FMDS0504
FMDS0504
TRANSFORMERS
Table of Contents
Page
©2019-2022 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.
5-4 Transformers
Page 2 FM Global Property Loss Prevention Data Sheets
List of Figures
Fig. 1A.Typical protective scheme for a two-winding, delta-wye connected transformer up to 10,000 kVA
in size using fuses on the primary of the transformers ..................................................................... 5
Fig. 1B. Typical protective scheme for a two-winding, delta-wye connected transformer up to
10,000 kVA in size using a circuit breaker on the primary of the transformers ............................... 6
Fig. 1C. Typical protective scheme for a two-winding, delta-wye connected transformer greater than
10,000 kVA in size .......................................................................................................................... 7
Fig. 1D. Typical protective scheme for transformers greater than 10,000 kVA in size with a secondary
selective system .............................................................................................................................. 8
Fig. 1E. Alternative protective devices that achieve the same level of protection as those shown in
Figures 1A, 1B, and 1C .................................................................................................................... 9
Fig. 2. Locations for transformer buildings and rooms ................................................................................ 20
Fig. 3. Minimum horizontal separation distance between outdoor liquid-insulated transformers and
exposed walls of main buildings ....................................................................................................... 23
Fig. 4. Use of 2-hour fire barriers and separation distances for protection of exposed main building walls . 24
Fig. 5. Determination of exposed roof area ............................................................................................. 25
Fig. 6. Fire barriers for multiple outdoor transformers .............................................................................. 26
Fig. 7. Nozzle and piping arrangement for transformer using nozzles of ordinary capacity ...................... 27
Fig. 8. Nozzle and piping arrangement for transformer using large-capacity nozzles ............................... 28
List of Tables
Table 1. Recommended Additional Protection for Specialty Transformers .................................................. 11
Table 2. Routine Offline (De-Energized) Tests ............................................................................................. 14
Table 3. Focused Off-Line Tests ................................................................................................................... 15
Table 4. Recommended Construction for Transformer Buildings and Rooms ............................................. 20
Table 5. Separation for Exposure Protection of Main Building Walls (also refer to Figure 3) .................... 23
Table 6. Separation and Extent of 2-hour Fire Barriers for Protection of Main Building Walls
(for dimensions refer to Figure 4) ................................................................................................. 24
Table 7. Transformer Fire Exposure to Noncombustible Building Roof Where Separation from Wall
is Based on a 3-hour Fire Barrier or Water Spray Protection ........................................................ 25
Table 8. Minimum Separation Distances Between Adjacent Transformers .................................................. 26
Table 9. General Criteria for Evaluating Thermographic Scans ................................................................... 31
Table 10A. Tests Typically Included in a Mineral Oil Screen Test ................................................................ 31
Table 10B. Tests Typically Included in an In-Service Natural Ester Oil Screen Test ................................... 31
Table 11. Gases Included in a Typical DGA Report ..................................................................................... 32
Table 12. Gases Represented by Chemical Formulas ................................................................................. 33
Table 13. 90th Percentile Gas Concentration Values for Transformers Filled with Mineral Oil .................. 33
Table 14. 95th Percentile Values of Gas Generation Rate for Transformers Filled with Mineral Oil ......... 34
Table 15. Typical Gas Values and Generation Rate for Transformers Filled with Silicon Oil with no
previous history ............................................................................................................................. 34
Table 16. Typical Dissolved Gas Concentration Value for Transformers with Ester Oil with no previous
sample history ............................................................................................................................... 34
Table 17. General Rules for Interpreting Power Factor Results .................................................................. 36
Table 18. Power Factor Limits for Transformers .......................................................................................... 36
Table 19. Interpretation of the Changes in Current Between Tests ............................................................. 37
1.0 SCOPE
This data sheet contains loss prevention recommendations related to the fire protection of ALL types of
transformers. It also contains loss prevention recommendations related to the electrical protection, electrical
testing, maintenance, and operation of large, critical distribution, power, and specialty transformers ONLY.
For loss prevention recommendations related to electrical protection, electrical testing, maintenance, and
operation of smaller and less-critical transformers, see Data Sheet 5-20, Electrical Testing.
See Data Sheet 5-20 for distribution transformers 5 MVA and below. These transformers step down the utility
voltage for use in a commercial or industrial property.
Specialty transformers include network transformers, rectifier transformers, arc furnace transformers,
induction furnace transformers, generator step-up transformers, converter transformers, smoothing reactors,
series reactors, and shunt reactors.
1.1 Changes
July 2022. Interim Revision. Added guidance for the use of FM Approved dissolved hydrogen on-line
monitoring system. Minor editorial changes were also made.
1.2 Hazards
Power and distribution transformers are reliable devices with low failure rates. Adequate electrical protection,
electrical testing, maintenance, and proper operation are key factors in reducing the risk of transformer
explosions and fires.
FM Global’s loss statistics show that it is about 10 times more likely for a transformer to fail without a resulting
explosion and fire. However, in some cases an internal fault may result in an overpressure of the transformer
sufficient to cause tank rupture. Tank rupture will release substantial quantities of oil, with a resulting oil mist
explosion and a long-burning, high heat release, oil pool fire. Fire protection and proper construction features
are critical to limiting the extent of damage to adjacent structures and equipment in the event of a transformer
fire.
Liquid
Device
Description Dry (MVA)
No.
1–5 5–10
26 Liquid thermal I I
device
49 Winding thermal I I
device
51N1 Time delay T T T
overcurrent relay.
(Ground fault
protection of
transformer wye
winding and
through faults.)
51TL Time delay T T T
overcurrent relay.
(Transformer
overload protection)
63 Sudden pressure T
relay
71 Liquid Level Device I I
Notes:
A—Alarm, I—Indication, T—Trip
1. Devices 50G, 50N/51N, 50NY/51NY, 67N and 87TN are
alternatives. (See Fig 1E.)
Fig. 1A.Typical protective scheme for a two-winding, delta-wye connected transformer up to 10,000 kVA in size using fuses
on the primary of the transformers
Liquid
Device
Description Dry (MVA)
No.
1–5 5–10
26 Liquid thermal I I
device.
49 Winding thermal I I
device.
50ND2 Instantaneous T T T
ground overcurrent
relay. (High
magnitude ground
fault protection of
transformer delta
winding and leads.)
50TF Instantaneous T T T
overcurrent relay.
(High magnitude
transformer internal
phase fault
protection.)
51N1 Time delay T T T
overcurrent relay.
(Ground fault
protection of
transformer wye
winding and
through faults.)
51ND2 Time delay ground T T T
overcurrent relay.
(Ground fault
protection of
transformer delta
winding and leads.)
51TF Time delay T T T
overcurrent relay.
(Phase through
fault protection.)
51TL Time delay T T T
overcurrent relay.
