and bus bars in an up-to-date plant load study and provide adequate overcurrent,
short circuit, and ground fault protection. Refer to DS 5-19 and DS 5-20 for electrical protection and system
studies.
2.4.1.3 Keep cables and bus bars clean, cool, and dry, and all connections tightened in accordance with
the relevant industry standard or to the manufacturer’s recommendations.

2.4.2 Maintenance of Cables and Bus Bars

2.4.2.1 For cables and bus bars that are key to the continuity of operations, establish and implement an
inspection, testing, and maintenance (ITM) program. See Data Sheet 9-0, Asset Integrity, for guidance on
developing an asset integrity program. The program should include monitoring of the insulation system. Use
this information to plan repair or replacement of compromised insulation systems prior to faulting in service.
Review the ITM program when the electrical system is modified, and when events occur that require the
program to be updated.
2.4.2.2 Perform an infrared survey on cable splices and connections every 1 to 3 years. Refer to DS 5-19
and DS 5-20 for general guidance on infrared surveys.

2.4.2.3 Cables and Bus Bars Operating Up to and Including 1000 V

2.4.2.3.1 Perform inspection, testing, and maintenance in accordance with Data Sheet 5-20.

2.4.2.4 Cables and Bus Bars Operating Over 1000 V

2.4.2.4.1 Visually inspect exposed sections of cables, splices, and connections on a regular basis, at minimum
quarterly, depending on their age and condition. Verify the following:
A. Cable supports are intact.
B. Where possible, ground connections for cable screen, shielding, and metallic jacketing, are intact.
C. There is no physical damage, such as overheating, corona, discoloration, crazing, cracking, corrosion,
contamination, or embrittlement on splice body and connector body insulation and/or the insulation along
the length of a cable.
2.4.2.4.2 Perform inspection, testing, and maintenance for bus bars in accordance with Data Sheet 5-19.
2.4.2.4.3 Perform the tests identified in Table 2 every 3 to 5 years for the most business-critical cables.
The frequency of testing will depend on previous results and, therefore, it is necessary to perform trending
analysis (i.e., more frequent testing for cable systems with a history of defects). DC high voltage withstand
testing (DC Hi-pot) can be destructive to extruded insulation systems (e.g., XLPE and EPR). A DC hi-pot
test can leave high-voltage DC charges in small voids of the insulation system and cause cables to fail while
in service just after having passed the DC hi-pot test. Very low frequency (VLF) testing and dampened AC
(DAC) testing or rated frequency testing are suitable for extruded insulation systems. See Section 3 for
additional information on Hi-pot testing.
Note: Insulation dissipation factor and partial discharge tests provide value for systems that operate over 4
kV.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 7

Table 2. Descriptions of Electrical Tests1, 2

Test Comments Acceptance Criteria

Insulation The test is typically performed at the systems Trend the tan δ value over time. Cable systems
Dissipation phase to ground voltage level with rated power that are in good condition have tan δ values that
Factor frequencies or at frequencies less than 0.1 Hz. are independent of time. Thus, changes in tan δ
(tan δ) with time can indicate that aging has occurred.
Offline PD will occur in the cable insulation and in cable Compare PD parameters to historical data for that
Partial accessories such as splice bodies and connector cable system or to cable systems under the same
Discharge insulation. Parameters of PD detection are or similar conditions. Trend the recorded data for
(PD) amplitude, pattern, decay of voltage, and future use. The measured PD characteristics, for
extinction of voltage. example phase resolved PD patterns (pulse
repetition rate vs phase angle vs magnitude), can
This test is typically performed at voltages higher be analyzed in various ways based on experience
than the rated phase to ground voltage. and knowledge of the types of defects anticipated
in the cable type and accessories to provide
insight into the condition of the cable system.
Insulation Perform this test between the outer-most metallic Compare the resistance to historical data for that
Resistance shield and earth. This test should be done at 1000 cable or to cables under the same or similar
(IR) V or more depending on the cable operating conditions. An increase in resistance over
voltage and the condition of the cable. historical readings indicates a deterioration of the
outer metallic shield, over time damage to the
outer shield will result in damage to the insulation.
1
Testing cables at lower voltage to collect data and then working up to the rated voltage in steps is useful in understanding the total health
of the insulation system.
2
IEEE 400-2012 and IEEE 400.2, 400.3, and 400.4 provide detailed guidance for performing high-voltage withstand testing. There are
too many scenarios of cable types and conditions to list the details for quantifying a healthy insulation system here. It is imperative that either
the cable owner or the contracted third party assessing and verifying the health of the insulation system has an in-depth knowledge of these
standards and the cable systems they are testing.

2.4.3 Subsea Cable

2.4.3.1 Protect newly installed subsea cable by burying it in the seabed or providing other mechanical means
to protect the cable.
2.4.3.2 Perform a subsea in-service survey on the cable annually to monitor the state (such as re-exposure,
free spanning) and route of the cable. The frequency of cable burial inspections depends on the seabed
dynamics. In the case of sand waves or scour hole developments due to strong tidal or wave action, increase
the frequency to twice per year. Compare the collected data through periodic subsea surveys with the
planned data and post-installation survey data.
2.4.3.3 Provide a distributed temperature measurement system (DTS) to monitor temperatures for hot-spots,
cold-spots, and other irregularities along the cable. The comparison of the temperature profile with previous
measurements can detect cable exposure due to wash-away of sediment or free-span.
2.4.3.4 Provide online partial discharge (PD) monitoring of subsea power cable systems to monitor PD
activities in the cable insulation system or in a cable joint due to cavity, water tree, contamination, and other
flaws.

