The Anatomy of a

Shermco Field NETA-Certified

Qualified to perform to

Service Engineer the highest standards

Reliable and Responsive

Part of a large international
Trusted Advisor
Offering independent Insulating Oils
Handbook
team available 24/7 service across all brands

Safety First
Back by the Best
On record as an industry
From the field, to our
leader in safe work practices
full-service rotating
apparatus services

Cutting-Edge
Trained from utility
transmission to
low-voltage distribution

Published by InterNational Electrical Testing Association

Odds are, you won’t find a job anywhere that a Shermco field service engineer can’t handle.
As one of the world’s most respected electrical maintenance and testing companies,
Shermco delivers A+ service from utility transmission to low-voltage distribution.

Our experience in commercial, industrial, generation and specialized sectors, such as oil
refineries, pulp and paper, steel and wind power, uniquely qualifies us to handle the
largest, most complex assignments. Our people are trained on the latest technologies and
safety practices. And, we support them with full-service rotating apparatus services and a
large fleet of service vehicles, ready to respond 24/7.

Corporate Headquarters | 2425 East Pioneer Drive, Irving, Texas 75061 Why trust your vital power services to anyone less? Count on an industry leader. Published by
p. 972.793.5523, 888.SHERMCO | f. 972.793.5542 | www.shermco.com Visit www.shermco.com and learn more today. InterNational Electrical Testing Association
Insulating Oils
Handbook

Published by
InterNational Electrical Testing Association
 THE “GO-TO”
STANDARDS FOR
ELECTRICAL SAFETY
AND RELIABILITY
KNOWLEDGE IS POWER

The ANSI/NETA Standards for

Acceptance and Maintenance Testing
Specifications for Electrical Power
Equipment and Systems!

ANSI/NETA MTS-2011 - New Edition

This standard should always be referenced when
writing maintenance specifications or performing
routine testing on electrical power systems.

When you harness the power of knowledge, you unleash the possibilities. ANSI/NETA ATS-2009
This standard should always be referenced in design
specifications or when performing acceptance testing
The industry’s knowledge leader is your trusted partner. on power system installations. ANSI/NETA MTS-2011

For over 90 years, Doble has partnered with our utility clients to improve ANSI/NETA ETT-2010
operations and optimize system performance. Through the dedication This standard ensures that your acceptance
of world-class staff, Doble provides products and services in 110 and maintenance tests are being preformed by
countries around the globe, delivering the solutions you need – from qualified technicians who are certified in accordance
laboratory testing services to hands-on, technical training seminars. It’s a with ANSI/NETA ETT requirements.
relationship we value and a partnership that will allow us take the industry
to the next level. Together. Available in Bound, CD ROM, or PDF
STANDARD FOR
MAINTENANCE
TESTING SPECIFICATIONS
FOR ELECTRICAL POWER
When you work with Doble’s laboratories, you’re Doble Laboratory Services Include: EQUIPMENT AND SYSTEMS
accessing the collective knowledge of our expert
Dissolved gas analysis
chemists and engineers. Whether you need routine
testing or help developing an entire testing program Furanic compounds in oil Order Your copy of the
for condition assessment, you’ll get the answers Water-in-oil analysis ANSI/NETA Standards Today!
and support you need from a team dedicated to
quality and accuracy. Degree of polymerization
Oil quality screens
Find more information at Metals-in-oil
www.doble.com Polychlorinated biphenyls (PCBs)
Become a NETA Affiliate and Save!
+1 617.393.2997 In-depth, laboratory seminars conducted
labservices@doble.com by Doble’s experts To order:
visit www.netaworld.org
DOBLE IS AN ESCO TECHNOLOGIES COMPANY
Or call 888-300-NETA (6382)
 Insulating Oils
Handbook

Table of Contents
Continuous Moisture-in-Oil Sensing: Field Applications for the Power Industry ..........1
Lance R. Lewand and Paul J. Griffin

Dissolved Gas Analysis — Past, Present and Future ......................................................8

Fredi Jakob, Ph.D.

New Insulating Oils: Alternative to Transformer Oil ................................................12

David W. Sundin, Ph.D.

Using Analytical Techniques to Determine Cellulosic Degradation in Transformers .....15

Lance R. Lewand

Understanding Water in Transformer Systems ..........................................................18

Lance R. Lewand

Which Insulating Oil Analytical Tests to Request and When.......................................22

Lance R. Lewand

Choosing a Sample Container for Transformer Oil Analysis .......................................26

Lance R. Lewand

OCB Diagnostics ...................................................................................................28

Fredi Jakob, Ph.D., Karl Jacob, P.E., Simon Jones, Rick Youngblood, and Alex Salinas

Sampling Transformer Oils —

Part 1 – How and Why to Take a Good Sample ............................................................34
Lance R. Lewand

Sampling Transformer Oils —

Part 2 – Sampling Practices and the Science of Sampling .............................................37
Lance R. Lewand

Sampling Transformer Oils —

Part 3 – Retrieving the Actual Sample ........................................................................40
Lance R. Lewand

Published by

InterNational Electrical Testing Association

3050 Old Centre Avenue, Suite 102, Portage, Michigan 49024
269.488.6382
www.netaworld.org
Insulating Oils Handbook

Table of Contents (continued)

Corrosive Sulfur in Transformer Systems .................................................................43

Lance R. Lewand

Sources of Sulfur in Transformer Systems ................................................................47

Lance R. Lewand

Hot Oil Reclamation: Why Is It Necessary? ..............................................................50

Scott D. Reed

Use of Gas Concentrations Ratios to Interpret LTC Dissolved Gas Data ......................58
Fredi Jakob, Ph.D., Karl Jakob, P.E., Simon Jones, and Rick Youngblood

The Negative Effects of Corrosive Sulfur on Transformer Components ........................62

Lance R. Lewand

Nomograph for LTC-DGA Data Interpretation ..........................................................65

Fredi Jakob, Ph.D., Karl Jakob, P.E., Simon Jones, and Rick Youngblood

Condition Assessment of Oil Circuit Breakers and Load Tap-Changers

by the Use of Laboratory Testing and Diagnostics ......................................................7
Lance R. Lewand and Paul J. Griffin

Natural Ester Dielectric Liquids ..............................................................................73

Lance R. Lewand

Laboratory Testing of Natural Ester Dielectric Liquids ..............................................77

Lance R. Lewand

Gassing Characteristics of Transformer Oil Under Thermal Stress ..............................80

Lance R. Lewand and Paul J. Griffin

Condition Assessment of Transformers — Analysis of Oil Data and Its Quality ............84
Lance R. Lewand

Recent Applications of DGA ...................................................................................86

Karl Jakob, P.E. and Fredi Jakob, Ph.D.

Passivators — What They Are and How They Work ....................................................90

Lance R. Lewand

What is Sludge? ....................................................................................................93

Lance R. Lewand

Metals Analysis in Transformers, Load Tap-Changers and Oil Circuit Breakers .............95
Lance R. Lewand

Testing for Corrosive Sulfur Effects .........................................................................98

Lance R. Lewand
 NOTICE AND DISCLAIMER
NETA Technical Papers and Articles are published by the InterNational Electrical Testing Association. Opinions, views,
and conclusions expressed in articles herein are those of the authors and not necessarily those of NETA. Publication
herein does not constitute or imply any endorsement of any opinion, product, or service by NETA, its directors, officers,
members, employees, or agents (hereinafter “NETA”).
All technical data in this publication reflects the experience of individuals using specific tools, products equipment, and
components under specific conditions and circumstances which may or may not be fully reported and over which NETA
has neither exercised nor reserved control. Such data has not been independently tested or otherwise verified by NETA.
NETA makes no endorsement, representation, or warranty as to any opinion, product, or service referenced in this
publication. NETA expressly disclaims any and all liability to any consumer, purchaser, or any other person using any
product or service referenced herein for any injuries or damages of any kind whatsoever, including, but not limited to,
any consequential, special incidental, direct, or indirect damages. NETA further disclaims any and all warranties, express
or implied including, but not limited to, any implied warranty or merchantability or any implied warranty of fitness for a
particular purpose.
Please Note: All biographies of authors and presenters contained herein are reflective of the professional standing
of these individuals at the time the articles were originally published. Titles, companies, and other factors may have
changed since the original publication date.

Copyright © 2009 by InterNational Electrical Testing Association, all rights reserved. No part of this publication may be
reproduced in any form or by any means, electronic or mechanical, without permission in writing from the publisher.
Insulating Oils Handbook
Continuous Moisture-in-Oil Sensing:
Field Applications for the Power Industry
PowerTest 2000
(NETA Annual Technical Conference)
Presenters
Lance R. Lewand and Paul J. Griffin
Doble Engineering Company
This paper discusses the application of a continuous
moisture-in-oil sensor that operates in dielectric fluids, such
as transformer mineral oils. The sensor can be used to help
in the assessment of the condition of transformers and for
oil processing.
Introduction
The principal reason for developing a continuous mois-
ture-in-oil sensor is that moisture continues to be a major
cause of problems in transformers and a limitation to their
operation. Excessive moisture is particularly problematic in
transformers as it affects the dielectric breakdown strength
of the insulation, the temperature at which water vapor
bubbles are formed and the aging rate of the insulating
materials [1]. In the extreme case, transformers can fail
because of excessive water in the insulation. The dielectric
breakdown strength of the paper insulation decreases sub-
stantially when its water content rises above 2 to 3 % by
weight. The dielectric breakdown voltage of the oil is af-
fected by the relative saturation (RS) of water in oil which
is constantly changing in a transformer environment. The
maximum loading that is possible while retaining reliable
operation (i.e., preventing the formation of water vapor
bubbles) is a function of the insulation water content. For
example, dry transformers (<0.5% water in paper) are much
less susceptible to bubble evolution. In this case emergency
loading at hot-spot temperatures below 180 C may be pos-
sible with little risk of bubble formation. In contrast, a wetter
transformer, with 2.0% moisture in the paper, runs the risk
of bubble formation with hot-spot temperatures as low as
139 C under the same conditions [1]. A more long-term
problem is that excessive moisture accelerates the aging of
the paper insulation, with the aging rate being directly pro-
portional to the water content. That is as the water in paper
content doubles the aging rate of the paper also doubles [2].
For these reasons it is important to have an assessment of the
moisture content of the insulation systems and to maintain
transformers in a reasonably dry state.
1
In addition, the success of oil or transformer drying
(dehydration) techniques is often measured by the amount
of water remaining dissolved in the oil after processing.
The historical approach to the detection of water in these
circumstances has been by the Karl Fischer titration, com-
monly referred to in North America as ASTM Test Method
D 1533B. This chemical analysis has three major drawbacks:
it cannot be developed into a continuous measurement; it
can be plagued with a number of introduced errors most of
which result from sampling and sample containers; and cer-
tain chemical interferences produce erroneous results [3].
Principles of Moisture Detection
The Doble moisture-in-oil sensor (DOMINO ) mea-
sures the capacitance of a thin-film polymer. The capacitance
changes proportionally to the change in RS of water in the
oil. Relative saturation, expressed in units of percent, is the
concentration of water (Wc) in the oil relative to the solu-
bility (S) or concentration of water the oil can hold at the
measurement temperature, as shown in Equation 1.
(Equation 1) RS = Wc /S (100%)
Where: Wc is in ppm wt./wt.
S is in ppm wt./wt.
The sensor converts the measured RS to a concentration,
stated in parts per million in mg/kg (ppm wt./wt.). To be
able to perform this calculation the instrument measures
both the temperature and the capacitance of the thin-film
polymer and converts it to RS. Equation 2 provides the
conversion for mineral oil:
(Equation 2) Log So = -1567/K + 7.0895
Where: So is the solubility of water in mineral oil
K is the temperature in Kelvin ( C + 273)
2 Insulating Oils Handbook
The structure of the moisture-in-oil sensor is shown 14 70

in Figure 1. The moisture-in-oil sensor is composed of an Relative Saturation

upper and lower electrode, a thin-film polymer and a sup-

Water Content, ppm
12 60
Bottom Oil Temperature

port base.

Relative Saturation, % and Water Content, ppm

10 50

Temperature, Degrees C
8 40

Upper
electrode 6 30

4 20

Active thin film

polymer 2 10

0 0

101
105
109
113
117
121
125
129
133
137
13
17
21
25
29
33
37
41
45
49
53
57
61
65
69
73
77
81
85
89
93
97
1
5
9
Hours

Lower
electrodes
Figure 2 — Continuous Moisture Detection in a Transformer

Supporting Excessive moisture in the oil may occur in relatively dry

glass substrate
Figure 1 — Structure of Moisture-In-Oil Sensor
Water vapor penetrates the upper electrode and reaches
the thin-film polymer. The amount of water vapor absorbed
is dependent on the RS of water in the oil. Because the sen-
sor relies on the movement of water molecules to and from
the thin-film polymer, oil flow around the sensor greatly
facilitates this process. Little or no oil flow around the sen-
sor will give low readings below the true water content. The
polymer used for this application is dispersive, that is, its di-
electric constant changes with changing water content. The
capacitance of the polymer, which is the dielectric constant
of the material divided by the dielectric constant of vacuum
(which is 1), is used to determine the RS of water in the
dielectric liquid. The moisture and temperature sensors are
mounted on the end of a probe, which is placed directly in
the oil. A signal cable transmits the capacitance changes
from the probe to a NEMA 4 weatherproof transmitter
transformers as the moisture migrates between the oil and
paper during extreme thermal transients. Moisture is forced
from the solid insulation while increasing in temperature.
Fortunately, the oil can hold more moisture with increasing
temperatures and therefore the percent saturation of water
in oil does not rise excessively. A problem can occur during
cooling because the moisture does not return to the cellulosic
materials quickly enough. The ability of the oil to hold mois-
ture decreases at the cooler temperatures during this cycle,
resulting in a high RS of moisture in oil and reduced di-
electric breakdown voltage [4]. Continuous moisture-in-oil
measurements made in a laboratory model experiment with
very wet paper insulation are shown in Figure 3. Note how
the RS of moisture in oil is driven up during the cooling
cycle and remains high. In effect the moisture is trapped in
the oil, as the diffusion into the paper is so slow. In a similar
model with dry paper insulation the water content as mea-
sured in RS quickly recovered to about the same baseline
value regardless of temperature (Figure 4).
100
%RS
220

housing which contains the electronics. 90 Temp. 200

Measured Relative Saturation in Oil, %

80 180

The Importance of a Continuous 70 160

Temperature, Degree C

Measurement in Transformers 60 140

There are significant advantages of having continuous 50 120

moisture measurements. Moisture is in a dynamic state, 40 100

migrating between oil and cellulosic materials and within 30 80

the cellulosic materials, varying with temperature, thermal 20 60

history, geometry and thickness of the solid insulation, mois- 10 40

ture content, and other factors (Figure 2). Taking a moisture 0 20
sample on a periodic basis without consideration of these 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360

factors can provide a false sense of reliability. Real time rat- Elapsed Time, Hours

ings and condition assessment of transformers requires real

knowledge of the moisture content of the solid insulation.
Figure 3 — Temperature Cycling in Very Wet Insulation Model
Figure 2 clearly demonstrates how the moisture content
in oil from an operating transformer can vary significantly
from hour to hour during the course of a day.
Insulating Oils Handbook 3
90 90 48

44
80 80

40
70 70

Dielectric Breakdown Voltage, kV

36
%RS
Measured Relative Saturation, %

60 60
Temp. 32

Temperature, Degree C
50 50 28

24
40 40

20

30 30
16

20 20 12

8
10 10

4
0 0
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 0
E la p s e d T im e , H o u r s 0 10 20 30 40 50 60 70 80 90 100
Water Content, % RS @ 22 C

Figure 4 — Temperature Cycling in Drier Insulation System Model Figure 6 — Dielectric Strength Versus Relative Saturation (RS)

To properly maintain and operate transformers, an under- The solubility of water in oil is an exponential function
standing of the effects of moisture on the dielectric break- of the oil temperature [4] as shown in Table 1. It is clearly
down strength of the electrical insulating liquids is necessary. apparent that, as the temperature of the oil increases, the
The moisture content reduces the dielectric breakdown volt- amount of water that can be dissolved in the oil increases
age of insulating liquids. It is common practice to measure tremendously.
moisture content as a concentration (ppm). The correlation
between the water content in new, filtered, mineral oils at TABLE 1
room temperature and the dielectric breakdown voltage
using ASTM method D 1816 (0.04 inch gap) is given in Water in Oil Solubility As a Function of Temperature
Figure 4. Of course, the dielectric breakdown voltage will Oil Temperature Water Content Water Content
also be a function of the number and type of particles and in Oil, ppm in Silicone, ppm
their conductivity, not just the water content. 0°C 22 88
Recall that the relative saturation of water in oil is the
water content at a given temperature divided by the solubil- 10°C 36 125
ity of water in the oil at the same temperature. Taking the 20°C 55 174
same dielectric breakdown voltage data as given in Figure 30°C 83 237
5 and converting it to RS provides a much straighter curve 40°C 121 316
except at the extremes. This is shown in Figure 6. It is evident
50°C 173 414
that there is a better correlation between RS and dielectric
breakdown voltage than with moisture concentration and 60°C 242 534
dielectric breakdown voltage. 70°C 331 678
80°C 446 850
90°C 592 1052
48

44

40
100°C 772 1287
Dielectric Breakdown Voltage, kV

36

32
A simple example can be used to illustrate that the di-
28
electric breakdown voltage of insulating oils is proportional
24
to the relative saturation of water in oil rather than the con-
20
centration in ppm. First, a dielectric breakdown test set and
16
cup is placed in a temperature and humidity-controlled envi-
12
ronment. The humidity is controlled so the concentration of
8
water in the oil is held constant at 30 ppm. The temperature
4
starts at 100 C and the first dielectric breakdown voltage
0
measurement is made. At 100 C the solubility of water in
0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 oil is about 772 ppm. The relative saturation of water in oil is
Water Content, ppm. wt./wt.
therefore about 4% and the dielectric breakdown voltage of
a well-filtered oil would be quite high. Now the temperature
Figure 5 — Dielectric Strength Versus Water Content of the chamber is reduced to room temperature or about
 4 Insulating Oils Handbook
22 C. The solubility of water in oil is about 60 ppm and the The Importance of an In-line
relative saturation is 50%. The dielectric breakdown voltage Moisture Measurement in Oil Processing
would be expected to be about one half of what it was when
the relative saturation was very low. If the temperature is Applications
cooled to 0 C and the dielectric breakdown voltage mea- DOMINOTM can be used for continuous monitoring
surement is made, the results should be quite low because of oil processing applications. It was determined that the
the solubility of water in oil at this temperature is about 22 following attributes would be required in a moisture-in-oil
ppm. As the water content in the oil is higher than this, sensor used for these applications:
the water forms an emulsion (the oil will appear cloudy2
with the suspended water) and begins to form condensation Provides a fast, continuous measurement
(Figure 7). During all this time the concentration of water
in oil has not changed (Table 2). Have the ability to withstand high flow rates
Can be mounted directly in the oil processing equip-
ment
Performance is not affected by pressure, vacuum, and
temperature
Water Content, 30 ppm Water Content, 30 ppm
Can be adapted to a variety of systems

Temperature, 50°C, Low RS Temperature, 0°C, High RS The application of a continuous moisture-in-oil sen-
Visual Appearance, Clear Visual Appearance, Cloudy
sor would greatly benefit the user in determining the end
point of any oil processing activity involving the removal of
dissolved water from the oil. However, additional benefits
could also be gained such as:
Determination of filter bleed-through
Figure 7 — Low RS Oil Versus High RS Oil Elimination of taking oil samples for water content by
chemical analysis
Determination of oil processing efficiency
TABLE 2 Determination of the amount of water removed by
Relationship Between Dielectric Strength filtering and/or dehydrating processes
and Water Content Determination of the effectiveness of field-drying a
transformer
Oil Water Content, Relative Dielectric
Temperature ppm Saturation Strength Determining the proper time to change filter cartridges
100°C 30 3.9% High due to filter bleed-through is a concern during transformer
22°C 30 50% About half
oil processing. Changing them is usually based on dif-
of high ferential pressure or a pressure increase, which can result
from excessive moisture, or particulate contamination.
0°C 30 >100% Very Low
This is often late in the process, when high amounts of
dissolved water is already in the processed oil. The purpose
Transformers are more complicated systems than given
of a continuous in-line moisture-in-oil sensor would be to
in this simple example. However, the same basic principles
accurately measure the amount of water in the processed
apply for the dielectric breakdown strength of the liquid
oil, clearly indicating when filters should be changed, and
dielectric. That is, it remains a function of the relative satu-
therefore speeding up the drying process. In most cases, a
ration of water in the oil. During the cool-down cycle of
moisture-in-oil sensor would indicate the need for a filter
a thermal transient in a transformer some of the moisture
change long before a pressure increase criterion would be
returns to the paper and some of the moisture remains in the
reached. Typically the water content in ppm remains fairly
oil. The relative saturation of water remaining in the oil will
constant at some low value while the filter is most effec-
influence its dielectric breakdown voltage. For this reason
tive, and then starts to increase gradually. A reasonable end
continuous measurement can be used to detect conditions
point can be chosen to change the filter before it becomes
in which moisture distribution is a problem.
ineffective. Monitoring of the differential pressure is still a
valuable practice as filters can be plugged from particulate
contamination and free water.
One of the most beneficial results from the use of an
in-line continuous moisture-in-oil sensor is the elimination
of the need to take periodic samples for Karl Fischer titra-
tion measurements (and the associated chemicals and their
Insulating Oils Handbook 5
disposal). The algorithm used in this particular moisture- appropriate actions can be taken to improve it. The moisture-
in-oil sensor converts relative saturation of water in oil and in-oil sensor can be used in a similar fashion to determine
temperature into the concentration of water in ppm, and the best operating conditions for the use of filters.
was determined initially by using the Karl Fischer titration. Calculating the total amount of water removed can be a
Figure 8 demonstrates how closely the two measurements useful measure of dehydrating effectiveness. It is quite simple
are related to each other. for oil in a storage tank. For instance, using Equation 3, it
can be determined that a filtering or dehydrating process
60
which removes 35 ppm of water (difference between inlet
and outlet measurements) on 10,000 gallons of oil would
50 remove 1.2 liters of water.
(Equation 3) Wt = 3.31X 10-6 Voil (Ci –Co)
40
Water Content, ppm

Where:
Wt = The total amount of water removed in liters
30

20
Voil = The volume of oil in gallons
Ci = The inlet concentration of water in oil in ppm wt./wt.
10
from the moisture-in-oil sensor
Co = The outlet concentration of water in oil in ppm
0 wt./wt. from the moisture-in-oil sensor
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

In the example the calculation would be as follows:

Samples
DOMINO Karl Fischer Titration

Figure 8 — Moisture-In-Oil Sensor Versus Karl Fischer Titration Wt = 3.31X 10-6(10,000gallons)(40–5)=1.2 liters of water

The accuracy of this particular sensor in ppm varies with
the temperature of the measurement because it is a conver-
sion from RS and not a direct ppm measurement. Table 3
below provides some useful values. For very aged oils the
relationship between RS and concentration (ppm) may need
to be determined experimentally.
TABLE 3
Accuracy of Moisture-In-Oil Sensor
TEMPERATURE % Accuracy, RS
0°C ± 1%
10°C ± 1%
20°C ± 1%
40°C ± 1%
60°C ± 1%
80°C ± 1%
The moisture-in-oil sensor can be used to estimate the
efficiency of an oil dehydrating process using vacuum or
filters. In the case of vacuum dehydrating, the inlet and outlet
oil can be monitored to determine the best flow rate, dwell
time, applied vacuum, temperature, and if other mechanical
aspects of the process are functioning adequately. This can
be done by comparing the difference in the water content of
the oil at the inlet and outlet of the processor. The greatest
difference in water contents will indicate optimum efficiency.
Once the most efficient processing conditions have been set,
the moisture-in-oil sensor can then be used to continuously
monitor the process. If the efficiency begins to decline, then
The effectiveness of field drying of the transformer,
whether the process is conducted out of service or in-ser-
vice, is much more difficult to assess than drying the oil
alone. This is because the large mass of cellulosic insulation
(paper, pressboard, wood structural members, etc.) contains
almost all of the water in a transformer and the moisture
distribution is uneven and unknown. A reasonable estimate
can be produced before and after processing if the following
variables are known:
Volume of oil
Accuracy, PPM Mass of paper
± 0.25
Concentration of water in ppm at a known temperature
± 0.40 (preferably one that is > 50 C)
± 0.60
At higher constant temperatures it is possible to estimate
± 1.25 the average moisture content of the cellulosic insulation
± 2.50 [6,7]. For example, if a transformer has 10,000 gallons
± 4.50 (37,850 liters) of oil, 9125 pounds (4148 kg) of paper, and
30 ppm of water in oil while maintained at 70 C (RS =
9.0), then the calculated amount of water in the oil and
paper is as follows:
The oil contains about 1.0 liter of water
The paper contains about 78.5 liters of water
The entire system would have an estimated total volume
of approximately 80 liters
The amount of water in oil was determined by Equation
3, where:
Ci = the water-in-oil content
Co = 0
6 Insulating Oils Handbook
The amount of water in the paper is estimated by multi-
plying the mass of paper times the concentration of water
in the paper in percent.
If the amount of paper is unknown it can be estimated
from the following:
Shell-form transformers, the paper mass = oil mass/3
Core-type transformers, the paper mass =oil mass/8
Core type transformers with enameled wire insulation,
the paper mass = oil mass/20
The amount of water in the paper can be calculated by
two means, using an equation or a family of curves. The
equation is as follows:
(Equation 4)
Cp = {2.17x10-5x[(RStox10-2)(Ps)]0.6685}xe4726 (Tto+273)
Where:
Cp = moisture content in the paper in %
RSto = relative saturation of water in oil at temperature,
Tto
Ps = The saturation vapor pressure at temperature, Tto
(from published data, see
Reference 8). The values are given in mm of Hg and need
to be converted to
atmospheres by dividing by 760 mm Hg/atmosphere.
e = natural log
Tto = top oil temperature
The amount of water was calculated as 1.9% of the mass
of cellulosic materials, which gives a mass of water of 173
pounds (78.5 kg). This mass can then be easily converted
to liters of water by multiplying by 0.454 liters/pound for a
volume of 78.5 liters. The other way to calculate the amount
of water in paper is to use equilibrium isotherm curves,
taking the concentration of water in oil and the top oil
temperature to find the equilibrium value [4].
Once the drying process has been completed the same
estimation process for moisture in the cellulosic materials
can be repeated. This should be done after the transformer
has been at an elevated temperature for at least three days
without processing. The difference in the amount of water
before and after filtering is the quantity removed.
Another way to determine how much water is removed
by processing each day is to use Equation 3. The flow rate
can be used to determine the volume of oil processed, and
the inlet and outlet water contents (while fairly constant)
provides the concentration of water removed. If the inlet
water-in-oil content changes significantly then the calcula-
tion should be restarted and the volumes of water removed
summed for each set of calculations. The efficiency of the
process can readily be observed by examining the difference
between the inlet and outlet water content [7].
In the example above, the paper contained 1.9% water
or 78.5 liters at the start of the processing. Ideally the water
content of the paper would be reduced to a final concentra-
tion of 0.5% or lower. This would require removal of at least
58 liters of water.
Conclusions
A new moisture-in-oil sensor, DOMINOTM, has been
developed to provide continuous in-situ measurements of
dissolved water in oil. The measurement of the capacitance
of a thin-film polymer is used to detect the relative satura-
tion (RS) of water in oil. The relative saturation of water in
oil along with the measured value for temperature are used
to calculate the concentration in ppm (wt./wt.).
Current applications in which this moisture-in-oil sensor
can be used are:
Continuous moisture measurement in transformers
On-line/off-line oil processing applications
Determination of filter bleed-through
An alternative to taking oil samples for water content
by chemical analysis
Determination of oil processing efficiency
Determination of the amount of water removed by
filtering and/or dehydrating processes
Determination of the effectiveness of field-drying a
transformer
References
[1] Griffin, P. J. “Water in Transformers – So What!”,
National Grid Condition Monitoring Conference,
May 1996.
[2] Lewand, L. R. and Griffin, P. J., “How to Reduce the
Rate of Aging of Transformer Insulation”, NETA
World, Spring 1995, pp. 6-11.
[3] Lewand, L. R. and Griffin, P. J., “Transformer Case
Studies”, Proceedings of the Sixty-Six Annual Interna-
tional Conference of Doble Clients, 1999, Sec.5-7.
[4] Griffin, P. J. “How to Prevent Rain in Power Trans-
formers”, ASTM Standardization News, November
1991, ASTM, Philadelphia, PA, pp. 30-33.
[5] Griffin, P. J., Bruce, C. M., and Christie, J. D. “Com-
parison of Water Equilibrium in Silicone and Min-
eral Oil Transformers”, Minutes of the Fifty-Fifth An-
nual International Conference of Doble Clients, 1988,
Sec. 10-9.1.
[6] “Estimating the Water Content of Cellulosic In-
sulation”, MKT-AB-12, Rev A, Doble Engineering
Company DOMINOTM Application Bulletin, No-
vember, 1999, 4 pp.
[7] “Transformer Oil Field Processing Applications”,
MKT-AB-16, Rev A, Doble Engineering Company
DOMINOTM Application Bulletin, November, 1999,
6 pp.
Insulating Oils Handbook 7
[8] “Vapor Pressure of Water Below 100 C” In CRC
Handbook of Chemistry and Physics, Ed. Robert C.
West, Boca Raton, FL. CRC Press Inc., 1978-1979,
P. D-232.

Lance Lewand received his Bachelor of Science degree from St.

Mary’s College of Maryland in 1980. He has been employed by the
Doble Engineering Company since 1992 and is currently the Laboratory
Manager for the Doble Materials Laboratory and Product Manager for
the DOMINO®. product line. Prior to his present position at Doble, he
was Manager of the Transformer Fluid Test Laboratory and PCB and
Oil Services at MET Electrical Testing in Baltimore, MD. Mr. Lewand
is a member of ASTM Committee D 27.