(Transformer
overload protection)
63 Sudden pressure T
relay
71 Liquid level device I I
Notes:
A—Alarm, I—Indication, T—Trip
1. Devices 50G, 50N/51N, 50NY/51NY, 67N and 87TN are alternatives. (See Fig 1E.)
2. Device 50GD is an alternative to 50ND/51ND. (See Fig 1E.)
Fig. 1B. Typical protective scheme for a two-winding, delta-wye connected transformer up to 10,000 kVA in size using a
circuit breaker on the primary of the transformers
Device
Description Action
No.
24 Volts/hertz relay (For unit T
connected transformers only)
26 Liquid thermal device A
49 Winding thermal device A
50ND2 Instantaneous ground T
overcurrent relay. (High
magnitude ground fault
protection of transformer delta
winding and leads.)
50TF Instantaneous overcurrent relay. T
(High magnitude transformer
internal phase fault protection)
51N1 Time delay overcurrent relay. T
(Ground fault protection of
transformer wye winding and
through faults.)
51ND2 Time delay ground overcurrent T
relay. (Ground fault protection
of transformer delta winding
and leads.)
51TF Time delay overcurrent relay. T
(Phase through fault
protection.)
51TL Time delay overcurrent relay. T
(Transformer overload
protection)
63 Sudden pressure relay T
71 Liquid Level Device I
87T Transformer differential relay T
(Fault protection for transformer
internal faults)
87TN3 Transformer ground differential T
relay (Ground fault protection of
transformer wye winding
internal faults)
Notes:
A—Alarm, I—Indication, T—Trip
1. Devices 50G, 50N/51N, and 50NY/51NY are alternatives. (See Fig 1E.)
2. Device 50GD is an alternative to 50ND/51ND. (See Fig 1E.)
3. Device 67N is an alternative to 87TN. (See Fig 1E.)
Fig. 1C. Typical protective scheme for a two-winding, delta-wye connected transformer greater than 10,000 kVA in size
Fig. 1D. Typical protective scheme for transformers greater than 10,000 kVA in size with a secondary selective system
Device
Description
No.
24 Volts/hertz relay (For unit connected
transformers only)
26 Liquid thermal device
49 Winding thermal device
50G1 Zero sequence instantaneous ground
overcurrent relay. (Ground fault
protection of wye winding and through
faults)
50GD2 Zero sequence instantaneous ground
overcurrent relay. (Ground fault
protection of transformer delta winding.)
50N Instantaneous ground overcurrent relay.
(High magnitude ground fault protection
of wye winding and through faults.)
50ND Instantaneous ground overcurrent relay.
(High magnitude ground fault protection
of transformer delta winding and leads.)
50NY Instantaneous ground overcurrent relay.
(High magnitude ground fault protection
of wye winding and through faults.)
50TF Instantaneous overcurrent relay. (High
magnitude transformer internal phase
fault protection)
51N Time delay overcurrent relay. (Ground
fault protection of transformer wye
winding and through faults.)
51ND Time delay ground overcurrent relay.
(Ground fault protection of transformer
delta winding and leads.)
51NY Time delay overcurrent relay. (Ground
fault protection of transformer wye
winding and through faults.)
51TF Time delay overcurrent relay. (Phase
through fault protection.)
51TL Time delay overcurrent relay.
(Transformer overload protection)
63 Sudden pressure relay
67N3 AC Directional neutral overcurrent relay.
(Ground fault protection of transformer
wye winding internal faults)
71 Liquid Level Device
87T Transformer differential relay (Fault
protection for transformer internal faults)
87TN Transformer ground differential relay
(Ground fault protection of transformer
wye winding internal faults)
Notes:
1. Devices 50G, 50N/51N, 50NY/51NY are alternatives.
2. Device 50GD is an alternative to 50ND/51ND.
3. Device 67N is an alternative to 87TN.
Fig. 1E. Alternative protective devices that achieve the same level of protection as those shown in Figures 1A, 1B, and
1C
2.1.1.2 Provide overvoltage transient protection for all transformers in accordance with Data Sheet 5-11,
Lightning and Surge Protection for Electrical Systems.
2.1.1.3 Provide differential protection to cover all the windings on multiple-winding power transformers,
including specialty transformers such as rectifier, induction, and arc furnace transformers when applicable.
Where the high magnetic fields of specialty transformers prohibit the use of conventional iron core current
transformers, use Rogowski coil current sensors to implement differential protection.
2.1.1.4 Provide overcurrent and ground fault protection for the tertiary winding of three-winding transformers
and autotransformers.
2.1.1.5 Provide an arc-monitoring system to detect arcing faults in transformer vaults with exposed energized
components. Provide this additional protection when the ground fault relay cannot be set low enough to detect
ground fault current due to neutral imbalance current flow. Table 1 contains a list of specialty transformers
and the additional protection recommended for them.
• Neutralization number
• Inhibitor content (only for inhibited oils)
2. For natural ester oil, include the following tests at a minimum:
• Color
• Dielectric breakdown strength
• Moisture
• Power factor
• Fire point
• Viscosity
2.1.3.2.3 Perform corrosive sulfur testing of transformers that were built or have had their oil processed or
replaced in 2000 or later, and if the oil has not been tested for corrosive sulfur. In particular, perform corrosive
sulfur testing of transformers that meet the following conditions:
• The transformer fluid is a mineral oil.
• The mineral oil is an uninhibited type.
• The oil preservation system is sealed (not free breathing).
• The transformer has a high sustained load factor.
• The transformer is operating in a hot climate.
• The transformer windings are bare copper.
Transformers that meet these conditions are at the highest risk of failure due to corrosive sulfur contamination.
2.1.3.2.3.1 Perform the IEC 62535 or Doble covered conductor deposition test in addition to the ASTM 1275
Modified Method B test.
2.1.3.2.3.2 When corrosive sulfur is detected, add copper passivators to the transformer oil to a concentration
of 100 ppm. Check the concentration of the passivators on an annual basis to determine if additional
passivators are needed. Add additional passivators as needed. Note: The addition of passivators needs to
be made known to the laboratory performing dissolved gas analysis, as passivators can affect the stray
gassing characteristics of the transformer oil.
2.1.3.2.4 In addition to the recommended standard fluid tests, perform the following tests on the transformer
fluid every three years:
• PCB
• Furan analysis
2.1.3.2.5 Perform the routine offline (de-energized) tests listed in Table 2 every three years.
2.1.3.2.6 Perform routine maintenance servicing for on-load tap changers per OEM guidance.
2.1.3.2.7 Perform the additional tests listed in Table 3 if routine tests indicate a problem with the transformer
to evaluate and investigation the fitness for continued use.
These additional tests are also useful in deciding whether to reenergize the transformer after it has tripped
due to the operation of protection relays.
2.1.3.3 Investigate the suitability of the bushing for continued use or replace the bushing when the power
factor value increases more than 2 times of the baseline value or capacitance increases to 10% or more over
the baseline value.