2.5 Contingency Planning

2.5.1 Equipment Contingency Planning

2.5.1.1 When a cable or bus bar breakdown would result in an unplanned outage to site processes and
systems considered key to the continuity of operations, develop and maintain a documented, viable cable
and bus bar equipment contingency plan per Data Sheet 9-0, Asset Integrity. See Appendix C of that data
sheet for guidance on the process of developing and maintaining a viable equipment contingency plan. Also
refer to sparing, rental, and redundant equipment mitigation strategy guidance in that data sheet.
In addition, include the following elements in the contingency planning process specific to cables and bus
bar:
A. Actions required to isolate damaged conductors, splice bodies and connectors to allow the rest of the
system to be put back into operation.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

5-31 Cables and Bus Bars
Page 8 FM Global Property Loss Prevention Data Sheets

B. Emergency replacement of custom-made or specialty cables, bus bars and splicing kits. This includes
long lead-time or special-order items.

2.5.2 Subsea Cable

2.5.2.1 For subsea cable, include the following elements in the contingency planning process specific to
subsea cable repair and replacement:
A. A contract to have a repair vessel available when repair is needed. In some cases, the contract might
not guarantee the availability of the repair vessel. The repair plan needs to include alternatives for such
a situation.
B. Appropriate electronic gear to locate the fault quickly. Time domain reflectometry pulses can be used
to determine the distance down the cable at which the fault occurred. However, this technique, if it is
intended to be used, must be considered in advance so the devices that can conduct such a test can be
located.
C. A contract to have an experienced repair team available, including an experienced jointer, when repair
is needed. If the repair team is traveling to/from different global regions to perform work, verify all local
working restrictions are understood and steps are taken to address these restrictions to permit the work
while the team is traveling.
D. Appropriate repair tools, such as cable jointing equipment, de-burial equipment, and burial equipment.

2.5.3 Sparing
2.5.3.1 Sparing can be a mitigation strategy to reduce the downtime caused by a subsea cable breakdown
depending on the type, compatibility, availability, fitness for the intended service, and viability of the sparing.
For general sparing guidance, see Data Sheet 9-0, Asset Integrity.

2.5.4 Equipment Breakdown Spares

2.5.4.1 Equipment breakdown spares for subsea cable are spares intended to be used in the event of an
unplanned outage of subsea cable to reduce downtime and restore operations. Provide the following
equipment breakdown spares for subsea cable:
A. Subsea cable with sufficient length for at least one repair operation. When the subsea cable is connected
to a land-based cable, provide spare land cable with sufficient length for at least one repair operation.
B. Accessories such as armor hang-offs, subsea repair splices (joints), sea-land power transition joints,
and armor anchoring devices.
2.5.4.2 Maintain the subsea cable equipment breakdown spares’ viability per Data Sheet 9-0.

3.0 SUPPORT FOR RECOMMENDATIONS

3.1 Cable Fire Behavior

3.1.1 Cable Tray Spacing

Cable trays vary in width and depth and may be run horizontally or vertically. They may be constructed of
aluminum, galvanized steel, stainless steel, polyvinyl chloride (PVC), epoxy coated aluminum, steel, or fiber
reinforced plastic (FRP).
Tray widths vary from 6 in. (0.15 m) to 3 ft (0.91 m) with loading depths of 3 to 6 in. (7.6 to 15.2 cm). A study
was completed to determine the vertical and horizontal spacing needed between horizontal cable trays to
prevent fire spread from one tray to another. Calculations for non-propagating cable were based on the
assumption that an area equaling the width of the cable tray would be involved. For self-propagating cable,
the assumption was that a fire could involve an appreciable length of the cable tray. Exposure to adjacent
cable trays was determined using heat release rates, flame heights, and critical heat flux for ignition
determined using the FM Global Fire Propagation Apparatus.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 9

3.1.2 Smoke Detection, Manual Response, and Nonthermal Damage Considerations

Smoke is a major factor in the ability of manual firefighting response to control a cable fire. Large-scale cable
fire tests were conducted at the FM Global Research Campus for the Electric Power Research Institute
(EPRI). A short time after the start of each test, the building filled with smoke, making it difficult to see the
fire. The test center is a large building similar in size to a small aircraft hangar (2,250,000 ft3 [63,700 m3]).
Accessibility and rapid detection are needed for effective manual response.
Locations not likely to have adequate access for manual response include the following:
• Cable tunnels
• Cable spreading rooms
• Underground cable vaults
• Areas above suspended ceilings
Rapid detection is also important. Smoke and heat detection are most commonly used. Smoke detection
systems in increasing order of sensitivity are: (a) photoelectric and ionization type smoke detectors, (b) beam
type smoke detectors, and (c) very early warning fire detection (VEWFD) systems, such as air-sampling
systems. Heat-detection systems in increasing order of sensitivity are (a) automatic sprinklers at ceiling levels,
and (b) line-type heat detectors located within each tray.
When polyvinyl chloride insulation is involved in a fire, one of the products of combustion is hydrogen chloride
gas, which forms aqueous hydrochloric acid with water vapor in the air. Sensitive relays, instruments, control
apparatus, copper bus bars, and metals such as iron, brass, aluminum, zinc, and alloys that are exposed
to the fumes are subject to corrosion and damage.
In reinforced concrete structures, chlorine contamination forms hygroscopic calcium chloride in the lime of
the cement structure and attacks the reinforcing rods. Concrete may spall months or years after the fire due
to corrosion of the reinforcing rods.

3.2 Electrical Field Tests

3.2.1 Insulation Resistance (IR) and Polarization Index (PI)

IR and PI measurements provide quantitative and relatively repeatable results. The test requires the cable
to be offline, and disconnecting the cable terminations for attachment of the testing apparatus. IR is very
sensitive to temperature and moisture; therefore, in addition to applied voltage, temperature and humidity
at the time of the test must be recorded and the results normalized to a base temperature. IR/PI results and
data trending should be considered along with the results from one or more other cable tests to assess the
condition and rate of degradation of cable insulation.

