Investigating Transformer Failures

William H. Bartley P.E.

The Hartford Steam Boiler Inspection & Insurance Co.
Hartford, CT

Major failures involving large oil-cooled transformers continue to occur worldwide on a frequent basis.
The deregulation of wholesale electricity supply has led to a number of changes and new challenges for
the global electric utility industry and the market participants. Increased equipment utilization, deferred
capital expenditures and reduced maintenance expenses are all part of today's strategies for transformer
owners. To make matters worse, world power consumption is increasing, and the load on each aging
transformer continues to grow.

Hartford Steam Boiler (HSB) has been insuring power equipment since its inception in 1866, when the
burgeoning industrial revolution in the U.S. used steam to drive the railroads, the riverboats and our
industrial machinery. Over the years, HSB has investigated thousands of transformer failures, and we
have conducted a number of studies on transformer claims. The results of our most recent study were
published at the 2003 Annual Technical Conference in Stockholm, for the International Association of
Engineering Insurers (IMIA). Excerpts from that study are included in this paper.1

FAILURE ANALYSIS
In early 2003, Hartford Steam Boiler sent out a request to all the national delegations of IMIA seeking
information on failures of transformers rated at 25 MVA and above, for the period 1997 through 2001.
Information was requested concerning year of loss, size in MVA, age at failure, application (such as
utilities, industrials), cause of failure, property damage portion, and business interruption portion. Data
was obtained on 94 cases. An estimate of the total population of power transformers would have been
useful, but it is impractical to obtain this information. Some of the insurance company contributors were
not able to identify the age of the transformers, and in some cases, the size of the transformer. All
amounts of losses were converted to U.S. dollars, using the following exchange rates: 0.9278 euros; 8.542
Swedish kronas; and 6.0858 French francs.
During this 5-year period, the number of large transformer claims reached a peak (25) in 1998. But, the
dollars paid out, reached a maximum in the year 2000, due to several claims in the multi-million dollar
range, plus one large business interruption loss. The largest transformer loss also occurred in 2000, at a
power plant, with a business interruption portion the equivalent of more than U.S. $86 million. Three of
the top four property damage claims were in industrial plants.
Table 1 displays the annual transformer claims, in cost per MVA. This is property damage only; it does
not include business interruption losses. Not all of the data contributed had size information. Therefore,
we could only analyze 78 claims for cost per size. The average cost was approximately $9,000 per MVA
(or $9 per kVA). Table 1 displays the annual transformer claims and cost per MVA.
Table 1 – Number and Amounts of Losses by MVA and Year
Total # of Failures
Year Failures w/data Cost /MVA
1997 19 9 $7,969
1998 25 25 $4,379
1999 15 13 $14,967
2000 20 19 $12,849
2001 15 12 $7,748

1
A complete copy of the IMIA paper is available at http://www.imia.com/
1
Application
During this period, the largest number of transformer claims occurred in the Utility Substation sector, but
the highest paid sector was Generator Step Up transformers. (This includes both property damage and
business interruption losses.) If the single largest business interruption claim is ignored, the Generator
Step Up transformer sector is still significantly higher (in dollars paid) than any other sector. This is to be
expected due to the very large size of these transformers.

Cause of Failure
For the failures reported, the leading cause of transformer failures is “insulation failure”. This category
includes inadequate or defective installation, insulation deterioration, and short circuits, but not exterior
surges such as lightning and line faults. Table 2 lists the number of failures and percentage of total costs
paid for each cause of failure. A description of each cause category is found below.

Table –2 Cause of Failures

Cause of Failure Number
Insulation Failure 24 52%
Design /Material/Workmanship 22 23%
Unknown 15 10%
Oil Contamination 4 4%
Overloading 5 3%
Fire /Explosion 3 3%
Line Surge 4 2%
Improper Maintenance /Operation 5 1%
Flood 2 1%
Loose Connection 6 1%
Lightning 3 <1%
Moisture 1 <1%
94

However, the risk of a transformer failure is actually two-dimensional: the frequency of failure and the
severity of failure. Figure 1 is a scatter plot, depicting the risk of failure. The number of failures for each
cause is on the horizontal axis, and the dollars paid for each cause is on the vertical axis. The higher risks
are in the upper right-hand corner. According to this analysis, the Insulation Failure is the highest risk for
all types of transformer failures.