2.1.3.4 Perform transformer paper insulation remaining useful life assessment when the following conditions
are detected:
A. CO2/CO ratio is greater than 10, and CO2 concentration is greater than 10,000 ppm.
B. DP levels (as indicated by furan analysis) are 400 or lower.
C. Transformer operating history indicates there has been accelerated deterioration of the paper insulation,
such as overloading or high moisture level.
C. Transformers for which the maintenance and testing intervals must be significantly extended due to
reasons such as the inability to remove the transformer from service, location in remote areas, or lack of
resources.
2.1.4.2 Provide online condition monitoring systems to monitor the following parameters (several separate
pieces of equipment may be needed to achieve the required level of condition monitoring):
A. Moisture-in-oil (with appropriate algorithms to translate this to a moisture-in-paper measurement)
B. Temperature
C. Dissolved gases
D. Power factor for bushing with testing tap
E. Partial discharge activity monitoring
F. On-load tap changer motor current
2.1.4.2.1 In addition to online condition monitoring systems, conduct periodic in-service testing using the
following equipment:
• Acoustic sensors to monitor partial discharge activity
• Corona cameras
2.1.4.3 Select online condition monitoring systems with the following features and capabilities:
A. The ability to either transmit data continuously to a manned location (SCADA), or the ability to allow
remote retrieval of the data
B. The ability to generate alarms (to a staffed location) when unusual conditions are detected.
C. The ability to analyze monitored parameters, generate a summary of the condition of the transformer,
and produce recommendations for basic action
D. The ability for the user to set caution and alarm levels
E. The ability to store and trend monitored parameters
F. Field calibration and self-diagnostic capabilities
G. A measurement of accuracy of +/-5% for multi-gas DGA monitoring systems and moisture in the oil.
For standalone H2 continuous on-line condition monitoring, provide a measurement accuracy of +/-20%
or 25ppm, whichever is the greater
H. A measurement accuracy of +/-1% for bushing leakage current monitoring and be able to accept leakage
current up to 300 mA
I. A sampling rate of at least once every 4 hours
J. Robustness and immunity to electromagnetic fields interference and radio frequency interference.
K. Security to prevent unauthorized changes to alarm conditions and limits
L. Environmental appropriateness for local site conditions
2.1.4.3.1 Online condition monitoring systems are not a substitute for electrical testing. Online condition
monitoring systems can be used to adjust the interval between electrical testing, but should not be used to
justify a lack of testing. FM Global views online condition monitoring systems as broadband, early warning
systems that trigger more focused electrical tests, including offline testing to confirm problems detected by the
online condition monitoring system. An effective online condition monitoring system requires at least the
following human elements:
A. Pre-defined action plans for various alert setting levels that are agreed by stakeholders.
B. Tracking of actions for response to alerts, including who responds to alerts, when, and with what
comments.
C. Clear ownership of the online condition monitoring systems with sufficient knowledge and experience
level. The data should be integrated with all available periodic offline testing and in-service testing
information, and operation history for condition assessment.
2.1.4.3.2 If continuous on-line H2 systems are to be installed, do all the following:
A. Use FM Approved continuous H2 monitoring systems
B. Install the system in accordance with manufacturer’s installation instructions.
2.1.4.4 For transformers with a bladder in the conservator tank, check the bladder for leaks once every two
years. This is done by swabbing the inside of the bladder to check for oil, and also checking the oil screen
and DGA results for indications of bladder leaks. Replace leaking bladders as soon as possible.
The bladder in a conservator tank generally has a 10-year life. A leaking bladder will allow oxygen and
moisture to enter the transformer, which accelerates its aging process. Proper maintenance of the bladder
is a critical part of a transformer life management program.
2.1.4.5 Test the electrical protection system (fuses, circuit breakers, batteries, surge arrestor, and electrical
and mechanical relays such as sudden pressure relay) in accordance with Data Sheet 5-19, Switchgear
and Circuit Breakers.
2.1.5 Operation
2.1.5.1 Install, operate, and maintain transformers in accordance with the manufacturers’ recommendations.
2.1.5.2 Follow the manufacturer’s instructions for filling new transformers with oil. Check all oil to ensure it
meets the specifications for new oil. In particular, make sure the moisture content, PCB content, and corrosive
sulfur content are within limits.
2.1.5.3 Immediately after commissioning, and periodically for several days after commissioning, inspect the
transformer thoroughly for indications of overheating, oil leaks, abnormal vibration, abnormal noise, and
malfunction. Verify proper operation and calibration of each monitoring and protective device. Perform
benchmark tests as recommended in Section 2.1.2.1.
2.1.5.3.1 At a minimum, perform dissolved-gas-in-oil analysis 18 to 24 hours after energization, one month
later, and six months later to determine if the transformer has any infant mortality problems.
2.1.5.4 Do not overload the transformer. If short-term overloading is required for emergency purposes, perform
studies to determine the economic, loss of life, and bubbling risks associated with overloading. Consider
the use of a dynamic transformer monitoring and control system as described in Section 2.1.4.3, or a cooling
system upgrade if frequent overloading of the transformer is expected.
2.1.5.5 Do not reenergize transformers after the operation of protective devices until it has been determined
that the transformer has not suffered internal damage and any external faults have been removed. At a
minimum, investigate the operation of the protective device and perform a dissolved gas analysis (DGA) test
to determine if the transformer has suffered internal damage. Additional testing may be needed depending
on the results of the DGA test (see Section 2.1.2).
2.1.5.5.1 Exercise care when taking a DGA sample immediately following a fault. It may be prudent to wait
several hours after the fault because fault gases may not have had enough time to migrate through the bulk
of the oil. Taking a DGA sample too early can produce misleading results.
2.1.5.5.2 Record the event and any subsequent tests results.
2.1.5.6 Do not perform online insulating fluid processing, such as degassing or regenerating, on a routine
basis or without a proper evaluation of the need for and consequences of processing the insulating fluid.
2.1.5.6.1 Take the following precautions when insulating fluid is processed or replaced:
A. Perform DGA and furan analysis before the insulating fluid is processed or replaced to preserve data
about the transformer’s condition. Repeat DGA analysis immediately after the fluid has been processed
or replaced. Perform DGA, oil screening test and furan analysis about six months after the fluid has been
processed or replaced.
B. Take proper precautions when handling insulating fluid with high levels of combustible gas as this may
present an explosion risk.
C. Check replacement mineral oil for corrosive sulfur and PCB contamination.
D. Ensure the equipment used to process the insulating fluid has been thoroughly cleaned to prevent
PCB cross-contamination.
E. Follow the manufacturer’s instructions regarding insulating fluid processing or replacement to prevent
moisture contamination of the transformer’s solid insulation. For example, if the transformer has to be
opened, most manufacturers limit this time to no more than 2 hours and recommend dry air be circulated
into the transformer at a specified flow rate.