3.2.2 Offline Partial Discharge

This test detects significant partial discharge sites above a specified detection level, magnitude or severity
of the defect at each partial discharge site, and the location of each of the significant partial discharge sites
within the insulation system. The test requires the cable to be offline, and disconnecting the cable terminations
for attachment of the testing apparatus. Interpretation of the results is complex and best performed by
experienced engineers.

3.2.3 Very Low Frequency (VLF) AC Withstand Voltage Test

This test involves the application of a low-frequency (0.1- 0.01 Hz) AC voltage across the cable insulation
that it must withstand for a specified amount of time without breakdown. The test voltage is typically higher
than operating voltage. VLF tests can be performed for diagnostics. Some diagnostic test methods are listed
below:
• VLF tangent delta measurement (VLF-TD). This test measures the tangent delta of the cable insulation
system at an elevated voltage.
• VLF differential tangent delta measurement (VLF-DTD). This test refers to the tangent delta tip-up
between measurements at 50% and 150% rated voltage.
• VLF tangent delta temporal stability (VLF-TDTS). This test measures the variation of tangent delta with
time at a particular voltage (TDTS), usually over a period of some minutes.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

5-31 Cables and Bus Bars
Page 10 FM Global Property Loss Prevention Data Sheets

The measured values of VLF-TD, VLF-DTD, and temporal stability (VLF-TDTS) are primarily influenced by
the condition (age, contamination, and moisture ingress) of the various cable system components
(accessories, cable insulation, and metallic shield). A high value measured by any of these tests would indicate
deterioration in the condition of the cable.

3.2.4 Withstand Testing

There are two approaches to withstand testing of cables and the accessories that make up the insulation
system:
A. Withstand testing is performed at the working voltage of the system or starting at a lower voltage and
stepping up the voltage until the working voltage is reached. Testing in this manner provides an opportunity
to gather information on the health of the insulation system with less risk of damage. The health of the
insulation system is gauged by performing advanced diagnostic testing and monitoring the dielectric
response during the withstand tests. The data from this type of testing is studied and compared to previous
results then trended to predict the health of the insulation system, and when to schedule maintenance
activities. There is a higher risk that an insulation system evaluated in this manner will fail while in service
causing non-scheduled outages.
B. Withstand testing is performed at voltages higher than the working voltage of the insulation system.
Testing voltages can change based on the age and accessories included in the insulation system, e.g. a
new system could be tested at 2 or 3 times the working voltage, where an aged system could be tested
at 1.5 times the working voltage. This testing is intended to cause the weak sections of the insulation
system to fail so they can be repaired during the scheduled maintenance and testing activities. Repair
staff and replacement components are required. This approach is intended to ensure the cable system is
suitable and reliable to operate at the working voltage, reducing the risk of in service failures.

3.2.5 Damped AC test (DAC)

DAC is used for benchmarking, maintenance testing and diagnostics testing on cable types operating from
5kv to 230kv. DAC is also used for cable operating over 230kv on a provisional basis and per IEEE for study
purposes only.
DAC testing can be performed in such a way to limit damaging effects to a cable and without reducing the
life expectancy of the insulation system. Typically, the DAC test consists of 50 sets of excitations to be applied
to a cable under test, however, this number may change based on the purpose of the test, the historical
data of the cable, and the age of the cable. The test consists of charging the cable to a predetermined voltage
level, generally higher than rated voltage for Benchmarking, however, the voltage level may be lower for
Maintenance and Diagnostic testing and then discharged into an inductance (see Figure 3 below).

3.2.6 Insulation Dissipation Factor (DF)

Insulation dissipation factor is the same as “tan δ”, the rate at which testing voltages discharge through an
inductor. DF will change as an insulation system ages. Determining the condition of an insulation system is
complex and will require comparison of multiple sets of data, including adjacent conductors of the same type
vintage and operating conditions, and historical data for the system. There are no fixed rules for determining
degradation of an insulation system, however, a common practice is to record and observe DF values over
time for trending purposes. Only trained personnel can assess the health of an insulation system after careful
analysis of the data generated while monitoring DF. This assessment can be valuable and influential in
determining the needed for partial maintenance of a cable system.

3.3 Online Condition Monitoring

3.3.1 Infrared Thermography (IR)

IR thermography involves accurately measuring infrared radiation emitted from thermally hot electric
equipment and converting infrared radiation into a visual image or thermogram. These devices can identify
hotspots, which could lead to accelerated degradation of electrical cable systems, even when temperature
differences are small. Additionally, several thermal images can be analyzed over a period of time and the
associated temperature trended with time. IR thermography can identify faulty cables, connectors, splices,

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 11

Fig. 3. Three stages of one DAC excitation; charging, switching, and discharging through inductance

or terminations that are generating excessive heat indicative of a degraded high resistance connection due
dirt, corrosion, or other contamination. Cracked or damaged insulating or jacket materials can also be
identified by these techniques.

3.3.2 Time domain Reflectometry (TDR)

Time-domain reflectometry (TDR) is a measurement technique used to determine the characteristics of
electrical lines by injecting a pulse down one end of the line and observing the reflected wave forms. It is
used to characterize and locate faults in cables. Basic TDR displays location of impedance change, which
could be a fault or water intrusion. Advanced TDRs can display the actual waveform or “signature” of the cable
on a screen, which will show the pulse transmitted down the cable from the instrument and any reflections
that come back to the TDR from discontinuities or impedance variations along the length of the cable. The
signature TDR waveform can be trended with a baseline to evaluate impeding concerns with the cable.