Description of Causes
Insulation Failures – Insulation failures were the leading cause of failure in this study. This
category excludes those failures where there was evidence of a lightning or a line surge. There
are actually four factors that are responsible for insulation deterioration: pyrolysis (heat),
oxidation, acidity, and moisture. But moisture is reported separately. The average age of the
transformers that failed due to insulation was 18 years

Design /Manufacturing Errors - This category includes conditions such as: loose or unsupported
leads, loose blocking, poor brazing, inadequate core insulation, inferior short circuit strength, and
foreign objects left in the tank. In this study, this is the second leading cause of transformer
failures.

2
 $1,000,000,000

Insulation
Failure
$100,000,000
Design /
Material

Cost of Failres
Unknown

$10,000,000

Increasing
Risk
$1,000,000

$100,000
0 5 10 15 20 25 30
Number of Failures by Cause

FIGURE 1 – Frequency - Severity of Transformer Failures

Oil Contamination – This category pertains to those cases where oil contamination can be
established as the cause of the failure. This includes sludging and carbon tracking.

Overloading - This category pertains to those cases where actual overloading could be established
as the cause of the failure. It includes only those transformers that experienced a sustained load
that exceeded the nameplate capacity.

Fire /Explosion - This category pertains to those cases where a fire or explosion outside the
transformer can be established as the cause of the failure. This does not include internal failures
that resulted in a fire or explosion.

Line Surge - This category includes switching surges, voltage spikes, line faults/flashovers, and
other T&D abnormalities. This significant portion of transformer failures suggests that more
attention should be given to surge protection, or the adequacy of coil clamping and short circuit
strength.

Maintenance /Operation - Inadequate or improper maintenance and operation were major causes
of transformer failures, when you include overloading, loose connections and moisture. This
category includes disconnected or improperly set controls, loss of coolant, accumulation of dirt &
oil, and corrosion. Inadequate maintenance has to bear the blame for not discovering incipient
troubles when there was ample time to correct it.

Flood – The flood category includes failures caused by inundation of the transformer due to man-
made or natural caused floods. It also includes mudslides.

Loose Connections - This category includes workmanship and maintenance in making electrical
connections. One problem is the improper mating of dissimilar metals, although this has
decreased somewhat in recent years. Another problem is improper torquing of bolted
connections. Loose connections could be included in the maintenance category, but we
customarily report it separately.

3
 Lightning - Lightning surges are considerably fewer in number than previous studies we have
published. Unless there is confirmation of a lightning strike, a surge type failure is categorized as
“Line Surge”.

Moisture - The moisture category includes failures caused by leaky pipes, leaking roofs, water
entering the tanks through leaking bushings or fittings, and confirmed presence of moisture in the
insulating oil. Moisture could be included in the inadequate maintenance or the insulation failure
category above, but we customarily report it separately.

Transformer Aging
Notice that we did not categorize "age" as a cause of failure. Aging of the insulation system reduces both
the mechanical and dielectric-withstand strength of the transformer. As the transformer ages, it is
subjected to faults that result in high radial and compressive forces. As the load increases, with system
growth, the operating stresses increase. In an aging transformer failure, typically the conductor insulation
is weakened to the point where it can no longer sustain mechanical stresses of a fault. Turn to turn
insulation then suffers a dielectric failure, or a fault causes a loosening of winding clamping pressure,
which reduces the transformer's ability to withstand future short circuit forces. Table 3 is the distribution
of transformer failures by age. The average age at failure for this group of transformers was 18 years.
Table 3 – Distribution of Losses by Age of Transformer
Age at failure Number of
Failures
0 to 5 years 9
6 to 10 …. 6
11 to 15 … 9
16 to 20 … 9
21 to 25 … 10
Over 25 years 16
Age Unknown * 35

FUTURE FAILURES
Due to the large number of aging transformers around the world, Hartford Steam Boiler developed a risk
model of future failures. Our latest model was included in a paper at the 2003 Weidmann ACTI Inc.
Second Annual Conference. The hazard rate model is shown in Figure 2, and the failure distribution
model for the fleet of transformers installed between 1964 and 1992 is shown in Figure 3.
Although we have not yet seen an
alarming increase in end of life failures,
such a rise must be expected eventually.
The most difficult task for the utility
engineer is to predict the future
reliability of the transformer fleet, and
the ability to replace each one as it fails.
Meeting the growing demand of the grid
and at the same time maintaining system
reliability with this aging fleet will
require significant changes in the way
the utility operates and cares for its
transformers.
FIGURE 2 – Hazard Rate