F. Replace all fluid-immersed tank gaskets when insulating fluid is replaced, regardless of the type of
oil.
G. When ester-based fluid is used to replace mineral oil in power transformers with voltage class of 69
kV or higher, consult with the transformer’s OEM for a design review for fitness of use.
2.1.7 Sparing
Sparing can be a mitigation strategy to reduce the downtime caused by a transformer 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.
ingress.
• The spare is owned/dedicated.
• The spare can supply 100% of production requirement.
• The spare is electrically and physically compatible with the existing installation (such as having been
installed in the vault in the past) and meets furnace operating permit requirements.
• The spare arc furnace transformer has spare secondary bus work or aligns with the existing bus.
• Spare inspection, testing, and maintenance meets OEM guidelines and FM Global recommendations.
• A viable equipment breakdown spare Joslyn switch is maintained.
Having a spare transformer that is not electrically compatible with the existing installation can negatively
impact furnace operations. This includes furnace emissions, depending on the operating permit requirements.
This needs to be considered in the ECP process.
See Section 3.4.1 for additional arc furnace transformer information.
Main Building
(plan view)
3
4
1 - Detached building
2 - Outside room with direct access from outside only
3,4 - Inside room with direct access from outside only
2.2.1.2 Arrange transformer rooms to allow for direct access from outdoors.
2.2.1.3 Where three-hour fire-rated construction is recommended for transformer rooms with non-FM
Approved fluids in Table 4, also protect exposed structural steel in accordance with Data Sheet 7-32, Ignitable
Liquid Operations.
2.2.1.4 Provide containment and emergency drainage in accordance with Data Sheet 7-83, Drainage and
Containment Systems for Ignitable Liquids. Where multiple oil-filled transformers are installed, provide
individual containment for the contents of each transformer to prevent spills from flowing to other transformers
or important equipment in the room.
2.2.1.5 Where a special protection system (e.g., foam-water sprinklers, compressed air foam, water mist)
is provided for transformer protection in accordance with Section 2.2.3, design containment to hold the
transformer liquid contents plus special protection system discharge for at least 20 minutes. Regardless of the
calculated curb height, provide no less than 3 in (7.6 cm) of containment.
2.2.1.6 Locate transformers a minimum of 3 ft (0.9 m) from walls, or more as needed for maintenance access
and ventilation requirements.
2.2.1.7 If transformer rooms contain liquid-filled transformers, arrange openings to be normally closed and
provide FM Approved fire doors and/or fire shutters that have the same fire rating as the rest of the room.
2.2.1.8 Where conductors penetrate fire-rated construction in transformer rooms and detached buildings, use
FM Approved fire stops with fire resistance equivalent to 1 hour or to the rating of the construction, whichever
is greater.
2.2.2 Occupancy
2.2.2.1 Use dry-type or gas-insulated transformers if suitable for the application.
2.2.2.2 Limit the loading of combustible materials, including transformer fluids, plastics, and insulation on
grouped electrical cables.
2.2.2.3 Where the surrounding occupancy could be exposed to nonthermal damage due to an indoor
transformer fire, do one of the following:
A. Locate the transformers in rooms with suitable construction so the surrounding occupancy will not be
exposed.
B. Equip the transformer room with a mechanical ventilation system designed to vent smoke to outdoors.
Provide power for the ventilation system from an emergency source that will not be deenergized as part
of the pre-fire plan.
2.2.2.4 Install FM Approved smoke detection in transformer rooms, with alarms arranged to sound at a
constantly attended location, regardless of any automatic sprinkler protection or heat detection that may also
be provided. The presence or absence of smoke detectors does not change the need for sprinklers. Arrange
smoke detection spacing in accordance with Data Sheet 5-48, Automatic Fire Detection.
2.2.2.5 In locations where dusty or corrosive atmospheres are or may be present, locate air-cooled
transformers in a pressurized room. Also filter the cooling air and remove corrosive contaminants.
2.2.3 Protection
2.2.3.1 Where sprinkler protection is recommended for transformer fluids per Table 4, provide one of the
following designs:
A. For FM Approved transformer fluids, provide sprinkler protection with a design of 0.2 gpm/ft2 (8 mm/min)
over the entire room.
B. For transformers using non-FM Approved fluids, provide sprinkler protection in accordance with Data
Sheet 7-32, Ignitable Liquid Operations. Also design containment and drainage in accordance with Data
Sheet 7-32.
2.2.3.2 An FM Approved foam-water sprinkler system, or a water mist system is acceptable as an alternative
to sprinkler protection for transformer liquids where recommended in Table 4.
A. Design the system per Data Sheet 7-32, Ignitable Liquid Operations, as well as Data Sheets 4-2, Water
Mist Systems, or 4-12, Foam Extinguishing Systems, as applicable.
B. Design the duration of system water flow for a minimum of 60 minutes.
C. Design the foam concentrate supply to provide flow for a minimum of 30 minutes.
D. Provide an FM Approved water mist system for machinery spaces. Ensure the size of door openings
into the room does not exceed the limitations of the Approval listing.
2.2.3.3 An FM Approved compressed air foam (CAF) system is acceptale to supplement a sprinkler/design
system or as an alternative to emergency drainage for transformer liquids. Design the system per Data Sheet
7-32, Ignitable Liquid Operations.
2.2.3.4 Provide sprinkler protection for transformer rooms and buildings where combustible grouped electrical
cables are present, per Data Sheet 5-31, Cables and Bus Bars.