3.3.3 Distributed Temperature Measurement System (DTS)

A DTS is an optical fiber-based temperature sensor incorporated into the power cable, or installed alongside
the power cable. Such a system can monitor the temperature along the power cable. A land-based device
evaluates the optical signals and displays a temperature profile over the cable length.

3.3.4 Strain Monitoring

Strain measurement on Subsea cables can detect in real-time physical actions on the cable such as the
impact of an anchor or fishing gear, or movement in the seabed. Monitoring strain and/or impacts on Subsea
cables, even if the cable has not been severely damaged at the time of impact or seabed movement, can
help to identify if a degradation of the insulation system has begun.

3.4 Subsea Cable Failure Modes

The historical loss incidents involving subsea power cables indicate the following failure modes:
A. Re-exposure of the buried cable due to seabed variation conditions such as sand waves or scouring

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

5-31 Cables and Bus Bars
Page 12 FM Global Property Loss Prevention Data Sheets

1. Sand waves are caused by current and wave action, resulting in the formation of dune systems
on the seabed that slowly travel across the seabed. They can assume heights from only a few inches
(centimeters) to several feet (meters). A cable buried in a sand wave area is expected to be exposed
and reburied at various sections along the cable route as time passes. If the dimensions of the sand
waves are sufficiently large, extensive spanning can also result. Local or global scouring can also result
in the exposure of cables, and, in case of very localized scouring, possible spanning of the cable.
Exposing the cable will render it vulnerable to other hazards such as shipping anchors or dropped
objects. Spanning introduces current-induced vibrations and additional tension in the cable. The added
tension in combination with vibration-induced fatiguing can result in breakage of the cable.
B. External threats due to fishing activities, ship anchors, dropped objects, or dredge activities
1. Based on a CIGRE survey in 2009, 85% of subsea cable failures were due to external threats. Almost
50% of damage was known to be caused by anchors.
2. Fishing equipment can typically penetrate the seabed up to 2 ft (0.6 m), and in some cases up to
5 ft (1.5 m) depending on its type, which can damage any cable exposed on the seabed. Ship anchors
can be dropped onto a cable due to mishaps or emergency anchoring. Depending on the ship size,
the penetration depth of ship anchors can be up to 16 ft (4.9 m) in a muddy seabed and up to 8 ft (2.4
m) in a sandy sea bed. The cable can be easily damaged when hit by ship anchors.
3. Dropped objects are mainly due to mishaps to an offshore installation where heavy lifting or other
marine operations take place. If these objects are heavy enough, the impact when they hit the seabed
can result in some penetration into the seabed. If a cable is present at such location, considerable
damage to the cable can result.
C. Insulation failure due to aging, fatigue cracks of lead sheath, water tree formation, partial discharge
activity, and electrical trees
1. Water trees are tree-like growths consisting of water-filled micro voids that can grow in most extruded
insulations when they are exposed to moisture and an electric field. The water trees that grow from
water-filled voids in the insulation are called bow-tie trees. Those trees initiated at interfaces are known
as vented or streamer trees. When water trees grow such that they bridge a considerable portion of
the insulation, and become possible sites for inception of partial discharge (PD). Overvoltages produced
by switching transients or lightning surges may increase electrical stress in a water tree to a level that
will cause initiation of PD.
2. Partial discharges (PD) are the local breakdowns of the gas in voids, cuts, cracks, fillers, and
contaminants, and delaminations at interfaces. Prolonged PD activity will cause pitting of the surfaces
first, and then concentrate in the pits to form electrical trees, which complete the failure.
3. An electrical tree is a network of fine gas-filled channels that propagate relatively quickly through
the insulation to cause failure by the action of PD. Electrical trees can be detected by PD
measurements; trees have characteristic PD patterns that vary with time.
D. Failure of joints due to inadequate design, poor joint assembly work or adverse weather conditions
during assembly
E. Movement of the sea bottom due to subsea landslides or earthquakes
F. Improper installation or handling of the cable
G. Missed features or inaccurate seabed characterization due to incomplete analysis of existing data or
where an existing sensor was not operating correctly.
H. Improper design of J-tubes
1. J-tubes should be carefully designed to account for the characteristics of the cable, including bending
radius, sufficient space for heat absorption, ensuring the that the inside path up the J-tube is free from
snags, and that sufficient planning is done to ensure the appropriate power in the winches.

3.5 Subsea Cable Protection Against External Threats

Subsea cable failures due to external threats, particularly due to fishing activities or ship anchors, are the
most common cause of subsea power cable failure.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 13

The best protection against external threat is to bury the subsea cable below the depth which the threat can
reach. Only a small depth (3.3 to 6.6 ft [1 to 2 m]) is required for the more frequent hazards caused by fishing
activities. However, 3.3 ft to 6.6 ft (1 to 2 m) burial depth is not a truly effective measure against anchors
dropped from large ships, as modern anchors can dig deep into the seabed. Selection of burial depth is a
complicated task and needs to be based on the following major factors:
A. From a cable design viewpoint, the burial depth of at least 3.3 ft (1 m) (6.6 ft [2 m] for HVDC cable)
is based on several requirements, but mainly from heat dissipation. Beyond this value, increased burial is
not recommended because the current capacity continues to degrade with depth.
B. From a cable installation viewpoint, the burial depth of 3.3 to 6.6 ft (1 to 2 m) is customary and achievable
with reasonably sized equipment without significant bottom disturbance, except in special circumstances.
C. From the navigation risk viewpoint, the industry developed the concept of a burial protection index
(BPI). This BPI recognizes that different seabed soils react differently to the penetration of fishing gear
and anchors (see Figure 4).