4
 Figure 3 - Failure Distribution
Model for Transformers installed in
U.S. between 1964 and 1992.

TRANSFORMER FAILURE MODES

The failure of a transformer can be a devastating and costly experience. However, it is an unfortunate fact
that despite even the most rigorous preventative maintenance program, failures can and will occur. The
preceding classification of failures is admittedly subjective in nature, but usually sufficient for our claims
process. However, in order to prepare an accurate failure scenario, and develop proper recommendations
to prevent a recurrence, a more in-depth analysis of the failure is necessary. To accomplish this, it is
necessary to first understand the different modes of failure for transformers.
A transformer can fail from any combination of electrical, mechanical or thermal factors. While it is
difficult to define a "typical" failure mode for a transformer due to its complexity, most actual failures
involve - and eventually result from - the breakdown of the transformer's insulation system. But, the
actual cause of the failure of a transformer may be due to electrical, mechanical or thermal factors. Of
course, a transformer failure may have more than one causal factor.

Electrically Induced Factors - An electrically induced factor typically results in damage to a transformer's
insulation system. Some of the more common electrically induced factors are:
Operation of a transformer under transient or sustained over-voltage conditions
Exposure to lightning surges and switching surges -
Partial discharge, which can be caused by poor insulation system design, by manufacturing defects,
and/or by contamination of the insulation system (both the solid insulation and oil).
Static electrification - wherein a static charge is developed between the insulating oil and metal
components of the transformer.
These failure modes may be discovered in combination with one another or in combination with other
mechanical or thermal evidence. It is important for all evidence to be evaluated together in order to
develop an accurate failure scenario.

Mechanically Induced Factors - A mechanically induced factor typically results in the deformation of a
transformer's windings, resulting in the abrasion or rupturing of its cellulose insulation. If the damage is
severe enough, the transformer can fail electrically. It is difficult to predict how long a transformer can
survive with this kind of damage and is entirely dependent upon its severity. Winding deformation
typically occurs in one of two ways: shipping damage or electromechanical forces.

5
Shipping or movement of the transformer: Although a transformer should be internally braced by the
manufacturer to withstand the forces associated with this type of movement, the bracing may not be
adequate; the shipper may not have followed the manufacturer's rigging instructions; or an accident may
have occurred during the transportation.
Magnetically induced electromechanical forces: Advances in materials and winding design have resulted
in much stronger windings, but many older transformers are still in service that have not benefited from
these advances. When a transformer experiences an internal fault the windings can be subjected to
magnetic forces that are significantly beyond their design capability.
Hoop (inward radial) buckling of the innermost winding - In this case, the conductor will buckle inward
toward the core between the axial spacers and will transmit the buckling to the core insulating cylinder at
the axial spacer locations.
Conductor tipping - In this case, axial forces that exceed the coil bundle’s compressive capability will
cause the bundle to tip. When the bundle tips, the paper insulation will tear open and expose the
energized conductor.
Conductor telescoping - When exposed to excessive axial forces, the individual conductors will telescope
over one another. This causes the entire layer to become mechanically unstable, as well as damaging the
paper insulation.
Spiral tightening - This is caused by radial forces that tighten the winding. This can be evidenced by a
spiral movement or shifting of the key spacers over the entire height of the winding.
End-ring crushing - This condition occurs when the mechanical strength of the radial end ring at the
bottom of the winding is exceeded by the winding's axial forces, resulting in mechanical instability of the
entire winding.
Failure of the coil clamping system - When a transformer is subjected to large sudden increases in current
flow the resulting magnetically induced electromechanical forces try to spread the winding coils apart
axially. If a failure of the clamping system occurs, its coils will spread apart.
A thorough failure investigation must consider that evidence indicating the presence of any of these
mechanical problems does not necessarily mean that the cause of the failure has been found. In some
cases, damage such as outlined above may simply have been the result of collateral damage due to a fault.

Thermally Induced Factors - The degradation of a cellulose insulation system is to be expected over time.
Thermal degradation results in the loss of physical strength of the insulation that will weaken the paper to
the point where it can no longer withstand the mechanical duty imposed on it by the vibration and
mechanical movement inside of a transformer. A well-designed (and properly operated and maintained)
transformer's insulation system should be able to provide reliable service for 30 years or more. The most
common thermally induced factors are summarized as follows:
Overloading of the transformer beyond its design capability for extended periods of time.
Failure of a transformer's cooling system. This can include blocking or fouling of the radiators or coolers,
the failure of the oil pumps, and the failure of a directed flow oil distribution system.
Blockage of axial oil duct spaces, limiting the amount of cooling oil to the windings in the immediate
area.
Operating a transformer in an overexcited condition (over-voltage or under-frequency). This can cause
excessive stray magnetic flux to severely overheat insulation in close proximity to the core or other
structural members.
Operation of the transformer under excessive ambient temperature conditions.