Paul J. Griffin received his BS degree at the American International

College and his MS at the University of Rhode Island. He has been em-
ployed by the Doble Engineering Company since 1978 and is currently
Vice President of Laboratory Services. He is secretary of the Doble Oil
Committee, a member of ASTM committee D 27, US Technical Advisor
to IEC TC10 for Fluids for Electrotechnical Applications, member of
the IEEE Insulating Fluid subcommittee of the Transformer committee,
and a member of the CIGRE Working Group 15.01 Fluid Impregnated
Insulating Systems.
8 Insulating Oils Handbook
Dissolved Gas Analysis —
Past, Present and Future
PowerTest 2000
(NETA Annual Technical Conference)
Fredi Jakob, Ph.D.
Analytical ChemTech International, Inc.
I. Introduction
It has been well over thirty years since dissolved gas analy-
sis, DGA, was introduced as a diagnostic tool for monitoring
mineral oil filled transformers. It is now universally accepted
as the method of choice to locate incipient thermal and
electrical faults. DGA methodology and applicability have
evolved significantly since its inception. The evolutionary
development includes new laboratory methods, on-line
DGA, application to additional types of fluid filled equip-
ment, application to dielectric fluids other than mineral oil
and new diagnostic interpretation protocols.
At this time the ASTM has approved two procedures
for laboratory DGA and a third method is on the verge of
approval. Devices, which periodically or continuously moni-
tor one or more gas in oil concentrations, are available for
on-line analysis. DGA was originally developed for trans-
formers, but it is now being applied to load tap changers, oil
filled circuit breakers, oil filled bushings and cable oil. Data
interpretation methods have been extensively developed for
transformers filled with mineral oil and IEEE is currently
developing a guide for silicone fluid filled transformers.
II. Laboratory Methods
ASTM procedure, D 36121, provides two different
methods for the extraction of dissolved gases from dielectric
fluid. Method D 3612-A involves an extraction process that
precedes the subsequent gas analysis. The dielectric fluid is
delivered into a previously evacuated glass apparatus and is
stirred for a sufficient time to extract most of the dissolved
gases. The volume of the evacuated glassware must be large
in comparison to the volume of dielectric fluid so that most
of the dissolved gases leave the fluid phase. The evolved gas is
compressed and the volume and temperature are measured.
Then, one or more aliquots of the gas are separated and
quantified with a gas chromatograph. Each gas concentra-
tion in the oil is reported in ppm by volume. ASTM 3612-A
is time consuming and labor intensive, but it has low detec-
tion limits and generates reproducible data. Mercury is used
in the extraction apparatus so precautions must be taken to
minimize worker exposure to mercury vapors.
ASTM D 3612-B uses direct injection of an oil sample
onto a heated gas stripping column that is in series with a
separation column, both of which may be contained within a
single gas chromatograph. The advantages of this method are
speed and elimination of potential mercury health hazards.
One disadvantage is that method B cannot be used with
Silicone fluids because excessive foaming will damage the
separation columns. Also, since fault gases are very soluble
in mineral oil the extent of gas extraction may not be as
complete as in the two stage method. The ASTM docu-
ment states, “The limit of detection for hydrogen specified
in Method B is higher than that specified for Method A.
This could effect the interpretation of results when low levels
of gases are present”.1 ASTM D 3612, which contains a
comparison of the detection limits of methods A and B, is
summarized in Table 1.
Table 1
Minimum Gas Detection Limits, ppm vol1
Gases Method D 3612-A Method D 3612-B
Hydrogen 5 20
Hydrocarbons 1 1
Carbon Oxides 25 2
Atmospheric Gases 50 500
New DGA methodology that involves a two stage pro-
cess2 has been developed and is being reviewed by ASTM
committee D-27 (Proposed Designation D 3612-C). The
first stage of the proposed “headspace” method involves a
partitioning of dissolved gases between oil and an inert gas,
argon. Oil samples are introduced into a sealed container
that has been purged with argon. Samples are then heated
and vigorously agitated for an extended period of time.
Ostwald partition coefficients are used to correlate gas
Insulating Oils Handbook
concentrations in the headspace of the sample container
with the initial gas concentrations in the oil. The second
stage involves analysis of the evolved gases with a gas
chromatograph. The advantages of this headspace method
include automation of the extraction and analysis process
and elimination of mercury. Detection limit comparisons
between the headspace method and D 3612-A are being
developed. One area of concern is potential variability of
Ostwald coefficients with oil type, though most believe the
difference of results due to this factor is not significant.
III. On-Line DGA
In-time maintenance of electrical equipment, performing
maintenance on only those units that require it, is a desirable
goal. In order to fully implement this objective, equipment
would have to be effectively monitored on an ongoing basis.
Since DGA is accepted as an excellent diagnostic monitor-
ing tool a great deal of effort has been expended to develop
on-line DGA. The two steps required to implement this
are automated separation of the fault gases from the oil
followed by quantitative analysis. One or more gases can
be extracted from the oil and subsequently analyzed with
a chromatograph, an infra-red spectrometer or a mass
spectrometer. Separation of the gases from the oil can be
achieved with a semi-permeable membrane or a mechanical
device. The small mobile hydrogen atoms, and to a lesser
extent other small molecules such as CO, are most readily
separated. The hydrogen can then be measured. Fortunately,
hydrogen is formed with any type of fault, partial discharge,
heating or arcing and the measurement of this gas can be
used to continuously monitor the condition of a transformer.
Hydrogen detection methods are best used as trigger devices
to indicate when a laboratory DGA is appropriate.
Mechanical extraction devices for separation of gases
from oil have also been developed. Oil is fed into a cylinder
fitted with a piston that is moved up and down. On the
down stroke headspace is created in the cylinder and the
gases partition between the oil and the headspace. During
the compression stoke the gas is pushed into a reservoir
and the cycle is repeated. This multiple stage extraction
procedure is potentially more effective than a single stage
extraction such as that used in ASTM D 3612-A . The
extracted gases can then be separated and analyzed by gas
chromatography or quantified without prior separation by
infra-red or mass spectroscopy .
On-line DGA is theoretically feasible but reliability,
detection limits and economic issues must still be resolved
before the method is widely implemented. Detection of a
limited number rather than all of the gases may be a suf-
ficiently viable compromise.
IV. Applicability of DGA
A. Fluid Types
Thermal or electrical faults release energy that will result
in the partial molecular destruction of dielectric fluids. The
extent of the molecular rearrangement depends on the
available energy and the nature of the dielectric fluid. Faults
9
will produce gases in mineral oils, high molecular weight
hydrocarbons, PCB’s, silicone fluids, perchlorethylene and
other dielectric fluids. Since DGA interpretation is empiri-
cal in nature it has only been effectively applied to dielectric
fluids that have been extensively studied. The current IEEE
guide3, 57.104, covers the application of DGA to mineral
oils only. High molecular weight hydrocarbons, also known
as less flammable hydrocarbons, contain similar molecules
to those present in conventional mineral oil and produce
the same gases under similar fault conditions. The IEEE
guide is applicable, without modification, for these fluids.
Silicone fluids produce the same fault gases as mineral oils
when they are thermally or electrically stressed, but relative
concentrations and fault gas concentration ratios are dif-
ferent. A separate IEEE trial-use guide, P1258, has been
developed for these fluids4. IEEE guides for the other fluids
are not currently available.
B. Equipment Types
DGA was initially developed to monitor transformers
and the success of this method is well known. Fault condi-
tions can exist in any type of electrical equipment and if
this equipment contains dielectric fluids, fault gases will be
produced. Table 2 shows the correlation between fault types
and the various fault gases.
Table 2
Fault Gases
Gases Indication
Hydrogen Partial discharge, heating, arcing
Methane, Ethane, Ethylene “Hot Metal” gases
Acetylene Arcing
Carbon Oxides Cellulose insulation degradation
1. Load Tap Changers
When a load tap changer, LTC, operates arcing occurs
and the expected fault gases, acetylene and hydrogen, are
produced. One might initially assume that the presence
of these gases masks the gases produced by other faults.
Coking and misalignment of contacts are the most com-
mon problems that occur in LTCs. Coking is a cumulative
problem that starts with an initial deposition on the contact
surfaces, which results in increased contact resistance, fol-
lowed by additional carbon build up on the contacts. This
process leads to exponentially increased heating or “thermal
runaway” and carbon build up. Youngblood 5 was one of the
first investigators to realize that the coking problem would
result in the production of the “hot metal gases,” methane,
ethane and especially ethylene. The concentration of these
gases depends on a number of variables including breath-
ing type, manufacturer, model type, etc. Generic fault gas
threshold values, similar to those in the IEEE guide for
transformers, have been developed by Youngblood and are
given in Table 3.
10 Insulating Oils Handbook

Table 3 2. Oil Filled Bushings

LTC Monthly Watch Criteria 5 John Stead presented a paper at the 1996 Doble confer-
ence7 showing how DGA could resolve conflicting Power
LTC Type Hydrogen Acetylene Ethylene Factor results for two bushings from the same manufacturer.
Free or Desiccant >1500 ppm >1000 ppm >1000 ppm The DGA data for these two bushings is given in Table 5.
Breather This data clearly indicates that partial discharge is occurring
Sealed >5000 ppm >9000 ppm >1200 ppm in bushing 2. The low ratio of CO2 / CO indicates severe
Vacuum >10 ppm >5 ppm > 100 ppm overheating of the paper in the bushing. Physical inspection
confirmed the interpretation of the DGA results.

Charles Baker6 and others have developed manufacturer Table 5

specific flag points and this approach, which is illustrated
in Table 4, is probably the most promising. DGA Data for Two Bushings, ppm7
Gas Bushing 1 Bushing 2
Table 4 Hydrogen 1705 19132
Equipment Specific Action Levels Oxygen 5546 4041
(McGraw Edison LTC 550)
Nitrogen 68216 50767
Hydrogen Methane Ethane Ethylene Acetylene CO CO2 Carbon Monoxide 441 537
LT1 100 100 100 500 100 100 150 Methane 146 1256
LT2 250 200 200 1200 200 500 300 Carbon Dioxide 710 1459
LT3 500 400 400 2000 400 1000 3000 Ethylene 1.8 11
Ethane 71 409
LT1 = Abnormal LT2 = High LT3 = Very High
Acetylene <0.1 0.2
A typical LTC case history is documented below:
Removal of oil from a bushing is not encouraged by most
AC TLH-21 138KV x 12KV 50 MVA Free Breather manufacturers, but the practice is common in Europe.
Date: February 25, 1993
3. Oil Filled Circuit Breakers
Date Mfr. Serial Number C2H2 CH4 C2H6 C2H4 H2 CO CO2
Application of DGA to oil circuit breakers, OCB’s, is
02/25/93 AC 018226580301 0 5 1 4 34 71 350 under investigation by ACTI in cooperation with several
utilities. Most OCB’s are free breathing and fault gas con-
This unit was determined to be operating properly. The centrations could be very dependent on the sampling time.
low concentrations of hydrogen and acetylene are considered Other tests such as particle size distribution, particle types
normal for a free breathing unit. The unit was scheduled and metals in the oil are being evaluated for the identifica-
for annual testing. tion of problem units.
Date: February 25, 1994
V. Supplementary Tests
Date Mfr. Serial Number C2H2 CH4 C2H6 C2H4 H2 CO CO2 Carbon oxides are produced by the degradation of cel-
02/25/94 AC 018226580301 44 1812 576 3143 149 33 645 lulose insulation or when oil is heated in the presence of
oxygen. The extent of cellulose degradation is a critical fac-
This unit was in “thermal runaway” when tested. Notice tor in estimating the condition and life expectancy of solid
the high level of ethylene, which is the key gas for overheat- insulation. Levels of CO and CO2 and the ratio CO2 / CO
ing. This unit was already heavily coked when the DGA can be used to indicate when further investigation of the
test was conducted. The unit was removed from service cellulose condition is warranted. The two available methods
and repaired. The reversing switch and some moveable dial are Degree of Polymerization, DP, and Furanic compound
contacts were replaced. analysis. DP is more definitive, but this invasive procedure
requires removal of paper samples from the equipment. Fu-
Date: February 27, 1995 ran analysis is a non-invasive procedure and is recommended
if the DGA results indicate cellulose decomposition.
Date Mfr. Serial Number C2H2 CH4 C2H6 C2H4 H2 CO CO2 A very interesting example of the effect of transformer
02/27/95 AC 018226580301 55 9 2 11 22 33 440 overheating was provided to ACTI by Mr. Charles Baker,
South Carolina Electric and Gas Co8. The transformer in
The unit is operating normally after completion of the question 33/13.8 kV was run for 4 days without operation
repairs. The LTC was placed on a six month test interval of the fans and pumps. The main winding temperature of
based on its previous failure history. this three and a half year old transformer reached 1500 C.
Insulating Oils Handbook 11
Rates of cellulose degradation double for each 6-8 degree 7. Stead, J. and Jakob, F., “Use of DGA to Confirm Un-
rise in temperature. If we assume a normal winding tem- satisfactory Doble Test Results of 115 KV Bushings”,
perature of 900, the sixty degree rise in temperature would Minutes of the Sixty-third Annual International Con-
correspond in a 210 (1024) increase in the rate of cellulose ference of Doble Clients, 1996, Sec. 3-4.
decomposition. Four days at the elevated temperature would 8. Charles Baker, Private Communication.
correspond to 4096 days, 11.2 years of normal operation. The
transformer insulation would then be 14.5 years old. The
laboratory analysis showed 0.45 ppm of 2-furfuraldehyde
which correspond, using the Chendong equation, to a DP of Dr. Fredi Jakob received his PhD at Rutgers, the State University of
New Jersey, in 1961. He is professor emeritus of analytical chemistry at
530 and an apparent operating time of 25 years. There was California State University-Sacramento and is the founder and laboratory
no additional data to indicate if the transformer was over- director of Analytical ChemTech International, Inc. (ACTI), which is a
heating during the first three and a half years of operation. wholly owned subsidiary of Weidmann Systems International. As a long-
Baker has recently reported that the transformer failed. term member of ASTM and IEEE and author of over fifty published
articles, Dr. Jakob is a traveling lecturer to private and governmental
agencies. He has been invited to speak at American Public Power meet-
VI. Future Developments ings, ASTM symposia, conferences held by Doble, NETA, and AVO
Proposed changes in DGA laboratory procedures have conferences, as well as other industrial organizations.
been noted above. IEEE Guide 57.104 was last revised in
1991 and is currently being revised once more. The commit-
tee is proposing a two step process for utilization of DGA
laboratory data. The current thinking is that there should be
one set of criteria for interpretation of the first DGA result
(proposed Table 1) and a second set of criteria for subsequent
tests (proposed Table 2). Current deliberations include the
concentrations of fault gases that will be classified as “nor-
mal” and levels that lead to a recommendation to monitor
the transformer at shorter intervals. Once it is determined
that one or more gases have exceeded these “normal” levels
then the rates of fault gas generation should be determined.
Table 2 will then tabulate IEEE recommendations for sub-
sequent test intervals and operational procedures.

References
1. ASTM, Annual Book of ASTM Standards, Volume
10.03, D 3612-96.
2. Jalbert, J. and Gilbert, R. “Comparison Between Head-
space and Vacuum Gas Extraction Techniques for the
Gas Chromatographic Determination of Dissolved
Gases from Transformer Insulating Oils”, 1994 Inter-
national Symposium on Electrical Insulation, Pitts-
burgh, PA, 1994.
3. IEEE C57.104-1991, “IEEE Guide for the Interpreta-
tion of Gases Generated in Oil-Immersed Transform-
ers”, 1991.
4. IEEE P1258, “IEEE Trial-Use Guide for the Inter-
pretation of Gases Generated in Silicone-Immersed
Transformers”, 1995.
5. Youngblood, Rick, et. al, “Application of DGA to De-
tection of Hot Spots in Load Tap Changers”, Minutes
of the Sixtieth Annual International Conference of
Doble Clients, 1993, Sec. 6-4.
6. Charles Baker, Private Communication.
12 Insulating Oils Handbook

New Insulating Oils:

Alternative to Transformer Oil
NETA World, Spring 2001

by David W. Sundin, Ph.D.

Kielectric Systems, Inc.

The choices confronting insulating oil users have never development of nonflammable, chlorinated fluids. Taken
been greater. Where conventional mineral oil has been the together, these nonflammable fluids are known as “askerals.”
staple diet of transformers for over a hundred years, today’s These fluids cooled and insulated well and allowed further
technology and market demands have produced a variety of use of transformers inside or near buildings because of their
new fluid options. This article discusses the new insulating nonflammable character.
fluids on the market and describes how each can be used But there were also problems with these fluids. Environ-
to solve specific problems. mental studies showed that askarel fluids were extremely
persistent in the environment. In addition, studies showed
It used to be easy.... that if the fluids were exposed to an electric arc or ex-
tremely high temperatures, highly toxic chemicals would
The first closed-core transformers were developed at the
be formed.
Abraham Ganz foundry in Budapest in 1884. These origi-
The use of askerals was phased out worldwide in the
nal units were air cooled. The drawbacks of using air as a
1970s and 1980s. Several fire-resistant fluids were developed
dielectric and cooling medium were soon apparent, and in
to take their place. The two main types of fire resistant fluids
1890 the first oil-cooled, oil-insulated transformer was build
used in the US are fire-resistant petroleum and silicone
by Brown, GmbH, in Germany. Since that time, napthenic
fluid. An example of fire resistant petroleum oil is shown
petroleum oil has been the standard cooling and insulating
in Table One.
medium in liquid-filled transformers. Oils with a napthenic
molecular structure were preferred over those with paraffinic
structure because they had better low temperature behavior Table One
and because napthenic oil was found in shallower wells, Typical Characteristics of Fire Resistant Petroleum Oils
which made it easier to extract from the ground. Characteristic & ASTM method Beta fluid Transformer
The specific petroleum fractions chosen for use in trans- Oil
formers were relatively light. This facilitated cooling flow Fire Point, ASTM D92, Deg.C 306 145
inside the transformer and the use of transformers in cold
Viscosity, D88, cSt. @ 100C 11.7 3.0
environments.
Napthenic petroleum continues to be the standard for Density @ 20ºC, g/cc 0.87 0.87
insulating oil used in the United States. In other parts of the Color, ASTM units L0.5 L0.5
world, scarcity of napthenic crude oil sources and improved Appearance Clear Clear
dewaxing processes have shifted the markets to the use of
Dielectric Breakdown, D877 55 55
paraffinic oils, but the characteristics of the final product
are roughly the same. Dissipation Factor, 40ºC, D924, % 0.1 0.1
Acid Value, D664, mg KOH/g 0.01 0.01
But then we wanted fire resistance…
In the early part of the 20th century, one of the disad-
vantages of using light petroleum fractions as insulating
fluids was their flammability. This was overcome with the
Insulating Oils Handbook 13

And now there are new choices…. Table Two

With the development of fire resistant fluids, the cat Typical Characteristics of ECO Fluid
was out of the bag. One size of transformer oil no longer Characteristic & ASTM method ECO fluid Transformer
fit everyone. Transformer users recognized that different Oil
characteristics of dielectric fluids could solve specific prob- Biodegradation, CEC-L33 >95% ~30%
lems that they encountered, for example:
Viscosity, D88, cSt. @ 100ºC 1.8 3.0
• Equipment operating at high temperatures Density @ 20ºC, g/cc 0.85 0.87

• Equipment operating in low ambient temperatures Color, ASTM units L0.5 L0.5
Appearance Clear Clear
• Equipment situated in environmentally sensitive areas.
Dielectric Breakdown, D877 59 55
In the past decade, new insulating fluids have been de- Dissipation Factor, 40ºC, D924, % 0.1 0.1
veloped to address each of these problems. These fluids are Acid Value, D664, mg KOH/g 0.01 0.01
grouped together under the heading of “functional fluids,”
as their characteristics influence the performance of the
equipment in which they are used. Functional fluids will Equipment at high temperatures
probably not replace the use of conventional transformer
mineral oil, at lest in the foreseeable future, but they are Specialty transformers are now being designed to operate
proving themselves to be an important tool to help equip- with much higher heat rise than would have been accept-
ment makers achieve specific performance. able a few years ago. Upgraded paper insulation systems can
withstand hot spot temperatures that would have severely
degraded old style Kraft insulation. Synthetic fluids are
Highly biodegradable fluids available that resist oxidation and aging better than con-
Because of rising environmental liability and spill cleanup ventional mineral oil, and they have enhanced heat transfer
costs, power utilities have requested transformer and fluid characteristics. Using these fluids in transformers that were
manufacturers to provide alternative dielectric fluids that built for mineral oil can yield temperature decreases of up
are more environmentally friendly than conventional to five degrees centigrade with no additional cooling sur-
transformer oil. face area. Because the synthetic oils are more stable than
In response to these requests, fluid manufacturers have mineral oil, service life can be extended. In addition, these
developed several types of highly biodegradable fluids. Some fluids have a significantly higher fire point than conventional
are made from synthetic hydrocarbons, some from vegetable transformer oil, which provides a higher safety margin at
oils, and some from synthetic chemicals called “esters.” these elevated operating temperatures. The characteristics
Most of these new fluids biodegrade far more completely of a fluid made for high temperatures are shown in Table
and rapidly than conventional transformer oil. For example, Three.
several of the “highly biodegradable” oils biodegrade more
than 95 percent when tested with standard methods. In Table Three
contrast, conventional transformer oil biodegrades to 30–50
Typical Characteristics of Alpha-2 Fluid
percent when tested with the same methods.
The different types of fluids available vary widely in Characteristic & ASTM method Alpha-2 Transformer
characteristics and price. This field is still developing, and fluid Oil
there have not been standards developed for fluid character- Fire Point, ASTM D92, Deg.C 250 145
istics, acceptable performance levels, or even test methods Viscosity, D88, cSt. @ 40ºC 3.9 3.0
to determine minimum levels of biodegradability. To date, Density @ 20ºC, g/cc 0.82 0.87
none of the fluids has found sufficient acceptance or use to
become a standard for highly biodegradable fluids. Color, ASTM units L0.5 L0.5
One example of a highly biodegradable oil is ECO Fluid, Appearance Clear Clear
which is made from synthetic hydrocarbon oils. Typical Dielectric Breakdown, D877 57 55
characteristics of ECO Fluid are shown in Table Two. Dissipation Factor, 40ºC, D924, % 0.1 0.1
Acid Value, D664, mg KOH/g 0.01 0.01
14 Insulating Oils Handbook

Low ambient temperature applications Conclusions

At temperatures below -45ºC, conventional transformer We have looked at the trend in insulating oil develop-
oils become too thick to effectively circulate in a transformer. ment: the realization that dielectric fluid characteristics can
Waxes in the oils form a crystalline structure, impeding help achieve specific performance attributes in equipment.
oil flow in the cooling circuit. Without flow, hot spots can Some of the newer applications are in environmentally sen-
develop in the transformer’s core, even at the lowest ambi- sitive areas, in low ambient temperatures, and in equipment
ent temperatures. These hot spots degrade insulating paper that operates at higher than normal temperatures.
and significantly shorten transformer life. Synthetic oils New applications will continue to drive the develop-
are now available that can remain fluid down to -65ºC. ment of improved functional fluids. The next century will
This effectively prevents this problem and greatly extends see changes in our industry that require highly specialized
the ambient temperature range in which transformers can fluids that will be used in equipment that we can only now
be situated. imagine.
OptiCool fluid is an example of an insulating oil made for
use at very low temperatures. Blended from synthetic oils,
it is stable at temperatures from -65º to 120ºC. OptiCool Dr. David Sundin is President of Dielectric Systems, Inc., a manu-
is miscible and compatible with conventional transformer facturer of fire-resistant and other specialty dielectric materials. He
is responsible for research and development on Dielectric System’s
oil. The Characteristics of OptiCool fluid are compared with transformer insulating fluids in addition to overseeing production and
those of conventional transformer oil in Table Four. marketing activities. Prior to joining DSI he held engineering and man-
agement positions at a major transformer manufacturer and was chief
chemist at a large oil refinery. Dr. Sundin is active in various industry
Table Four standards committees. He holds a BA in chemistry, an MBA, and a PhD
Typical Characteristics of OptiCool Fluid in Engineering. He is recognized expert in electrical insulating fluids and
fire resistance in transformers; he has presented his research in profes-
Characteristic & ASTM method OptiCool Transformer sional forums worldwide.
fluid Oil
Pour Point, ASTM D97, ºC -65 -40
Viscosity, D88, cSt. @ 100C 1.7 3.0
Density @ 20ºC, g/cc 0.82 0.87
Color, ASTM units L0.5 L0.5
Appearance Clear Clear
Dielectric Breakdown, D877 59 55
Dissipation Factor, 40ºC, D924, % 0.1 0.1
Acid Value, D664, mg KOH/g 0.01 0.01

The next generation

The next generation of dielectric fluids will almost cer-
tainly follow the path outline here. Functional fluids will
be used to help equipment designers get the most out of
their equipment. Some of the specialized applications that
we can see on the horizon are:

• Supercooled devices will need distinct insulating fluids

that can be used at very low temperatures and with ma-
terials compatible with cryogenic application.
• The proliferation of cogeneration equipment will re-
quire new insulating oils that can be used with these
devices.
 THE “GO-TO”
STANDARDS FOR
ELECTRICAL SAFETY
AND RELIABILITY
KNOWLEDGE IS POWER

The ANSI/NETA Standards for

Acceptance and Maintenance Testing
Specifications for Electrical Power
Equipment and Systems!

ANSI/NETA MTS-2011 - New Edition

This standard should always be referenced when
writing maintenance specifications or performing
routine testing on electrical power systems.

When you harness the power of knowledge, you unleash the possibilities. ANSI/NETA ATS-2009
This standard should always be referenced in design
specifications or when performing acceptance testing
The industry’s knowledge leader is your trusted partner. on power system installations. ANSI/NETA MTS-2011

For over 90 years, Doble has partnered with our utility clients to improve ANSI/NETA ETT-2010
operations and optimize system performance. Through the dedication This standard ensures that your acceptance
of world-class staff, Doble provides products and services in 110 and maintenance tests are being preformed by
countries around the globe, delivering the solutions you need – from qualified technicians who are certified in accordance
laboratory testing services to hands-on, technical training seminars. It’s a with ANSI/NETA ETT requirements.
relationship we value and a partnership that will allow us take the industry
to the next level. Together. Available in Bound, CD ROM, or PDF
STANDARD FOR
MAINTENANCE
TESTING SPECIFICATIONS
FOR ELECTRICAL POWER
When you work with Doble’s laboratories, you’re Doble Laboratory Services Include: EQUIPMENT AND SYSTEMS
accessing the collective knowledge of our expert
Dissolved gas analysis
chemists and engineers. Whether you need routine
testing or help developing an entire testing program Furanic compounds in oil Order Your copy of the
for condition assessment, you’ll get the answers Water-in-oil analysis ANSI/NETA Standards Today!
and support you need from a team dedicated to
quality and accuracy. Degree of polymerization
Oil quality screens
Find more information at Metals-in-oil
www.doble.com Polychlorinated biphenyls (PCBs)
Become a NETA Affiliate and Save!
+1 617.393.2997 In-depth, laboratory seminars conducted
labservices@doble.com by Doble’s experts To order:
visit www.netaworld.org
DOBLE IS AN ESCO TECHNOLOGIES COMPANY
Or call 888-300-NETA (6382)
Insulating Oils Handbook
Using Analytical Techniques
to Determine Cellulosic Degradation
in Transformers
NETA World, Winter 2001-2002
by Lance R. Lewand
Doble Engineering Company
Insulating materials used in power transformers have been
selected because of their abundance, low cost, and longev-
ity under normal operating conditions. Oils in the U.S. are
expected to last 30 or more years before forming excessive
amounts of acids and sludges and can then be rejuvenated
by treatments with absorbents such as clay. They can also be
easily replaced. Modern oil preservation systems are designed
to minimize exposure of the insulating oil to air thus retard-
ing its oxidation. The solid insulation (paper and pressboard)
is the main dielectric in transformers and also serves as
mechanical support. Localized severe degradation in those
materials must be considered most serious as this can result
in loss of adequate dielectric strength. In addition, cellulosic
materials cannot be easily replaced; therefore, their longev-
ity, which is primarily a function of temperature, becomes
a limiting factor in the operation of transformers. The end
of life criteria, tensile strength, or degree of polymerization
(DP) are physical characteristics of the paper insulation. If
paper insulation is maintained in a dry state, its good elec-
trical properties will be retained even as it becomes quite
brittle. However, mechanically weakened paper can break
especially as windings vibrate and move, particularly during
through faults thus reducing insulating capability. Dielectric
breakdown is then more likely to occur.
Fortunately, as cellulosic materials are degraded, byprod-
ucts such as carbon oxide gases (carbon monoxide and carbon
dioxide) and furanic compounds are formed which can serve
as indicators of the aging process. Cellulosic materials, most
often paper samples, can be tested directly for DP, a measure
of its average molecular weight that correlates well with
mechanical properties.
Cellulose is a long straight chain polymer (polysaccha-
ride) of glucose molecules (monomers), and is the major
constituent of paper and pressboard. Glucose is a sugar
that has six carbons and is typically in the more stable ring
structure called a pyranose. The glucose rings are linked by
15
an oxygen atom in what is referred to as a glycosidic link-
age. The long-chain cellulose molecules interact with each
other due to hydrogen bonding resulting in strands, mats
and paper sheets.
Much of the mechanical strength of paper and pressboard
comes from the long-chain cellulose polymer. As the cel-
lulose ages, the polymers are cleaved and become shorter,
resulting in reduced mechanical strength. The primary forms
of degradation of the cellulose polymer are hydrolytic, oxida-
tive, and thermal. In the case of each of these mechanisms
free glucose is generated and the ring structure tends to be
opened to form chains. Although temperature is likely to
be the most important factor, oxygen and water have been
clearly shown to have a significant effect on the degradation
of Kraft paper. The degradation of cellulose molecules results
in the formation of gases, primarily carbon monoxide and
carbon dioxide, furanic compounds, and other byproducts.
The carbon oxide gases often provide early warning of
excessive damage. However, other materials such as paints
and gaskets can outgas carbon oxide gases when exposed
to excessive temperatures and, therefore, are not always
attributable to the degradation of the cellulosic insulation.
Confirmatory and complementary tests have been developed
which detect oil soluble breakdown products of the cellulose
chain (called furanic compounds) with the primary indica-
tor being 2-furfural.
Furanic Compounds
Furanic compounds are five-membered ring structures
that are formed in a manner in which the open-chain glu-
cose molecule goes through a series of dehydration reactions
(elimination of water molecules) and then recycles into
a five-membered ring structure. The furanic compounds,
unlike sugars such as glucose, are oil soluble and, therefore,
are detectable.
16
High concentration of 2-furfural is a clear indication of
cellulose degradation as this is the only type of material in
transformers which yields this byproduct. Under some con-
ditions where carbon oxides may be lost, such as when a leak
occurs in the gas space of a nitrogen blanketed transformer
or from the conservator tank for those that are free breath-
ing, the furanic compounds will continue to accumulate
and provide a gross indication of the relative aging of the
cellulosic insulation or a thermal incipient-fault condition
involving cellulosic materials. Conversely, when cellulosic
materials are exposed to extreme temperatures which result
in charring, furanic compounds can be destroyed and the
carbon oxides may be the only byproducts remaining in
significant quantities.
Experience is required in evaluating the furanic com-
pound data since there are factors such as the type of insula-
tion preservation/oil expansion system, type of conductor-
wrapped insulation, and family of transformer, all of which
influence the interpretation. For example, the treatment of
the oil or the transformer can result in the removal of sig-
nificant amounts of furanic compounds. Not knowing this
information may lead to a misdiagnosis of the actual con-
dition of the transformer. In addition, furanic compounds
are generated from thermal events, not electrical discharge
activity and therefore can be useful in the assessment of
failure mode and incipient-fault conditions leading to the
failure. Tests for furanic compounds should be performed
initially for all power transformers to establish a baseline, for
important or older transformers, when high carbon oxides
are generated, for highly loaded transformers, and when
other tests indicate accelerated aging.
In order to detect the degradation of cellulosic materials,
sufficient quantities must be degraded to increase the con-
centration of indicator gases and furanic compounds in the
oil to thresholds considered to be problematic. Experience
has shown that significant damage, including charring of
the cellulosic insulation, when limited to isolated hot spots
due to incipient-fault conditions, will produce carbon oxides
and furanic compounds below thresholds used to indicate
problems involving the cellulosic insulation.
The analysis of data for furanic compounds should be
based on the type of insulating paper used and the preserva-
tion system employed. For Kraft paper insulation, suitable
guidelines are as follows:
• For normal aging <50 ug/L/year of 2-furfural should be
generated.
• Generation rates >50 ug/L/year of 2-furfural is consid-
ered accelerated aging
• Values of 2-furfural > 1000ug/L should raise a flag for
further study
For thermally-upgraded (TU) Kraft paper insulation
using the dicyandiamide process, practical guidelines are
as follows:
Insulating Oils Handbook
• For normal aging the rate of 2-furfural generation
should be much less than 50 ug/L/year and usually in
the vicinity of 10-20 ug/L/year
• If estimating insulation quality from the 2-furfural con-
tent, use these guidelines:
• Normal <100 ug/L
• Midlife (examine rate) > 100 <1000 ug/L
• Last third of life? > 1000 ug/L (flag for
further study)
Degree of Polymerization (DP)
The degree of polymerization test is used to assess in-
sulation aging and is performed on paper samples taken
directly from the transformer so it is an intrusive test. The
DP provides an estimate of the average polymer size of the
cellulose molecules in materials such as paper and press-
board. The DP correlates well with mechanical properties
such as tensile strength but has the advantage that it can
be performed on used materials that have taken a set dur-
ing service life. Generally, paper in new transformers has a
DP of about 1000. Aged paper with a DP of 150-200 has
little remaining mechanical strength, therefore making the
windings more susceptible to mechanical damage during
physical movement, which can cause the paper to tear or
crumble. This may occur when transformers are moved or
during events such as through faults. Since paper insulation
does not age uniformly due to thermal, water, oxygen and
byproduct concentration gradients, samples from several
distinct locations provide the best diagnosis. The DP test
provides the most reliable indication of the overall aging of
the paper insulation as it is a direct measurement. This test
should be performed:
• when there is other evidence of very accelerated aging
of the insulation
• when an internal investigation is being performed and
the transformer is more than 20 years old
• for condition assessment of older transformers for pos-
sible refurbishment
• for consideration of a partial rewind
• for failure assessment
• for condition assessment of insulation when purchasing
a service-aged transformer
• to assess the condition of a transformer after an extreme
overheating event such as loss of cooling
Insulating Oils Handbook 17

Conclusions
The combination of analyses of furanic compounds in oil,
DP, along with routine dissolved gas-in-oil analysis is a very
powerful set of tools to assess the condition of the cellulosic
insulation. The more specific information known about a
transformer and its family, the better the diagnosis that can be
provided.