Fig. 4. Illustration of burial protection index

BPI = 1 is considered suitable for water depths greater than 330 ft (100 m) where anchoring of ships is unlikely
or where shipping and anchoring are prohibited.
BPI = 2 would provide protection from vessels with anchors up to about 4400 lb (2,000 kg). This may be
adequate for normal fishing activities, and small merchant ships, but would not be suitable for larger ships’
anchors.
BPI = 3 would be sufficient to protect from anchors of all but the largest ships. It would be suitable for
anchorages, at the entrances of harbors where ships have been known to accidentally deploy an anchor,
and heavily trafficked shipping channels, with adjustments made to suit known ship anchor sizes.

3.6 Repairing Subsea Cables

Repairing subsea cables can be very challenging. There are few repair vessels for subsea power cables in
the world. Those may already be under contract with utilities to standby in specified locations, and as a result
the ability to obtain such a vessel may involve significant delays. (see Figure 5).

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

5-31 Cables and Bus Bars
Page 14 FM Global Property Loss Prevention Data Sheets

Fig. 5. Cable ships by market

Thus, it is a good strategy to have an equipment contingency plan to repair the subsea cable so that a vessel
is available when needed. It is generally easy to find a vessel that can unbury the cable, and to re-bury it
after repair, but the repairing itself may take a special arrangement that should be worked out ahead of time.
Since almost all the subsea cables are non-standard products, it is important to have equipment breakdown
spare cable as part of the equipment contingency plan of the same type manufactured with the original cable
and stored for potential repairs.

3.7 Loss History

An analysis was conducted on grouped cable losses over a recent 10-year period. The causes of fire loss,
in terms of frequency, are shown in Figure 6. The results of a similar analysis of electrical breakdown losses
are included in Figure 7.

3.8 Illustrative Losses

3.8.1 Lack of Adequate Cable Fire Stopping

The main circuit breaker for a 4 kV bus failed “violently” as the ID fan speed was changed. Operators were
switching ID fans from slow to fast when they heard the circuit breaker fail. This ignited a fire in the switchgear
cabinet, connected bus, and nearby power and control wiring. The fire spread through unprotected cable
openings into the computer room above containing boiler logic controls. The circuit breaker was a 2,000
amp air blast type that controlled one of the ID fans for a coal fired boiler powering a turbine generator. This
is one of three turbine generators at an electric utility generating station. Also damaged were five other
breakers, 15-20 cable trays, and a steel truss and section of metal pan floor. No cause was given for the
electrical failure in the switchgear. The public fire department was not called until 15 minutes after the fire was
discovered.

3.8.2 Oil-Filled Cable Fire at a Hydro-Electric Plant

A phase-to-ground fault occurred on an oil-filled cable of a turbine-generator unit. The cable, located in the
upper portion of a cable shaft, ignited the oil-impregnated layers and adjacent cables in the 860 ft (260 m)
vertical shaft. Two sets of three, 245-kV oil-filled cables and auxiliary equipment in and around the cable shaft

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 15

Fig. 6. Fire causes by frequency

Fig. 7. Electrical breakdown by frequency

were damaged. Intense heat damaged all cables in the shaft. Even though the entire 860 ft (260 m). length
of cabling was not fire damaged, these cables could not be spliced, so the entire length of the six cables
had to be replaced.
This incident resulted in several units being taken offline; however, within hours, only two turbine-generator
units were inoperable. One of the units was restarted about four months later, with the last unit expected
to be restarted seven months after that.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

5-31 Cables and Bus Bars
Page 16 FM Global Property Loss Prevention Data Sheets

3.8.3 Ground Fault Shuts Down Paper Mill

A paper mill was totally shut down by a sequence of events after an electrical ground fault took place at a
cable connection in a disconnect switch cabinet. The fault was apparently due to insulation failure at a stress
cone (a cone shaped insulation applied over cable insulation at the connection point). The cable involved
was fed from the main bus No. 1, which supplied power to the steam plant at the mill. The ground fault relay
for the circuit breaker closest to the fault did not operate. Instead a breaker further upstream operated. This
breaker interrupted utility power for bus No. 1. Fans for all but one of the plant boilers were shut down. A
significant additional factor was that backup battery power supply for the boiler control room panel went dead.
Operators were unable to see the panel and could not keep the remaining boiler on line. Loss of steam shut
down the plant.

3.8.4 Cable Fire in a Paper Mill

An integrated paper mill lost all power due to a fire involving cables. The fire started in a pipe tunnel, when
wood planking from scaffolding left over from maintenance was ignited by an uninsulated steam line. The
pipe tunnel had large steam lines as well as power and communication cables running through it to other parts
of the mill. The fire spread to the cable tray below the wood planking and through the pipe tunnel to an
electrical control room.
The original smoldering fire was found by plant personnel and reportedly extinguished, without notifying the
mill fire brigade or mill management. A few hours after initial notification, operators began receiving numerous
process trips and alarms. It was determined that a cable fire had continued in the electrical control room.
After notification of the fire brigade and mill management, the operators manually initiated the Halon fire
protection system as the automatic system had been improperly impaired. The electrical systems to the
affected area were sequentially disconnected resulting in a full black plant condition (entire mill was shutdown
with no power).
Damage resulted in approximately 3 miles (5 km) of cables being replaced, as splicing was not considered
cost effective due to the inaccessibility of the cables within the cable tunnel. Minimal heat damage was done
to the mill’s buildings. Some minor damage to other equipment occurred due to the immediate shutdown of
mill power. The three paper machines were shut down for over 2 days, and the sale of power from the utility
was interrupted. Improper emergency communication, inadequate scaffold maintenance protocols, lack of
adequate impairment handling for the automatic Halon system, and failure to fire stop wall and floor openings
were all negative factors that increased the loss.