X
Containment
Fig. 3. Minimum horizontal separation distance between outdoor liquid-insulated transformers and exposed walls of main
buildings
Table 5. Separation for Exposure Protection of Main Building Walls (also refer to Figure 3)
Minimum Horizontal Distance from Containment to Exposed Building
Wall (Dimension X in Figure 3)
2-hour fire-rated wall, Non-combustible Combustible Wall,1
1
Fluid or Transformer Fluid Volume, gal ft (m) wall, ft (m)
Type (m3) ft (m)
FM Approved Per Approval Listing 3 (0.9)
transformer
FM Approved Liquid <10,000 (38) 5 (1.5) 25 (7.6)
in non-Approved >10,000 (38) 15 (4.6) 50 (15.2)
transformer
Non-Approved <500 (1.9) 5 (1.5) 15 (4.6) 25 (7.6)
transformer liquid ≤5,000 (19) 15 (4.6) 25 (7.6) 50 (15.2)
>5,000 (19) 25 (7.6) 50 (15.2) 100 (30.5)
Note 1. For definition of combustible and noncombustible construction materials, see Appendix of Data Sheet 1-1, Firesafe Building
Construction and Materials
X
Plan view Elevation view
Table 6. Separation and Extent of 2-hour Fire Barriers for Protection of Main Building Walls (for dimensions refer to
Figure 4)
Separation and Extent of 2-hour Fire Barrier
Fluid Volume Dimension1,2 Noncombustible Wall3 Combustible Wall3
Fluid Type gal (m3) (See Fig. 4) ft (m) ft (m)
FM Approved ≤10000 X 5 (1.5) (1.5)
transformer fluid (38) Y 5 (1.5) 25 (7.6)
Z 25 (7.6) 25 (7.6)
>10000 (38) x 15 (4.6) 15 (4.6)
Y 15 (4.6) 50 (15.2)
Z 50 (15.2) 50 (15.2)
Non-Approved <500(1.9) X 5 (1.5) 5 (1.5)
transformer fluid Y 15 4.6) 25 (7.6)
Z 25 (7.6) 25 (7.6)
≤5000(19) X 15 (4.6) 15 4.6)
Y 25 (7.6) 50 (15.2)
Z 50 (15.2) 50 (15.2)
>5000 (19) X 25 (7.6) 25 (7.6)
Y 50 (15.2) 100 (30.5)
Z 100 (30.5) 100 (30.5)
Note 1. The X distances refer to minimum separation between the closest inside edge of the spill containment barrier area and the 2 hour
fire barrier. These are the same as Table 5 for 2-hour fire-rated walls. Dimension Y is the horizontal extent of the barrier starting
from the respective edge of containment.
Note 2. Barrier vertical extent is dimension Z in the Table or the building height plus 30 in. (0.75 m) parapet, whichever is less.
Note 3. For definitions of combustible and noncombustible construction materials, see Appendix A of DS 1-1, Firesafe Building Construction
and Materials.
2.3.1.1.2 Where exposure protection is provided per alternatives C, D, or E of 2.3.1.1.1, determine the extent
of roof area exposed to excessive radiant heating, if any, using Figure 5 and Table 7.
X
Main building
with
noncombustible X
roof Containment
W
Table 7. Transformer Fire Exposure to Noncombustible Building Roof Where Separation from Wall is Based on a 3-hour
Fire Barrier or Water Spray Protection
Maximum Liquid Volume gal Building Height, W (see Fig. 5)
Liquid Type (m3) ft (m) ft (m)
Non-Approved transformer <1000 (3.8) Any Not exposed
fluid Any ≥50 (15) Not exposed
1000-5000 (3.8-19) ≥25 (7.5) Not exposed
<25 (7.5) 15 (4.5)
>5000 (19) <50 (15) 25 (7.5)
2.3.1.1.3 Provide a Class A rating for external fire resistance over at least the section of the roof, if any, that
is determined to be exposed per 2.3.1.1.2 and Table 7. (See Data Sheet 1-42, MFL Limiting Factors, for
description of Class A roof installations).
2.3.1.3.2 Where the separation distances in Table 8 cannot be met, provide 2-hour fire-rated barriers between
transformers as shown in Figure 6 and as follows:
A. Extend barriers 1 ft (0.3 m) vertically and 2 ft (0.6 m) horizontally beyond transformer components
that could be pressurized as the result of an electrical fault, including bushings, pressure-relief vents,
radiators, and tap changer enclosures.
B. Use of concrete block or reinforced concrete construction is preferred.
Transformer
Containment
2.3.1.3.3 For installations where the separation distances between transformers in Table 8 are not met and
barriers are not provided, install water spray protection on each transformer in accordance with Section 2.3.2.
2.3.1.4.4 Provide a system for removal of rainwater from the containment area.
2.3.1.4.5 Extend the containment perimeter at least 5 ft (1.5 m) beyond fluid-containing components, including
any external coolers, for transformers with up to 1,000 gal (3.8 m3); and 8 ft (2.4 m) for transformers with
more than 1,000 gal (3.8 m3).
Fig. 7. Nozzle and piping arrangement for transformer using nozzles of ordinary capacity. Upper-level nozzle piping shown
by solid line, lower level by dashed line. Both levels are the same distance from the transformer.
Fig. 8. Nozzle and piping arrangement for transformer using large-capacity nozzles. Upper-level nozzle piping shown by
solid line, lower level by dashed line. Both levels are the same distance from the transformer.
C. Locate components of the water spray system (piping, spray nozzles, etc.) a minimum of 18 in.
(45 cm) from the transformer.
D. Do not locate piping over the top of the transformer or tank relief vents.
E. Do not let nozzles spray onto bushings.
F. For multiple transformer installations, design the water spray system and water supply based on
simultaneous operation of the water spray systems for the transformer of fire origin and adjacent
transformers on each side.
2.3.2.2 If automatic water spray is provided on the building wall for exposure protection (per Section
2.3.1.1.1[E]), design protection as follows:
A. Provide a density of 0.2 gpm/ft2 (8 mm/min) over the coverage area.
B. Determine coverage (demand) area per Section 2.3.1.1.1(E).
C. Design for direct impingement application, or provide rundown application with a maximum distance
of 10 ft (3 m) between levels.
D. Spacing between adjacent nozzles on each level per the manufacturer’s design tables.
E. Axial distance between nozzles and wall per the manufacturer’s design tables.
F. Water spray equipment designed in accordance with Data Sheet 4-1N, Water Spray Fixed Systems,
and installed per Data Sheet 2-0, Installation of Sprinkler Systems.
2.3.2.3 Where transformers present an exposure to buildings or equipment, provide a hose stream allowance
as follows:
A. Adequate for 1 hour hose stream demand of 250 gpm (950 L/min) for transformers holding FM Approved
liquids or up to 1000 gal (3.8 m3) of mineral oil.
B. Adequate for 2 hour hose stream demand of 500 gpm (1900 L/min) hose stream for greater than 1000
gal (3.8 m3) of mineral oil in an individual transformer.
2.3.3 Occupancy
Adhere to the recommendations in Section 2.2.2.
3.1.1 Thermography
Thermography is performed with a thermographic camera with the transformer online. It can be used very
effectively to detect the following problems:
• Low fluid level in transformer or bushing4•Blocked oil flow in radiators4•Shut radiator banks
• Eddy current heating of the tank
• Overheating tap changer contacts4•Loose connections at bushing terminals
• Moisture contamination of surge arrestors
Table 10B. Tests Typically Included in an In-Service Natural Ester Oil Screen Test
Test and ASTM method IEEE Value for Voltage Class
≤69kV 69 kV to 230 kV More than 230 kV
Dielectric strength, 23440 28 30
ASTM D1816 47 50
1 mm gap
2 mm gap
Dissipation factor, Per OEM guidance Per OEM guidance Per OEM guidance
ASTM D924,%, maximum
25°C
100°C
Water content, ASTM 450 350 200
D1533, mg/kg, maximum
Fire point (ASTM D92) 300 300 300
Viscosity (ASTM D445) ≥10% ≥10% ≥10%
The interfacial tension and neutralization number (acidity) are usually used as criteria for determining when
the oil needs to be processed or replaced. At certain interfacial tension and acidity levels, the oil is prone
to forming sludge that can coat the windings and lead to decreased cooling efficiency. In addition to increasing
the oil-sludging tendency, acid is also a catalyst for cellulose aging and can accelerate the deterioration of
the paper insulation.