Lance Lewand received his BS degree at St. Mary’s College of Mary-

land in 1980. He has been employed by the Doble Engineering Company
for the past seven years and is currently Project Manager of Research
in the materials laboratory and Product Manager for the DOMINOTM
product line. Prior to his present position at Doble, he was the Manager
of the Transformer Fluid Test Laboratory and PCB and Oil Services at
MET Electrical Testing in Baltimore, MD. Mr. Lewand is a member
of ASTM committee D 27.
18
Understanding Water
in Transformer Systems
The Relationship Between Relative Saturation
and Parts per Million (ppm)
NETA World, Spring 2002
by Lance R. Lewand
Doble Engineering Company
Water content in transformer oil in parts per million
(ppm) is a familiar concept to most in our industry, and lim-
its of 30 to 35 ppm are generally referenced. However, these
simple concentration limits have limited value in diagnosing
the condition of transformer systems and, thus, the concept
of relative saturation (RS) of water in transformer oil has
been re-introduced over the past 15 years. The concept of
relative saturation of water in transformer oil is not a new
one and was originally championed by Frank Doble as early
as the mid 1940s.Thus, this article discusses and details the
relationship between RS and ppm.
It is well known that moisture continues to be a major
cause of problems in transformers and a limitation to their
operation. Particularly problematic is excessive moisture
in transformer systems, as it effects both solid and liquid
insulation with the water in each being interrelated. Water
affects the dielectric breakdown strength of the insulation,
the temperature at which water vapor bubbles are formed,
and the aging rate of the insulating materials. In the extreme
case, transformers can fail because of excessive water in the
insulation. The dielectric breakdown strength of the paper
insulation decreases substantially when its water content
rises above two to three percent by weight. Similarly, the
dielectric breakdown voltage of the oil is also affected by
the relative saturation (RS) of water in oil. The maximum
loading that is possible while retaining reliable operation
(i.e., preventing the formation of water vapor bubbles) is a
function of the insulation water content. For example, dry
transformers (<0.5 percent water in paper) are much less
susceptible to water bubble evolution. In this case, emer-
gency loading at hot-spot temperatures below 180°C may
be possible with little risk of bubble formation. In contrast,
a wetter transformer, with 2.0 percent moisture in the pa-
per, runs the risk of water bubble formation with hot-spot
temperatures as low as 139°C under the same conditions.
A more long-term problem is that excessive moisture ac-
Insulating Oils Handbook
celerates the aging of the paper insulation, with the aging
rate being directly proportional to the water content. For
example, as the water content in the paper doubles so does
the aging rate of the paper. The deterioration of the paper
insulation results from the weakening of the hydrogen bonds
of the molecular chains of the paper fibers. For these reasons
it is important to have a means of assessing the moisture
content of transformer systems and to maintain transformers
in a reasonably dry state.
In order to fully understand water and its dynamics in
transformer systems, a short explanation of the different
types of water encountered and the concepts of solubility
and relative saturation are provided.
Types of Water in Oil
Water can exist in several different states within the
transformer. There are three basic types of water found as-
sociated with transformer oil:
• Dissolved water is hydrogen bonded to the hydrocarbon
molecules of which oil is composed.
• Emulsified water is supersaturated in solution but has
not yet totally separated from the oil. It usually gives oil
a milky appearance.
• Free water is also supersaturated in solution but in a
high enough concentration to form water droplets and
separate from the oil.
In most cases, when one is analyzing or discussing the
amount of water in oil, dissolved water is being referred to
as emulsified, and free water is visually apparent.
Insulating Oils Handbook
What is Water in Oil (ppm), Solubility
of Water in Oil, and RS of Water in Oil?
The detection of water in oil performed in the laboratory
is most often performed by an analytical technique called
Karl Fischer titration described in ASTM Test Method
D 1533 or IEC Method 60814. Both methods are very
comparable and involve a coulometric titration technique
involving the reduction of an iodine-containing reagent. The
methods are used to determine the amount of water in an
oil sample on a weight-to-weight (mg/kg) basis or what is
commonly known as ppm (parts per million).
The concepts of solubility and relative saturation can
sometimes be difficult to understand, but it is an impor-
tant concept when trying to assess the dryness or wetness
of a transformer system. Solubility is defined as the total
amount of water than can be dissolved in the oil at a spe-
cific temperature. The solubility of water is not constant
in oil but changes due to temperature. As the temperature
increases, the amount of water that can be dissolved in oil
also increases. The increase is not linear but exponential in
function. For example, at 10°C only 36 ppm of water can
be dissolved in the oil, whereas when the temperature in-
creases to 90°C, the amount of water that can be dissolved
in the oil increases tremendously to almost 600 ppm. The
table shown lists the calculated solubility limits for oil at
various temperatures. These levels are the greatest amount
of water that can be dissolved at the temperatures listed. If
the concentration of water in oil is greater than that shown
for that specific temperature then, in all likelihood, the oil
is supersaturated with water, and free or emulsified water
could exist.
Table 1 — Water in Oil Solubility as a
Function of Temperature
Oil Temperature Water Content in Oil, ppm
0°C 22
10°C 36
20°C 55
30°C 83
40°C 121
50°C 173
60°C 242
70°C 331
80°C 446
90°C 592
100°C 772
The solubility for mineral oil can be calculated using
Equation 1:
(Equation 1) Log So = -1567/K + 7.0895
19
Where: So is the solubility of water in mineral oil
K is the temperature in Kelvin (°C + 273)
Relative Saturation (RS) is the actual amount of water
measured in the oil in relation to the solubility level at that
temperature. Relative saturation, expressed in units of per-
cent, is the concentration of water (Wc) in the oil relative to
the solubility (So) or concentration of water the oil can hold
at the measurement temperature, as shown in Equation 2.
(Equation 2) RS = Wc /So (100%)
Where: Wc is in ppm wt./wt.
So is in ppm wt./wt.
For example, a sample of oil was taken for determination
of the water content. The temperature of the oil at the time
of sampling was 62°C. The laboratory performed the analy-
sis and determined the water content to be 11 ppm. From
Equation 1, it is calculated that the solubility level at 62°C
is 259 ppm. As discussed previously, relative saturation is
the actual measured value compared to the solubility value.
In this case it is 11 ppm divided by 259 ppm resulting in a
relative saturation of 4.25 percent.
Effects of Relative Saturation on Dielectric
Strength
To properly maintain and operate transformers, an
understanding of the effects of moisture on the dielectric
breakdown strength of the electrical insulating liquids is
necessary. Increasing moisture content reduces the dielectric
breakdown voltage of insulating liquids. The correlation be-
tween the water content in new, filtered, mineral oils at room
temperature and the dielectric breakdown voltage using
ASTM method D 1816 (0.04 inch gap) is given in Figure
1 (water content, ppm). Of course, the dielectric breakdown
voltage is also a function of the number and type of particles
and their conductivity, not just the water content.
Taking the same dielectric breakdown voltage data
and converting it to RS (Figure 1, %RS graph) provides a
much straighter curve except at the extremes. It is evident
that there is a better correlation between RS and dielectric
breakdown voltage than with moisture concentration and
dielectric breakdown voltage.
20 Insulating Oils Handbook
48
Dielectric Break down Voltage, k
44
The water
High RS

Decreasing Relative
40
concentration was
36
constant at 30 ppm.

Saturation, %
32 The temperature was
28 changed to change the
Medium RS relative saturation.
24

20

16
Low RS
12

4 Increasing Dielectric Strength

0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60
Figure 2 — Relationship between Dielectric Strength and RS
Water Content, ppm, wt./wt.
48 Transformers are more complicated systems than this
simple example. However, the same basic principles apply
Dielectric Break down Voltage, k

44

40 for the dielectric breakdown strength of the liquid dielectric.

36 That is, it remains a function of the relative saturation of
32 water in the oil. During the cool-down cycle of a thermal
28 transient in a transformer some of the moisture returns to
24 the paper and some of the moisture remains in the oil. The
20 relative saturation of water remaining in the oil will influ-
16 ence its dielectric breakdown voltage.
12

8 What Does This All Mean for a

4
0 10 20 30 40 50
RS, %@22°C
Figure 1— Dielectric Strength Versus Water Content
and Relative Saturation (RS)
A simple example illustrates that the dielectric break-
down voltage of insulating oils is proportional to the rela-
tive saturation of water in oil rather than the concentration
in ppm. The humidity is controlled in this example so the
concentration of water is held constant at 30 ppm. The
first dielectric breakdown measurement is made at 100°C.
At this temperature the solubility of water in oil is about
772 ppm (Table 1). The relative saturation of water in oil
is therefore about four percent (30 ppm/772 ppm x 100),
and the dielectric breakdown voltage of a well-filtered oil
would be quite high. The temperature is now reduced to
room temperature or about 22°C. The solubility of water
in oil is about 60 ppm (Table 1), and the relative saturation
is 50 percent.
The dielectric breakdown voltage would be expected to
be about half of what it was when the relative saturation
was very low. If the temperature is cooled to 0°C, the re-
sults of a dielectric breakdown voltage should be quite low
because the solubility of water in oil at this temperature is
about 22 ppm (Table 1). As the water content in the oil is
higher than this, the water forms an emulsion and begins
to condense. During all this time the concentration of
water in oil has not changed. This relationship is shown in
Figure 2.
Transformer System?
60 70 80 90 100 Water does not remain at the same concentration in
insulations but, rather, it is continuously migrating between
the solid and liquid insulation. In order to understand
the significance of the water-in-oil value, the operating
temperature of the transformer at the time of sampling
must be known. Most of the water in a transformer sys-
tem resides in the solid insulation (paper and pressboard)
and not in the oil. As temperature increases the water is
forced from the paper into the oil. Although the amount
of water in the paper will change relatively little, the
concentration in the oil may change by an order of mag-
nitude or more, depending upon the initial water content
of the paper and the temperature increase. Fortunately, as
described previously, the solubility of water in oil
increases with temperature such that the relative saturation
may not change much under such conditions, even though
the absolute water values in ppm can increase tremendously.
In fact, the normal suggested limits of 30 to 35 ppm may
be indicative of a wet transformer if the insulation was at
equilibrium at temperatures of 25°C or below since this
represents a relative saturation of 50 percent or greater
in the oil. To maintain reasonable dielectric breakdown
strength of oil, it should remain below 50 percent satura-
tion of water in oil.
References
Doble, F. “The Doble Water Extraction Method,” Minutes
of the Thirteenth Annual conference of Doble Clients, 1946,
Sec. 10-401.
Insulating Oils Handbook 21
Griffin, P. J. “Water in Transformers – So What!,” National
Grid Condition Monitoring Conference, May 1996.
Lewand, L. R. and Griffin, P. J., “How to Reduce the Rate
of Aging of Transformer Insulation,” NETA World,
Spring 1995, pp. 6-11.
Moser, H.P. “Part II. Aging of Insulating Materials,”
Transformerboard, Special Print of Scientia Electrica,
translated by W. Heidemann, EHV-Weidmann Lim.,
1979, pp. 12-15.
Griffin, P. J., Bruce, C. M., and Christie, J. D. “Compari-
son of Water Equilibrium in Silicone and Mineral Oil
Transformers,” Minutes of the Fifty-Fifth Annual Inter-
national Conference of Doble Clients, 1988, Sec. 10-9.1.

Lance Lewand received his BS degree at St. Mary’s College of Mary-

land in 1980. He has been employed by the Doble Engineering Company
for the past seven years and is currently Project Manager of Research
in the materials laboratory and Product Manager for the DOMINOTM
product line. Prior to his present position at Doble, he was the Manager
of the Transformer Fluid Test Laboratory and PCB and Oil Services at
MET Electrical Testing in Baltimore, MD. Mr. Lewand is a member
of ASTM committee D 27.
22
Which Insulating Oil Analytical Tests
to Request and When
NETA World, Summer 2002
by Lance R. Lewand
Doble Engineering Company
Transformer oil testing laboratories receive hundreds of
thousands of samples per year for analysis. Some samples
only require one test, such as Polychlorinated (PCB) con-
tent or dissolved gas-in-oil analysis (DGA). Other samples
require a multitude of tests. Choosing the correct tests to
perform in a given circumstance can be a daunting task.
Testing is of increasing importance today as new and bet-
ter tests become available. This is especially true in the utility
and industrial environments where out-of-service testing of
apparatus is becoming less frequent. The in-service testing,
primarily performed on easily sampled electrical insulating
liquids, can help assess the condition of the insulating mate-
rials and, more importantly, serve as diagnostic procedures to
detect and identify incipient faults in apparatus. In addition,
some of these same tests can be used to verify that a failure
has occurred and then to aid in the identification of the cause
of failure. Advances in testing have provided some tools to
help estimate the condition of the solid insulation, which is
often a difficult task. Some tests can be performed to help
decide which remedial actions might be most effective, such
as reclamation or reconditioning of the oil. Finally, testing
also provides a means to check the condition of oil in storage,
whether it be new or used, to determine if it complies with
specifications such as ASTM D3487, IEC 60296, IEEE
C57.106, or company specifications.
This article attempts to provide some basic information to
users of testing laboratories so they can select the right tests
to diagnose the condition of the solid and liquid insulation
in electric apparatus. Oil testing is not limited to samples
from transformers but also includes bulk transports, drums,
and other electrical apparatus. Sampling and testing can
also help to determine:
• If accidental mixing of different dielectric liquids has
taken place.
• If the method of transportation contaminated the di-
electric liquid.
• If the handling equipment to transfer the dielectric liq-
uid contaminated the product.
Insulating Oils Handbook
Some of the situations discussed in which specific ana-
lytical tests are employed include:
• Qualification of a refiner and large bulk shipments of
oil.
• Small shipments of new oil and new oil shipped in large
power transformers.
• Shipments of new oil in drums and small distribution
transformers.
• Base line tests for power and distribution transformers.
• Tests to perform on in-service oil.
Qualification of a Refiner and Large Bulk
Shipments of Oil
It is usually recommended that if a large quantity of new
oil is to be purchased from a refiner during the course of
year, the product (oil) from the refiner be qualified prior
to initiation of shipments. Product qualification is usually
accomplished by having the refiner submit samples for the
tests referenced in Table 1. The purpose of the tests in Table
1 is to provide a comprehensive analysis of the oil to make
certain that the oil meets certain minimum standards for
electrical, physical and chemical properties. Once a product
has been qualified it is absolutely necessary to confirm that
a large shipment (tanker) of oil arrives uncontaminated.
The tests in Table 2 are recommended tests that should be
performed once tankers arrive on site prior to being placed
into a transformer. It is assumed that the oil will be filtered
(processed) in the transfer from the tanker to the transform-
er. In performing these tests, the test results from delivered
oil should compare very closely to that of the qualification
sample. It is well understood that some of these tests may
take a day or more to complete and demurrage charges on
the tanker(s) will be incurred. However, removal, flushing,
and replacement of contaminated oil would be much more
costly then the demurrage charges.
Insulating Oils Handbook 23

TABLE 1 Small Shipments of New Oil and New Oil

Tests to Qualify a Product Shipped in Large Power Transformers
TEST ASTM TEST ASTM For small shipments of oil, a subset of the complete quali-
METHOD METHOD fication tests such as those given in Table 3 can be used to
Aniline Point D 611 Oxidation Stability D 2440
check the quality of the oil. As long as there are no major
discrepancies, there is reasonable assurance that the other oil
Carbon Type D 2140 Polychlorinated D 4059
Composition Biphenyls (PCBs)
properties should be acceptable. It should be noted that once
oil has been placed in equipment it is no longer considered
Color D 1500 Pour Point D 97
to have the same properties as oil in bulk containers, such
Corrosive Sulfur D 1275 Power Factor at 25°C D 924
as tankers, and is evaluated based on different limits.
Dielectric D 877 Power Factor at 100°C D 924
Breakdown
Dielectric D 1816 Power Factor Valued Doble
Breakdown Oxidation
TABLE 3
Tests on Small Shipments of New Oil (Tanker or Less)
Flash Point D 92 Relative Density 60/60 D 1298
Furanic Compounds D 5837 Rotating Bomb D 2112
Tests of New Oil Shipped in Large Power Transformers
Gassing Tendency D 2300 Sludge-Free Life Doble TEST ASTM METHOD TYPE OF TEST
at 80°C Dielectric Breakdown D 877 Electrical
Impulse Breakdown D 3300 Viscosity at 40°C D 445 Voltage
Voltage Dielectric Breakdown D 1816 Electrical
Interfacial Tension D 971 Visual Examination D 1524 Voltage
Neutralization D 974 Water Content D 1533 Power Factor at 25°C D 924 Electrical
Number Power Factor at 100°C D 924 Electrical
Oxidation Inhibitor D 2668 Interfacial Tension D 971 Physical
Content
Relative Density 60/60 D 1298 Physical
Viscosity at 40°C D 445 Physical
In certain cases, additional tests may be requested. For Color D 1500 Physical
instance, analysis for benzotriazole (BTA), a metal deactiva- Visual Examination D 1524 Physical
tor in the oil, might be requested. This is especially true for Oxidation Inhibitor D 2668 Chemical
those receiving transformers and oil internationally where Content
the use of BTA is not prohibited. Water Content D 1533 Chemical
Neutralization Number D 974 Chemical
Polychlorinated Biphenyls D 4059 Chemical
TABLE 2 Furanic Compounds D 5837 Chemical
Tests on Large Shipments of New Oil
(Two or More Tankers)
TEST ASTM TEST ASTM As shown in Tables 1, 2, and 3 there is a requirement for
METHOD METHOD analysis of polychlorinated biphenyls (PCBs). The analysis
Color D 1500 Polychlorinated D 4059 for PCBs needs to performed only once during the life
Biphenyls (PCBs) span of a transformer, as it is a characteristic that should
Corrosive Sulfur D 1275 Power Factor at D 924 not change. However, if after the test the oil is handled in a
25°C manner that might cause it to be contaminated with PCBs,
Dielectric Breakdown D 877 Power Factor at D 924 then PCB analysis should be conducted again to verify the
100°C concentration for regulatory purposes. New oil should not
Dielectric Breakdown D 1816 Relative Density D 1298 have any PCBs present, but the methods and equipment
60/60 used to transport and transfer the oil may result in accidental
Furanic Compounds D 5837 Rotating Bomb D 2112 contamination and is very important that the concentration
Gassing Tendency D 2300 Viscosity at 40°C D 445 be measured prior to use.
at 80°C
Interfacial Tension D 971 Visual D 1524
Examination
Neutralization D 974 Water Content D 1533
Number
Oxidation Inhibitor D 2668
Content
24 Insulating Oils Handbook

Shipments of New Oil in Drums and TABLE 5

Small Distribution Transformers Tests after Energizing New Transformers
It is impractical to perform all the tests listed in Tables TEST METHOD Perform on Perform on
2 or 3 on oil from drums. In some cases, tests should be Distribution Large Power
Transformers Transformers
made on a composite sample from five to 10 (or more)
drums performing the tests listed as ‘composite’ in Table 3 Dissolved Gases in Oil D 3612 YES YES
to provide an indication of the overall quality of oil received. Furanic Compounds in D 5837 YES YES
Tests on oil in small distribution transformers are usually Oil
not as critical as in power transformers. Metals (Dissolved) in Oil D 5185 NO YES
Metals (Particulate) Doble NO YES
in Oil

TABLE 4 Particle Count Doble NO YES

Tests for Oils Received in Drums and Small Water Content and D 1533 YES YES
Relative Saturation
Distribution Transformers
TEST ASTM Perform on Perform on
METHOD Drums Transformers
Color D 1500 Composite YES
Tests to Perform on In-Service Oils
Dielectric Breakdown D 877 YES YES
Since the oil and solid insulation deteriorates in a trans-
Voltage former while it is in service, periodic sampling and testing
Dielectric Breakdown D 1816 NO YES
is necessary to ensure that deterioration is detected before
Voltage it becomes excessive. The rate at which the insulating ma-
Interfacial Tension D 971 Composite YES
terials will degrade depends on several factors, such as the
type of oil preservation system (amount of oxygen present),
Neutralization Number D 974 Composite YES
operating temperature, water content of the insulation, and
Oxidation Inhibitor D 2668 Composite YES the amounts and types of contaminants. The frequency at
Content
which testing is conducted varies with the test, importance
Polychlorinated D 4059 Composite YES of the equipment, whether an incipient-fault condition
Biphenyls
is known to be present or if a problem exists, and when
Power Factor at 25°C D 924 Composite YES a family of transformers has been identified as having a
Power Factor at 100°C D 924 Composite YES history of problems. All large power and large distribution
Relative Density 60/60 D 1298 Composite YES transformers should be tested immediately after installation
Visual Examination D 1524 YES YES and periodically thereafter. Those units experiencing rapid
Water Content D 1533 YES YES deterioration of the oil or paper or those units for which
an incipient-fault condition has been identified should be
subjected to more frequent testing. Table 6 provides some
guidance as to which tests to choose.
Baseline Tests for Power
and Distribution Transformers
In addition to the testing specified in Tables 3 and 4, there
are certain tests that should be performed immediately after
a transformer is placed in service or energized (if there will
probably be a long duration before actually being placed
into service). This is recommended in order to provide a
baseline analysis to enable trending over the service life of
a transformer and to assure that the transformer does not
have any gross contamination present. These additional tests
are indicated in Table 5.
Insulating Oils Handbook 25

TABLE 6
Tests for In-Service Oils in Transformers

TEST ASTM METHOD Perform on Perform on Perform on

69 kV Units >69 kV 288 kV Units >288 kV Units
Color D 1500 YES YES YES
Dielectric Breakdown Voltage D 877 YES YES NO
Dielectric Breakdown Voltage D 1816 YES YES YES
Dissolved Gases in Oil D 3612 YES YES YES
Furanic Compounds in Oil D 5837 YES YES YES
Interfacial Tension D 971 YES YES YES
Neutralization Number D 974 YES YES YES
Oxidation Inhibitor Content D 2668 YES YES YES
Particle Count Doble NO YES/NO YES
Pour Point D 97 NO YES/NO YES
Power Factor at 25°C D 924 YES YES YES
Power Factor at 100°C D 924 YES/NO YES YES
Relative Density 60/60 D 1298 YES YES YES
Viscosity at 40°C D 445 NO YES/NO YES
Visual Examination D 1524 YES YES YES
Water Content and % Saturation D 1533 YES YES YES

YES/NO indicates that it is a useful test to perform but it is not always necessary.

It has become common practice to group some of the tests

listed in Table 6 into “screen packages” consisting of five to
10 tests which are routinely employed to check the condi-
tion of transformer oil. For example, physical tests such as
visual examination/color and relative density are combined
with electrical tests such as dielectric strength and power
factor and also with chemical tests such as water content and
neutralization number to form a screen package. Be advised
that each laboratory has its own screen package tests and
there is no standardization throughout the industry. Make
sure the screen package that a laboratory is offering are the
tests that are required.

Lance Lewand received his BS degree at St. Mary’s College of Mary-

land in 1980. He has been employed by the Doble Engineering Company
for the past ten years and is currently Project Manager of Research in
the materials laboratory and Product Manager for the DOMINOTM
product line. Prior to his present position at Doble, he was the Manager
of the Transformer Fluid Test Laboratory and PCB and Oil Services at
MET Electrical Testing in Baltimore, MD. Mr. Lewand is a member
of ASTM Committee D 27.
The Anatomy of a
Shermco Field NETA-Certified
Qualified to perform to

Service Engineer the highest standards

Reliable and Responsive

Part of a large international
Trusted Advisor
Offering independent Insulating Oils
Handbook
team available 24/7 service across all brands

Safety First
Back by the Best
On record as an industry
From the field, to our
leader in safe work practices
full-service rotating
apparatus services

Cutting-Edge
Trained from utility
transmission to
low-voltage distribution

Published by InterNational Electrical Testing Association

Odds are, you won’t find a job anywhere that a Shermco field service engineer can’t handle.
As one of the world’s most respected electrical maintenance and testing companies,
Shermco delivers A+ service from utility transmission to low-voltage distribution.

Our experience in commercial, industrial, generation and specialized sectors, such as oil
refineries, pulp and paper, steel and wind power, uniquely qualifies us to handle the
largest, most complex assignments. Our people are trained on the latest technologies and
safety practices. And, we support them with full-service rotating apparatus services and a
large fleet of service vehicles, ready to respond 24/7.

Corporate Headquarters | 2425 East Pioneer Drive, Irving, Texas 75061 Why trust your vital power services to anyone less? Count on an industry leader. Published by
p. 972.793.5523, 888.SHERMCO | f. 972.793.5542 | www.shermco.com Visit www.shermco.com and learn more today. InterNational Electrical Testing Association
26
Choosing a Sample Container
for Transformer Oil Analysis
NETA World, Fall 2002
by Lance R. Lewand
Doble Engineering Company
The last several articles discussed understanding water in
transformer systems and which tests to choose and when.
The results obtained from the analysis of a sample can
provide unparalleled information concerning the condi-
tion of the insulating materials within electrical apparatus,
life assessment, and the operating condition of the device.
However, a sample that is not representative of the bulk
oil insulation can provide erroneous information that can
easily mislead personnel to incorrectly assess the condition
of the oil or the electrical apparatus. The art of sampling is
very important and will be covered in a future article. One
aspect of the sampling function that is often overlooked is
the sample container itself. Unfortunately, there are many
instances in which the sample container plays a major role
in the quality of the sample taken.
Purpose of a Sample Container
Sample containers are used to retrieve and store samples.
It is important for them to prevent contamination and to
provide the best sample for a specific test. ASTM Practices
D 923 and D 3613, the Doble Reference Book on Insulat-
ing Liquids and Gases, or IEC Methods and Guides 60475
and 60567 are to be consulted when choosing a sample
container. Sample containers should have the following
characteristics. They must:
• Be large enough to hold the volume of liquid necessary
for analysis.
• Not impart any contamination (chemicals or particles)
to the sample from the container itself.
• Seal the sample from external contamination.
• Shield the sample from direct sunlight to prevent pho-
todegradation. This can be done either by having a dark
container or by having a covering for that container.
Insulating Oils Handbook
• Prevent the loss or gain of gases or water when testing
for these properties.
The volume of the sample is of the utmost importance, as
various analytical tests require very different sample volumes.
For example, a test for inhibitor content (ASTM D 2668)
requires only a few milliliters of oil whereas the test for
impulse breakdown (ASTM D 3300) may require as much
as two to four liters. In general, it is good practice to provide
the sample volume required for each test plus ten percent.
If unsure of the sample volume required for specific test or
tests, laboratories such as the Doble Materials Laboratory
have reference lists that detail such information.
Which Sample Container to Use
For general oil quality tests, glass bottles — either amber
or clear — function well. Amber bottles provide protection
against photodegradation while clear bottles enable visual
inspection of the sample. To prevent photodegradation
in clear bottles, shielding from direct sunlight by storing
them in cardboard or some other type of container works
well. Bottle caps must be constructed from a compatible
material that will not contaminate the sample. For example,
bottle caps with liners composed of paper or having glue
that is soluble in the oil are not appropriate. Liners made
out of foil, Teflon® or polyethylene are usually safe to use.
If sealed tightly, the glass bottle is an appropriate container
from which to draw a sample for water analysis. However,
problems have been encountered with caps working loose
over time. If not properly protected, glass bottles are more
apt to break than other containers, and they are not suitable
containers for dissolved gas-in-oil analysis, as gases such as
hydrogen and carbon monoxide will be lost.
Metal cans have become popular because they are more
resistant to breakage than glass bottles. Cans with soldered
seams prepared with some fluxes will contaminate the
sample; therefore, welded seams are preferred. Cans made
Insulating Oils Handbook
of tin, aluminum, and stainless steel have been used and are
especially good containers to hold samples for oil quality
tests. Metal bulbs constructed out of stainless steel are also
used and can be invaluable when sampling in areas that
have a lot of airborne environmental contamination such
as coastal areas where salt spray is common.
The use of plastic bottles has grown significantly in the
past decade. Like cans, they resist breakage and shield the
sample from sunlight when dark plastics are used. They are
appropriate containers for samples in which oil quality tests
are to be performed. Samples being tested for water should
not be stored in plastic bottles as water ingress or egress can
occur in just a few hours. Not all plastics are compatible with
oil so selection of the construction material is important,
and containers made of substances such as polypropylene,
polyvinyl chloride (PVC), and especially polystyrene are
not appropriate. Tests such as power factor will be affected
by incompatible plastic bottles because of dissolved com-
ponents transferred to the sample that will cause increases
in dielectric loss. Bottles made of high density polyethylene
have been found suitable and are one of the preferred sample
containers when electrostatic charging tendency or furanic
compound analysis is to be performed. In the case of furanic
compounds, high-density polyethylene does not have the
silanol groups found in glass that attracts the semipolar
furanic compounds to the glass walls and thus removes
them from the sample. Samples with a low concentration
of furanic compounds are not as affected as samples with
larger amounts of furanic compounds.
The most appropriate container for taking samples for
dissolved gas-in-oil and water content analyses, and the easi-
est to manipulate, is the ground glass syringe whose barrel
and plunger have extremely tight tolerances. This type of
syringe has been found to satisfactorily prevent the ingress
of gases and water into the sample, and in turn prevent the
egress of water and dissolved gases from the sample over
a period of time. Care should be taken with glass syringes
to ensure the stopcock is tight and in the closed position
once the sample is taken. Also, the samples collected in
syringes must be quickly protected from photodegradation
(degradation by light) by immediately placing them in the
dark or in their shielding containers. Stainless steel cylinders
may also be used but can be more difficult to manipulate or
to determine when all the air has been removed from the
cylinder. Metal cylinders will also add significant shipping
costs because of weight but are definitely more durable in
transit.
Sample identification is an extremely vital aspect of the
sampling process, and the identification is usually attached
to the container itself to avoid confusion. Many laboratories
receive samples that can not be related to a specific device.
Even if the sample was taken correctly, the lack of proper
identification makes the sample useless, and is therefore
a bad sample. When sampling personnel retrieve samples
from electrical apparatus, it should be done in a prescribed
sequence so nothing is forgotten.
27
Sample Storage and Transport
Samples should not be stored longer then a few days
before shipping to the laboratory for analysis. The properties
of the dielectric liquid tend to decay over time, and gases
for dissolved gas analysis (DGA) can be lost or gained if
the sample is held too long, as there is a finite time in which
the sample container can maintain the sample integrity.
Remember that the sample is supposed to represent the bulk
liquid insulation and storing it too long will change those
properties. Store samples in the dark to prevent photodeg-
radation, and keep them away from temperature extremes
and humid environments.
Package the sample containers to protect from leakage
and breakage, especially if the samples are in glass. The
United States Environmental Protection Agency Resource
Conservation and Recovery Act (RCRA) listed in 40 CFR
Part 261.4(d) requires that all samples shipped must be
packaged so they do not leak, spill, or vaporize from their
packaging. Other countries may have similar requirements
as well. Use methods to prevent sample container breakage
or leakage such as the use of cardboard, Styrofoam, and other
protective materials. Package the samples in such a way as to
avoid container-to-container contact. Also, some shipments
become too heavy and unwieldy, and the shipping company
may not be able to handle it safely. In this case, it is advisable
to separate the shipment into several packages.
Lance Lewand received his BS degree at St. Mary’s College of Mary-
land in 1980. He has been employed by the Doble Engineering Company
for the past seven years and is currently Project Manager of Research
in the materials laboratory and Product Manager for the DOMINOTM
product line. Prior to his present position at Doble, he was the Manager
of the Transformer Fluid Test Laboratory and PCB and Oil Services at
MET Electrical Testing in Baltimore, MD. Mr. Lewand is a member
of ASTM committee D 27.
28 Insulating Oils Handbook

OCB Diagnostics
PowerTest 2002
(NETA Annual Technical Conference)

Presenter
Fredi Jakob, Ph.D., Weidmann-ACTI
Co-Authors
Karl Jakob, P.E., and Simon Jones, Weidmann-ACTI
Rick Youngblood, Cinergy
Alex Salinas, Southern California Edison

I. Introduction III. Diagnostics

During the B.D. era, before deregulation, run to failure, Development of our diagnostic program involved four
time based and operation count based maintenance meth- sequential steps: 1) Identification of fault mechanisms 2)
ods were widely employed. These methods were effective Selection of appropriate physical and chemical tests to detect
in maintaining the power delivery system but were labor problem units 3) Correlation of test data with physical in-
intensive and not cost effective. Time based and operation spection of problem units and 4) Development of algorithms
count based maintenance methods could not identify units to classify OCB condition.
that developed problems between scheduled inspections.
Units were often inspected on a time basis and no problems A. Fault Types
were identified. After deregulation, A.D., fiscal requirements
OCB contacts can become coated with oxides or sulfides,
led to decreases in trained maintenance personnel, deferral
which result in increased contact resistance and increased
of capital expenditures, equipment life extension programs,
contact operating temperatures. Contacts also erode because
and efforts to maximize uptime and minimize maintenance
of mechanical wear or arcing. Arc suppression grids, which
costs. Reliability centered and condition based maintenance
are made of cellulose, deteriorate to some degree whenever
are key A.D. concepts that have been implemented in
an arc is suppressed. Each of these OCB problems can be
the power industry. The ultimate goal of condition based
detected with one or more diagnostic tests. Figures 1 shows
maintenance would be to perform maintenance “just in
contacts that have been severely eroded. Figure 2 illustrates
time”, before the equipment fails in service. Condition
a new arc suppression grid and one that shows degradation.
based maintenance requires periodic or continuous (on-
A wide range of chemical and physical test methods is avail-
line) equipment monitoring. Equipment is scheduled for
able to detect these types of faults.
inspection and/or maintenance only when diagnostic test
results indicate a potential problem.