3.8.5 Cable Fire in Super Computer Manufacturing, Product Testing Lab

A fire occurred in electrical cabling under a raised floor in the product testing laboratory of a 25,000 ft2 (2,300
m2) super-computer manufacturing facility. The facility engaged in the assembly and testing of cluster based
super-computing systems primarily for academic, governmental and commercial uses. Functional testing
of all computer systems manufactured occurred in an approximately 4500 ft2 (420 m2) specially designed
testing laboratory in a separated area within the facility.
The day of the event, a smoke alarm in the testing laboratory area of the facility went off notifying plant security
officials. The smoke alarm tripped the HVAC units within the test lab. Plant officials arrived to find light smoke
and a strong smoke odor in the test lab, however, most computer systems under test continued to operate.
Plant personnel also restarted the room HVAC systems.
Further investigation revealed fire damage to electrical cabling and a power distribution unit located under
the raised tile floor near the middle of the room. Two 30 amp fuses within the power distribution unit were found
to be fused. The damaged cabling and the power distribution unit were subsequently replaced. Thermal
damage was limited to the electrical cabling itself however, the resulting smoke spread throughout the testing
laboratory. This caused damage to five computer systems under test at the time of loss, as well as various
test equipment, electrical distribution equipment and the buildings walls, ceilings, floors, etc.
Repair/replacement of the computer systems took several months to complete and caused a delay in product
delivery to customers. Additionally, the inability to perform product testing while building repair and clean-up
activities were underway delayed other product deliveries. This location was sprinklered, however, no
sprinklers operated and no impairment existed at the time of the fire.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 17

4.0 REFERENCES

4.1 FM Global
Data Sheet 2-0, Installation Guidelines for Automatic Sprinklers
Data Sheet 4-1N, Fixed Water Spray System for Fire Protection
Data Sheet 4-9, Halocarbon and Inert Gas (Clean Agent) Fire Extinguishing Systems
Data Sheet 5-20, Electrical Testing
Data Sheet 9-0, Asset Integrity

4.2 Other
American National Standards Institute (ANSI). ANSI/IEEE 242, Protection and Coordination of Industrial and
Commercial Power Systems. ASTM International.
ASTM E-84, Standard Test Method for Surface Burning Characteristics of Building Materials.
CIGRE. Report 680, Implementation of long AC HV and EHV cable systems.
Institute of Electrical and Electronics Engineers, Inc. (IEEE). IEEE 1202, Standard for Testing Flame-
Propagation and Smoke Generation of Cable or Splices/Connectors.
Insulated Cable Engineers Association (ICEA). ICEA T-29-520, Conducting Vertical Cable Tray Flame Tests
with Theoretical Heat Input Rate of 210,000 B.T.U./Hour.
National Fire Protection Association (NFPA). NFPA 70/ANSI C1, National Electrical Code (NEC).
National Fire Protection Association (NFPA). NFPA 262, Standard Method of Test for Flame Travel and Smoke
of Wires and Cables for Use in Air Handling Spaces.

APPENDIX A GLOSSARY OF TERMS

FM Approved: Products and services that have satisfied the criteria for Approval by FM Approvals. Refer
to the Approval Guide for a complete listing of products and services that are FM Approved.
FM Specification Tested: Products that have been evaluated by FM Approvals according to recognized
standards and are subject to examinations and followup by FM Approvals. The cables listed are classified
as FM Approvals GP-1 based on test results to determine fire propagation and a corresponding Fire
Propagation Index. FM Approvals GP-1 is a non-self-sustained flame propagation having a Fire Propagation
Index less than l0. The classification may be used as an aid in determining fire protection techniques
applicable to given field conditions.
Grouped cable: cable trays or cable bundles installed adjacent to each other. Cable tray systems normally
consist of two or more trays either stacked or adjacent to each other. Group cable may also include cable
bundles installed without confinement of a tray or tray system.
J-tube: A tube used in offshore installations to enable subsea cable to pulled up from the seabed to an
offshore platform, including a wind turbine support structure or platform.
Non-propagating cable:
A. Cable with a fire propagation index (FPI) of less than 10 when tested in the FM Global Fire Propagation
Apparatus.
B. Cable that, when tested in accordance with the FM 3972 Vertical Tray Test, does not have flame spread
of more than 5 ft (1.5 m) beyond the 60 kW fire exposure. See FM Specification Tested.
C. Cable that has passed the NFPA 262 test.
Propagating cable:
A. Cable that has not been tested by the FM Global Fire Propagation Apparatus
B. Cable that has been tested in the FM Global Fire Propagation Apparatus and has an FPI of 10 or more.
Plenum: A compartment or chamber to which one or more ducts are connected and that forms part of the
air distribution system.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

5-31 Cables and Bus Bars
Page 18 FM Global Property Loss Prevention Data Sheets

Re-exposure: Exposure of a buried cable. Scour and sand wave can cause the seabed to change and expose
a buried cable.
Sand wave: Movement of seabed sediments due to wave action and/or water currents.
Scour: Removal or movement of seabed or lakebed soils by currents or waves caused by structural or other
elements interrupting the natural flow regimen above the sea/lake floor.