The tests described in the following sections may also be performed on the oil to determine if incipient faults
or deterioration are present in the transformer.
Other key gases sometimes included in DGA results are propane and propylene.
There are many factors that will affect the DGA results for a transformer. The interpretation of DGA results
is a highly specialized field requiring an understanding of both oil and paper chemistry as well as transformer
design, construction, and operation.
Factors such as the oil preservation system (sealed, nitrogen blanketed, or free breathing) will influence the
dissolved gas analysis results. Sealed transformers keep most of the fault gases within the oil. Free breathing
transformers allow some fault gases to escape to atmosphere, while the dissolved gases in nitrogen
blanketed transformers will partition between the oil and the gas blanket depending on the individual gas
partial pressures. These transformers will all show different DGA patterns for the same fault.
Having the tap changer in a separate compartment or within the main tank will also have an effect on the
DGA pattern. Tap changers generate gases at different rates depending on factors such as the diverter switch
type, number of tap changer operations, and the tap changer design. Faults with the tap changer will also
generate gases. These gases can mix with the oil in the main tank and may either mask faults in the main tank
or lead to a wrong conclusion that there is a fault in the main tank.
Table 13 lists the 90th percentile value of dissolved gas concentration values for a mineral oil filled transformer
as a reference. They are statistic values based on IEEE C57.104: IEEE guide for the interpretation of gases
generated in oil-immersed transformers. Users may decide to use different dissolved gas concentrations
for a transformer based on experience with other similar transformers.
Table 14 provides reference guidance for gas generation rate when there is no industrial reference or
manufacturer guidance available. When any gas has the generation rate above those indicated in Table 14,
further investigation is needed to assess the root cause and corrective action based on the assessment
results.
For silicon oil filled transformers, refer table 15. Table 16 is used for ester oil filled transformers.
Table 13. 90th Percentile Gas Concentration Values for Transformers Filled with Mineral Oil
Gas Concentration in Oil in µL/L (ppm)
O2/N2 Ratio ≤ 0.2 O2/N2 Ratio > 0.2
Transformer Age in Years Transformer Age in Years
Gas Unknown 1-9 10-30 >30 Unknown 1-9 10-30 >30
Hydrogen(H2) 80 75 100 40 40
Methane(CH4) 90 45 90 110 20 20
Ethane(C2H6) 90 30 90 150 15 15
Ethylene(C2H4) 50 20 50 90 15 25 60
Acetylene(C2H2) 1 2
Carbon Monoxide(CO) 900 500
Carbon dioxide (CO2) 9000 5000 10000 5000 3500 5500
Table 14. 95th Percentile Values of Gas Generation Rate for Transformers Filled with Mineral Oil
Gas Generation Rate µL/L/year (ppm/year)
O2/N2 Ratio ≤ 0.2 O2/N2 Ratio > 0.2
Period between first and last point of the series
Gas 4-9 Months 10–24 Months 4–9 Months 10–24 Months
Hydrogen(H2) 50 20 25 10
Methane(CH4) 15 10 4 3
Ethane(C2H6) 15 9 3 2
Ethylene(C2H4) 10 7 7 5
Acetylene(C2H2) Any increasing rate Any increasing rate
Carbon 200 100 100 80
Monoxide(CO)
Carbon Dioxide(CO2) 1750 1000 1000 800
Table 15. Typical Gas Values and Generation Rate for Transformers Filled with Silicon Oil with no previous history
TDCG
(Total
CO2 dissolved
H2 CH4 C2H2 C2H4 C2H6 CO(Carbon (Carbon combustible
Status (Hydrogen) (Methane) (Acetylene) (Ethylene) (Ethane) Monoxide) Dioxide) gas)
Typical <200 <100 <1 <30 <30 <3000 <30,000 <3360
value
(ppm)
Generation <20 <10 <1 <3 <3 <300 <1500 na
rate G1
limit1 (ppm
per month)
Generation <100 <50 <1 <15 <15 <1,500 <15,000 na
rate G2
limit2 (ppm
per month)
1
G1 limit means: if one or more gases generation rates exceed G1 limits, the unit needs to be paid attention.
2
G2 limit means: if one or more gases generation rate exceed G2 limits, the unit is considered in critical condition.
Table 16. Typical Dissolved Gas Concentration Value for Transformers with Ester Oil with no previous sample history
Ester Fluid H2 CH4 C2H6 C2H4 C2H2 CO (Carbon CO2 (Carbon
Type (Hydrogen) (Methane) (Ethane) (Ethylene) (Acetylene) Monoxide) Dioxide)
Soybean- <115 20 <235 20 <2 <170 na
based, (ppm)
High-oleic <35 <25 <60 <20 <1 <500 na
sunflower-
based (ppm)
Synthetic <65 <105 <125 <150 <15 <1400 na
(ppm)
It is important to ensure that oil sampling practices follow ASTM D3613 or equivalent standards. This is
especially true if oil sampling is done by the transformer owner or a third party rather than a transformer oil
testing laboratory. Improper oil sampling practices will render oil screen testing and dissolved gas analysis
completely useless and, even worse, it can give misleading results that cause a bad decision to be made. For
example, when taking oil samples for dissolved gas analysis, an oil sampling syringe should be used so
dissolved gases are not lost to atmosphere. The syringe should also be protected from sunlight and extreme
temperatures that can also cause the concentration of dissolved gases to change. Taking an oil sample for
DGA from the same bottle that has been used to collect oil for oil screen tests will result in a DGA result
that has higher concentrations of oxygen and nitrogen. The concentration of hydrogen and some of the other
lighter gases will also be lower as these gases are lost when the oil sample is exposed to atmosphere.
A proper flushing of the oil valve on the transformer is needed to ensure a good oil sample is collected. The
body of the oil valve frequently collects debris, rust, and free water, which will result in misleading oil screen
test results.
It is important to note that, as with all general guidelines for test results, better interpretation can be obtained
from trends and comparing test results to similar transformers or nameplate data. Other relevant test
information should also be consulted and complementary tests performed to confirm a diagnosis.
The power factor test set can also be used to determine the condition of the transformer bushings. Condenser
or capacitive bushings will have either a potential test tap or a power factor test tap to allow power factor
and capacitance testing to be performed.
The GST-ground mode is used to measure the overall power factor and capacitance of the bushing. In this
test mode all the terminals of the winding under test are short circuited and the other windings are grounded.