II. OCB Maintenance

Oil filled circuit breakers; OCB’s have traditionally been
serviced on a time based, operational count or fault current
basis. We believe that OCB condition can be ascertained
by non-invasive diagnostic tests that can reliably indicate
when an internal inspection and/or maintenance or repair
is required. Cinergy, Southern California Edison and
Weidmann-ACTI have cooperated in a project to identify
a battery of chemical and physical tests to assess OCB
condition.

Figure 1
Insulating Oils Handbook 29
indicates that the ratio of heating to arcing gases should be
indicative of problems such as increased contact resistance
or contact erosion.
Arc suppression grids deteriorate every time that they
quench an arc. When the grids are new the arc suppression
time is low. The particles generated from the grid degrada-
tion are numerous but small in size. As the grid opening
enlarges the arc suppression time increases and larger sized
degradation particles are produced. A fault is limited only by
the maximum current amplitude of the source impedance
and the interruption time of the breaker. Maximum fault
power (I2T) determines time to failure. Particle size and
count continues to grow until the distance from the point
source of heat and the maximum grid hole diameter cause
a cooling and blast zone buffer. At this point the produc-
tion of larger particles decrease and smaller particles again
“Good” Grid increase. A plot of large particle size production with time
seems to follow a bell shaped curve. The difference in the
leading edge to the trailing edge of the curve is noted by
the measurable increase in contact metals present in the
oil. Arc suppression grids are constructed of cellulose and
these particles can be distinguished from carbon and shiny
metal particles by chemical microscopy. At temperatures
above 300o C cellulose is destroyed and the resulting carbon
particles are observed in the oil.
Arc tip and arc shaft erosion can be measured by determi-
nation of the characteristic metals in oil and with chemical
microscopy, which can distinguish between shiny metal and
carbon particles. Oil quality assessment tests are also useful
in identifying OCB problems. Dielectric breakdown voltage
measurements, ASTM D-1816, are effected by moisture,
metals, carbon particles and cellulose particles in the oil.
Particles, especially carbon, also effect oil color. Since OCB’s
Close up of a Degraded Grid.
are free breathing devices the moisture level is higher than
that found in sealed components.
Figure 2 Table 1 is a summary of the tests that we have incor-
porated in our OCB diagnostic program and the type of
problem that can be determined with these technologies.

B. Laboratory Tests
Dissolved Gas Analysis, DGA, has been extensively
applied to locate incipient thermal or electrical faults in
transformers. Normal OCB operation will produce the
“key gases” associated with arcing under oil. These key
gases include acetylene and hydrogen, which are produced
at the very high temperatures associated with arcing. The
“hot metal gases”, methane, ethane and ethylene are pro-
duced whenever the oil is overheated from any cause. The
temperature required to produce acetylene is considerably
higher than that required to produce the hot metal gases.
Insulating fluids absorb and distribute fault energy. Thus the
temperature of the oil is very high in the vicinity of an arc
and decreases with distance away from the arc. This varia-
tion of temperature results in the production of acetylene
close to the arc and hot metal gases further removed from
the arc location. Very little heating occurs in a healthy
OCB so the amount of hot metal gases generated should
be small. This analysis of key gas production mechanisms
30 Insulating Oils Handbook

Table 1 Table 2
Diagnostic Tests th
90 Percentile Gas Concentrations in OCB Oil Samples
OCB Problem Test(s) Result(s) Fault Gas 90th Percentile Concentration
Increased Contact DGA Increased Hot Hydrogen 62
Resistance Metal Gases, In-
creased Heating to Carbon Monoxide 136
Arcing Gas Ratios
Methane 28
Contact Tip Erosion DGA, Metals, Hot Metal Gases,
Chemical Microsco- Metals Observed Ethane 14
py, Dielectric-1816 in Oil, Metal
Particles Observed Ethylene 71
by Chemical Mi- Acetylene 173
croscopy, Lowered
Dielectric Break- Total Dissolved Combustible Gas 530
down Voltage
Arc Suppression DGA, Particle Increased Fault Diagnostic software has been developed to evaluate OCB
Grid Degradation Count, Chemical Gas Levels, Large samples according to our established norm values. OCB
Microscopy, Dielec- particles in the Oil,
tric-1816, Color Cellulose Fibers,
sample results are placed into three broad categories of
Decreased Dielec- Normal, Caution and Warning. Southern California Edison
tric Breakdown designates OCB condition numerically using a 3, 2, and 1
Voltage scale. Normal indicates that the sample should be retested
according to the initial utility criteria. Caution indicates
that the sample should be tested more frequently, again at
IV. Diagnostic Protocols a rate determined by utility protocols. Warning indicates
that an internal inspection is appropriate. Figures 3, 4 and
Once the appropriate analytical procedures were selected
5 are examples of test reports for samples that fit in each of
to ascertain the effect of OCB problems it became neces-
these evaluation categories.
sary to determine normal values for each of the measured
parameters. We used a statistical approach to evaluate test
data from problem free units in order to establish the norms.
We then determined the ninetieth percentile values for each
parameter, for several thousand OCB samples. For example,
the 90th percentile individual fault gas concentration and the
total dissolved combustible gas concentration for all samples
evaluated are given in Table 2. Data for selected heating to
arcing fault gas ratios was calculated in a similar manner.
Similar calculations combined with field observations were
used to establish norms for particle size distributions and
metals in oil. Relevant IEEE guides are used to evaluate
results of oil condition assessment tests. These values are
all generic in nature and no attempt has made at this time
to develop unit specific flag points.
Insulating Oils Handbook 31

Figure 3
OCB Condition: Normal (3)

Serial number 20695 TK1 Counter #

Model Gallons 267 Condition Code:
Pole C Fluid Type Mineral Normal (3)
Manufacturer FPE KV Rating 69
Date Interrupt

Hydrogen 2 Between 2 and 5: 3374895 Moisture 28

Methane 1 Between 5 and 15: 749850 Color Number L2.0
Ethane 0 Between 15 and 25: 4084 Dielectric 1816 20 1-24
Ethylene 1 Between 25 and 50: 571 Silver L0.50
Acetylene 0 Between 50 and 100: 90 Chromium L0.50
Carbon Monoxide 84 Greater than 100: 0 Copper L0.50
Carbon Dioxide 518 ISO Code: 23/20/13 Nickel L0.50
Oxygen 24309 Fibers: 20 Phosphorous L0.50
Nitrogen 53664 Metals: 10 Lead L0.50
Total Dissolved Gas 78579 Carbon: 50 Tin L0.50
Total Combustible Gas 88 Other 20 Zinc L0.50
Equivalent TCG Percent 0.0973 Opacity 2.0 Tungsten L0.50

Narrative: No problems found. No action taken. Unit stayed in service, reset maintenance schedule.

Figure 4
OCB Condition: Caution (2)

Serial number 28760 Counter # 120

Model CG-38 Gallons 220
Condition Code:
Pole Fluid Type Mineral Caution (2)
Manufacturer MGE KV Rating 38
Date 1987 Interrupt 40

Hydrogen 1 Between 2 and 5: 2199309 Moisture 30

Methane 1 Between 5 and 15: 3891922 Color Number L3.0
Ethane 1 Between 15 and 25: 98949 Dielectric 1816 10 1-24
Ethylene 2 Between 25 and 50: 2252 Silver 2.38
Acetylene 2 Between 50 and 100: 60 Chromium L0.50
Carbon Monoxide 17 Greater than 100: 0 Copper L0.50
Carbon Dioxide 382 ISO Code: 23/22/17 Nickel L0.50
Oxygen 25182 Fibers: 10 Phosphorous L0.50
Nitrogen 50557 Metals: 5 Lead L0.50
Total Dissolved Gas 76145 Carbon: 70 Tin L0.50
Total Combustible Gas 24 Other 15 Zinc L0.50
Equivalent TCG Percent 0.0229 Opacity 4.0 Tungsten L0.50

Narrative: Analysis indicated mild contact erosion, and poor oil quality.
Remedial Action: Unit was put on ½ maintenance schedule and remained in service.
32 Insulating Oils Handbook

Figure 5
OCB Condition: Warning (1)

Serial number 318027-A Counter # 608

Model FZO-69 Gallons 855 Condition Code:
Pole A Fluid Type Mineral Warning (1)
Manufacturer Allis- KV Rating 57
Date 1953 Interrupt 21

Hydrogen 7280 Between 2 and 5: 901411 Moisture 23

Methane 6624 Between 5 and 15: 3453273 Color Number L6.5
Ethane 729 Between 15 and 25: 2645616 Dielectric 1816 10 1-24
Ethylene 5424 Between 25 and 50: 221381 Silver L0.50
Acetylene 9356 Between 50 and 100: 2462 Chromium L0.50
Carbon Monoxide 91 Greater than 100: 0 Copper 1.63
Carbon Dioxide 437 ISO Code: 29/28/24 Nickel L0.50
Oxygen 44520 Fibers: 10 Phosphorous L0.50
Nitrogen 24023 Metals: 10 Lead L0.50
Total Dissolved Gas 94686 Carbon: 80 Tin L0.50
Total Combustible Gas 25706 Other 0 Zinc L0.50
Equivalent TCG Percent 20.6182 Opacity 5.0 Tungsten L0.50

Narrative: Oil sample obtained approximately 4 hours after a 30000 AMP, 3 cycle fault.
Contacts evaluated at 25 percent degraded. Copper present due to shaft wear on movable contact.
Grid not significantly degraded.

Remedial Action:
Rotated movable contact ¼ turn.
Filtered and processed oil.
General maintenance. Returned to service.

V. Sampling Protocols
The overall objective of our OCB diagnostic program is
to minimize the number of required internal inspections and
thus to realize considerable maintenance cost savings. Utili-
ties must decide when and at what frequency to sample their
OCB’s. A sampling strategy has been developed by Mr. Alex
Salinas at Southern California Edison, SCE, to determine
when to draw the initial sample, when to draw subsequent
samples and when to conduct an internal inspection. The
entire sampling process is incorporated in their system soft-
ware. The software, which maintains OCB operations data,
generates an initial work order when any of three conditions
are met: twenty interruptions, three hundred operations or
five years from the last test. This first work order will be for
an initial oil sample. The results of the initial test generate
a code: 1=Schedule internal inspection, 2=Reset oil sample
trigger at half the initial values (10 interruptions or 150
operations) and 3=Reset sample trigger at 20 interruptions
or 300 operations. The software maintains all of the test data,
resets the triggers and prints the work orders. This closed
loop operation requires no operator intervention. A flow
chart for the SCE protocol is given in Table 3.
Maintenance cost savings for the SCE program can be es-
timated considering the number of reduced internal inspec-
tions. SCE currently inspects about three hundred breakers
a year, both oil filled and others. The oil filled breakers range
in voltage class from 7.2 kV to 220 kV. Their largest number
of breakers operates at 69 kV. Inspection costs range from
2K$ - 20K$ on OCB’s in the voltage range from 7.2 kV to
220 kV. Based on statistical evidence less than ten percent
of the units tested will require an internal inspection. The
savings promise to be considerable.
Cinergy Corporation currently uses both condition based
(CB) and fault adjusted operation count (FAO) as the
initial triggers for breaker maintenance. The non-invasive
condition based triggers used are thermography, ultrason-
ics, oil quality and dissolved gas analysis. The main invasive
tests are ductor and power factor. The invasive tests are
used minimally due to limited outage request acceptance.
Fault adjusted operation count is a method that employs
knowledge of the maximum fault amplitude based on source
and circuit impedance and fault duration. This is used to
Insulating Oils Handbook 33

Table 3
SCE Flow Chart

determine the estimated number of breaker operations (300) criteria are not met. Can this time interval be extended
before maintenance is needed. Unless the breaker employs without increasing system outages due to breaker failures?
the use of some form of I2T monitors it is still very dif-
ficult to determine true fault duty cycle. Additionally, due
to deregulation and customer commitment, outages and Dr. Fredi Jakob received his PhD at Rutgers, the State University of
internal inspections will be kept to a minimum. Although New Jersey, in 1961. He is professor emeritus of analytical chemistry at
California State University-Sacramento and is the founder and laboratory
each test gives different positive information to determine director of Analytical ChemTech International, Inc. (ACTI), which is a
maintenance interval, Cinergy has found the addition of wholly owned subsidiary of Weidmann Systems International. As a long-
oil particle analysis to determine grid and contact health term member of ASTM and IEEE and author of over fifty published
invaluable to determine overall maintenance interval. articles, Dr. Jakob is a traveling lecturer to private and governmental
agencies. He has been invited to speak at American Public Power meet-
ings, ASTM symposia, conferences held by Doble, NETA, and AVO
VI. Conclusions and Future Work conferences, as well as other industrial organizations.
Normal or threshold values for OCB test parameters are
at this time generic in nature. Because of design variations
unit specific normal values may be more appropriate. Unit
specific test values will have to be empirically determined.
We are maintaining a very detailed database so that we can
develop unit specific normal values in the future.
We are cooperating with utility clients to further evalu-
ate testing frequency protocols. For example SCE tests the
OCB oil every five years if the interruption (20) or operation
34 Insulating Oils Handbook

Sampling Transformer Oils

Part One – How and Why to Take a Good Sample
This three-part article deals with the art of sampling transformer oil

(See page 37 for Part Two)

NETA World, Winter 2002-2003

by Lance R. Lewand
Doble Engineering Company

The results obtained from the analysis of an insulating
oil sample can provide unparalleled information concerning
the condition of the insulating materials within electrical
apparatus, life assessment, and the operating condition of
the device. However, a sample that is not representative of
the bulk oil insulation can provide erroneous information
which can easily mislead maintenance personnel to incor-
rectly assess the condition of the oil or the electrical ap-
paratus. In one case, samples taken from two transformers
showed very high concentrations of hydrogen and no other
gases, prompting maintenance personnel to give these units
priority for diagnostic surveys. It was later found that the
cause of the high hydrogen was a galvanic reaction occur-
ring in the drain valves in which water was converted to
hydrogen because of the interaction of a galvanized fitting
with a dissimilar metal.
Why Sample?
For in-service oil-filled electric apparatus, sampling of
the oil provides a method to determine the condition of
the solid and oil insulation as well as the operating condi-
tion of the apparatus without opening or de-energizing
the apparatus. This is especially important in the present
utility and industrial climates, as equipment outages for
out-of-service testing have become very limited. Sampling
provides a means to check the condition of oil in storage,
whether it be new or used, and to determine if it complies
with specifications such as TOPS, ASTM D 3487, IEC
60296, IEEE C57.106, or company specifications. Sampling
can also help to determine:
1) If accidental mixing of different dielectric oils has taken
place;
2) If the method of transportation contaminated the oil;
3) If the handling equipment to transfer the oil contami-
nated the product.
What is a Good Sample?
Simply put, a good sample is one that is representative
of the content of the bulk oil insulation. Since samples are
usually retrieved from a drain valve or the attached sam-
pling cock, preparation of that area is important to obtain
a good sample. Cleaning the drain valve and the sampling
cock inside and out is the first step in avoiding sample con-
tamination. Cleaning the outside of the drain valve is just as
important as cleaning the inside. The dirt and debris falling
off the outside of the valve into the sample container during
the sampling process can contaminate many samples.
Most of the contamination in the apparatus consists of
water and particles (paper fibers, metal particles, etc.), which
over time will settle out on the bottom of the apparatus near
the drain valve. This material needs to be flushed out of the
system to get to the bulk oil insulation. It is necessary to
remove at least one to two liters of oil from the drain valve,
cap the drain valve, and then flush out the sampling cock
before proceeding with sampling. On occasion, two liters
will not be sufficient, especially when sampling a nonen-
ergized transformer or certain OCBs and LTCs. Specific
sampling techniques and precautions, especially those deal-
ing with low volume electric equipment, are detailed in the
Doble Reference Book on Insulating Oils and Gases and ASTM
Practices D 923 and D 3613.
Good Samples Versus Bad Samples
It is sometimes very clear to the laboratory performing
the analysis on the oil that the sample was taken improperly.
For example, the presence of free water or foreign objects
such as insects, pipe sealing tape or putty are strong indica-
tors that the drain valve was not adequately flushed out prior
to sampling. Once analysis has begun and it is determined
that there is a high or free water content coupled with a low
dielectric strength, with all the other test results being ac-
ceptable, then it strongly indicates that the proper sampling
Insulating Oils Handbook 35
technique was not adhered to. It may even imply that some These materials increase dielectric loss, which increases the
chemical reactions were taking place in the drain valve that power factor. Incompatible inorganic and organic materials
were not representative of the bulk oil insulation. from the drain-valve stem packing or drain-plug sealants
can also have the same effect on the power factor.
Lab Tests Most Easily Affected
As indicated previously, the analytical tests most eas- Costs Associated with a Bad Sample
ily affected by sampling are dielectric strength and water In the case of a single sample, the costs for routine oil
content. This is due to the fact that drain valves are usually quality analysis and DGA testing are just a very small frac-
at very low points in the tanks where debris and water ac- tion of the total costs associated with taking and analyz-
cumulation occurs. Water can also be present as a result ing a sample. Some of the items and costs associated with
of condensation that occurs in the drain valve, which is sampling and analysis are:
due to the position of the drain valve on the tank. In most
cases the drain valve protrudes 15 to 30 centimeters (six ITEM COST ($)
to 12 inches) away from the main tank. From experiments Labor to take sample 275
performed at Doble Engineering, it was found that the oil
Materials to take sample 15
in many of these valves varies in temperature from eight
Packaging and shipping cost 8
degrees Celsius to 15 degrees Celsius cooler than the bulk oil
insulation. When oil or air has an elevated relative saturation Analysis cost 70
or humidity and there is a significant cooling, condensation Engineering evaluation of the data (10-15 min) 35
of water will occur. This is exactly what happens in a drain TOTAL 403
valve. Other analytical tests easily affected by sampling are
dissolved metals, particulate metals, particle counts, dis-
solved gases-in-oil, and power factor. Every situation is different, but in many cases the analysis
The concentration of metals, whether dissolved or in a cost is only about 17 percent of the entire sampling and data
particulate state, are especially impacted by the amount of review process. In a situation where the sample has been
cleaning performed on the drain valve and the amount of determined to be nonrepresentative of the bulk oil insula-
flushing that is performed. Debris that settles to the bottom tion, the following additional costs may also be incurred:
of the apparatus and subsequently into the drain valve can
consist of metal particles. In addition, just the simple fact of ITEM COST($)
removing the drain-valve plug or opening the sampling cock
Labor to take original sample 275
will create particulate metals. This is due to the grinding of
Materials to take original sample 15
the surfaces between the valve body and the drain plug or
sampling cock. In fact, it is becoming more apparent that Packaging and shipping cost for original sample 8
that these types of samples should only be retrieved after a Analysis of original sample 70
minimum of two (and sometimes three to four) liters of oil Engineering evaluation of the data (10-15 min)
have been passed through the drain valve. of original sample 35
The same is true of retrieving a sample for particle count Additional engineering time to confirm
sample was nonrepresentative 35
where valve debris, whether inside or outside, can severely
skew the results. The debris, soot, and grime that exist on Labor to take 2nd sample 275
the outside of the drain valve are of serious consequence, Materials to take 2nd sample 15
especially in industrial locations. This debris can be easily Packaging and shipping cost for 2nd sample 8
transferred to the sample bottle while the sampling process Analysis cost of 2nd sample 70
is taking place. This validates the importance of cleaning the Engineering evaluation of the data
outside of the valve prior to taking the actual sample. (10-15 min) of 2nd sample 50
Dissolved gas-in-oil analysis is another test impacted by TOTAL 856
sampling, drain valve components, and sampling materials.
When galvanic fittings (zinc coated) are used in the drain
valve assembly — such as the drain plug — a galvanic reac- The cost of taking a bad or nonrepresentative sample has
tion with water can cause very high levels of hydrogen to be more than doubled from the original total. This is in part
produced. If this residue is not flushed out adequately then it due to the fact that review of data from the second sam-
will be transferred to the sample and included in the analysis, pling takes longer as there is a more critical and thorough
causing a level of concern that is not warranted. Galvanic review.
plumbing fittings such as nipples can also have the same If the original sample was not recognized as bad, the costs
effect. Brass, bronze, stainless steel or black iron should be associated with that sample can be staggering. For example,
the only materials used. In addition, drain valve assemblies a bad sample could cause a customer to try a remedial effort
should not be composed of dissimilar metals as corrosion can in an attempt to improve the condition of the insulating oil
result, which may end up in the sample. Debris, water and — such as processing the oil through clay or vacuum-pro-
other ionic contaminants also affect the power factor test.
36
cessing a transformer to remove moisture — then associ-
ated costs may skyrocket to between $10,000 and $30,000.
This is one reason why Doble always recommends taking a
second sample to confirm the results of the first before any
remedial activities begin. Other factors, such as accidental
sample switching or misidentification, can also be the source
of an erroneous assessment.
Part two of this series will cover the sampling practices
to follow and the science of sampling.
Insulating Oils Handbook
“ASTM D 3613: Standard Practice for Sampling Electri-
cal Insulating Oils for Gas Analysis and Determination
of Water Content” in Electrical Insulating Oils and Gases;
Electrical Protective Equipment, Annual Book of ASTM
Standards, Vol. 10.03, ASTM, West Conshohocken,
PA, 2001.
Griffin, P. J. “Water in Transformers – So What!” National
Grid Condition Monitoring Conference, May 1996.

Lance Lewand received his BS degree at St. Mary’s College of

References
“Items of Interest” in The Doble Exchange, The Doble En-
gineering Company, Watertown, MA, USA, Volume
11, Number 3, September 1993, Page 4.
Transformer Oil Purchase Specification (TOPS), edited by
the Doble Oil Committee, Rev. TOPS-884, Doble En-
gineering Company, Watertown, MA.
Maryland in 1980. He has been employed by the Doble Engineering
Company since 1992 and is currently Project Manager of Research in
the materials laboratory and Product Manager for the DOMINOTM
product line. Prior to his present position at Doble, he was the Manager
of the Transformer Fluid Test Laboratory and PCB and Oil Services at
MET Electrical Testing in Baltimore, MD. Mr. Lewand is a member
of ASTM committee D 27.

“ASTM D 3487: Standard Specification for Mineral Insu-

lating Oil Used in Electrical Apparatus” in Electrical In-
sulating Oils and Gases; Electrical Protective Equipment,
Annual Book of ASTM Standards, Vol. 10.03, ASTM,
West Conshohocken, PA, 2001.

“IEC 60296: Specification for Unused Mineral Insulating

Oils for Transformers and Switchgear” International
Electrotechnical Commission, 3, rue de Varembe, Ge-
neva, Switzerland, 1982.

“IEEE C57.106-1991: IEEE Guide for Acceptance and

Maintenance of Insulating Oil in Equipment”, IEEE,
345 East 47th Street, New York, NY, 1992

Reference Book on Insulating Oils and Gases, edited by the

Doble Client Committee on Oil Insulation, 1993,
Doble Engineering Company, Watertown, MA.

“ASTM D 923: Standard Practice for Sampling Electrical

Insulating Oils” in Electrical Insulating Oils and Gases;
Electrical Protective Equipment, Annual Book of ASTM
Standards, Vol. 10.03, ASTM, West Conshohocken,
PA, 2001.
Insulating Oils Handbook 37

Sampling Transformer Oils

Part Two– Sampling Practices and the Science of Sampling
This three-part article deals with the art of sampling transformer oil
(See page 40 for Part Three)

NETA World, Spring 2003

by Lance R. Lewand
Doble Engineering Company

The first part of this three-part series defined how and
why to take a good sample and explored the costs associ-
ated with taking a bad sample. Part two covers sampling
technique and factors influencing the sampling process.
There are a number of industry-recognized standards that
define the correct way to retrieve samples from electrical
apparatus or storage containers. Some of the sources that
list these sampling techniques include:
• Doble: Reference Book on Insulating Liquids and Gases
• ASTM D 923: Standard Practice for Sampling Electrical
Insulating Liquids
• ASTM D 3613: Standard Practice for Sampling Electri-
cal Insulating Oils for Gas Analysis and Determination of
Water Content
• IEC 60475: Method of Sampling Liquid Dielectrics
• IEC 60567: Guide for the Sampling of Gases and of Oil
from Oil-filled Electrical Equipment and for the Analysis
of Free and Dissolved Gases
These techniques have been developed over a number
of years and have incorporated the expertise of many indi-
viduals. However, sampling technique involves much more
than just taking the sample. It involves a more thorough
knowledge of the information to be gained from taking a
proper sample and includes sample site preparedness and
site cleanup after sample retrieval. Some of the items that
are a part of sample technique are:
1. Materials used to aid in retrieval of a sample.
2. Safety precautions to adhere to.
3. Environmental concerns.
4. Identification of the sample and apparatus information.
5. Final checks prior to sampling.
6. Taking the sample (cleaning and preparation of valves).
7. Cleanup after sample has been retrieved.
Materials Used
Whoever takes the samples must be fully prepared for any
eventuality that could occur at the sample site. For example,
items such as sheet plastic, plastic bags, absorbent materi-
als, flush oil containers, and catch pans are all important
materials to have to prevent or clean up liquid spillage. It
must be remembered that, in order to take a correct sample,
some liquid waste will be generated. Of course, bottles and
syringes will be needed as sample containers and must be of
sufficient size to hold the volume of dielectric liquid neces-
sary for the desired tests. Labels are required to sufficiently
and correctly identify those containers. Make-up oil may
be necessary to add oil to low volume devices. Bottled ni-
trogen may be needed to pressurize a transformer to relieve
a negative pressure in order to get the dielectric liquid out
of the apparatus.
Tygon™ tubing or other compatible tubing is also nec-
essary to direct the dielectric liquid from the drain valve
to the flush container, sample bottle, and syringe. Tubing
should only be used once and then discarded as the walls of
the tubing have memory (gases, water, and other chemical
compounds are held in the walls of the tubing) which can
be transferred to the next sample. Incompatible tubing such
as natural rubber or polyvinyl chloride will contaminate a
sample.
The appropriate tools and plumbing accessories must be
on-site to manipulate the drain valve in order to retrieve the
sample. Personal protective equipment such as nitrile gloves
is used to protect personnel from the liquid dielectric and/
or polychlorinated biphenyls (PCBs). Personal protective
equipment and safety practices to protect against electrical
or physical hazards must also be present and observed.
38
Safety Precautions
There are several safety precautions that must be followed
in order to secure the well-being of the equipment as well as
the personnel retrieving the sample. Death is not normally
associated with sample-taking, but it has happened when
electrical hazards have not been observed. Routines and
complacency often contribute to a lax adherence to safety
precautions. Some of the more critical safety precautions
are:
• Make sure there is positive pressure on the electrical ap-
paratus.
• Take into consideration the remaining volume in low
oil volume apparatus (this may require de-energizing
the equipment).
• De-energize instrument transformers before sampling.
• Secure electrical dangers.
• Make sure Occupational Safety and Health Adminis-
tration (OSHA) requirements are met.
Making sure there is positive pressure on the electrical
apparatus prior to sampling is the single most critical fac-
tor in assuring that the equipment survives the sampling
procedure. Sampling of electrical equipment while under
negative pressure will allow atmospheric air to be drawn
into the equipment through the drain valve, which will rise
through the transformer as bubbles. These bubbles are areas
of weak dielectric strength and can easily cause failure of
the apparatus through flashover. Most transformers have
pressure gauges that allow determination of the actual pres-
sure. If positive or negative pressure cannot be determined,
then follow the procedure in ASTM D 923 to determine
the pressure condition. This involves using a slug of oil in
clear tubing attached to the sampling cock. If negative pres-
sure does exist then no samples are to be drawn until that
negative pressure is relieved. Sometimes this is as simple as
adding dry nitrogen to the headspace of a transformer to
pressurize the unit, or waiting until ambient temperature
has increased to a sufficient degree to cause the expanding
dielectric liquid to pressurize the apparatus.
The remaining volume in some electrical apparatus —
especially oil circuit breakers, bushings, load tap-changers,
and small instrument transformers — is of serious concern.
Electrical components are positioned at critical clearances
and that distance is determined with liquid insulation pres-
ent. If the liquid insulation level is too low the insulation
between the energized components is now gas instead of
the insulant, and the dielectric integrity becomes compro-
mised. This is why it is important to check the liquid level
not only before sampling but after as well, and this will aid
in maintaining a safe operating environment. Instrument
transformers must also be de-energized prior to sampling
in order to secure the electrical hazards.
Electrical hazards are especially prevalent in small
distribution pole and pad-mounted transformers. Pole
transformers do not usually have a sample valve so the lid
Insulating Oils Handbook
of the transformer must be removed to take a sample. In
addition, the primary and secondary voltage terminations
are extremely close to personnel. This is also true of pad-
mounted transformers where the secondary and sometimes
primary cable or bus bar is within feet of the sample valve.
Any wrong move by sample personnel and serious injury
or death can result. Because of such concerns, OSHA has
instituted lockout/tag procedures that must be adhered to
secure against such dangers.
Environmental Concerns
Dielectric liquid spillage as a result of sampling is a main
environmental concern, as some of these liquids may still
contain PCBs. The United States and many other countries
have very strict guidelines for spill cleanup and notification
of PCB materials. In the United States even one drop of
liquid containing more than 50 parts per million of PCB
is considered “improper disposal.” Even if the oil does not
contain any PCBs or the spill is small and accidental there
may be regulations dealing with the dielectric liquids’ release.
It is, therefore, easier to prevent against spillage then to clean
up after the spill has occurred. This is why many sampling
personnel lay down plastic and absorbent materials under
the drain valve prior to sampling and then use a catch pan
to trap larger volumes of liquid.
Identification of the Sample and Appara-
tus Information
Sample identification is an extremely vital aspect of the
sampling process. Many laboratories receive samples that
cannot be related to a specific device. Even if the sample
was taken correctly, the lack of appropriate identification
makes the sample useless. When sampling personnel retrieve
samples from electrical apparatus, it should be done in a
prescribed sequence so nothing is forgotten.
The lack of information concerning an apparatus severely
limits the laboratory in its ability to provide an in-depth
diagnosis. Apparatus information such as the age, type of
preservation system, and any previous incipient fault condi-
tions or oil reclamation activities can alter a diagnosis. For
instance, the Doble Materials Laboratory relies heavily on
the type of preservation system that is part of the trans-
former to provide a diagnosis based on oil quality, dissolved
gas analysis, and furanic compound results. In transformers
that have a sealed conservator preservation system, oxygen
and nitrogen values are expected to be below certain levels. If
test values are above those levels, there may be several causes
for this such as a breach in the bladder or diaphragm, another
leak elsewhere on the transformer, or poor sampling. All
these items would concern the operator of the equipment.
However, if no information is provided or the information
is incorrect then no diagnosis or an incorrect one will be
provided by the laboratory. Laboratories also use the age and
the type of preservation system of a transformer as exceed-
ingly pertinent information when providing a diagnosis
based on furanic compound results.
Insulating Oils Handbook 39