APPENDIX B DOCUMENT REVISION HISTORY

The purpose of this appendix is to capture the changes that were made to this document each time it was
published. Please note that section numbers refer specifically to those in the version published on the date
shown (i.e., the section numbers are not always the same from version to version).
January 2021. Interim revision. Updated contingency planning and sparing guidance.
January 2020. The following changes were made:
A. Broadened the scope to include guidance for higher voltage cables and bus bars.
B. Expanded recommendations for construction and location.
C. Added guidance on when to provide automatic fire protection for grouped cables and cable trays.
D. Updated the reference information section.
E. Added recent illustrative losses and updated loss history.
F. Improved description of FM Global Group 1 cable, and made terminology consistent throughout the
document.
December 2004. The fire protection section of this standard has been revised with the following changes
made:
1. The fire protection section has been reformatted. The loss experience section has been removed as this
is covered by an Understanding the Hazard (UTH) publication Grouped Cable (P0218).
2. Cable that has been plenum rated in accordance with UL-910, or cable that has a maximum flame spread
distance of 5 ft (1.5 m) or less when tested in accordance with NFPA 262, Standard Method of Test for Flame
Travel and Smoke of Wires and Cables for Use in Air Handling Spaces, is considered equivalent to FM
Approved Group 1 (nonpropagating) cable.
3. The FM Global Fire Propagation test is compared to other cable test standards.
4. Fire protection recommendations for cable above suspended ceilings is included.
January 2001. The recommendation for smoke detection for electrical rooms was revised to provide
consistency within 5-series data sheets.
September 2000. This revision of the document was reorganized to provide a consistent format.

APPENDIX C SUPPLEMENTARY INFORMATION

C.1 Insulation Failure

Insulation failure is due to many different causes, the most common of which include mechanical damage,
excessive temperatures, corona, ozone, overvoltage, chemical environment, and damage from rodents and
termites.
Conductors supplying power to motors and other electrical apparatus are especially subject to mechanical
abuse and are exposed to many deteriorating influences, such as vibration, moisture, heat, oil, corrosive
liquids and gases, and miscellaneous solvents.
Poor housekeeping is a common contributing factor to insulation failure. Oil soaked insulation breaks down
easily, ground faults or short circuits result, and a severe fire may follow. Cable trays should be kept free
of foreign materials that can build up and cause overheating and failure of the conductor insulation.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 19

Combustible materials such as paper dust, dried paper pulp, wood chips, oily rags, lunch papers, and similar
materials should not be allowed to accumulate in the trays; these materials ignite easily, sometimes
spontaneously, and aid in the propagation of flame.

C.2 Cable Flammability Tests

C.2.1 General
Full-scale fire tests are the most reliable method of determining whether a cable propagates fire. Due to
the large numbers of cable and cost of testing, small scale tests have been developed to predict cable
flammability. These tests attempt to duplicate, under small scale test conditions, what is likely to happen under
large scale conditions. Examples are NFPA 262, CSA-FT4, ICEA T-29-520, UL 1581, IEEE 1202, and IEC
332-3. The cable is typically arranged in a vertical configuration and exposed to a flame. The severity of the
test depends on the number of cables and the energy of the ignition source. The quantity of cable could
range from one cable to a cable bundle; the ignition source could range from a Bunsen burner to a ribbon
burner rated at 300,000 Btu/hr (87.9 kW).

C.2.2 FM Specification Tested Cable

FM Approvals tests cable using the ASTM E-2058 Fire Propagation apparatus (see Figure 8), which
determines two basic flammability properties of cable insulation: propensity for ignition and heat release rate.
The cable sample is exposed to heat flux exposures between 10 and 65 kW/m2, and time to ignition is
measured. The square root of the inverse of the time to ignition is plotted against heat flux exposures. Thermal
Response Parameter (TRP) is the slope of the line.
The heat release rate test is then performed under simulated large-scale fire exposures. A 24 in. (61 cm)
long sample is placed vertically in the quartz tube. The bottom 5 in. (12.7 cm) of the cable is exposed to 50
kW/m2. A gas concentration of 40% oxygen is used to simulate a large scale exposure fire. The heat release
rate is measured by thermocouples and gas concentration is monitored in the exhaust duct.
The TRP and the chemical heat release rate are combined to produce the Fire Propagation Index (FPI) which
is used to classify the cable as non-propagating or self-propagating. Cable with an FPI of less than 10 is
considered a non-propagating cable and is listed in the Approval Guide, an online resource of FM Approvals.
Additional information on this testing is provided in FM Approval Class Number 3972, Test Standard for Cable
Fire Propagation.
To validate the test, fifteen full scale cable tests were conducted at the FM Global Research Campus using
the FM Global Fire Products Collector. Two 16 ft (4.9 m) long and 2 ft (0.61 m) wide cable trays were arranged
vertically 1 ft (0.3 m) apart.
A single layer of cable with no air space between was arranged in the trays. A 210,000 Btu/hr (61 kW) propane
air burner with a flame height of 2 ft (0.6 m) was used between the trays as the exposure fire. By using two
trays in a parallel tray arrangement, as described above, flame radiation is increased by 50%. There is
self-sustained fire propagation beyond the heat flux zone of the propane burner.
The fire products generated during combustion were collected in the hood and exhaust duct, and
measurements of flammability properties were made. As predicted using the large-scale Fire Propagation
apparatus, fire spread was limited in non-propagating cable (i.e., cables with an FPI of less than 10 when
tested in the ASTM E-2058 apparatus). The flame extinguished when the burner was turned off after a
20-minute exposure. Conversely, self-propagating cables resulted in flame spread the full length of the tray.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

5-31 Cables and Bus Bars
Page 20 FM Global Property Loss Prevention Data Sheets

Fig. 8. ASTM E-2058 Fire propagation apparatus.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 21

In the FM Approvals Class Number 3972 test, the fire exposure is a 205,000 Btu/hr (60 kW) propane diffusion
flame from a 2 x 1 ft (0.6 x 0.3 m) burner between parallel, facing cable trays. This flame is approximately
2 ft (0.6 m) high and produces a maximum surface heat flux of 12,700 to 14,300 Btu/hr ft2 (40 to 45 kW/m2)
1 ft (0.3 m) above the burner.
The nature of the diffusion flame from the sand burner and the facing, parallel arrangement of the simulated
cable trays is also critical in creating a representative fire exposure. This arrangement results in a large
fraction of the flame energy transferred to the cables in the same way energy would be transferred during
an expected fire. For this reason, the FM 3972 prototype is a more realistic method for determining cable
flammability than other cable flammability test methods.