6
If evidence of thermally induced problems is found, it must be considered and combined with any other
evidence discovered of mechanical or electrical problems to develop a complete failure scenario.
CONDUCTING THE FAILURE INVESTIGATION
As an aid to the investigation, we recommend the IEEE Standard C57.125 “Guide for Failure
Investigation, Documentation, and Analysis for Power Transformers and Shunt Reactors” [2]. This guide
can be used to ensure that all important factors are considered and examined in a failure investigation.
The IEEE document is an excellent source of information for an Investigator to help develop a general
“battle plan” prior to arriving at the site, and the various checklists are very useful. The IEEE guide also
contains comprehensive appendices on transformer construction, diagnostic testing, and sample failure
investigation case histories.
A failure investigation typically begins after a transformer has been tripped off-line by a protective
device, or when someone suspects that the transformer is no longer suitable for service. The decision of
what to do next (after a transformer has been tripped by a protective device) will vary depending on the
circumstances and the operating procedures of the owner-user. But, when a transformer fails, time is of
the essence and a prompt investigation is paramount. Work crews are almost always on site before the
Investigator arrives. Valuable information can be innocently destroyed by work crews attempting to
restore service. Therefore, upon notification of a failure, instruction should be given to the work crew to
minimize any disruptive impact on the investigation. Cooperation at all levels of the owner-user can
speed up the investigation at the site and improve the accuracy of the diagnosis.
The investigation is comprised of four major components: Preparation, Testing, Inspection, and
Conclusions. The preparation begins as soon as the Investigator is notified. The Testing and the On-Site
Inspection may occur simultaneously. In some situations, the Investigator may be able to prescribe a
series of tests, before arriving on site. The Inspection may involve three steps – an external examination,
an internal examination, and a teardown inspection. Once sufficient data has been gathered from the
inspection and test, the engineering analysis can hopefully lead to contributing causal factors.

PREPARATION /INFORMATION GATHERING

Some preparation, prior to arrival at the site may prove to be valuable, and save considerable time and
effort. Information that should be requested or gathered prior to the Investigator’s arrival would include:
All historical dissolved gas analysis (DGA) test results
All historical oil screen test results (e.g., dielectric breakdown, acidity, interfacial tension)
All historical diagnostic field test reports (e.g., Doble power factor, frequency response analysis)
A copy of the transformer nameplate
All maintenance and other work records
A copy of the factory test data sheets
A record of local weather conditions at the time of failure
A record of the approximate loading at the time of failure
A system one-line diagram showing the locations and types of protective devices
A record of all relay operations (“flags” or “targets”) at the time of the failure
Any system dispatcher’s logbook pages which pertain to the failure
A list of any faults or switching events in the system just prior to the failure
A list of any recent ancillary equipment problems, such as CT or PT failures, or blown fuses on capacitor banks.
If the transformer is equipped with surge arrestors that have operation counters, did any of the arrestors operate
just prior to the failure?

This information will provide a history of the condition of the transformer and may give some indications
of system conditions that may have contributed to or caused the failure.

7
ON-SITE INSPECTION
The on-site inspection is the logical next step in the process. Upon arrival at the site the Investigator
should interview any employees who may have been in the vicinity at the time of the failure. Try to get
as much information about what they saw and heard. When all interviews are concluded, a visual
inspection should be performed.
xternal Inspection –
Photographs are an essential part of an external inspection and help to record or document valuable
information for later analysis. The experienced investigator will look for the following visible
abnormalities in the transformer:
Bulging or rupture of the external tank Evidence of a foreign object or animal contact
Evidence of an oil spill or a fire (burn marks or debris on top of the tank or
Evidence of the operation of a pressure relief nearby bus)
device Evidence of vandalism or sabotage
Damage to the bushings or surge arrestors Position of the LTC – as found
Damage to the radiators, fans, or pumps Position of the DETC - as found
Damage to the conservator Oil Level in the LTC compartment
Low oil level in main tank Damage to the Control Cabinet
Evidence of overheating (e.g. blistered paint)

Careful notes should be taken to supplement the photographs. Any items in the Preparatory list (such as
historical test data, or relay “flags”) that were not available in advance should be gathered at this time. If
no visible damage is found externally, the next step in the investigation is usually a prescribed set of
diagnostic tests, which are outlined in the next section.