Final Checks The next and final part of this series will deal with taking
Adhering strictly to safety and environmental concerns the actual sample, sample storage, and transport.
will assist personnel in adequately preparing the site around
the electrical apparatus to be sampled. However, specific
attention should be given to some final checks before the References
sampling commences. These include: 1. Reference Book on Insulating Liquids and Gases, edited
by the Doble Client Committee on Liquid Insulation,
• Confirmation of positive tank pressure — again. Doble Engineering Company, Watertown, MA, 1993.
• Using the correct sample containers, both in size and
compatibility. 2. “ASTM D 923: Standard Practice for Sampling Elec-
trical Insulating Liquids” in Electrical Insulating Liquids
• Labeling the sample information completed. and Gases; Electrical Protective Equipment, Annual Book
• Protecting the sample from outside contamination. of ASTM Standards, Vol. 10.03, ASTM, West Con-
shohocken, PA, 2001.
• Relative humidity of ambient air less than 50 percent
(avoid rain or snow conditions). 3. “ASTM D 3613: Standard Practice for Sampling Elec-
• Temperature of dielectric liquid higher than or equal to trical Insulating Oils for Gas Analysis and Determina-
the ambient air. tion of Water Content” in Electrical Insulating Liquids
and Gases; Electrical Protective Equipment, Annual Book
Checking for positive pressure is so critical that it requires of ASTM Standards, Vol. 10.03, ASTM, West Con-
a second check and must not be overlooked. As mentioned shohocken, PA, 2001.
in Part one of this series, it is very difficult to perform the
requested analysis when not enough sample volume exists. 4. IEC 60475: Method of Sampling Liquid Dielectrics,
Therefore, it is very important to make sure the correct size International Electrotechnical Commission, 3 rue de
sample container is used. If unsure of the sample volume Varembe, Geneva, Switzerland, 1974.
requirements, it is better to provide more sample than to
return to that apparatus at a later date to re-sample. Labeling 5. IEC 60567: Guide for the Sampling of Gases and of Oil
samples with complete sample information prior to actually from Oil-filled Electrical Equipment and for the Analysis of
taking of samples is recommended. In this way, sample con- Free and Dissolved Gases, International Electrotechnical
tainers will not go unmarked and confusion can be avoided Commission, 3 rue de Varembe, Geneva, Switzerland,
when multiple pieces of equipment are being sampled. 1992.
Some geographic locations are inherently inhospitable
to providing an environment where a good sample can be
taken. Inhospitable locations include salt spray areas and Lance Lewand received his BS degree at St. Mary’s College of Mary-
high wind gusting areas that naturally kick up sand, dirt, land in 1980. He has been employed by the Doble Engineering Company
for the past seven years and is currently Project Manager of Research
soot, and other debris that can easily deposit foreign matter in the materials laboratory and Product Manager for the DOMINOTM
into an open sample container while it is being filled. In product line. Prior to his present position at Doble, he was the Manager
these instances protecting the sample from outside con- of the Transformer Fluid Test Laboratory and PCB and Oil Services at
tamination may require special precautions. Really adverse MET Electrical Testing in Baltimore, MD. Mr. Lewand is a member
environments may require totally enclosed sampling systems of ASTM committee D 27.
such as stainless steel cylinders to protect against outside
contamination.
The ambient environment in which a sample is taken
can also contaminate the sample with excessive amounts
of moisture. Taking samples during rain, hail, or snow
conditions should be avoided. If the situation is completely
unavoidable then necessary precautions must be enacted to
keep the sample from absorbing any external moisture. It is
recommended that sampling be performed only when the
relative humidity conditions are less than 50 percent so as
to minimize the amount of external moisture the sample
will absorb. It is recognized that this may be unavoidable
in certain geographical locations and during certain peri-
ods of the year, but every effort must be made to minimize
external contaminants and have a sample that reflects the
bulk liquid insulation.
40 Insulating Oils Handbook

Sampling Transformer Oils

Part Three — Retrieving the Actual Sample
This three-part article deals with the art of sampling transformer oil

NETA World, Summer 2003

by Lance R. Lewand
Doble Engineering Company

The first part of this three-part series defined how and Table 1
why to take a good sample and explored the costs associ- Sampling Points for Various
ated with taking a bad sample. Part two covered sampling Dielectric Liquids in Apparatus
technique and factors influencing the sampling process. Part
3 will discuss the technique of taking an actual sample and
Sample from Bottom Sample from Top
methods for correctly storing and transporting a sample
for analysis. Mineral oil Polychlorinated Biphenyls
Different dielectric liquids require sampling from differ- Silicone Trichlorobenzene
ent locations based on their relative density (specific grav- R-Temp Tetrachlorobenzene
ity). In general, dielectric liquids with a relative density less Midel 7131 Wecosol
than one should be sampled from the bottom drain valve Reoloec 138 Perchloroethylene
whereas dielectric liquids with a relative density greater than
one should be sampled from the top fill valve as long as it Beta Fluid
is below the liquid level. There are exceptions to this, and Shell Diala HFX
the sampling point can change throughout the life of the WEMCO-FR
transformer. For instance, mineral oil transformers that have
MEPSOL
no drain valve are usually accessed and sampled through the
top. Another example involves retrofilling of askarel trans- Opticool
formers. When a transformer is filled with askarel, sampling ALPHA-1 FLUID
should be performed from the top fill valve because the Polyalphaolefins (PAOs)
relative density is greater than one. However, many askarel
transformers have been retrofilled and the askarel fluid BIOTEMP
replaced with silicone which has relative density less than BIOTRANS
one. In this case, the transformer should now be sampled ECO Fluid
from the bottom. Table 1 is a list of sampling points for EDISOL TR
various dielectric liquids in apparatus during routine test-
ing. Sampling of drums, tankers, and other types of storage ENVIROTEMP® FR3
containers are performed in a different manner. Consult the ENVIROTEMP 200
previously referenced guides for specific procedures.
Once the correct valve from which to retrieve the sample
has been determined, that valve should be prepared for
taking of the sample. As mentioned previously, check for
positive pressure on the apparatus before opening the drain
valve. Adequate preparation of the valve for sampling con-
sists of the following:
Insulating Oils Handbook 41
Clean the outside of the valve to remove any loose de-
bris that may fall into the sample.
Make sure the valve and sampling cock are closed be-
fore removing the drain plug.
Prepare the area under the valve with absorbent materi-
als and a catch pan. Valve Seat

Transformer Tank
Slowly remove the drain plug.
Clean the inside of the valve with a lint-free cloth. Drain Plug

Reinsert the drain plug and then purge the sampling

cock.
Close the drain valve and remove the drain plug again,
remembering to be prepared to catch left over oil from Sampling Cock Water and Debris Accumulation

the sampling cock purge.

Clean the inside of the valve again. Figure 2 — Debris and Water in Sampling Cock

Install brass, bronze, black iron, or stainless steel adapt-

ers to the drain valve and then to a hose barb so that
tubing can be attached.
A diagram of a two-inch globe valve is shown in Figure 1.
Globe valves are used most often in transformer construc-
tion as they provide the best seal against pressure and
vacuum.
the main drain valve. Once the adapters are all installed
with the hose barb, the final assembly may resemble the
shown in Figure 3.
Once the valve is totally prepared, sampling can com-
mence. The practices as referenced previously all provide very
detailed information concerning taking the actual sample.
However, listed below are additional points to remember:

Hand Wheel

Valve Stem

Packing Nut

Sampling Cock Valve Seat

Valve Body Sampling Cock

Valve Body
Valve Opening
Adapter

Side View – Cutaway Front View

Hose Barb

Figure 1 — Globe Valve Diagram Tubing

Although the procedure listed above sounds like a lot of

extra work, it is necessary in order to retrieve a sample free
from outside contamination. Doble Engineering recom-
mends that samples be retrieved from the main drain rather Figure 3 — Drain Valve with Adapter, Hose Barb and Tubing
than the sampling cock. Although convenient, the sampling
cock is connected by a very small hole to an area between
the drain plug and the valve seat. This is the area that ac-
cumulates all the debris and water as shown in Figure 2. Flush at least two to four liters of dielectric liquid
Special care must be taken to purge this area. through the valve prior to taking a sample.
Even after repeated flushings, the sampling cock is rarely If taking both syringe and bottle samples, take bottle
totally clean, and water and debris will break free and sub- samples first and syringe samples last.
sequently contaminate the sample. However, flushing of the
sampling cock is important as it does remove a large portion Rinse bottles two to three times with about one third of
of the water and debris prior to taking the sample through their volume prior to taking actual samples.
42
Rinse syringes two to three times prior to taking actual
samples.
Fill the bottles without causing aeration or turbulence
to the oils.
If using glass bottles, fill to about 2 to 3 cm of the top
and secure the caps.
If using metal cylinders, metal cans or plastic bottles fill
to overflowing and close or cap.
The flushing procedure is very important in order to re-
move debris and water from the valve to get a sample that
reflects the bulk liquid insulation. Cast iron valves tend to
retain more moisture on valve walls then do brass, bronze,
or stainless, so more flush liquid may be required.
De-energized equipment may require more flush liquid
(eight to 15 liters) to clean the valve, as more condensation
of water occurs and settles to the bottom as the apparatus
cools. On low-volume apparatus this should be monitored
closely.
As mentioned earlier, when multiple samples are re-
quired, Doble recommends taking the bottle sample first and
syringe sample last for several reasons. One reason is that the
water content is usually performed on the syringe sample
so in addition to the original flushing that is performed,
taking the bottle samples first provides additional flushing.
The other reason is the syringe sample is used for the DGA
test, which is the most critical of all the tests, as it provides
information on the operating condition of the transformer.
Hopefully, the additional flushing caused by the filling of
the bottle sample will provide a syringe sample that is best
representative of the bulk liquid insulation.
Rinsing a bottle several times removes any debris remain-
ing from the bottle manufacturing process and conditions
the container to receive the sample by warming the walls of
the container so water condensation does not occur during
sampling. The same is true of the syringe, where flushing and
purging helps to remove any debris and moisture, coats the
plunger to create an adequate seal, and helps to remove air
bubbles. Once the syringe is filled, any air bubbles remaining
must be quickly removed. However, if gas bubbles appear
after the dielectric liquid has cooled then do not release those
bubbles, as they are gases that have just come out of solution
but still comprise the sample. Syringe samples must also be
shielded from the sunlight to prevent photo-degradation
of sample.
When filling bottles with the dielectric-liquid sample,
aeration and turbulence must be avoided. Aeration and tur-
bulence will cause air and water to be trapped in the sample,
thus increasing the water content and possibly affecting
some of the other properties of the oil. Glass bottles are not
filled to the very top to avoid breakage due to the expansion
or the contraction of the liquid. Metal cylinders, metal cans,
and plastic bottles do not suffer from this problem and,
therefore, may be filled to overflowing and sealed.
Insulating Oils Handbook
Cleanup
Cleanup is a necessary step of the sampling activity. The
area should be left cleaner than found so possible hazards
can be minimized for the next sampling crew. Make sure
that drain valves and sampling cocks are wiped clean of oil
and closed tightly. Replace drain plugs using the correct
type of pipe sealant so that they are easily removed the next
time the apparatus is sampled. Remove all debris and tools
from the area, and clean up any liquid spillage. In addition,
record the top oil temperature of the apparatus so that the
relative saturation can be calculated, and make sure that the
samples are adequately labeled.
Observations and Recommendations
Sampling of the liquid dielectrics from electrical ap-
paratus is a nonintrusive way to permit access to helpful
information regarding the operating condition of the ap-
paratus. Physical, electrical, and chemical tests performed
on the sample can permit an educated determination of the
condition of the solid and liquid insulation in the apparatus
or the existence of an incipient fault condition.
Tests of the liquid dielectric usually supplement tests
performed on the apparatus proper, often supporting or
assisting in the interpretation of such tests. These tests are
especially helpful in cases where apparatus cannot be re-
moved from service for complete testing or where complete
testing can only be performed infrequently.
The first requirement of liquid dielectric testing is a
representative sample of the material in question, what has
been referred to as “good” samples. A “bad” sample is not a
sample. A “bad” sample represents the loss of time and ex-
pense and the possible overlooking of a hazardous condition
in its incipient or developing stage. Bad sampling practices
resulting in bad samples are to be avoided.
This article has attempted, sometimes repetitiously, to
summarize reasonable and effective sampling procedures.
It has discussed sampling containers, their labeling, sam-
pling care, sampling techniques, cleanliness before and
after sampling, sample storage and transport, and safety,
and environmental concerns. A review of practices on each
client system, in light of these recommendations, may result
in some “tightening up” of routine, familiar practices and
lead to fewer “bad” and hopefully all “good” samples in the
future.
Lance Lewand received his BS degree at St. Mary’s College of
Maryland in 1980. He has been employed by the Doble Engineering
Company since 1992 and is currently the Laboratory Manager in the
Doble Materials Laboratory and Product Manager for the DOMINOTM
product line. Prior to his present position at Doble, he was the Manager
of the Transformer Fluid Test Laboratory and PCB and Oil Services at
MET Electrical Testing in Baltimore, MD. Mr. Lewand is a member
of ASTM committee D 27.
Insulating Oils Handbook 43

Corrosive Sulfur
in Transformer Systems
NETA World, Fall 2003

by Lance R. Lewand
Doble Engineering Company

Corrosive sulfur and the effect that it has in transformer Table 1

systems can be significant. The extent of the corrosion
Sulfur and Sulfur Compounds Found in Crude Oil
damage caused by sulfur can be so severe as to cause failure
of the apparatus. The problems with corrosive sulfur have GROUP CHEMICAL REACTIVITY
been recognized for quite some time. As early as 1948, FORMULA
F.M. Clark and E.L. Raab issued a report on the subject Elemental (Free) S Very Reactive
for method development within ASTM. Sulfur is found Sulfur
in many materials of transformer construction, including Mercaptans (thiols) R-SH Very Reactive
the copper, paper insulation, gaskets, and oil. Not all sulfur Sulfides (thio-ethers) R-S-R1 Reactive
compounds are considered corrosive, but the tendency to
operate transformers at substantially higher temperatures Disulfides R-S•S-R Stable
can aggravate an already present corrosive sulfur condition Thiophenes Five-membered ring Very Stable
or convert stable compounds into reactive ones that will containing sulfur
cause damage.
R = paraffin with straight or branched chain hydrocarbon or cyclic
hydrocarbon
Presence of Sulfur in Mineral Oil
There are different types of sulfur compounds found in
refined transformer oil. Not all types, however, are consid- process is to either remove or convert many of the corrosive
ered to be corrosive or reactive. Elemental sulfur and sulfur and reactive sulfur species (i.e., elemental sulfur, mercaptans,
compounds in concentrations up to 20 percent are present and sulfides) to more stable compounds such as thiophenes
in the crude oil used to make transformer oil. There are five in an unsaturated ring and disulfides in a saturated form.
basic groups of sulfur and sulfur compounds found in crude The steps in the refining process that aid in this effort are
oil (see Table 1). atmospheric distillation at various temperatures, vacuum
As shown in Table 1, elemental sulfur and the sulfur- distillation, catalytic reaction, and hydro-treating and
containing mercaptans are very reactive followed by sulfides. hydrogeneration. It should be recognized that the refin-
Reactive sulfur is mainly in the form of organic sulfur com- ing process is not always totally successful, as incomplete
pounds like R-SH, where the sulfur is attached at the end of refining may leave small quantities of mercaptans behind
an organic molecule. When the molecule is more complex, or the hydrogenation process may produce elemental sulfur
for instance when the sulfur is surrounded or contained as opposed to hydrogen sulfide.
within the molecule, then the sulfur compounds are more After refining, there is some sulfur, left but the total sulfur
stable and less reactive, as in R-S S-R. Thiophenes are the (comprised of the five groups listed in Table 1) remaining
most stable of all these sulfur compounds. in new oil product is expected to be from 0.02 percent to 1
Some sulfur compounds can actually aid in the oxidation percent. This information was slightly dated so the Doble
stability of the transformer oil and may also act as metal Materials Laboratory analyzed several samples and found
passivators and deactivators reducing the catalytic effect that most oils had a very low total sulfur content as shown
on oil oxidation in transformers. The goal of the refining in Table 2.
44
Table 2
Total Sulfur Content in some US Oils
Oil Sulfur Content
Calumet Caltran 60-08 0.006%
Cross Oil CrossTran 106 0.012%
Ergon Hyvolt II 0.006%
San Joaquin Hytrans 61 <0.001%
Shell Diala AX from Deer Park 0.073%
What Is Corrosive Sulfur?
Corrosive sulfur species are defined as all organic sulfur
compounds that will react with mercury to form sulfides,
such as mercaptans. Elemental (free) sulfur is very reactive
and will react to form corrosive acids. It has been suggested
that low elemental sulfur levels (low ppm range) can cause
a corrosive condition. Doble has determined that elemental
sulfur concentrations as low as 1 ppm can be corrosive to
copper.
Reactions of Corrosive and Reactive Sulfur
Corrosive and reactive sulfur compounds can react on
contact with copper and other metals. Copper is, by far, the
least resistant metal to a sulfur attack. Effects of elemental
sulfur are even more problematic as its ability to combine
with copper does not require heat to promote the reaction.
In oxygen deficient environments such as those found in
sealed, gas blanketed, and sealed conservator transformers,
corrosive and reactive sulfur species combine with copper,
aluminum, and other metals to form copper or cuprous sul-
fide (Cu2S), aluminum sulfide (Al2S3) and other inorganic
sulfides. Copper sulfide is black, gray, green, blue, or violet
in color and is sometimes confused with carbon. Aluminum
sulfide is a yellowish-gray material that can become very
gray in the presence of oxygen and water.
In the presence of an oxygen environment such as that
available in sealed transformers that have a significant leak,
free breathing transformers, free breathing conservator
transformers, and other free breathing apparatus such as
OCBs and LTCs, different types of compounds are formed
from the reaction of sulfur with metals. Oxygen can also
become available from the copper itself. The copper used for
the windings is usually CDA-110 (UNS-C11000) material
for ETP copper which is termed electrolytic tough pitch.
This specification has a minimum copper purity requirement
of 99.90% and it is not considered an oxygen free material
(<5 ppm). There is not an oxygen specification for CDA-
110 but it is usually contains about 500 ppm of oxygen or
less. Copper winding samples that the Doble Materials
Laboratory has tested contained around 200-250 ppm of
oxygen. Different grades of copper can contain much higher
amounts of oxygen that is then available for use in a reaction
with sulfur. Therefore, the manufacturer of the transformer
Insulating Oils Handbook
must be careful in selecting the correct grade of material
for construction. Reactions involving oxygen, sulfur, copper,
aluminum, or other metals can produce copper or cuprous
sulfite (Cu2SO3), copper sulfate (CuSO4), aluminum sulfate
[Al2(SO4)3], and other inorganic sulfates. Copper sulfite is
usually white to pale yellow in color whereas copper sulfate
is white or pale brown in color and aluminum sulfate is a
white material.
Noncorrosive to Corrosive
One of the major questions is if noncorrosive sulfur
species can be converted to corrosive and reactive species
in a transformer. Experience has shown that noncorrosive
sulfur can become corrosive after being exposed to elevated
temperatures on hot metal surfaces and thus produce metal
sulfides. This attack would corrode the metal surfaces. To
make matters worse, the corrosion material could detach
and become nuclei for discharge and gas inception [3]. This
may not be of concern with oils with low sulfur contents
that pass the corrosive sulfur test as the quantity of corrosive
sulfur compounds produced may not be sufficient to cause
extensive damage.
The published literature does not detail if arcing in a
transformer can change stable sulfur species into reac-
tive or corrosive forms. The Doble Materials Laboratory
performed experiments in which a voltage of 25 kV was
applied to sustain an arc through a Cross Oil CrossTrans
106 transformer oil sample in a test cell equipped with a
needle to sphere with a 0.1 inch gap. The CrossTrans 106
sample was found to be noncorrosive prior to testing. The
arcing produced the following gases:
Table 3
Combustible Gases Produced
from Arcing in CrossTrans 106
Gas Concentration
Hydrogen 2370 ppm
Methane 255 ppm
Carbon monoxide 32 ppm
Ethane 38 ppm
Ethylene 496 ppm
Acetylene 4375 ppm
The oil was tested for total sulfur, elemental sulfur and
organo-sulfur species before and after the experiment was
conducted. Tables 4, 5, and 6 document this information.
Insulating Oils Handbook 45

Table 4 Table 5
Total Sulfur Content of CrossTrans 106 Elemental Sulfur Content of CrossTrans 106
Before Arcing Experiment 0.0118% Before Arcing Experiment <1 ppm

After Arcing Experiment 0.0117% After Arcing Experiment <1 ppm

Table 6
Organo-Sulfur Content of CrossTrans 106
COMPOUND Before After COMPOUND Before After
Arcing Arcing Arcing Arcing
hydrogen sulfide <1 ppm <1 ppm ethyl methyl disulfide <1 ppm <1 ppm
sulfur dioxide <1 ppm <1 ppm 2-ethyl thiophene <1 ppm <1 ppm
carbonyl sulfide <1 ppm <1 ppm 2,5-dimethyl thiophene <1 ppm <1 ppm
methyl mercaptan <1 ppm <1 ppm 3-ethyl thiophene <1 ppm <1 ppm
ethyl mercaptan <1 ppm <1 ppm 2,4 & 2,3-dimethly thiophene <1 ppm <1 ppm
methyl sulfide <1 ppm <1 ppm 3,4-dimethyl thiophene <1 ppm <1 ppm
carbon disulfide <1 ppm <1 ppm methyl ethyl thiophenes <1 ppm <1 ppm
isopropyl mercaptan <1 ppm <1 ppm trimethyl thiophenes <1 ppm <1 ppm
ethylene sulfide <1 ppm <1 ppm tetramethyl thiophenes <1 ppm <1 ppm
tert-butyl mercaptan <1 ppm <1 ppm benzothiophene <1 ppm <1 ppm
n-propyl mercaptan <1 ppm <1 ppm methyl benzothiophenes <1 ppm <1 ppm
ethyl methyl sulfide <1 ppm <1 ppm dimethyl benzothiophenes <1 ppm <1 ppm
thiophene <1 ppm <1 ppm trimethyl benzothiophenes <1 ppm <1 ppm
sec-butyl mercaptan <1 ppm <1 ppm tetramethyl <1 ppm <1 ppm
benzothiophenes
isobutyl mercaptan <1 ppm <1 ppm dibenzothiophene <1 ppm 2 ppm
ethyl sulfide <1 ppm <1 ppm methyl dibenzothiophenes 11 ppm 19 ppm
n-butyl mercaptan <1 ppm <1 ppm dimethyl dibenzothiophenes 46 ppm 47 ppm
methyl disulfide <1 ppm <1 ppm trimethyl dibenzothiophenes 54 ppm 59 ppm
2-methyl thiophene <1 ppm <1 ppm tetramethyl dibenzothiophenes 17 ppm 26 ppm

3-methyl thiophene <1 ppm <1 ppm Unidentified volatile sulfur <1 ppm 11 ppm
tetrahydrothiophene <1 ppm <1 ppm Total 128 ppm 147 ppm
46 Insulating Oils Handbook
The arcing did not reduce the total amount of sulfur left
in the oil nor did it convert any of the sulfur compounds
already present into free sulfur. However, conversion of some
of the thiophene compounds did occur (see Table 6). The
presence of sulfates and sulfites cannot be determined by this
method and it may be that those were the compounds that
were converted. This may explain the differences between the
before and after test results. What is of significant interest is
the amount of unidentified volatile sulfur compounds that
were created through the arcing process. No mercaptans
or sulfides (corrosive or reactive sulfur) were formed. The
lack of these sulfur species may be a result of the energy
applied through arcing that could have been so severe that
any reactive or corrosive species that were produced were
instantaneously degraded and thus none remain.
The next article will detail the sources of sulfur in a
transformer system and how it impacts the insulation of a
transformer.

References
1. Clark, F.M. and Raab, E.L. “The Detection of Corro-
sive Sulfur Compounds in Mineral Transformer Oil”,
ASTM Publication, Presented at the Society Meeting,
June 21-25, 1948, pp. 1201-1210.
2. Lipshtein, R.S. and Shakhnovish, M.I. Transformer Oil,
2nd Edition, Israel Program for Scientific Translations,
Jerusalem, 1970.
3. Wilson, A.C.M. Insulating Liquids: Their Uses, Manu-
facture and Properties, IEEE, New York and UK, 1980
4. Oommen, T.V. “Corrosive and Non-corrosive Sul-
fur in Transformer Oils”, Electrical/Electronics Insu-
lation Conference, Chicago, October 4-7, 1993.

Lance Lewand received his BS degree at St. Mary’s College of Mary-

land in 1980. He has been employed by the Doble Engineering Company
for the past seven years and is currently Project Manager of Research
in the materials laboratory and Product Manager for the DOMINOTM
product line. Prior to his present position at Doble, he was the Manager
of the Transformer Fluid Test Laboratory and PCB and Oil Services at
MET Electrical Testing in Baltimore, MD. Mr. Lewand is a member
of ASTM committee D 27.
Insulating Oils Handbook 47

Sources of Sulfur
in Transformer Systems
NETA World, Winter 2003- 2004

by Lance R. Lewand
Doble Engineering Company

A previous article provided an introduction to sulfur ments present in the sample and can provide the percentage
in transformer systems and some of the negative effects concentrations of each of these elements, thus providing an
that corrosive or reactive sulfur can have on a transformer. elemental breakdown of the material or particles. Results
However, transformer mineral oil is not the only material from two gaskets are shown in Tables 1 and 2.
containing sulfur, and this article will explore some of the
other sources. Sulfur compounds are also present in gaskets, TABLE 1
some water-based glues, copper, and paper insulation used Elemental Composition of an O-ring
in the manufacture of transformers. Sulfur can also be
introduced into the transformer through accidental means Radiator Gasket
such as through the use of incompatible hoses. ELEMENT Outside Surface Inside Surface
It is generally accepted that older gaskets used in trans-
Organic Component ≈80% ≈80%
former applications were made from cork, cork/glyptal, and
corkprene. In more recent years, the most oil-compatible Silicon 8.0% 2.1%
gaskets have been nitrile rubbers such as BUNA-N, fluo- Zinc 5.4% 8.3%
roelastomers such as VITON®, or fluorosilicone rubbers. Sulfur 4.3% 8.2%
Properly made nitrile rubbers of the correct grade (butadiene
acrylonitrile) and fluoroelastomers (fluorinated hydrocar- Titanium 1.2% 0.0%
bon) are excellent gaskets for use in transformers. In the Calcium 1.1% 1.2%
manufacture of some of these materials such as the nitriles, Aluminum 0.0% 0.2%
sulfur is used in the curing process when the formulations
are being developed into a hardened material. The curing
process is supposed to eliminate all sulfur from the finished TABLE 2
product. Most gasket manufacturers assume the sulfur is
eliminated after the curing process. In some cases, the con- Elemental Composition
centration of sulfur contained in the final gasket product of a Butterfly Valve Flat Gasket
is not monitored. ELEMENT Outside Surface Inside Surface
Doble Engineering performed scanning electron mi-
Organic Component ≈80% ≈80%
croscopy/energy-dispersive x-ray analysis (SEM/EDX)
on numerous gaskets taken from recently manufactured Silicon 9.5% 9.0%
transformers. Each gasket was prepared for analysis by Aluminum 4.9% 5.7%
cleaning the outside surface with a sulfur-free hydrocar- Zinc 2.4% 2.3%
bon solvent. The gasket was then cut lengthwise to reveal
the inside surface. The outside and inside surfaces of each Sulfur 2.1% 2.1%
gasket were coated with evaporated graphite. The samples Copper 0.4% 0.0%
were then subjected to SEM/EDX analysis in which an Calcium 0.3% 0.3%
electron beam of the scanning electron microscope enters
Titanium 0.3% 0.3%
the bulk of a sample producing an x-ray emittance. The
x-ray peak positions along the energy scale identify the ele- Iron 0.1% 0.3%
48 Insulating Oils Handbook
Both gaskets contained a large amount of sulfur, especially Note: SEM/EDX analysis cannot quantify elements
the O-ring gasket. The SEM/EDX analysis was performed such as fluorine, nitrogen, oxygen, carbon, hydrogen, and
on the inside surface of the gasket to determine if the out- boron but can sometimes give a qualitative indication of
side surface had possibly been contaminated with corrosive the amount in high enough concentrations.
sulfur from the oil. It is clear that sulfur is a component of As shown in the two tables, the fluoroelastomer shows
both original gaskets. The original formulations for a nitrile no sulfur on the inside surface and very little on the outside,
rubber, fluoroelastomer, or a fluoro-silicone rubber (a fluoro- suggesting that material was cured correctly. The opposite is
polydimethylsiloxane) do NOT contain any sulfur. true of the Parker nitrile material, which shows a very high
In discussions with elastomer manufacturers, it was found sulfur content on both surfaces, suggesting that the sulfur
that very few manufacturers (except for E.I. DuPont) were was not removed after the curing process.
performing any chemical testing on the finished product to Water-based glues, used to secure the paper insula-
determine what amount of sulfur, if any, remained. There also tion during manufacture, may sometimes contain sulfur
does not appear to be any standard on what percentage of compounds. There has been at least one known instance in
sulfur should remain in the final product. It then becomes which the glue used in the manufacture of the windings has
obvious that the onus is on the final user of the material to contributed to a corrosive sulfur condition.
specify a sulfur-free or low-sulfur material for use or to test Most coppers used in manufacturing transformer wind-
the material prior to use. ings contain some impurities. Sulfur happens to be one of
In light of this information, additional SEM/EDX analy- those impurities, along with silver, arsenic, phosphorous,
sis was performed on gasket material available in the Doble tellurium, and oxygen. The amount of sulfur allowed in most
Materials Laboratory. One sample was a fluoroelastomer, of the electrical grades of copper is 15 parts per million or
and another was a nitrile rubber produced by Parker. The less. Analyses performed on random copper samples from
results are shown in Tables 3 and 4. windings showed that the sulfur contents were very low at
five parts per million or less. However, care still has to be
taken in the selection of materials used in construction so
TABLE 3 that copper with a high sulfur content is not used.
The pulping process for coated transformer paper, such
Elemental Composition of a Fluoroelastomer Gasket as electrical Kraft paper, converts wood chips to cellulose
ELEMENT Outside Surface Inside Surface by removing the majority of lignin (95-98.5 percent) and
other impurities. There are two basic processes:
Organic, Fluorine >90% >90%
Component
• The sulfite process is considered an acidic process and
Calcium 6.88% 7.40% uses sulfur dioxide, sulfuric acid, and calcium bisulfite.
Magnesium 1.06% 2.47% • The main process used today and the one that is used
Chlorine 0.31% 0.13% to produce electrical grade coated transformer papers is
Silicon 0.40% 0.0% the sulfate process, which is also called “alkaline pulp-
ing.” Sodium hydroxide and sodium sulfide are used
Phosphorous 0.40% 0.0%
in what is termed the “cooking process.” The cooking
Aluminum 0.36% 0.0% process under conditions of heat, pressure, and chemi-
Sulfur 0.33% 0.0% cals (pulping liquors) removes the lignin and impuri-
Potassium 0.26% 0.0% ties from the wood chips in order that only cellulose
remains. The pulping liquor is removed and recycled for
use again, and the remaining cellulose pulp is washed
several times to remove as much of the pulping liquor
as possible from the cellulose pulp.
TABLE 4
Elemental Composition of a Parker Nitrile Gasket The Kraft process is slightly different, in that the same
ELEMENT Outside Surface Inside
chemicals are used, but the pulp is intentionally undercooked
Surface
and results in the darker color of the paper as well as ex-
ceptional mechanical strength. The pulp fibers in the Kraft
Organic Component >95% >95% process do absorb some of the sulfur compounds that can-
Sulfur 2.46 2.73 not be removed via the washing/rinsing process. Tests were
Zinc 1.79 1.85 performed to determine how much total sulfur remains in
the finished paper products. The first analysis performed was
Aluminum 0.40 0.21
SEM/EDX analysis of new Kraft and thermally-upgraded
Silicon 0.35 0.21 (TU) Kraft papers from United States manufacturers. These
results are listed in Table 5 and are for the surface of the
paper only.
Insulating Oils Handbook 49
In addition, several different samples of Kraft paper corrosive or reactive. In these cases, appropriate material
insulation were analyzed for total sulfur and total sulfate compatibility testing should screen out these materials
content. The results are present in Table 6. before they are used in transformer construction.