Table 3. Comparison of FM Large-Scale Cable Flammability Test with Other Flammability Tests
Test Designation Exposure Heat Exposure Fire Type
Release Rate
Btu/hr (kW)
UL 1666 Riser Cable 530,000 Propane Diffusion Flame
(155)
UL 910 Plenum Cable 300,000 Methane Diffusion Flame
(88)
FM 3972 Vertical Cable Tray 205,000 Propane Diffusion Flame
(60)
Insulated Cable Engineers Association (ICEA) T-29-520 205,000 Propane Flame from a rich
Vertical Cable Tray (60) air-gas mixture
IEC 332-3, Category AF, greater than 35-mm2 cable cross 137,000 Propane flame from a rich
section, Vertical Cable Tray (40) air-gas mixture
IEEE 1202 (nearly identical to UL 1581, CSA C22.2, UL 70,000 Propane flame from a rich
1685, ASTM D 5424) Vertical Cable Tray (20) air-gas mixture
IEEE 383 72,000 Propane flame from a rich
Vertical Cable Tray (21) air-gas mixture
IEC 332-3 (all categories except AF, > 35-mm2) 70,000 Propane flame from a rich
Vertical Cable Tray (20) air-gas mixture

Table 3 provides a comparison of the types and intensities of the exposure fires used in various U.S. and
IEC cable standards. The most widely used source strength for testing the fire performance of cables is 70,000
Btu/hr (20 kW). In most cases, the burner includes a venturi mixer to draw in air at a rate of about 5 times
the fuel flow for the flame. Although the resulting flame is far from premixed, this significant addition of air is
likely to eliminate flame luminosity, reducing flame heat transfer from the burner to levels well below the
critical flux for ignition of even commodity polymers (50,000 to 70,000 Btu/hr [15 to 20 kW/m2]), let alone
highly halogenated cable insulation.
The 70,000 Btu/hr (20 kW) fire exposure is not considered sufficient to validate a cable as non-propagating.
Additionally, while some of these test protocols appear to have comparable fire exposures to the FM 3972
test (e.g., IEC 332-3), the exposure fire type is also critical. The IEC 332-3 test, for example, is deceptive since
little of the energy from the partially-premixed burner flame is transferred to the cable surface due to the
small flame area and low flame emissivity.
The ICEA T-29-520 Guide for Conducting Vertical Cable Tray Flame Tests also uses a fire type that does
not create a realistic fire exposure. While this test uses a 205,000 Btu/hr (61.5 kW) source, it consists of a
10 in. (25 cm) wide ribbon burner supplied with air to propane in a volume ratio of 1.25:1. Even this small
amount of air is likely to reduce flame luminosity, and hence flame heat transfer. Furthermore, there is only
one ladder type cable tray used in this test, with no confinement of heat from the horizontal source flame
whatsoever, so that heat can escape to the ambient environment.
The horizontal ribbon burner aggravates this heat loss by aiming the source flame toward the open area
behind the cables.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

5-31 Cables and Bus Bars
Page 22 FM Global Property Loss Prevention Data Sheets

C.3 Subsea Cable

C.3.1 Figure 9 shows a typical structure of HVAC subsea export cable. Subsea cables generally use extruded
insulation such as XLPE (cross linked polyethylene) consisting of 3 layers. The first around the conductor
transforms the cable into a smooth surface using a semi-conductive layer and thus produces a stress-reducing
layer in case of bending of the cable. A second layer provides an insulation layer, and then a third layer
provides an insulation screen.
Armor layer is used for protecting against faults due to abrasion and from vessel damage. More armor will
provide more protection from external damage. However, more armor will result in a heavier cable, which will
be more difficult to place, restricting the number of vessels available and making repairs more problematic.
One layer of armor is common for subsea cables.

1- Copper conductor
2- Inner semiconductive layer
3- XLPE insulation
1 4- Outer semiconductive layer
2 5- Swelling tape
3 6- Lead sheath
4 7- Outer sheath
5 8- Filler
9- Fibre optical element
6 10- Bedding layer
11- Armor
7 12- Outlet serving

10

11

12

Fig. 9. Typical HVAC subsea power cable: 3 phases bundled with fiber-optic elements

C.3.2 Main accessory items for subsea cable are described below:
A. Armor hang-off device (see Figure 10)
B. Armor anchoring device: used to prevent any possible cable movement at the sea-land transition
location
C. Sea-land power transition joint
An armor hang-off device is used to lock a cable on top of a J-tube at off-shore platforms.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.

Cables and Bus Bars 5-31
FM Global Property Loss Prevention Data Sheets Page 23

Fig. 10. Accessory armor hang-off

C.4 Online PD Measurement

Partial discharge activity can be monitored on a continuous basis during the operating life of the equipment.
Sensors attached to key points in the cable are required for this system. The terminations and joints of the
cable system are typical PD measurement points. Online PD also helps to detect weak points in the
insulation system while the cable is operating. Alternative to sensors, handheld devices can also be used
to locate PD. Some acoustic devices (sniffers) make it possible to observe corona discharges along the
surface of the insulator.

©2020-2021 Factory Mutual Insurance Company. All rights reserved.