The high voltage lead is attached to the center conductor of the bushing and the bushing flange ground is
connected to ground terminal of the test set. All currents flowing through the main capacitive insulation of
the bushing, the upper porcelain, the lower porcelain, the sight glass, and the liquid or compound filler in the
bushing are being measured.
The UST test mode is used to measure the power factor and capacitance of the bushing’s main insulation
(C1) only. The high voltage lead is attached to the center conductor of the bushing, the low voltage lead is
connected to the test tap, and the bushing flange ground is connected to ground terminal of the test set and
is guarded. Only currents flowing through the main capacitive insulation of the bushing are being measured.
The GST-guard mode is used to measure the power factor and capacitance of the test tap (C2). In this test
the high voltage lead is connected to the test tap (it is important to make sure the applied test voltage is
appropriated for the test tap so as not to damage the test tap insulation), the low voltage lead is connected
to the center conductor of the bushing and guarded, and the bushing flange ground is connected to the
ground terminal. Only currents flowing through the test tap insulation, the ground flange, and the immediate
area of the porcelain around the test tap, are measured.
• Overall and C1 Power Factor: In general, if the value doubles original nameplate value or more,
the bushing is questionable and needs to be replaced or refurbished.
• C2 Power Factor: Should closely match nameplate and in general be less than 1.0%.
• Capacitance: Should match nameplate with changes of more than 10% being cause for concern.
Otherwise the bushing is questionable, and needs to be replaced or refurbished.
Non-condenser-type bushings (e.g., solid porcelain bushings, epoxy bushings, compound filled, oil-filled and
gas-filled bushings) do not have a test tap. It is not possible to measure the power factor and capacitance
for these bushings. A hot collar test can be performed on these bushings to measure the watts loss and
current. Increase of the watts loss is an indication of contamination of the outer surface of the bushing (typical
Watts loss is between 0.02 and 0.04 Watts, with a Watts loss greater than 0.1 Watts being a cause for
concern). Decrease of the current is an indication of internal voids, low oil or compound levels (typical
currents of 40 to 120 mA).
This test measures the frequency response of the transformer to either a variable frequency sinusoidal voltage
input (frequency response method) or to an impulse voltage input (impulse response method). In both cases,
the voltage is very low and the test is nondestructive.
The test protocol is as follows:
For each winding, inject the signal at one end of the winding and measure the response at the other end
of the same winding. For star connected windings, inject the signal at one end of the winding and measure
the response at the neutral.
Inject the signal into the high voltage winding and measure the response at the low voltage winding for each
phase.
For windings with tap changers, perform the test with the tap changer in 2 positions: the neutral position
and the position that puts the entire winding in the circuit.
The response (or transfer function) is displayed as a Bode Plot (amplitude vs. frequency) from power
frequency to 2 MHz.
The shape of the frequency response is analyzed. In particular, the resonances of the frequency response
curve provide information about the mechanical disposition of the transformer windings, leads, and core. This
test is sensitive to very small changes in transformer winding, lead, and core geometry and can pick up
problems such as axial winding collapse, hoop buckling, movement of windings relative to each another, and
shorted turns.
To gain the most value from frequency response analysis, it is useful to have a previous frequency response
plot to compare the readings to. However, the skill of knowledgeable experts as well as the experience in
frequency response analysis has advanced to the stage where it is possible to obtain good information from
a standalone test by comparing this against a database of expected results.
This test does require a knowledgeable person to interpret the results. Some companies are developing
expert systems to perform this analysis as well as general rules for reading the results. This will enable less
knowledgeable persons to provide a preliminary diagnosis. This is particularly valuable if FRA testing is used
to determine if a transformer should be reenergized after seeing a fault.
R minimum = 0.8 x E
√kVA
Where:
E is the voltage rating of the winding under test.
kVA is the rating of the transformer.
0.8 is a factor used for oil filled transformers at 20°C. A factor of 16 should be used when windings are
untanked.
Because of the temperature dependence of resistance measurement, dielectric absorption ratios are used
to evaluate insulation cleanliness and dryness. The dielectric absorption ratio is the ratio of the 60-second
insulation reading to the fifteen second insulation reading. A dielectric absorption ratio in the range of 1.3
to 3 indicates a clean and dry transformer.
Another insulation ratio used is the polarization index, which is a ratio of the 10-minute insulation reading
to the 60-second insulation reading. A polarization index less than 1 indicates a wet or contaminated
insulation. A polarization index greater than 2 indicates dry and clean insulation.
It is important to discharge the windings after an insulation resistance test because any residual charge will
interfere with other electrical tests and generate bad results. There is also a potential for insulation resistance
testing to magnetize the core and this can also result in bad test results. For these reasons it is good practice
to make insulation resistance testing the last test performed on the transformer.
Electromagnetic interference measurement detects radio frequency noise using radio frequency antennae
or high frequency current transformers. It is possible to apply triangulation or other methods to localize the
source of radio frequency noise generated by the transformer. This test is relatively new and is often used as
a complement to partial discharge measurements.
A. Single gas or composite gas monitors: these sensors detect hydrogen primarily. These monitoring
systems give a single readout in respond to abnormal operation of transformers and provide warning signal
once the pre-set threshold value is exceeded. It is a relatively economical system.
B. Multi-gas monitors: these systems can monitor up to eight or nine dissolved gasses individually. There
are main two technologies used: gas-chromatography system or photo-acoustic spectroscopy. These
systems provide a lot more information about the condition of transformers. They significantly increase
the usefulness of the monitoring system for real time diagnostics on transformers.
Installation of dissolved gas monitors on existing transformers requires connection oil valves that are
pre-existing.
Because aging of transformers depends on these four factors, the nameplate age alone cannot be solely
relied upon to determine when a transformer needs to be replaced.
Degradation of paper causes the long cellulose chains to break. Shorter cellulose chains provide less tensile
strength. The length of cellulose chains in paper is measured in degrees of polymerization (DP). New
transformers have paper insulation with a DP of 800. The IEEE has set the transformer end-of-life criterion
at a DP of 200, which represents a loss of 50% tensile strength.
FM Global recommends the use of oil screen tests, DGA, and furan analysis as screening tools to indicate
when transformers are approaching their end of life. Oil screen tests will identify transformers with high
moisture in paper concentrations. The ratios and concentrations of CO and CO2 in DGA analysis will identify
transformers where thermal degradation of the paper is a serious concern. And, furan analysis will provide
an approximation of the DP of the paper insulation.