TABLE 5 Conclusions
Many internal and external sources of sulfur exist within a
Surface Composition of Kraft and TU Kraft
transformer besides the oil. Internal sources include copper,
ELEMENT KRAFT TU-KRAFT paper, gaskets, glues, and possibly other materials. External
Organic Component >95% >95% sources usually include inadvertent contamination from
incompatible materials such as oil-transfer hoses.
Calcium 3.6% 2.5%
Sulfur 0.4% 1.1%
Silicon 1.1% 1.4%
References
Casey, James P., ed. Pulp and Paper: Chemistry and Chemi-
cal Technology, Interscience Publishers, Inc., New York,
TABLE 6 1952.
Sulfur Composition in Various Electrical Papers
PAPER Total Sulfur Total Sulfate Lance Lewand received his BS degree at St. Mary’s College of
Content* Content* Maryland in 1980. He has been employed by the Doble Engineering
Kraft Paper-1 700 ppm 205 ppm Company since 1992 and is currently the Laboratory Manager for the
Doble Materials Laboratory and Product Manager for the DOMINO®
Kraft Paper-2 300 ppm <7.5 ppm product line. Prior to his present position at Doble, he was the Manager
of the Transformer Fluid Test Laboratory and PCB and Oil Services at
TU-Kraft 700 ppm 158 ppm MET Electrical Testing in Baltimore, MD. Mr. Lewand is a member
Kraft Crepe Paper-1 600 ppm 93 ppm of ASTM committee D 27.
Kraft Crepe Paper-2 500 ppm 30 ppm

*Total sulfur analysis was performed by ASTM Method 129,

and total sulfate analysis was performed by EPA Method 300.0

As shown in Table 6, the amount of sulfur varies between

electrical paper manufacturers — sometimes considerably.
The amount of sulfur present is fairly significant in most
of the samples. The amount of reactive or corrosive sulfur
in relation to the total is unknown, although it is assumed
that the amount of sulfates in the sample is at least the
minimum amount.
Accidental contamination of the transformer oil with
corrosive and reactive sulfur compounds can occur by use of
incompatible materials or contaminated processing equip-
ment to transfer oil. For example, hoses made from natural
rubber or gasoline hoses both contain high amounts of sulfur
easily transferred to the oil being pumped through them.
Extra care must be exercised in the selection of hoses so that
no incompatibility exists. Oil-processing equipment runs the
risk of being contaminated from processing a transformer
with corrosive and reactive sulfur, and thus contaminating
the next transformer to be processed. The best safeguard is
to check the remaining oil left in the processing equipment
prior to its next use.
As described above, several materials in the transformer
will contain sulfur such as copper, paper, and oil. In some
cases, the sulfur species in question are stable or are so tightly
bound in the material that they would not be available for
reactions. In other cases some of the sulfur compounds are
50 Insulating Oils Handbook

Hot Oil Reclamation:

Why Is It Necessary?
PowerTest 2003
(NETA Annual Technical Conference)
Scott D. Reed
Environmental Protection Services

Introduction transformer consists of two components: oil and cellulose

paper. Although the combination of oil and paper offers
Maintaining transformers is not a new idea, but during
an excellent dielectric medium for the transformer, it is the
times of tough economic conditions, it is even more impor-
combination of the oil and the paper that will ultimately
tant to protect expensive equipment and prevent extensive
destroy the transformer. The interaction of oil with insulat-
outages. What makes this even more critical is that, unlike
ing paper creates byproducts that decompose the insulation
transformers manufactured over forty years ago, new trans-
system. These byproducts will increase the moisture in
formers are computer designed with minimal engineering
the transformer as well as create an environment that will
margins. As a result, new transformers are designed with
overheat the transformer. Heat and moisture will cause a
tight operating clearances. Any prolonged overloading or
transformer to fail. In time, the formation of these byprod-
overheating conditions can dramatically reduce the lifespan
ucts can cause the failure of the transformer.
of the transformer. Thus, it is important for transformer
owners and service companies to monitor and test trans-
former oil on a regular basis. The testing frequency should Insulation Analysis
be based on the oil condition. The question is, as a service In order to treat the “problem” within a transformer, it is
company, are you promoting transformer oil testing for your important to understand how the “problem” comes to exis-
customers, and more importantly, how do you help your tence. Moisture can either enter the oil externally through
customers with bad transformer oil? condensation or internally through chemical reaction.
Given a choice between air or transformer oil, oxygen
Transformer Oil will leave the air and dissolve into the transformer oil. Thus,
a free breathing transformer will begin to transfer oxygen
Oil is the lifeblood of a transformer. In the United States,
from the air to the oil. Oxygen is also a problem for sealed
it is estimated that there are over 30,000,000 transformers in
transformer tanks that underwent a vacuum-filling proce-
service with over 1,000,000,000 gallons of mineral oil. There
dure. A transformer that is vacuum-filled will still contain
are many reasons why mineral oil is used as an insulating
small quantities of oxygen within the oil. This is a problem
fluid in transformers. One, it is an excellent dielectric me-
because oxygen, in the presence of a catalyst, will chemically
dium for insulating the components within the transformer
react with the hydrocarbons within the transformer oil. This
and two; it is a good heat transferring agent to dissipate the
chemical reaction forms byproducts that attack the paper
heat away from the windings to the tank walls and radia-
insulation system of the transformer. The decomposition of
tors. The radiators are designed to maximize the surface area
the paper will release water and peroxide into the oil. Perox-
to effectively cool the oil. The circulation of cool oil back
ide is the first oil decaying byproduct formed. Just as oxygen
through the windings is a continuous process.
has a preference for oil, water and peroxide will leave the oil
But transformers will self-destruct over time. If the
and adsorb into the paper. Thus, paper has an affinity for the
transformer is properly maintained, the destruction process
very byproducts that contribute towards the destruction of
can take many years. But if the transformer is neglected, the
the paper. In addition, the adsorption of water and peroxide
transformer aging process can accelerate and the lifespan
into the paper further contributes to the release of even more
can be quickly shortened. This is because of the insulation
oxygen into the oil. Thus, the problem becomes compounded
system within the transformer. The insulation system of a
Insulating Oils Handbook 51
and in time, creates other decaying byproducts such as acids,
alcohols, and sludges. Initially the paper will adsorb these
decaying byproducts and in effect, act as a filter. Thus, the
Summary of Oil Tests
oil testing results will indicate no breakdown is occurring, and the Test Objective
but in fact, the deterioration process has already begun—it’s
just not detectable. Once the paper is saturated with acids, ASTM Test Test Type Test Significance
sludges and other harmful byproducts, these byproducts will Method
precipitate out and contaminate the oil.
D 791 Interfacial The IFT test measures the pres-
Unfortunately, these byproducts will destroy the insula- Tension ence of soluble contaminants and
tion system and lead to the failure of the transformer. Thus, it oxidation products. A decreasing
is a cyclical “problem” that can never be stopped; it can only value indicates an increase in
be slowed down. By treating the symptoms of the “problem”, contaminants and or oxidation
you can prolong the lifespan of the transformer. products within the oil.
So, knowing that we can never eliminate the “problem” D974 Acid Number The acid neutralization number is
attacking the insulation system within the transformer, the a measure of the amount of acid
only thing to do is to monitor the symptoms. This can be materials present in the oil. As
accomplished by analyzing the oil. By testing the oil, the the transformer ages, the oil will
results can give you a strong indication of what is occurring oxidize and increase in acidity. The
acid value can also increase from
inside the substation transformer. contamination of other foreign
material such as paint, varnish,
Oil Testing etc…

Before you begin oil testing, it is important to identify D1533 Oil Moisture The Karl Fischer Moisture Test
which transformers need tested, how often, and which tests measures water ppm within the
are necessary for analyzing the oil. It is recommended that transformer oil. Note, it does not
measure the water content in the
all substation class transformers have the following tests paper insulation.
performed at least annually. Listed below is a summary of
recommended oil tests that should be performed as well as D-924 Power Factor The power factor test measures
a description of the test11. This list is by no means a com- the dielectric losses of the oil, or
energy that is dissipated as heat. A
plete list of all the tests that can be performed. Rather, it is low value indicates low losses. It is
a recommended list. The decision is up to each customer to a useful test for measuring changes
determine which tests should be conducted and analyzed. within the oil resulting from con-
It is important that when sampling, the individual verifies tamination or deterioration.
that there is positive pressure on the pressure gauge so that
D-877 Dielectric The dielectric test measures the
no air bubbles enter the transformer. Otherwise, serious D-1816 Breakdown ability of the oil to withstand
injury and failure of equipment may result. Voltage electric stress without failure. The
The intention of this paper is not to provide an in depth higher the value, the lower the
chemical analysis on how each test can measure the chemical presence of contaminants such
changes taking place within the oil and the insulation. For as water, dirt, or other conductive
particles.
the purposes of this paper, the information provided about
each test is to serve as a guide for the decision maker. D 3612 Dissolved Gas The DGA test measures various
Analysis gas ppm levels that are present.
Different gases will dissolve in
the oil that indicate various types
of thermal and electrical stress oc-
curring within the transformer.

D-1500 Oil Color The color test is a simple test that

indicates oil quality. The higher
the color number is, the higher
the probability of contamination
or deterioration of the oil.

EPA 8082 PCB Test the PCB ppm level of oil.

52 Insulating Oils Handbook

Oil Analysis Suggested Limits for In-Service Oils

Although it is important to test the oil in transformers, Group II & Group III
it is more important to know how to interpret the data
from the results. No one test can be used independently to Test Group II Group III ASTM Test
Method
determine the oil condition. Rather, all of the results should
be reviewed simultaneously to give a full understanding Method Neutralization
Number, 0.20 0.50 D-974
about what is occurring in both the oil and the transformer. Mg KOH/g (max)
This will allow the customer to review the options and
make a decision as to how to treat the oil. The IEEE Guide Interfacial Tension,
for the Reclamation of Insulating Oil and Criteria for Its Use mN/m (min) 24 16 D-971
(IEEE Std 637-1985) has four group classifications for oil
evaluation. In terms of what is actually occurring in the transformer,
listed below is a simple table that provides a general sum-
Group I: Oils that are in satisfactory condition for con- mary of what’s happening in the transformer based on the
tinued use. neutralization number and the interfacial tension value33.
Group II: Oils that required only reconditioning for fur-
ther service. Transformer Oil Results Analysis
Group III: Oil in poor condition. Such oil should be re- Oil Color Acid (NN) IFT Oil Status Transformer Condi-
claimed or disposed of depending upon eco- tion
nomic considerations. Water White 0.03-0.10 30-45 Excellent Good
Group IV: Oils in such poor condition that it is techni- Yellow Tint 0.05-0.10 27-30 Good Sludge Dissolved in
cally advisable to dispose of them. Oil Solution
Yellow 0.11-0.15 24-27 Marginal Acid coating Insula-
The difference between reconditioned and reclaimed tion, Sludge ready to
oil will be discussed later. Listed below is a table with the deposit in Trans-
recommended threshold levels allowed for the oil to remain former
as a Group I category22. Orange 0.16-0.40 18-24 Bad Sludge in Radiators,
Coil & Core
Reddish-Brown 0.41-0.65 14-18 Very Bad Sludge hardening &
“layering”, insula-
Suggested Limits for In-Service Oils tion is shrinking &
Group I by Voltage Class weakening
Brown 0.66-1.5 9-14 Extremely Radiators blocked
Test Voltage ASTM Bad with sludge, increased
< 69kV 69kV-288kV > 345kV Method operating temperature
Dielectric Breakdown Black Over 1.5 Below 9 High Risk Transformer failure
Voltage, 60Hz, 26 26 26 D-877 is likely
0.100gap(min)
Neutralization Number,
mg KOH/g (max) .20 .20 .10 D-974 It is important to not just consider these two tests as indi-
cators as to whether or not oil treatment is necessary. Other
Interfacial Tension, 24 26 30 D-971 oil quality results should be reviewed as well as dissolved gas
mN/m (min)
analysis (DGA) results should also be considered.
Water ppm (max) 35 25 20 D-1533
Dissolved Gas Analysis
As discussed earlier, the insulation and the oil will
For oil that does not meet the recommended thresholds decompose over time. The rate of decomposition will in-
above, there are two options. One, the oil can be utilized in crease significantly if there are faults present that increase
a lower voltage application, assuming it was utilized above the operating temperature. Degradation will vary based on
a 69 kV application. Two, the oil can be reconditioned or the severity and nature of the fault. Fault processes vary in
reclaimed to meet the Group I classification. Listed below energy dissipation. As such, the gas composition will also
are the thresholds recommended for oil treatment. Upon vary. By drawing a sample and having the gas composition
your own evaluation, you may establish your own threshold analyzed, it is possible to distinguish between three fault
values for treating the oil. processes, which greatly differ in energy characteristics.
These processes are arcing, corona (partial discharge) and
Insulating Oils Handbook 53
pyrolysis (thermal decomposition). For example, gases such Hot Oil Reclamation
as H2, CH4, C2H2, C2H4 and C2H6 are the result of oil The IEEE Draft: Guide for Acceptance and Maintenance
decomposition of the oil while CO and CO2 are the result of Insulating Oil in Equipment, (C57.106/D7) has the fol-
of insulation decomposition44. The IEEE Guide for the In- lowing definitions66:
terpretation of Gases Generated in Oil-Immersed Transformers
(IEEE Std C57.104-1991) has four categories for total Reclamation of Oil: The restoration of usefulness by the
dissolved combustible gases (TDCG). removal of contaminants and products of degradation
such as polar, acidic, or colloidal materials from used
Condition 1: TDCG below this level indicated the trans- electrical insulating liquids by chemical or adsorption
former is operating satisfactorily. means such as Fuller’s Earth.
Condition 2: TDCG within this range indicates greater
than normal combustible levels. Any indi- Reconditioning of Oil: The removal of insoluble con-
vidual combustible gas exceeding specified taminants, moisture, and dissolved gases from used,
levels should prompt additional investiga- electrical insulating liquids by mechanical means, such
tion. as vacuum processing or filtering.
Condition 3: TDCG within this range indicates a high From a technical position, these are different processes
level of decomposition. Any individual to treat different problems. The first is chemical and the
combustible gas exceeding specified levels second is mechanical. However, from a practical position, it
should prompt additional investigation. is necessary to remove all types of contaminants that destroy
Condition 4: TDCG within this range indicates exces- the insulation characteristics of a transformer. Since the
sive decomposition. Continued operation purpose of a hot oil reclamation process is to prolong the
could result in failure of transformer. life of the transformer, a hot oil reclamation process must
have the ability to perform both chemical and mechanical
Below is a table with recommended thresholds for gas oil treatment.
concentration levels55. Specifically, the reclamation process must have the abil-
ity to remove:
Dissolved Gas Concentrations 1. Water from the oil and the windings
STATUS DISSOLVED KEY GAS CONCENTRATION LIMITS (ppm)
2. Contaminants from the oil
H2 CH4 C2H2 C2H4 C2H6 CO CO2 TDCG
3. Acids, alcohols, oxides and other oil
Condition 1 100 120 35 50 65 350 2500 720
decaying byproducts
Condition 2 101-700 121-400 36-50 51-100 66-100 351-570 2501-4000 721-1920
Condition 3 701-1800 401-1000 51-80 101-200 101-150 571-1400 4001-10000 1921-4630
4. Dissolved Gases resulting from over-
heating or electrical stress
Condition 4 > 1800 > 1000 > 80 > 200 > 150 > 1400 > 10000 > 4630
5. Sludge “layered” within the coil/core as-
sembly as well as in the radiators

Oil Maintenance For on-site energized oil treatment, the Hot Oil Rec-
lamation Processing (HORP) rig needs mobilized and set
Thus far we have developed an understanding of what is up at the customer’s substation. Because it is important to
occurring inside the transformer, along with which tests to test the oil initially, as well as monitor the oil as it is being
perform, as well as how to interpret the results. Next, it is reclaimed, the HORP rig needs to be equipped with an
important to consider the options available to protect the on-board laboratory to analyze the oil as it is processed.
transformer. Listed below are different options depend- The process begins by connecting two sets of high-pres-
ing on the specific data for each transformer that needs sured hoses to the transformer and circulating the hot oil
processed. in a closed loop path. The oil needs to be pulled from the
bottom of the transformer and return to the top.
a. Hot Oil Reclamation
b. Alternative Options There are six key components to the removal of harm-
i. Retrofilling ful contaminants within the oil. The six components are as
ii. Vacuum Filling follows:
iii. PCBs
1. Heat
2. Centrifuge
3. Fuller’s Earth
4. Vacuum Degassing/Dehydration Column
54
5. Inhibitor
6. Filter stages
Different companies have different oil treatment process-
ing rigs. Below is a general layout of the EPS HORP rigs.
Heat
Heat is accomplished by literally heating the oil when
it enters the HORP rig. Heating the oil has two purposes:
one, to help remove water adsorbed in the paper insulation
of the transformer; two, to dissolve the sludge layered in the
transformer into a soluble form where it can be removed
from the oil after entering the HORP rig. This is why it
is advantageous to perform the reclamation process on an
energized transformer. The heat generated from the load
will aid in transferring the moisture from the paper to the
oil as well as dissolving the sludge buildup internal to the
core and coil assembly. In addition, the mechanical vibration
at 60 Hertz also aids in the circulation of the oil through
the windings. This circulation is important to remove the
sludge within the windings.
As the temperature increases in the transformer, the water
will transfer from the paper to the oil. This will continue
until the moisture reaches a new equilibrium between the
oil and the insulation. As the oil circulates through the
processing rig, the water is removed before re-entering the
transformer. Below you will find a graph that compares
Insulating Oils Handbook
water absorption between paper and oil. Note the higher
the temperature, the higher the water ppm is in the oil, and
the lower the moisture content is in the paper. Remember,
water has an affinity for paper, but when you introduce
heat, it transfers the moisture into the oil. This is important
because water in the paper will decompose the insulation
integrity77.
The second benefit of heat is at approximately 180 F,
the oil reaches the aniline point and becomes a solvent for
its own decaying products within the oil. As the hot oil
circulates through the transformer, it begins to dissolve the
sludge that has built up within the core and coil as well as
the radiators. The insoluble sludge that is carried out by the
oil can be removed via filtering stages within the rig, while
the soluble sludge can be removed via fullers earth, which
will be explained later. The amount of sludge accumulated
in the transformer will affect how many circulation passes
are required. Transformers in poor condition may require as
many as 20 passes to desludge the core and coil assembly.
A pass is defined as the total gallons processed through the
HORP rig divided by the nameplate gallon rating. Thus, if a
transformer has 2,000 gallons, and the HORP rig circulates
40,000 gallons, then 20 passes were performed on the oil.
Centrifuge
A centrifuge is not always necessary, but if there is too
much moisture or other contaminants in the oil, then it is
beneficial to utilize a centrifuge to aid in the hot oil reclama-
tion process of the transformer. A centrifuge is an excellent
means to separate out insoluble contaminants such as water
and carbon from the oil. A centrifuge does not remove dis-
solved water or other soluble contaminants (e.g. acid), but it
is important to have this option on a HORP rig because it
can remove free water and other suspended contaminants.
Eventually, water and other contaminants can be removed
in multiple stages via vacuum dehydration, fullers earth,
and filters, but a centrifuge can dramatically decrease the
processing time by removing the excess contaminant in one
Insulating Oils Handbook
stage. The other stages are critical to the reclamation process,
but by utilizing a centrifuge, the oil treatment process can
be faster and therefore cheaper for the customer.
Fullers Earth
Fullers earth is a naturally mined adsorbent clay. Adsorp-
tion is the attraction of one substance to another substance.
Fullers earth will adsorb water, acids, sludge, alcohols and
other oil decaying products. Thus, as the oil passes through
the fuller’s earth, the adsorption process will remove the dis-
solved acid, sludge and water from the oil. In addition, fullers
earth will also adsorb microscopic particles and suspended
carbon created from internal arcing. As a result, fullers earth
improves both the power factor and the interfacial tension
of the transformer oil while subsequently reducing the
acid content. The continuous circulation of hot, dry clean
oil through the energized transformer will also clean the
transformer as it continues to dissolve sludge, oil decaying
products and moisture from the insulation. Upon return-
ing the “dirty” oil through the HORP rig, the adsorption
process will again remove the contaminants from the oil
before returning to the transformer. This process is repeated
until all of the incoming oil properties meet the minimum
standards as outlined by the customer and the contractor.
Attached is a graph that outlines the effectiveness of fuller
earth in relationship to the acid content of the oil89.
Vacuum Degassing-Dehydration
Before the oil returns to the transformer, it is important
to have the oil pass through a vacuum degassing/dehydra-
tion column. Although there are several ways to achieve a
high vacuum on the oil, the general idea is to maximize the
surface area of the oil in order to remove moisture, oxygen
and other gases. The vacuum degassing column will also
55
remove volatile forms of acids, but fullers earth will be
the primary method to extract acid from the oil. Since the
circulation process will continue to remove moisture from
the insulation, it is important to continuously operate the
dehydration column so that wet oil does not return to the
top of the transformer.
Inhibitor
Most transformers have an inhibitor additive to the oil.
An inhibitor is important because it aids in the resistance
of oxidation. Most inhibitor utilized in transformer oil is
2,6 di-tertiary-butyl para-cresol (DBPC). Because of the
reclamation process, an inhibitor is removed from the oil
through the fullers earth bed and the vacuum degassing
column. Thus, during the last pass, it is important to add
DBPC to the oil as it is returning to the transformer for
protection against oil oxidation.
Filter Stages
Because of the dangers of metallic particles and other
foreign particles found inside the transformer, it is critical
to have multiple filter houses throughout the HORP rig to
guard against particles from re-entering the transformer.
Multiple filters throughout the HORP rig ensure the pro-
tection of the transformer. Contaminants and particles in
the oil not only degrade oil qualities such as dielectric and
power factor, but contaminants such as iron particles also
introduces the risk of failure. Thus, the very last step before
the oil exits the HORP rig is to pass through a 0.5 micron
particle filter to trap any remaining particles that have not
been previously captured.
Safety
The advantage of the hot oil reclamation process is that, in
most cases, it can be performed on energized transformers.
Thus, the customer never has to take an outage and incur
additional expenses of setting up a mobile transformer. In
addition, the energized process helps to aid in the clean-
ing of the transformer. Typically, 230kV is considered the
ceiling for performing hot oil reclamation on energized
transformers. But not all transformers qualify. Performing
a hot oil reclamation process on energized transformers is
safe as long as:
1. Properly designed oil processing equipment is utilized
on energized transformers.
2. The crew is well trained and experienced with process-
ing on energized transformers.
3. The transformer size meets the acceptance criteria for
processing energized.
4. The oil condition is safe to process while energized.
The oil processing system must have the ability to elimi-
nate bubble formation. Bubbles are stress points and have
low dielectric strength. As a result, a bubble or a conglomer-
56
ate of bubbles can cause arcing and or failure. Static charges
can also develop when mineral oil flows through pipes and
hoses so it is important to have the system grounded. In ad-
dition to the equipment, the crew must be very well trained
to perform this service on energized transformers — both
for the protection of the personnel and the equipment.
If a transformer is too small, or the voltage rating is
above 230 kV, then it is recommended that the transformer
be processed de-energized. Likewise, if the condition of
the transformer oil is too poor because of high acids, high
moisture, high dissolved gases, low dielectric or low inter-
facial tension, then the transformer cannot undergo a hot
oil reclamation energized. Instead, the poor conditions force
the transformer to be processed de-energized. Every time
a transformer is de-energized and re-energized, there is a
significant amount of electrical and mechanical stress on the
transformer. There is always the added risk that this stress
could lead to failure of the transformer. This is especially
true in older transformers. In addition, the added benefits
of performing the reclamation process energized are lost
when the transformer is reclaimed de-energized. Thus, it is
advantageous to monitor and prevent the transformer from
reaching a point where it cannot be reclaimed energized.
Alternative Options
Retrofilling
In some cases, retrofilling is the preferred option. A
straightforward process, retrofilling is where the oil is
drained and transported for disposal; the transformer is
flushed, and then refilled with replacement oil. Small trans-
formers or transformers without valves are good candidates
for retrofilling. Another good retrofill candidate is oil with
really poor oil qualities. However, it is recommended that the
transformer undergo a hot oil reclamation process following
the retrofill. This is because only 10% of the interior surface
of the windings will be reached during the retrofill process.
Retrofilling cannot remove or clean the sludge deposits
located within the transformer. In addition, the remaining
oil that was not drained will still be oxidized. Like PCBs, a
small quantity of oxidized oil can contaminate a large batch
of oil. Within a year, the transformer oil can deteriorate to
the previous poor condition if the transformer does not
receive a hot oil reclamation process.
It is critical to make sure that there is enough settling
time after a retrofill is performed. Every time a transformer
is drained and refilled, there is an increased probability of
transformer failure upon re-energizing because of the risk
of an air bubble entrapped within the windings.
Vacuum-Filling
In some cases, the primary voltage is too high to perform
a retrofill. Instead, the transformer must be vacuum-filled.
The vacuum-filling process will remove air and water, the
biggest contaminants of an insulating system. In addition, it
is also important to remove the air so that no voids are left
within the insulation. These voids can create corona and elec-
trically stress the windings. How long a transformer needs
Insulating Oils Handbook
vacuum-filled depends on the size of the transformer and
the recommendation of the manufacturer. After a sufficient
amount of vacuum time has passed, then the transformer
is ready to be filled with oil while still under vacuum. This
allows the oil to fill all of the voids. There are cases when
it is important to circulate the oil through the transformer.
This will allow any remaining moisture or dissolved gases
to be removed through the oil circulation process.
PCBs
Because of transformers that have a PCB contamination
problem, most people assume that retrofilling is the only
option for PCB reduction or reclassification. This is not cor-
rect. What makes the reclamation processing rigs owned by
EPS unique is that they are the only mobile rigs permitted
by the United States Environmental Protection Agency to
destroy PCBs. The advantage is that the customer cannot
only receive all of the benefits provided from the reclamation
process, but also offer the added benefit of PCB destruction.
EPS calls this process the PCBXSM process.
Because of the benefits of the PCBXSM process, leach
back rates of PCBs are typically 3-5% of the original PCBs
concentration compared to 8-12% normally associated
with retrofilling. This is the result of the circulation of the
hot oil through the coil and core assembly, which flushes
the PCBs out of the transformer and into the HORP rig
for destruction. The other benefit is, depending on the size
of the transformer; the PCBXSM process is often cheaper
than retrofilling the oil. This is because of the labor costs,
tanker demurrage and quantity of oil required for both the
retrofill and disposal.
Conclusion
It is important to test and monitor transformers. Future
transformer designs will continue to reduce performance
margins, making testing even more important. Relatively
speaking, testing is inexpensive and very informative as to
what is occurring inside the transformer. Oil testing allows
the customer to prevent minor “problems” from turning into
major “problems”. Purchasing a substation transformer is a
major a financial investment. If the customer is proactive
in maintaining the transformer oil, it can prolong the life
of the transformer and delay the inevitable need to replace
the unit. However, when the need arises to treat the oil, it is
important to understand the options available and the how
the processes work to treat the symptoms of the problem.
References
1 1
Dr. Fredi Jakob, “W-ACTI Diagnostic Reference
Guide”, 2002
2 2
IEEE Guide for Reclamation of Insulating Oil and Cri-
teria for Its Use (IEEE Std 637-1985)
3 3
S. Myers, “Transformer Life Can Be Extended”, TMI
Reference Guide
Insulating Oils Handbook 57
4 4
Dr. Fredi Jakob, “W-ACTI Diagnostic Reference
Guide”, 2002
5 5
IEEE Guide for the Interpretation of Gases Generated in
Oil-Immersed Transformers (C.57.104-1991)
6 6
IEEE Draft: Guide for Acceptance and Maintenance of
Insulating Oil in Equipment, (C57.106/D7)
7 7
C. Manger, “High Voltage Equipment Maintenance
101, Mineral Insulating Oil”, TechCon 2002
8 8
C. Bandt, “Filtering Insulating Oil”, Presentation for
American Society for Testing and Materials, 2002

Scott Reed received his BS in electrical Engineering from North

Carolina State University in 1989, where he specialized in power engi-
neering. Since 1999, Scott has served as Vice President of Environmental
Protection Services, a company that focuses on all three phases of a
transformer’s life span: installation (vacuum filling of substation and
power transformers), maintenance (hot oil reclamation), and disposal
(obsolete non-PCB and PCB equipment). Prior to joining EPS, Scott
worked as an engineer for Baltimore Gas & Electric in their substation
group, focusing on relay system protection of transformers. Scott also
serves as a member of the IEEE Power Engineering Society and the
ASTM Electrical Insulating Liquids and Gases Group.
58 Insulating Oils Handbook

Use of Gas Concentrations Ratios

to Interpret LTC Dissolved Gas Data
PowerTest 2003
(NETA Annual Technical Conference)

Presenter
Fredi Jakob, Ph.D., Weidmann-ACTI
Co-Authors
Karl Jakob, P.E., and Simon Jones, Weidmann-ACTI
Rick Youngblood, Cinergy Corporation