Internal Inspection –

In some situations, an internal inspection may be performed on-site to assess the extent of damage.
Sometimes the tank is completely drained of oil, while in some cases the oil is just lowered enough to
expose the top of the core and coil assembly. However, an internal inspection requires great care and
carries considerable risk. Entering a transformer without proper knowledge of transformer construction
may result in unintentional additional damage.
CAUTION: Before entering the tank, all SAFETY rules must be observed. In most cases, safety officials
will consider the transformer tank a “Confined Space”, and a Confined Space Entry program must be
followed. Do not attempt this alone. The tank must be vented to the atmosphere for a reasonable amount
of time to expel the combustible gases. Combustible gases in the tank are heavier than air, and present an
asphyxiation hazard. After the tank’s atmosphere has been tested and determined to be safe, any
personnel entering the tank should wear a calibrated Confined Space Monitor for the duration of the
internal inspection. If the gas monitors indicate a low oxygen level at any time during the inspection, all
personnel inside the tank must be evacuated.
In addition, prior to entering the tank, all personnel must remove loose personal items, such as jewelry,
wallets, combs, paper clips, pocket change, and the like. The only items that should be carried inside the
tank are a camera, pen and paper, and a flashlight. These items should be recorded before entry, and then
verified after leaving the tank. Leaving a flashlight inside a high voltage transformer could later be
catastrophic. (It’s possible to find almost anything inside of a failed transformer, including screwdrivers,
pliers, welding rods, and safety glasses from a previous work crew’s carelessness. In one true story, a
wooden ladder was found inside a large power transformer that had been in service for over twenty years
before it suffered a failure.)
Internal abnormalities to look for inside the transformer include:
The odor of burnt insulation
The appearance of burnt oil or insulating fluid (“good” oil has a clear amber color; burnt oil is dark and

8
 opaque, similar to burnt coffee, with an unmistakable acrimonious odor.)
Evidence of metal deposits (splatter or “BB’s”) *
Evidence of broken insulating material (paper, wood, barrier board)
Obvious physical damage (broken porcelain, burnt or charred leads)
Displacement of the coils
Damage to the coil clamping system
Evidence of arcing between turns, or between windings
Evidence of partial discharge activity (carbonized tracking marks)
Evidence of any free water in the bottom of the tank (Since water is
heavier than oil, the presence of water in the tank, will cause rust at the bottom.)

* Molten droplets of metal form tiny spheres, similar to “BB’s”. These deposits can be found almost
anywhere in the transformer, including the top of the yoke, the tops of the coils, on internal ledges of the tank,
and on the bottom of the tank.

If the internal inspection does not provide sufficient information as to the cause of the failure, it may be
necessary to de-tank the transformer and perform a diagnostic teardown to search for the damage and
clues. A diagnostic teardown is a painstaking, meticulous process that is usually accomplished in the
shop of a manufacturer or repair firm. If the transformer is large enough, it is quite common for a
teardown to involve personnel from the owner, the insurer, the manufacturer, and one or more consultants
to represent their various interests.

CONCLUSIONS
Final conclusions on the transformer failure cannot be formulated until all the Preparatory Data is
collected, the inspections are completed, the diagnostic tests are performed, and the test results
interpreted. Often a hasty conclusion to meet an arbitrary deadline can lead to the wrong diagnosis. After
all the information is gathered, an in-depth analysis of the failure will allow you to prepare an accurate
failure scenario and, most important, develop proper recommendations to prevent a recurrence.

REFERENCES

[1] W. Bartley, "Analysis of Transformer Failures", International Association of Engineering Insurers

36th Annual Conference – Stockholm, 2003 ( http://www.imia.com/)
[2] R. James, W. Bartley, “Transformer Asset Management”, Weidmann – ACTI Inc. Second Annual
Conference, New Diagnostic Concepts for Better Asset Management, November 2003.
[3] IEEE Std C57.125-1991, IEEE Guide for Failure Investigation, Documentation, and Analysis for Power
Transformers and Shunt Reactors

The Author
William Bartley, P.E. is the Principal Electrical Engineer in the Engineering
Department of The Hartford Steam Boiler Inspection and Insurance Co. Mr. Bartley
earned a B.S. degree in Electrical Engineering from the University of Missouri at
Rolla, and has been employed by Hartford Steam Boiler since 1971. He is responsible
for developing electrical standards, OEM relations, large failure investigations, repair
procedure development, and new testing technologies. He is a Senior Member of
IEEE and serves on both the Transformer Committee and the Electric Machinery
Committee.