4.0 REFERENCES
4.1 FM Global
Data Sheet 4-1N, Fixed Water Spray Systems for Fire Protection
Data Sheet 5-11, Lightning and Surge Protection for Electrical Systems
Data Sheet 5-19, Switchgear and Circuit Breakers
Data Sheet 5-48, Automatic Fire Detection
Data Sheet 9-0, Asset Integrity
FM Approval Standard 6520, Dissolved Gas Analysis Monitoring Systems in Liquid Immersed Transformers
4.2 Other
ANSI/IEEE Std. 493, Design of Reliable Industrial and Commercial Power Systems
ANSI/IEEE Std. C57.91, Guide for Loading Mineral Oil-Immersed Transformers
ANSI/IEEE Std. C57.92, Guide for Loading Oil-Immersed Power Transformers
ANSI/IEEE C57.109, Transformer Through-Fault Current Duration
ANSI/IEEE C37.100, Thermal Evaluation of Oil-Immersed Distribution Transformers
International Council on Large Electric Systems (CIGRE). Life Management Techniques for Power
Transformers. WG A2.18. 2003
from 5 to 10,000 kVA. The transformer includes electrical protection to clear high current as well as sustained
low current faults. A pressure relief device and tank discharge strength prevent tank rupture under a low-level
electrical fault.
FM Approved transformer fluid: A transformer insulating fluid that exhibits a reduced fire hazard, proven
through testing by FM Approvals. The transformer is not necessarily FM Approved.
Higher secondary voltages: Secondary voltages equal to or greater than 480 volts, including 480/277 volt
systems.
In or near commercial buildings: within the interior of, on the roof of, attached to the exterior wall of, in
the parking area serving, or within 30 meters of a non-industrial non-substation building. Commercial buildings
are typically accessible to both members of the general public and employees, and include: 1) Public
assembly properties, 2) educational properties, 3) institutional properties, 4) residential properties, 5) stores,
6) office buildings, and 7) transportation centers (e.g., airport terminal buildings, subway stations, bus
stations, or train stations).
Lower secondary voltage: Transformers with secondary voltages below 480 volts.
Navigable waterway: Navigable waterway is defined by 40 CFR Part 112 as:
A. All waters that are currently used, were used in the past, or may be susceptible to use in interstate
or foreign commerce, including all waters subject to the ebb and flow of the tide.
B. All interstate waters, including interstate wetlands, mudflats,and sandflats.
C. All other waters such as intrastate lakes, rivers, streams (including intermittent streams), wetlands,
mudflats, sandflats, sloughs, prairie potholes, wet meadows, playa lakes, or natural ponds, the use,
degradation, or destruction of which could affect interstate or foreign commerce including any waters that
could be used for recreational purposes, or from which fish or shellfish could be taken and sold in interstate
or foreign commerce; or that are used or could be used for industrial purposes by industries in interstate
commerce.
Network transformer: These transformers are located in vaults in buildings or adjacent to buildings. The
vaults contain two or more power transformers. These transformers are supplied from different transmission
or distribution lines and are paralleled on their low voltage side through circuit interrupting devices called
“network protectors”. Typically, high voltage current interrupting devices have not been used in the network
vault. The low-voltage bus of a network vault may be electrically tied to a number of other vaults to form
a network secondary distribution system, called a low-voltage network grid.
Primary winding: The winding into which energy normally flows. The primary winding can be energized
from the secondary winding under abnormal conditions.
Radial transformer: A transformer that can only be energized from the primary winding.
Secondary winding: The winding from which energy flows during normal operation.
2. Section 2.3 Outdoor Transformer, recommendation 2.3.1.2.3 for open pit containment system. Where
automatic sprinkler protection is used, the sprinkler density has been increased to 0.3 gpm/ft2 (15 mm/min)
from 0.15 gpm/ft2 (6 mm/min). Where a flame arrestor is used to increase the amount of time before burning
oil enters the pit Size No 5 rather than Size No 2 ASTM D448 Standard Classification for Sizes of Aggregate
for Road and Bridge Construction should be used. Size No 5 is closer to the 1.5 in (3.8 cm) washed and
uniformly sized rock tested at the FM Global Research Campus.
3. Section 3.1.1, Approved and Equivalent Transformer. A transformer equivalent to an FM Approved
transformer is defined as a transformer with a UL listing per NEC Section 450-23 with electrical protection
to clear sustained low current faults. The reference to NEC 450-23 was left out of the previous edition.
Transformers complying with NEC 450-23 include four of the five safeguards requested for an FM Approved
transformer. Low current fault protection is the remaining safeguard.
May 2003. The following changes were done for this revision:
1. Tables 2a and 2b Separation Distances. The change allows medium sized transformers containing FM
Approved less flammable fluids to be located as close to buildings and to other transformers as small
transformers were in the previous standard. This provided there is adequate space for inspection and
maintenance. Medium sized transformers may contain up to 10,000 gal (37.9 m3) of fluid.
2. Section 2.3.1.2 Containment. The change increases the quantity of FM Approved less flammable fluid
in a transformer before a containment system is recommended. It increases the size of the transformer to
1320 gal (5 m3) for transformers containing all FM Approved less flammable fluids. It further increases the size
to 2640 gal (10 m3) if the fluid is certified as biodegradable and if a release does not expose navigable
waterways. A definition is included for navigable waterways. The fluid would have to be certified as
biodegradable by the responsible governmental authority.
3. Minor editorial changes were made to Section 2.3.1.2.1.4.
January 2001. The recommendation for the 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.
The following major changes have been made:
a) Addition of emergency power supply recommendation for mechanical ventilation (Section 2.2.1.1.3).
b) Change requirement for smoke detection to fire detection (Section 2.2.1.1.4).
c) Add recommendation for location of rooms containing network transformers to outside wall where
possible (Section 2.2.1.2).
d) Addition of fire protection recommendations for multiple indoor oil insulated transformers (Section
2.2.1.3.2).
e) Open pits without protection acceptable containment for FM Approved less flammable fluid insulated
transformer (Section 2.3.1.2.1.2 and 2.3.1.3.1).
December 1998. Editorial changes were made.
APPENDIX C BIBLIORGRAPHY
ANSI/IEEE Std. C57.12.80
Modern Power Transformer Practice, Edited by R. Feinberg, Wiley, 1979, pgs. 114, 115.
ANSI/IEEE Std. C57.12.00
ANSI/IEEE Std. 979
Electric Utility Systems and Practices, 4th Edition, Edited by Homer M. Rustebakke, Wiley, 1983, pg. 150.
Distribution Transformer Tank Pressure Study, EPRI 325 Final Report, February 1976, Prepared by Allis-
Chalmers Corporation.
The J&P Transformer Book, 11th Edition, A. C. Franklin, D. P. Franklin.RTE/ASEA Transformer Instruction
Book.
Johnson, D. L., Conserving Inherent Transformer Life Expectancy, Bonneville Power Administration, 51st
Annual International Conference of Doble Clients.
NFPA 70, National Electrical Code (NEC).
NFPA 850, Electric Generating Plants.
US Code of Federal Regulations, Part 761: Polychlorinated Biphenlys (PCBs).
European Standard EN 50195, Code of Practice for the Safe Use of Fully Enclosed Askarel-Filled Electrical
Equipment.