I. Introduction
Dissolved Gas Analysis, DGA is based on the fact that
the release of energy in oil filled electrical equipment results
in a partial destruction of insulating fluids and/or solid
insulation. The number of molecules destroyed is relatively
small but detectable amounts of low molecular weight gases
are produced. The quantification of these low molecular
weight “fault gases” is the basis of all DGA.
In the main tank of a transformer very little, if any fault
gas other than carbon oxides should be produced. Carbon
oxides are continuously produced by unavoidable decompo-
sition of cellulose insulation. In contrast, equipment that in-
volves conductor switching operations produces these same
fault gases during normal operation. Initially it was assumed
that the gases produced by the arcing process associated
with switching would “mask” gases associated with equip-
ment problems such as carbon deposition on the contacts,
which results in excessive heating. Youngblood1 was one of
the first investigators to recognize that the gassing pattern
for a healthy load tap changer, LTC, would be significantly
different from that observed in a problem LTC. Specifically,
he suggested that acetylene and hydrogen are generated
during the normal arcing process and the hydrocarbons,
methane, ethane and especially ethylene are generated
when overheating occurs in a problem LTC. These three
hydrocarbons are often called ‘hot metal gases’ since they
are produced when a heated conductor is in contact with
mineral oil based dielectric fluid. Correlation of fault gasses
with specific gas producing conditions is known as the “key
gas” method. Key gas correlations are given in Table 1.
Table 1
“Key” Fault Gasses
Gasses Indication
Hydrogen Partial Discharge, Heating Arcing
Ethylene, Ethane, Methane “Hot Metal” Gasses
Acetylene Arcing
Carbon Oxides Cellulose Insulation Degradation
II. Data Interpretation
A. Concept
The interpretation of DGA data for transformers, LTC’s
and OCB’s is empirical in nature. The development of in-
terpretation protocols for OCB’s and LTC’s parallels the
development of DGA diagnostics for the main tanks of
transformers. Key gases associated with heating problems
are methane, ethane and ethylene. These gases are listed by
Halstead2 in order of increasing energy required for their
production. Initially Youngblood ignored the levels of arcing
gases, acetylene and hydrogen that developed whenever an
LTC operated. Subsequent work by Youngblood5 indicated
that arcing gases are also diagnostically significant. For
example, increased acetylene levels were often followed by
increased heating gases. The increased acetylene is due to
changes in the arc duration and/or characteristics as the
contacts are eroded or covered with carbon.
The next step in the development of diagnostic protocols
for LTC’s was the empirical determination of normal or
threshold values. The gas retention rate in an LTC is very
dependent on breathing configuration, so this is a major
Insulating Oils Handbook 59
factor in determining threshold levels. Free breathing LTC’s
rapidly lose gases to the environment while sealed LTC’s Table 2
retain most of the gas produced. Gas Formation as a Function
Threshold levels have been determined for specific models
and types of LTC’s by Doble3 and Baker4. Generic levels of Operation Count. (Duval5)
have been set by Youngblood5, and are useful for specific
Operations: 500 3600 49000 Gas produced/operation
LTC models where the threshold values have not yet been
determined. Fault gas ratios, which are discussed below, Gas/500 Gas/3600 Gas/49000
should be considered applicable for unit evaluation only
when threshold values are reached. Hydrogen 6870 12125 14320 13.74 3.37 0.29

B. Fault Gas Concentration Ratios Methane 1028 5386 10740 2.05 1.50 0.22

1. Normal Operation Acetylene 5500 35420 53670 11.00 9.84 1.10

Arcing, which generates very high temperatures, occurs
Ethylene 900 6400 35839 1.80 1.78 0.73
with each switching operation in an LTC. Arcing pro-
duces acetylene and hydrogen. When arcing gases are being Ethane 79 400 3944 0.16 0.11 0.08
produced, ‘heating gases’ are also produced since the oil
temperature varies with distance from the arcing contacts.
At the lower oil temperatures, heating gases are produced *Note: Some gas is always lost with time. Therefore, the
more abundantly than acetylene. gas concentration per operation is expected to decrease with
If one accepts this hypothesis, one would expect that operation count. Duval did not provide breathing configura-
the ratios of heating to arcing gases in a problem free unit tions for this data.
would remain fairly constant with operational count. Since
the contact surface is being eroded and additional deposits Heating to Arcing Ratios
are being formed with each operation, these ratios will pre-
dictably increase slightly with operation count, providing Ethylene Ethylene Methane+Ethylene+Ethane
———————— ———— ———————————
there isn’t a large change in the load current. Although the Acetylene+Hydrogen Acetylene Acetylene+Hydrogen
load on a distribution circuit in a sub station does remain
nearly constant over a period of time with minor ups and
Temperature Dependant Ratios
downs, anything that causes a significant change in average
constant load will change the ratios. The data presented in Ethane Ethylene
a paper by Duval5 supports this hypothesis. Note that all ———— —————
Methane Ethane
of the significant ratios in these problem free units are well
below those found in a LTC that has developed a heating
or other problem.
C. Application of Gas Concentration Ratios
As is the case with main tank DGA, ratios are not valid
2. Heating Problems unless at least one of the fault gases exceeds its threshold
Initially a resistive film develops on contacts, which re- value. Threshold values used by Weidmann-ACTI are model
sults in an increase in contact resistance, increased heating specific, whenever this data is available. If model specific
and an increase in heating gas concentrations. Since more data is not available, we use the “monthly watch” levels de-
heating gases are now being produced, the ratio of heating veloped by Youngblood5 as our threshold values. “Normal”
to arcing gases increases. This change in gas concentra- values we use for the ratios represent the 90th percentile of
tions and gas concentration ratios indicates problems. As fault gas concentrations from a very large number of DGA
the resistance increases, the amount of energy dissipated results (~2500 units). These 90th percentile values which are
increases. The ratios of ethane to methane and ethylene to listed in Table 3 are generic and do not take into account
ethane both increase with increasing temperature. Thus these the difference in gassing rates of specific units. Ratios are,
two heating gas ratios should also reflect increased contact as expected, model specific and we hope to refine our results
resistance and heating. as more data becomes available. Table 4 lists the ratios
calculated for a Westinghouse LTC, which according to
Bakers5 threshold levels requires attention.
60 Insulating Oils Handbook

Table 3 Another concept under investigation is normalization of

th
Generic 90 Percentile Fault Gas Ratios fault gas data. We believe that during normal switching op-
erations, the ratio of ethylene to acetylene is fixed. This ratio
R1 R2 R3 R4 R5 should remain constant for different numbers of operations.
Ethylene Ethylene Ethylene+ Ethane+Methane Ethane Ethylene Furthermore, since these two gases have approximately the
———— ———————— —————————— ———— ————
Acetylene Acetylene+Hydrogen Acetylene+Hydrogen Methane Ethane same solubility in mineral oil and approximately the same
escape rates from the oil, the ratios should remain fairly
0.3378 0.5000 0.9157 0.2067 4.83 constant even for free breathing units. For example if the
concentration of both ethylene and acetylene are 100 ppm
Note the difference between these model specific ratios then the ratio is one. If a heating problem is superimposed
listed in Table 4, and the generic values which are listed on the normal arcing process and the gas levels are 175 ppm
in Table 3. for ethylene and 150 ppm for acetylene there would be an
“additional” 25 ppm of ethylene due to the heating problem.
Table 4 One could thus normalize our results using a ratio of:
Unit Specific Ratio Comparison Ethylene Total Ethylene
(Based on values set by C. Baker) (due to heating) (Ethylene produced during normal arcing)
——————— —————————————————
Acetylene Acetylene
WESTINGHOUSE LTC UTT
(due to arcing)

Ethylene Ethane Methane Acetylene Hydrogen CO CO2

LT3 LTC Case Study I
Gas
Level 3,000 250 1,000 5,000 5,000 3,000 10,000
Westinghouse UTT-A 138KV x 69KV x 13.8KV Sealed
Ratio 1 Ratio 2 Ratio 3 Ratio 4 Ratio 5
Date Comments C2H2 CH4 C2H6 C2H4 H2
0.60 0.30 0.43 0.25 12 8/31/92 Removed from Service 8527 3279 1135 9606 9083
McGRAW EDISON LTC 550-BL 12/17/93 Post Repair 501 387 16 375 2883
5/1/94 Normal 541 534 9 313 3800
LT3 2,000 400 400 400 500 1,000 3,000 8/17/95 placed on 6 month watch 648 590 52 836 3995

Ratio 1 Ratio 2 Ratio 3 Ratio 4 Ratio 5

5.00 2.22 3.11 1.00 5 Date Comments Ratio 1 Ratio 2 Ratio 3 Ratio 4 Ratio 5
8/31/1992 Removed from Service 1.13 0.55 0.8 0.35 0.15
12/17/1993 Post Repair 0.75 0.11 0.23 - -
* LT3 is Baker’s designation of a unit requiring inspection. 5/1/1994 Normal 0.58 0.07 0.2 - -
8/17/1995 Placed on 6 Month Watch 1.29 0.18 0.32 0.06 0.09

III. Case Studies Ratio 1 C2H4/C2H2

Two case studies for Cinergy data are presented below. Ratio 2 C2H4/(C2H2+H2)
This data, from Cinergy’s initial work, was interpreted Ratio 3 (CH4+C2H4+C2H6)/(C2H2+H2)
without consideration of fault gas concentration ratios.
Ratio 4 C2H6/CH4
Case history II is of particular interest since serious damage
occurred in the six month interval between scheduled tests. Ratio 5 C2H4/C2H6
The data indicates that critical gas ratios such as the ratio of
ethylene to acetylene doubled from March ’92 to Feb ’93.
This points out the necessity of trending both ratios and 31-Aug-92
gas concentrations. We believe that the extent of the coking Based on the DGA result this unit was immediately re-
would have been less severe if the unit had been inspected moved from service. The fault was determined to be due
in Feb ’93 based on the doubling of the significant ratios, to contact misalignment.
rather than to have waited for six additional months.
17-Dec-93
Typical levels of gasses for a sealed, unit, annual monitoring
IV. Future Work was indicated.
The empirical analysis of DGA data for LTC’s is well
developed. Gas concentration levels and gas concentration 1-May-94
ratios can differentiate between normal and problem units. Again gas levels are typical for a sealed unit, annual moni-
We believe that both the concentration and ratio values will toring continued.
work best if they are model specific. The compilation of these
values require user feedback on problem units. Trending of
both the gas concentrations and ratios is always the best
method to identify incipient problems.
Insulating Oils Handbook 61
17-August-95 formers,” Journal of the Institute of Petroleum, Vol. 59,
Again the Gas levels are indicative on normal operation, Sept. 1959, pp. 239-241.
however the fifty percent increase in the level of Acetylene
indicated increased surveillance. 3. Doble Client Transformer Committee Subcommittee
Report on Transformer Load Tap Changer Dissolved
Gas Analysis September 24, 2001.

LTC Case Study II 4. Baker, Charles. Personal correspondence. 2002.

Federal Pacific TC-25 69KV x 12KV 20MVA 5. Youngblood, R., Baker, C., Jacob, F., Perjanik, N. “Appli-
Desiccant Breather cation of Dissolved Gas Analysis to Load Tap Chang-
ers”.
D a te C o m m e n ts C2H2 CH4 C2H6 C2H4 H2

3 /1 2 /1 9 9 2 A n n u a l D G A T e s t C yc l e 589 60 2 89 144 6. Duval, Michel. “A Review of Faults Detectable by gas-

2 /1 / 1 9 9 3 6 M o n th T e s t C y c l e 1625 342 70 534 3099
in-Oil Analysis in Transformers”. IEEE Electrical In-
sulation Magazine May/June 2002, Vol. 18 no. 3, pp.
8 /1 2 /1 9 9 3 T h e rm a l R u n a w a y 1633 53434 55535 253024 2217

8-17.
D a te C o m m e n ts R a ti o 1 R a ti o 2 R a ti o 3 R a ti o 4 R a ti o 5

3 /1 2 /9 2 A n n u a l D G A T e s t C yc le 0 .1 5 0 .1 2 0 .2 1 - 0 .1 5

2 /1 /9 3 6 M o n th T e s t C y c l e 0 .3 2 0 .1 1 0 .2 0 .2 1 0 .3 3 Dr. Fredi Jakob received his PhD at Rutgers, the State University of
8 /1 2 /9 3 Th e rm a l R u n a w a y 155 66 94 1 .0 3 155 New Jersey, in 1961. He is professor emeritus of analytical chemistry at
California State University-Sacramento and is the founder and laboratory
director of Analytical ChemTech International, Inc. (ACTI), which is a
Ratio 1 C2H4/C2H2 wholly owned subsidiary of Weidmann Systems International. As a long-
term member of ASTM and IEEE and author of over fifty published
Ratio 2 C2H4/(C2H2+H2) articles, Dr. Jakob is a traveling lecturer to private and governmental
Ratio 3 (CH4+C2H4+C2H6)/(C2H2+H2) agencies. He has been invited to speak at American Public Power meet-
ings, ASTM symposia, conferences held by Doble, NETA, and AVO
Ratio 4 C2H6/CH4 conferences, as well as other industrial organizations.
Ratio 5 C2H4/C2H6

12-Mar-92
This unit indicated the early stages of mechanical difficul-
ties. While the Acetylene and Hydrogen levels are elevated,
the level of Ethylene is less than 100 ppm. Indicating a
continuance of annual monitoring

1-Feb-93
At this time the unit was placed on a 6 month monitoring,
due to the elevated, Acetylene, Hydrogen, and Ethylene
levels. At 534 ppm Ethylene immediate removal from
service was not indicated.

12-Aug-96
Too late, by August, the unit was in thermal runaway.
As indicated by the extremely high level of Ethylene, at
253,024 ppm. Repairs included a Tap Shaft board, Slip
Rings, and a New Reversing Switch assembly.

References
1. Youngblood, R., Jacob, F., Haupert, T.J. “Application of
DGA to Detection of Hot Spots in Load Tap-Chang-
ers”, Minutes of the Sixtieth Annual International Con-
ference of Doble Clients, 1993, Sec. 6-4.1.

2. Halstead, W. D., “A Thermodynamic Assessment of the

Formation of Gaseous Hydrocarbons in Faulty Trans-
62 Insulating Oils Handbook

The Negative Effects of Corrosive

Sulfur on Transformer Components
NETA World, Spring 2004

by Lance R. Lewand
Doble Engineering Company

The previous two articles discussed the differences be-

tween corrosive sulfur and stable sulfur and some of the reac-
tions that may occur. This article will discuss and show the
effects of corrosive sulfur on different parts in a transformer
system and the negative impact it can have. Corrosion of
the metal surfaces, especially exposed copper surfaces, is
one of the primary reactions of a corrosive/reactive sulfur
condition. Figure 1 shows NLTC contacts that are overly
contaminated with what appears to be a buildup of carbon
but which is, in reality, copper sulfide. Figure 2
The NLTC contacts shown in Figure 1 were discovered
by accident. A maintenance function involving a power
transformer led to the belief that a piece of hardware had Metal is not the only surface that is affected by corrosive
fallen into the main tank of the transformer. A visual in- and reactive sulfur. Sulfur reactions with paper-wrapped
spection was conducted through the manhole of the unit, conductors can cause some of the conductor, such as copper
and the blackened NLTC contacts were apparent. This or aluminum, to be transferred to the surface of the paper.
caused the unit to be removed from the station, drained, and In some instances, the transfer can be extensive, as shown
inspected by entering through a manhole. After draining in Figure 3.
and entering the unit, personnel wiped the NLTC contacts
with a cloth and discovered that the plating of the contacts
was removed along
with the black mate-
rial. This, of course,
caused concern, and
an investigation was
initiated in which
it was determined
that copper sulfide, Figure 3
not carbon, was the
black, shiny mate-
rial found on the Layer 1: Next to Conductor Side facing conductor
Figure 1
contacts. This type Element Conc.
of corrosion, as shown in Figure 1, could easily lead to
overheating and arcing of these components, thus severely Copper 78.4%
damaging or causing failure of the apparatus. The effects of Sulfur 17.4%
a corrosive/reactive sulfur attack on a metal conductor do Silicon 2.8%
not always result in a black coating. In some cases, a silver
Calcium 4.4%
coating or a silver discoloration of the copper conductor
surface occurs, as in Figure 2.
Insulating Oils Handbook
The analysis on the paper surface shown in Figure 3 was
performed by SEM/EDX. When this type of analysis is
performed on paper, it usually indicates a very high organic
content due to the cellulose composition. In this particular
case (Figure 3), the copper/sulfur contamination that had
been transferred to the paper insulation had been signifi-
cant enough to mask out most of the organic component.
In analyzing this paper, a closer examination of the surface
was conducted and SEM micrographs were produced at a
magnification of 200 times. Figure 4 is a SEM micrograph
of uncontaminated Kraft paper and crepe paper.
Figure 4A — Kraft
Figure 4B — Crepe
The individual paper fibers are clearly visible in both the
Kraft and crepe papers in Figure 4 with no signs of foreign
material present. Even the crimping of the crepe paper is
clearly visible. When copper/sulfur or other metal/sulfur
contamination of the paper surface occurs, the results can
be profound as shown in Figure 5.
63
All the fiber surfaces and the gaps between the fibers
are encrusted with a contaminating material in Figure 5A.
The contamination present on the surface of the paper in
Figure 5B has not yet progressed to the same point as in
Figure 5A, but there is significant contamination present
and consists of mainly copper sulfides and sulfates. The
small spherical structures in the left-hand portion of Figure
5B are composed mainly of copper and aluminum sulfides
and sulfates.
Figure 5A
Figure 5B
In an effort to evaluate the effect that this type of con-
tamination has on paper insulation, dielectric breakdown
strength testing by ASTM Method D 149 was performed
on the three layers of insulation that surrounded the copper
conductor. The first layer, or the one closet to the conduc-
tor, was the insulation shown in Figure 5A. The other two
layers of paper insulation were also contaminated but not
to the degree of the first layer. The results of this testing is
shown in Figure 6.
64
Figure 6
Layer 1, shown in Figure 6, had an extremely low dielec-
tric strength at 80 volts/mil. As the contamination is reduced
with each paper layer further out from the conductor, the
dielectric strength increases significantly to almost as new
condition (1800 volts/mil oil impregnated). Of significant
interest is the fact that the mechanical strength was not
impacted by the contamination as the DP values for all three
layers ranged from 903 to 938. The silvery appearance of
the paper in Figure 3 is due to the alteration of the sulfur-
bearing compound due to exposure to higher temperatures,
as copper sulfide is usually gray-black in appearance. This
reaction of copper and sulfur created deposits on the first
and second layers of paper on the same sample. Because
of the deposition of the copper and sulfur, the first inside
wrap became a poor insulator and, instead, was more con-
ductive. When the corrosive sulfur contamination is this
severe, a failure of the transformer is almost inevitable, as
in this case.
Processing Oil To Remove Corrosive
Sulfur
Several methods were examined to see if they could be
used to reduce the corrosive sulfur content in transformer
oil. An oil that clearly failed the corrosive sulfur ASTM test
D 1275 was used for the experiments. The oil was treated
with Fuller’s earth (clay) at a ratio of 30 grams of clay per
liter of oil (0.25 pounds of clay: 1 gallon of oil) at two tem-
peratures, 45 and 85 C. This type of treatment proved to be
futile in that the treated oils failed the ASTM D 1275 test.
Another type of process using a synthetically-manufactured
absorptive also proved to be unsuccessful in removing the
corrosive sulfur. The last test involved the use of activated
carbon. Three hundred milliliters of contaminated oil were
processed through 128 grams of carbon activated at 120 C
for 24 hours. The sample was then tested according to
ASTM D 1275. The treated sample failed the test, indicating
that it was still corrosive.
Currently, the only successful way to remove corrosive
sulfur from transformer oil is to remove it from the trans-
former and dispose of it. Adequate flushing of the trans-
former should be effected to remove or dilute any remaining
contaminated oil in the system. If the oil has been present in
the transformer system for some time then damage to the
Insulating Oils Handbook
insulation or the metal surfaces may have already occurred
and should be taken into consideration. However, removal
and replacement of the contaminated oil may help mitigate
any future damage.
Conclusions
The effects of corrosive/reactive sulfur not only adversely
affect the conductor material and other metal surfaces but
also can have drastic effects on insulating materials such as
paper. The information presented clearly demonstrates that
serious contamination due to sulfur-containing surfaces
can drastically reduce the dielectric strength of the paper
insulation while no or very little change has taken place in
the mechanical strength of the paper.
Identifying the sources of sulfur contamination and be-
ing able to monitor concentrations of corrosive and reac-
tive sulfur will help detect problem areas earlier. Therefore,
a multifaceted approach is employed in which different
tests are used to provide complementary and meaningful
information. Other than draining and flushing with new
oil, a commercially-viable remedial process has not been
developed which can remove corrosive and reactive sulfur
species from the oil. Hopefully, a method for treating the
contaminated oil can be developed, and Doble is continu-
ing this research.
References
1. ASTM D 1275: “Standard Test Method for Corrosive
Sulfur in Electrical Insulating Oils” in Electrical Insu-
lating Liquids and Gases; Electrical Protective Equip-
ment, Annual Book of ASTM Standards, Vol. 10.03,
ASTM, West Conshohocken, PA, 2001.
2. DIN 51353: Testing of Insulating Oils, Detection of
Corrosive Sulfur, Silver Strip Method, Auslandsnor-
menvermittlung im DIN, Burggrafenstrabe 6, D-1000
Berlin 30, Germany, 1985.
Lance Lewand received his BS degree at St. Mary’s College of
Maryland in 1980. He has been employed by the Doble Engineering
Company since 1992 and is currently the Laboratory Manager for the
Doble Materials Laboratory and Product Manager for the DOMINO®
product line. Prior to his present position at Doble, he was the Manager
of the Transformer Fluid Test Laboratory and PCB and Oil Services at
MET Electrical Testing in Baltimore, MD. Mr. Lewand is a member
of ASTM committee D 27.
Insulating Oils Handbook 65

Nomograph for
LTC-DGA Data Interpretation
PowerTest 2004
(NETA Annual Technical Conference)
Presenter
Fredi Jakob, Ph.D., Weidmann-ACTI
Co-Authors
Karl Jakob, P.E., and Simon Jones, Weidmann-ACTI
Rick Youngblood, Cinergy

I. Introduction energy requirements were calculated by Halstead2. Initially

Deregulation of power producers has put extreme pressure
on utilities to cut their operating expenditures. Maintenance
is always a target for budget cuts. Maintenance dollars are
limited but system reliability must not be infringed. Condi-
tion based maintenance is an effective methodology to both
reduce maintenance and increase system reliability.
Transformers require little field maintenance. The
dielectric fluid in transformers must be monitored but
internal components are rarely examined. DGA has been
a universally accepted tool for transformer diagnostics, but
its major function is to locate incipient faults. In contrast to
transformers, load tap changers and oil filled circuit break-
ers are routinely maintained. Maintenance cycles are based
on time, operation count or a combination of these factors.
Extension of maintenance intervals is possible and desirable
and can be achieved with diagnostic testing. Maintenance
dollar savings and increased reliability are both achievable
with an effective testing program. Application of DGA
to LTCs and OCBs was initially viewed with skepticism
since these devices produce “fault” gases during their normal
operation. Table 1 compares both the normal and abnormal
gases produced in transformers and devices that make and
break contacts during their operation. Youngblood1 was one
of the first engineers to point out that observation of the
“heating” gases methane, ethane and especially ethylene can
be the basis of condition assessment for LTCs and OCBs.
II. Data Interpretation
A. Concept
The interpretation of DGA data for transformers, LTCs
and OCBs is empirical in nature. The development of in-
terpretation protocols for OCBs and LTCs parallels the
development of DGA diagnostics for the main tanks of
transformers. Key gases associated with heating problems
are methane, ethane and ethylene. These gases, in the order
listed, require increased energy for their production. The
Youngblood1 ignored the levels of arcing gases, acetylene
and hydrogen that developed whenever an LTC operated.
Subsequent work by Youngblood5 indicated that arcing gases
are also diagnostically significant. For example, increased
acetylene levels were often followed by increased heating
gas concentrations. The increased acetylene is due to changes
in the arc duration and/or characteristics as the contacts are
eroded or covered with carbon.
The next step in the development of diagnostic protocols
for LTCs was the empirical determination of normal or
threshold values. The gas retention rate in an LTC is very
dependent on breathing configuration, so this is a major
factor in determining threshold levels. Free breathing LTCs
rapidly lose gases to the environment while sealed LTCs
retain much of the gas produced.
Threshold levels have been determined for specific models
and types of LTCs by Doble3 and Baker4. Generic levels
have been set by Youngblood5, and are useful when threshold
values have not yet been determined. Fault gas ratios, which
are discussed below, are applicable for unit evaluation when
threshold values are reached or exceeded. We are investi-
gating the value of these ratios even if individual gases are
below the threshold values.
B. Fault Gas Concentration Ratios
1. Normal Operation
Arcing, which generates very high temperatures, occurs
with each switching operation in an LTC. Arcing produces
mainly acetylene and hydrogen. When arcing gases are be-
ing produced, “heating gases” are also produced since the oil
temperature varies with distance from the arcing contacts.
At lower oil temperatures, heating gases are produced more
abundantly than acetylene.
If one accepts this hypothesis, one would expect that
the ratios of heating to arcing gases in a problem free unit
would remain fairly constant with operational count. Since
66
the contact surface is being eroded and additional deposits
are being formed with each operation, these ratios will
predictably change slightly with operation count, providing
there is not a large change in load current. Although the load
on a distribution circuit in a substation does remain nearly
constant over a period of time with minor ups and downs,
anything that causes a significant change in average constant
Insulating Oils Handbook
load will change the ratios. The data presented in a paper by
Duval6, shown in Table 1, supports this hypothesis. Note
that all of the significant ratios in these problem free units
are well below those found in a LTC that has developed
a heating or other problem. Gas ratios that are useful in
detecting LTC problems are listed in Table 2.

Table 1. Gas Formation as a Function

of Operation Count (Duval6)

Operations: 500 3600 49000 Gas produced/operation

Gas/500 Gas/3600 Gas/49000

Hydrogen 6870 12125 14320 13.74 3.37 0.29

Methane 1028 5386 10740 2.05 1.50 0.22

Acetylene 5500 35420 53670 11.00 9.84 1.10

Ethylene 900 6400 35839 1.80 1.78 0.73

Ethane 79 400 3944 0.16 0.11 0.08

Ratio: Ethylene 0.16 0.18 0.66

Acetylene

*Note: Some gas is always lost with time. Therefore, the gas concentration per
operation is expected to decrease with operation count. Duval did not provide
breathing configurations for this data.

The ratio is fairly consistent and independent of operation
count. When the operation count is very high, 49,000, the
ratio increase is probably related to changes in the contact
condition.
2. Heating Problems
Initially a resistive film develops on contacts, which
results in an increase in contact resistance, increased heat-
ing and an increase in heating gas concentrations. Since
more heating gases are produced, the ratio of heating to
arcing gases, ethylene to acetylene, increases. This change
in gas concentrations and gas concentration ratios indicates
problems. The concentration ratios of ethane to methane
and ethylene to ethane are temperature dependent, so both
are expected to increase with increasing temperature in a
problem unit. These two heating gas ratios should also reflect
increased contact resistance and heating.

Table 2. Important Gas

Concentration Ratios

Heating to Arcing Ratios

Ethylene Ethylene Methane + Ethylene + Ethane Methane + Ethylene + Ethane
Acetylene + Hydrogen Acetylene Acetylene + Hydrogen Acetylene
Temperature Dependent Ratios
Ethane Ethylene
Methane Ethane
Insulating Oils Handbook
C. Application of Gas Concentration Ratios
As is the case with main tank DGA, ratios of fault gas
concentrations are most valid if at least one of the fault
gases exceeds its threshold value. Threshold values used by
Weidmann-ACTI are model specific, whenever this data is
available. If model specific data is not available, we use the
“monthly watch” levels developed by Youngblood5 as thresh-
old values. “Normal” values used for the ratios represent the
90* percentile of fault gas concentrations from a very large
67
number of DGA results (-2500 units). These 90* percentile
values, which are listed in Table 3, are generic and do not
take into account the difference in gassing rates or breath-
ing configurations of specific units. Ratios are, as expected,
model specific and we hope to refine our results as more data
becomes available. Table 4 lists the ratios calculated for a
Westinghouse LTC and for a McGraw Edison LTC, which
according to Baker’s* threshold levels require attention.

Table 3. Generic 90th Percentile Fault Gas Ratios

Rl R2 R3 R5 R5 R6

Ethylene Ethylene Ethylene + Ethane + Methane Ethane Ethylene Methane + Ethylene + Ethane
Acetylene Acetylene + Hydrogen Acetylene + Hydrogen Methane Ethane Acetylene
0.36 0.24 0.38 0.31 7.51 0.56

Note the difference between the model specific ratios listed in Table 4 and the generic values which are listed in Table
3. The differences in critical ratios for the Westinghouse and McGraw Edison units illustrate how model specific the gas-
sing characteristics can be.

Table 4. Unit Specific Ratio Comparison

(Based on values set by C. Baker)

WESTINGHOUSE LTC UTT

Ethylene Ethane Methane Acetylene Hydrogen
LT3 3,000 250 1,000 5,000 5,000
Ratio 1 Ratio 2 Ratio 3 Ratio 4 Ratio 5
Ethylene Ethylene Ethylene + Ethane + Methane Ethane Ethylene
Acetylene Acetylene + Hydrogen Acetylene + Hydrogen Methane Ethane

0.60 0.30 0.43 0.25 12.00

McGRAW EDISON LTC 550-BL

LT3 2,000 400 400 400 500
Ratio 1 Ratio 2 Ratio 3 Ratio 4 Ratio 5
Ethylene Ethylene Ethylene + Ethane + Methane Ethane Ethylene
Acetylene Acetylene + Hydrogen Acetylene + Hydrogen Methane Ethane

5.00 2.22 3.11 1.00 5.00

LT3 is Baker’s designation of a unit requiring inspection.

68 Insulating Oils Handbook
D. Graphical Methods be above these values before the ratios are considered valid.
Graphical methods have been widely used as an aid for As previously stated, this concept requires further study.
the interpretation of DGA data for transformer main tanks. Figure 2 is a copy of the proposed nomograph for the
A nomograph developed by Church7 et al. has been success- empirical interpretation of LTC - DGA data. The threshold
fully used to graphically calculate and interpret fault gas ratio values indicated are generic in nature. Unit specific infor-
data for power transformers. A similar nomograph can be mation should be used whenever sufficient data is available.
applied to interpretation of DGA data for both LTCs and The relative position of each scale is based on a statistical
OCBs. The nomograph consists of adjacent parallel loga- determination of the 90th percentile gas ratios from our
rithmic scales on which fault gas concentration in parts per extensive database. These threshold values are shown in
million (ppm) is plotted. A simple example will illustrate the Table 5. These values are also generic in nature and should be
concept, shown in Figure 1. Rose8 uses the ethane/methane replaced if model specific data is available. The nomograph
ratio as an indicator of a heating problem. He has stated that is constructed so that any line with a positive slope indicates
the critical value of this ratio occurs when it exceeds unity. a LTC heating problem.
Each of these gases is plotted below on a logarithmic scale.
The two scales are aligned, without any offset. Table 5. 90th Percentile Values for
Different Breathing Configurations
C2H2 C2H4 C2H6 CH4 H2
Configuration
Free Breather 2733 851 107 379 1418
FB/Desiccant 2811 473 130 141 467

Sealed 3744 1337 178 572 1781

Vacuum 42 49 79 61 72

Figure 1. Logarithmic Representation of Ethane / Methane Gas

Concentration Ratios.

A straight line connecting the methane concentration of

800 ppm to the ethane concentration of 1200 ppm, a ratio
of 1.5, would have a slope greater than one. Any set of equal
gas concentrations would have a zero slope. A sample with
less ethane than methane will be connected by a line with a
negative slope. The slope of the line is thus a clear indication
of the LTC condition. In cases where the transition for a
selected ratio, from normal to a problem condition occurs at
any value other than one, the relative position of the scales is
adjusted so that a horizontal line indicates the transitional
ratio. Lines with positive slopes indicate ratios greater than
the transitional value. Lines with negative slopes correspond Figure 2. Proposed Nomograph for the 90th Percentile
to ratios below the critical values. Gas Thresholds.
Threshold levels are also indicated on each scaled gas
concentration. At least one of the gas concentrations should
Insulating Oils Handbook
III. Indicating Problem Severity
The severity of a problem, or condition of the unit, may
be designated as Normal, Caution or Warning. A proposed
nomograph has been developed by taking the 90th percentile
values as a baseline, and aligning the concentration scales so
that any connecting line with this ratio will have a slope of
zero. Units having a zero slope and/or a slope that resides
below these threshold values are diagnosed as Normal. A
slope of the line residing within one standard deviation of
the threshold values is diagnosed as a Caution, and at two
standard deviations is a Warning. A section of the proposed
nomograph is illustrated in Figure 3.
Figure 3. Nomograph Diagnosis of LTC Condition.
IV. Future Work
The empirical analysis of DGA data for LTCs is well
developed. Gas concentration levels and gas concentration
ratios can differentiate between normal and problem units
We believe that both the concentration and ratio values will
work best if they are model specific The compilation of these
values requires user feedback on problem units. Trending
of both the gas concentrations and ratios is always the best
method to identify incipient problems.
Another concept under investigation is normalization
of fault gas data. We believe that during normal switching
operations, the ratio of ethylene to acetylene is fixed. This
ratio should remain constant for different numbers of opera-
tions. Furthermore, since these two gases have approximately
the same solubility in mineral oil and approximately the
69
same escape rates from the oil the ratios should remain
fairly 100 ppm then the ratio is one. If a heating problem
is superimposed on the normal arcing process and the gas
levels are 175 ppm for ethylene and 150 ppm for acetylene
there would be an “additional” 25 ppm of ethylene due to
the heating problem. One could thus normalize our results
using a ratio of:
Ethylene Total Ethylene
(due to heating) (Ethylene produced during normal arcing)
————————— = ——————————————————
Acetylene (due to arcing) Acetylene
References:
1. Youngblood, R., Jakob, F., Haupert, T.J. “Application of
DGA to Detection of Hot Spots in Load Tap Chang-
ers,” Minutes of the Sixtieth Annual International
Conference of Doble Clients, 1993, Sec. 6-4.1.
2. Halstead, W. D., “A Thermodynamic Assessment of the
Formation of Gaseous Hydrocarbons in Faulty Trans-
formers,” Journal of the Institute of Petroleum, Vol. 59,
Sept. 1959, pp. 239-241.
3. Doble Client Transformer Committee Subcommittee
Report on Transformer Load Tap Changer Dissolved
Gas Analysis, September 24, 2001.
4. Baker, Charles. Personal correspondence. 2002.
5. Youngblood, R., Baker, C, Jakob, R, Perjanik, N.,
“Application of Dissolved Gas Analysis to Load Tap
Changers.”
6. Duval, Michel., “A Review of Faults Detectable by Gas-
in-Oil Analysis in Transformers,” IEEE Electrical In-
sulation Magazine, May/June 2002, Vol.18 No. 3, pp.
8-17.
7. Church, J.O., Haupert, T.J., and Jakob, R, Electrical
World, Vol. 201, No. 10, October 1987, pp. 40-44.
8. Rose, Don. Personal communication.
Dr. Fredi Jakob received his PhD at Rutgers, the State University of
New Jersey, in 1961. He is professor emeritus of analytical chemistry at
California State University-Sacramento and is the founder and laboratory
director of Analytical ChemTech International, Inc. (ACTI), which is a
wholly owned subsidiary of Weidmann Systems International. As a long-
term member of ASTM and IEEE and author of over fifty published
articles, Dr. Jakob is a traveling lecturer to private and governmental
agencies. He has been invited to speak at American Public Power meet-
ings, ASTM symposia, conferences held by Doble, NETA, and AVO
conferences, as well as other industrial organizations.
70
Condition Assessment of Oil Circuit
Breakers and Load Tap-Changers by the
Use of Laboratory Testing and Diagnostics
NETA World, Summer 2004
by Lance R. Lewand and Paul J. Griffin
Doble Engineering Company
Introduction
The use of laboratory testing is of increasing importance
today as it provides the data to diagnose the condition of
oil-filled electrical apparatus. This information can be used
to detect and identify incipient faults in apparatus, provide
an indication of their severity, and identify long-term aging
trends. In today’s environment, where out-of-service testing
of apparatus is not always possible, being able to acquire
easily-sampled electrical insulating liquids in service is an
advantage.
Current-day practices in electrical substations are migrat-
ing more towards condition-based, as opposed to time-based,
maintenance in order to protect valuable assets, extend their
lives, and use budgeted money effectively. Condition-based
maintenance used for transformers for many years is now
being applied for load tap-changers (LTCs) and bulk oil
circuit breakers (OCBs). Since LTCs and OCBs can fail
mechanically, electrically, and from deterioration because of
local overheating, it is reasonable to assume that by-products
of the deterioration and overheating could be found in the
oil. In recent years dissolved gas-in-oil analysis and other
insulating liquid tests have been used as effective tools to
detect problems in LTCs and OCBs. Specific guidelines
and algorithms have been developed for evaluating the
results from normal and abnormal LTCs and OCBs to aid
in condition assessment. The goal of a laboratory diagnostic
program is to provide a consistent, reliable analytical tech-
nique to detect problems and provide a ranking or relative
health index.
Laboratory Tests Used for Condition
Assessment of LTCs and OCBs
Oil tests were chosen to be able to diagnose the condi-
tion of LTCs and OCBs, based on previous experience with
transformers, knowledge of the operation of these apparatus,
and empirical evidence from failures and problems. Al-
though the same or nearly the same tests are used for LTCs
Insulating Oils Handbook
and OCBs, the methods of evaluation and the diagnostic
approach is significantly different. The tests currently used
are described below:
Dissolved gases in oil (DGA): Great emphasis is placed
on this test, as it is an important diagnostic tool for detect-
ing localized overheating or excessive arcing as well as
other abnormalities. Localized overheating of conductors
and surrounding insulation may lead to a thermal runaway
condition which causes carbonization and by-product
polymeric films to form on conductors. This increases the
surface resistance of the contacts, thereby causing increased
heating to the point of failure unless the cycle is interrupted
by maintenance. This can be detected by observing the gen-
eration of hydrogen and hydrocarbon gases such as methane,
ethane, ethylene, and acetylene. It has been determined that
as an overheating event develops into a thermal runaway
condition, the ratio of the hydrocarbon gases changes and
can be used as an additional diagnostic tool. The analysis is
complicated because:
• Gassing characteristics of oils vary to some degree with
their composition, the amount of oxygen present, and
the temperature.
• Less soluble gases such as hydrogen and carbon mon-
oxide will escape to the atmosphere more quickly than
the more soluble gases when the apparatus is venting
properly.
• A temperature distribution exists around an incipient-
fault area, which varies with the source of heating and
cooling available.
• More than one problem type may coexist in the same
apparatus.
• The progression from normal to a fault condition may
vary in time and with operating conditions such as
load.
• Catalytic surfaces and the composition of those surfaces
may influence the results.
Insulating Oils Handbook 71
Regardless of the complications listed above, DGA has meaningful test. It provides an indication of the amount of
become an invaluable tool in helping to detect and identify material that has been worn or sublimated from the moving
problems and their severity in OCBs and LTCs. and/or stationary contacts and is now present in the oil. It
Particle count and size: The total number of particles also provides a quantitative analysis as to composition of
by size groupings is used to detect abnormal quantities of the metals found in the oil.
by-products and wear materials. The ratio(s) of the size
groupings provides information as to the extent a detrimen- Diagnostic Methods for LTCs
tal condition has progressed. Larger particles are especially
The dissolved gas-in-oil test is most often the key test
important, as the dielectric strength of the insulating oil is
for detecting and diagnosing problems with LTCs. LTCs
more adversely affected by these particles, and formation of
are a crucial element of utility networks, as they must oper-
larger particles are an indicator of advanced deterioration.
ate repeatedly in a precise fashion in order to maintain a
Particle typing: Particle typing has been used suc-
constant voltage output. This must be achieved regardless
cessfully in other fields such as lubrication and hydraulic
of variations on input or load. LTCs have been a weak link
systems for quite some time. In OCBs and LTCs, particles
in many networks, since they deteriorate over time due to
are formed from three main mechanisms: wear, arcing and
mechanical problems or contact wear from repeated opera-
overheating.
tions. Erosion of the contacts over time is expected due to
the nature of their function. Coking of the contacts causes
• Metal particles are formed by wear as two metallic sur-
overheating which can cause thermal runaway.
faces move against each other. Roylance and Hunt have
There are three main types of LTCs: reactive with arcing
determined that the size, morphology, and types of par-
contacts in oil, resistive with arcing contacts in oil, and arcing
ticles are dependent upon the severity of the pressure
contacts in a vacuum bottle. The resistive LTC mechanism
applied to the surface and the angle at which the sur-
is a European design. It has been determined that the dif-
faces intercept.
ference between resistive and reactive types is important, as
• Arcing can also form metal particles, but these particles
the shorter time of arc extinction of the resistive type (five to
are quite different in morphology and topography as
six milliseconds after contact separation) should lower the
they are not wear-induced. Arc-produced metal par-
concentrations of gases generated (see Griffin). The vacuum-
ticles are formed from molten metal being quenched
type models break under vacuum and so the arcing gases
by the cooler surrounding oil and, thus, are produced as
should not be present at all or in very small quantities.
a somewhat teardrop-shaped particle. Arcing will also
Doble has developed guidelines for testing of LTCs
produce organic particles such as carbon fines and larger
which are model-specific on the basis of empirical evidence
conglomerations from the breakdown of the oil. (See
of gassing behavior and problems detected. (See Asche.) Lau
Peelo.)
et al determined that on-line filters can change the relative
• Overheating increases the rate of decay of other materi-
composition and quantities of gases generated. The on-line
als and induces the formation of by-products such as
filters are used by many utilities to remove carbon and other
polymerized oil films. The examination of the filter pro-
particles from the oil, hence lengthening maintenance in-
vides a qualitative identification of the types of particles.
tervals as coking is reduced.
In the process of particle typing there is an attempt to
The LTC diagnostics are then used to determine a condi-
relate the particles to specific materials of construction.
tion code providing a relative ranking that can be used for
Along with microscopic examination of the particles
condition-based maintenance. A similar system used for
trapped on a filter, a Doble carbon-coding process has
OCBs is described later in this article.
been introduced to aid in quantifying the carbon load-
ing in the sample.
Diagnostic Methods for OCBs
Oil Quality: Several oil quality tests can be performed OCBs, consisting of moving and stationary contacts
to check for dielectric strength and aging. The program and ancillary components involved in making and break-
described here presently uses water content, dielectric ing the circuit, can wear out, losing the ability to perform
breakdown voltage (D 1816, 1 mm gap), and neutralization their intended function. This could happen because of mis-
number. The three oil-quality tests provide the essential alignment, poor contact surfaces, wear, improper timing of
information necessary to aid in a diagnosis without being contact movement, and a thermal runaway condition as well
overly complicated. The dielectric breakdown voltage test as other factors. Dielectric failure may occur from excessive
provides information on the insulating capability of the oil. localized moisture and/or excessive amounts of conductive
The water content aids in determining how wet or dry the particles. In addition, even stationary components such as
system is and if free water exists. The neutralization number the arc-chute materials can breakdown which results in
provides information on the extent of the degradation of the inadequate arc quenching and carbon buildup.
oil and is important since high concentrations of organic
acids can exacerbate an already deteriorated condition.
Total Metals: The metals test, consisting of both particu-
late metals and those dissolved in the oil, is an extremely
72
How a Condition Assessment is Determined
In order to provide a condition assessment for OCBs and
LTCs, a numeric ranking is determined through the use
of the four separate sections of the analytical data: DGA,
oil quality, particle count, and metals. The rankings from
the four groups are summed or otherwise manipulated to
provide a numeric ranking. The ranking system is further
reduced to a “Condition Code” from which specific main-
tenance functions are recommended. This ranking system
is shown in Table 1.
TABLE 1
Condition Codes and Maintenance Assessment
Condition Assessment
Code
1 Remove from service now — Remedial action
needed
2 INVESTIGATE now to determine problem
3 Monitor — Resample in 3 months or 1 month
depending on loading
4 Monitor — Resample in 12-15 months (the
resampling interval for LTCs would be less)
5 Of No Concern — Resample in 3 years (the
resampling interval for LTCs would be less)
A Condition Code 1 indicates an OCB or LTC in the
worst possible condition. A Condition Code 5 would in-
dicate an apparatus in good condition. A large emphasis
is placed on the ethylene to acetylene ratios above certain
levels, since it clearly distinguishes the severity of the over-
heating. In general, the gassing results and ratios are given
more weight than the oil quality, particle count, or metal
results, as it most often detects the problems in the earlier
stages, detects a wide range of problems, and is the most
reliable indicator.
Conclusions
In economic terms, condition-based maintenance of
LTCs and OCBs makes practical sense. It focuses resources
on intervention to prevent LTC and OCB failures and can
save a substantial amount of money not only in terms of
replacement and installation costs but also in terms of lost
revenue. The cost of the program is much less than continu-
ing to service LTCs and OCBs on a prescribed maintenance
schedule, which is currently being further extended in most
utilities and industrial sites. This type of diagnostic approach
using analytical data from insulating-oil test results helps
to identify quickly apparatus in poor condition and then
focuses attention in a more timely fashion. It facilitates rank-
Insulating Oils Handbook
ing of apparatus as to condition so that a priority hierarchy
can be established. This permits maintenance activities to
be more thoughtfully developed and managed, thus saving
time, money, and unplanned outages due to failure.
References
Asche, R. G., “Dissolved Gas Analysis of Load Tap Chang-
ers, Subcommittee Project Report,” Proceedings of the
2002 Annual International Conference of Doble Clients,
Transformer Test and Maintenance, 2002, Sec. 13E.
Griffin, Paul J., “Field Testing — Laboratory Diagnos-
tics, Acquiring Information on Transformer Health,”
The Life of A Transformer Seminar, Doble Engineering
Company, Clearwater, Florida, USA, February 16-20,
2003.
Lau, M., Horn, W., Schellhase, H., Dominelli, N., and
Ward, B., “Successful Application of Filtration System
on On-Load Tap-Changers,” Proceedings of the 2002
Annual International Conference of Doble Clients, Trans-
former Test and Maintenance, 2002, Sec. 13C.
Peelo, D., “Tutorial T3: Circuit Breaker Basics,” presented
at the Sixty-Ninth Annual International Conference of
Doble clients, 2002.
Roylance, Brian J. and Hunt, Trevor M., Wear Debris Anal-
ysis, 1999.
Lance Lewand received his Bachelor of Science degree from St.
Mary’s College of Maryland in 1980. He has been employed by the
Doble Engineering Company since 1992 and is currently the Laboratory
Manager for the Doble Materials Laboratory and Product Manager for
the DOMINO®. product line. Prior to his present position at Doble, he
was Manager of the Transformer Fluid Test Laboratory and PCB and
Oil Services at MET Electrical Testing in Baltimore, MD. Mr. Lewand
is a member of ASTM Committee D 27.
Paul J. Griffin received his BS degree at the American International
College and his MS at the University of Rhode Island. He has been
employed by the Doble Engineering Company for the past 24 years
and is currently Vice President of Laboratory Services. He is secretary
of the Doble Oil Committee; a member of ASTM committee D 27,
subcommittee chair of Analytical Tests, section chair of Gases in Oil;
US Technical Advisor to IEC TC10 for Fluids for Electrotechnical Ap-
plications; member of the IEEE Insulating Fluid subcommittee of the
Transformer committee, and a member of the CIGRE Working Group
15.01 Fluid Impregnated Insulating Systems.
Insulating Oils Handbook
Natural Ester Dielectric Liquids
NETA World, Fall 2004
by Lance R. Lewand
Doble Engineering Company
The newest versions of natural ester dielectric fluids have
been in use in the electrical apparatus industry since ap-
proximately 1998 and are becoming more popular. Unlike
transformer mineral oil, which is refined from petroleum
derived from compressed plant and animal tissue (mostly
microorganisms from 70 to 440 million years ago), these
liquids are produced from renewable resources such as
vegetable oils and seeds.
Reference is given to these liquids as “the newest ver-
sion” because vegetable oils have been experimented with
and used as dielectric liquids since the late 1880s. George
Westinghouse and his team of engineers used castor oil and
linseed oil as dielectric liquids and insulation impregnants.
This helped to reduce the size of the transformers by reduc-
ing dielectric clearances and allowed the solid insulation to
last longer by minimizing the effects of oxygen. Until Elihu
Thomson patented mineral oil for use in transformers in
1892 and its acceptance in the mid-1890s, vegetable oils
(natural esters) were the dielectric liquids used.
In fact, many vegetable oils have been tested over the
years, including linseed, cod liver, hemp, walnut, poppy,
sunflower, cottonseed, sesame, peanut, rapeseed, olive, and
others. Two oils in particular, linseed and tung, were very
important in the early years of transformer manufacturing.
These oils were considered to be “drying oils” and, as such,
polymerized easily when exposed to oxygen. Because of this
characteristic they were used extensively in the formulation
of insulating varnishes. [See Clark (1962).]
These same characteristics, however, presented problems
and were objectionable when used as a liquid dielectric.
Formation of organic peroxides occurred and eventually
resulted in the formation of aldehydes and acids. These
vegetable oils polymerized easily when exposed to oxygen.
Such by-products are unwanted in transformer systems as
they accelerate the degradation process of the liquid and
solid insulation. In addition, vegetable oils have much higher
viscosity and pour point characteristics and could not be
manufactured consistently. Since early transformers were all
free-breathing, oxygen played a major role. Thus, vegetable
73
oils were quickly supplanted by the introduction of Western
Pennsylvania paraffinic mineral oils which were in turn
eventually succeeded by naphthenic-based mineral oils.
Beginning in the early 1990s, there was renewed interest
in developing natural ester dielectric liquids able to over-
come some of the deficiencies that plagued the vegetable
oils used in the 1890s and 1900s. ABB Inc. and Cooper
Power both began intensive research into the formulation
of natural esters, and both now have commercially available
products. In addition, Waverly Light and Power, a small
city-owned utility in Iowa, has used a soy-based dielectric
liquid developed by the University of Northern Iowa’s Ag-
Based Industrial Lubricants Research Program in pole top
transformers since about 1998. This product is now produced
by Cargill. Several patents were issued for these products
beginning in 1999. [See Oomen and Claiborne (1999),
McShane et al (2000), and Cannon and Honary (2000).]
The main purpose for development of many of these
dielectric liquids was to create an environmentally friendly
product that was not only stable when used as an insulating
liquid in electrical apparatus but also readily biodegraded
when exposed to the environment. This would allow spills
and leaks to be handled more easily than transformer min-
eral oils. Since the first production and use of these liquids,
other positive attributes have been discovered. As listed in
Lewand (2001), potential users of these types of dielectric
liquids would expect some of the following qualities to be
present.
High level of biodegradability
Nontoxic
Material compatibility with electrical apparatus
components
Good dielectric strength and insulation properties
Similar dielectric constant as the solid insulation
Long-term oxidative and thermal stability
Relatively low pour point
74
Can be consistently produced
Long service life
Compatibility with presently used liquids (mineral
oil, etc.)
Reasonable cost
Examples of current, commercially-known, natural ester
dielectrics are shown in Table 1.
Table 1 - Natural Ester Dielectric Liquids
NAME TYPE MANUFACTURER
BIOTEMP Comprised mostly of mono-un- ABB Inc.
saturated high oleic acid triglyc-
eride vegetable oils. The oleic
acid group is defined as having
one carbon double bond, part of
the eighteen carbon atoms in the
hydrocarbon chain of a carbox-
ylic acid. Examples of high oleic
oils are sunflower, safflower, and
rapeseed (canola).
BIOTRANS A mixture of partially hydroge- Cargill
nated soybean oil high in oleic
acid content, methyl esters pro-
duced from soybeans, palm or
coconut oils used to thin the
dielectric liquid
Envirotemp FR3 Edible-seed oil based dielectric Cooper Power Systems
liquid. It is a natural ester (tri-
glyceride - fatty acid ester) con-
taining a mixture of saturated
and unsaturated fatty acids with
14 to 22 carbon length chains
containing one to three double
bonds. Suitable vegetable oils,
which may be used indepen-
dently or combined, include:
soya, sunflower, and rapeseed
(canola).
Coconut Oil Coconut Oil University of
Moratuwa, Sri Lanka
What is a Natural Ester?
Esters can be natural, such as those derived from veg-
etable oils as discussed in this article, or they can be made
synthetically from a group of chemicals chosen to yield
certain properties. “Ester” is a term applied to chemical
compounds with a certain structure. The structure of a fatty
ester is represented by the following formula:
Insulating Oils Handbook
The O represents oxygen, C represents carbon, R repre-
sents an alkyl or aryl group, OR’ represents an alkyl or aryl
group attached to an oxygen, a single dash represents a single
bond, and a double dash represents a double bond. For those
wishing to know even more chemistry, an alkyl group is one
that contains carbon and hydrogen in the following pattern
CnH2n + 1. The n can be any number and simply refers to the
length of the molecule. An alkyl group is derived from its
alkane parent. An alkane is a hydrogen-to-carbon saturated
compound in which all the groups attached to the carbon are
single-bonded. An aryl group or aryl radical is an aromatic
hydrocarbon in which there is one hydrogen less than its
parent form.
All these ester oils have a triglyceride component of
which many of the properties of the oil are based on the
fatty acid content of the oil. A formula for a triglyceride is
shown below.
The R, R’, and R” are organic groups (carbon, hydrogen,
oxygen) consisting of chains of eight to 22 carbons and are
the fatty acid component. Fatty acids include oleic, linoleic
and linolenic (unsaturated containing one to three double
bonds), and palmitic and stearic (saturated fatty acids, no
double bonds).
The type of bonding has a marked impact on the proper-
ties. For example, oils with multiple double bonds tend to
oxidize more easily and potentially polymerize — not a qual-
ity wanted in a liquid dielectric in a transformer. Saturated
fatty acids (single bond) offer better oxygen stability but also
raise the pour point possibly to an unacceptable level such as
20 degrees Centigrade (68 degrees Fahrenheit). In countries
like Sri Lanka this may not be a problem, but in the United
States the ambient temperatures are often much lower. At
temperatures below the pour point, the liquid turns into a
solid and, thus, must function as a solid dielectric.
In some of the other natural ester dielectric liquids, a
compromise was made where pour point considerations
and oxidation stability were weighed against each other. For
example, all three US-manufactured natural esters contain
an oleic acid component which only has one double bond,
allowing some oxygen stability and reducing the pour point
to about -12 to -18 degrees Centigrade (10 to 0 degrees
Fahrenheit) before the use of pour point depressants. It must
Insulating Oils Handbook
be said that compromises are not only made in the refining
of natural esters but also in transformer mineral oils as well.
Not all properties are maximized since they may negatively
influence other properties, and, thus, the refining process is
based on meeting a specification with the refining equip-
ment available at a cost a customer is willing to pay.
Refining of a Natural Ester Dielectric
Liquid?
Natural esters are refined in a totally different manner
than transformer mineral oils. The first major difference is
the source of the material for refining. In the case of trans-
former mineral oils, crude oil is extracted from the ground
by drilling and goes through a series of air and vacuum
distillation steps, followed by treatment with hydrogen,
pressure, and catalyst. In the case of natural esters, the source
materials are crops that are grown and then harvested. In
this respect, the seed oils are more attractive than mineral
oils as they are a renewable resource.
Natural ester dielectrics are manufactured in multi-step
processes involving several techniques. Two techniques
are used for obtaining the crude oil from oil seeds. Batch
pressing is the first technique and has been used for some
time. Batch processing consists of applying either hydraulic
pressure to the vegetable seed or pulp via hydraulic press
or through the use of a rotating screw or worm and is best
suited for softer crops like sunflower and rapeseed. Another
refining process involves crushing the seeds and extracting
the oil with a solvent such a hexane and is better suited for
harder crops such as soybeans.
A degumming step is necessary to remove materials other
than oil, such as chlorophyll. This is performed either by
physical separation, which takes time, or by combining the
material with water and caustic material to accelerate the
separation process.
The next step, a bleaching process that is part of a neutral-
ization process, also subjects the oil to clay treatment to re-
move polar compounds. Deodorizing the oil is accomplished
via steam distillation under vacuum up to 200 degrees
Centigrade to remove unwanted volatile compounds. The
last step, winterizing, which may be optional and depends
on the starting material and the degree of refining, involves
chilling the oil to remove excessive saturates.
Because these refining techniques can be carefully
controlled, a more consistent product is produced. In ad-
dition, the new natural ester dielectrics differ from their
predecessors not only in the refining process but also in
the additives used. Whereas the early natural esters had no
additives, the new ones have a variety of additives enhanc-
ing performance.
Additives
The natural ester dielectric liquids contain additive pack-
ages consisting of chemicals to reduce the pour point, aid
in oxygen stability, and in some cases have an antimicrobial
agent or copper deactivators. This is in contrast to mineral
oil, which has either no additives or simply oxidation inhibi-
75
tors. Mineral oil produced to Doble TOPS or ASTM D
3487 specifications are only allowed to contain DBPC (2,6-
ditertiary-butyl paracresol, BHT) or DBP (2,6-ditertiary-
butyl phenol) in concentrations up to 0.3 percent. DBPC
and DBP have had a lengthy history of use in transformer
oil and no adverse effects have been documented.
The U.S. patents for BIOTEMP®, BIOTRANS® and
Envirotemp® FR3 incorporate enough variation into the
descriptions of each liquid that the exact combination and
concentration cannot be determined. Most of these addi-
tives have been well-established in the chemical and food
industries for some time, but it is not fully known if there
are any adverse characteristics when used in transformers
over a long period. The table below provides a listing of the
additives and their described use. It must be emphasized that
not all of these additives are used but the possibility for some
or combination thereof to be used does exist. In some of the
dielectric liquids listed in the table, the additive package can
make up as much as three percent of the liquid.
Possible Additives in BIOTEMP®,
BIOTRANS® and Envirotemp® FR3
Liquid Additive and Function
BIOTEMP Oxidation Inhibitors: Phenolic antioxidants such as:
BHA (butylated hydroxy anisole), TBHQ (mono-
tertiary butyl hydroquinone), DBPC (BHT, 2,6-
ditertiary-butyl paracresol/butylated hydrotoluene), and
aklyated diphenylamines
Copper Deactivator: Benzotriazole derivative
Pour Point Depressant: PMA (polymethacrylate)
BIOTRANS Oxidation Inhibitors: citric acid (mostly used as
sequester of metals to avoid catalytic effect of those
metals), TBHQ (mono-tertiary butyl hydroquinone)
Pour Point Depressant: diethylhexyl adipate, polyalkyl
methacrylate
Envirotemp Oxidation Inhibitors: Phenolic antioxidants such as:
FR3 BHA (butylated hydroxy anisole), TBHQ (mono-
tertiary butyl hydroquinone), DBPC (BHT, 2,6-
ditertiary-butyl paracresol/butylated hydrotoluene),
THBP (tetra hydro butro phenone)
Pour Point Depressant: (polyvinyl acetate oligomers
and polymers and/or acrylic oligomers and polymers)
Antimicrobial agent: (BHA, potassium sorbate, sorbic
acid, monoglycerides and/or Vitamin E)
Conclusions
Natural esters have been used as a dielectric liquid since
the invention of the oil-filled transformer. Because of their
chemistry these liquids have some limitations in their use.
Recent advances in research and refining and additive
packages have produced a new breed of natural esters that
try to address these limitations. A later article will discuss
the physical, chemical, and electrical properties of natural
esters, differences as compared to transformer mineral oils,
and some of the advantages of these oils.
76 Insulating Oils Handbook

References
Abeysundara, D.C., Weerakoon, C., Lucas, J.R., Guna-
tunga, K.A.I., and Obadage, K.C. “Coconut Oil as an
Alternative to Transformer Oil,” in ERU Symposium,
November, 2001.
Cannon, G.S., and Honary, L.A.T., “Soybean Based Trans-
former Oil and Transmission Line Fluid,” United States
Patent 6,159,913, issued December 12, 2000.
Clark, Frank M. Insulating Materials for Design and En-
gineering Practice, John Wiley and Sons, New York,
1962, 1218 pages.
Lewand, Lance R. “Laboratory Evaluation of Several Syn-
thetic and Agricultural-Based Dielectric Liquids” in
Proceedings of Sixty-Eighth Annual International Confer-
ence of Doble Clients, 2001.
McShane, C. P., Corkran, J.L., Harthun, R.A., Gauger, G.
A., Rapp, K.J. and Howells, E., “Vegetable Oil Based
Dielectric Liquid,” United States Patent 6,037,537, is-
sued March 14, 2000.
Oommen, T.V., and Claiborne, C.C., “Electrical Trans-
formers Containing Electrical Insulation Fluids Com-
prising High Oleic Acid Oil Compositions,” United
States Patent 5,949,017, issued September 7, 1999.

Lance Lewand received his Bachelor of Science degree from St.

Mary’s College of Maryland in 1980. He has been employed by the
Doble Engineering Company since 1992 and is currently the Laboratory
Manager for the Doble Materials Laboratory and Product Manager for
the DOMINO®. product line. Prior to his present position at Doble, he
was Manager of the Transformer Fluid Test Laboratory and PCB and
Oil Services at MET Electrical Testing in Baltimore, MD. Mr. Lewand
is a member of ASTM Committee D 27.
Insulating Oils Handbook 77

Laboratory Testing
of Natural Ester Dielectric Liquids
NETA World, Winter 2004-2005
by Lance R. Lewand
Doble Engineering Company

A previous article presented information on a new breed
of liquid dielectrics, namely natural esters. This article
provides information on the tests that can be performed
to evaluate the chemical, physical, and electrical properties
of natural ester oils. After the last article, I became aware
that Dielectric Systems Incorporated also manufactures a
natural ester dielectric liquid called ECO Fluid which is
based on a high-oleic acid content canola oil (see Sundin).
Information concerning ECO Fluid is presented in this
article along with the other natural esters. In addition, two
oils have been removed from the market, BITORANS
1000 and 5000.
Throughout this article, many comparisons are made to
the oil quality test results of transformer mineral oil, as this
is something we are all familiar with. It must be emphasized,
however, that the properties of a natural ester cannot be
correlated directly to that of a mineral oil as the chemistries
of the two types of liquids are very different. They are so
different, in fact, that ASTM has produced a new specifi-
cation just for natural esters, referenced as ASTM D 6871
“Specification for Natural (Vegetable Oil) Ester Fluids Used
in Electrical Apparatus.”
For the most part, the same tests
used to evaluate mineral oil are used Table 1 – Physical Tests of Natural Ester Dielectric Liquids
to evaluate natural esters. Results and
the meaning of the tests are different ASTM TEST
between the two oils. Doble evaluates
natural esters using five categories Data Source
then, once they are in use, the extent of degradation due to
thermal, oxidative, and hydrolytic actions and contamina-
tion. As oils age while in service, acidic and polar compounds
are formed and ultimately develop into a sludge or polymer-
ized material if remedial action is not taken. The oil can be
replaced or rejuvenated by clay treatment or possibly other
processes such as activated alumina to remove the acidic and
polar contaminants before they induce excessive aging of the
solid insulation and corrosion of metal structures or result in
sludge formation. The generation of sludge and polymerized
material can hamper cooling in the transformer, resulting
in an increased rate of aging of the insulation.
Physical Tests
Below are the results of testing of new natural esters.
Where Doble did not perform the testing, the manufac-
turer’s data is provided, if available.
As shown in Table 1, significant differences exist between
natural esters and typical mineral oil – especially aniline
point, flash point, interfacial tension, pour point, and vis-
cosity. Certain tests such as aniline point and interfacial
Mineral Oil BIOTEMP Coconut Oil ECO Fluid Envirotemp
Typical Values FR3
Doble Producer Producer Doble

identified as follows: Aniline Point, D 611, °C ≤ 80 25.0 2.4

Color, D 1500 ≤ 0.5 L0.5 1.0 L0.5 (Green)
Flash Point, D 92, °C > 145 328 225 321 326
• Physical tests
Fire Point, D 92, °C 180a 358 341 362
• Electrical tests Interfacial Tension, D 971, mN/m > 40 26 26 24
• Chemical tests Pour Point, D 97, °C ≤ (-40) -12a 20 -12 -21
Relative Density, D 1298, 60/60 ≤ 0.910 0.919 0.917 0.920 b 0.923
• Accelerated oxidation tests Viscosity at 40°C, D 445 ≤ 11.0 41.4 29 92c 33.8
(discussed in the next article) Refractive Index at 20°C, D 1807 1.4808 1.4708 1.46 1.4750
• Dissolved gas analysis Visual, D 1524 Clear/Bright Clear/Bright Clear Clear/Bright
(discussed in the next article) a
Without pour point depressants added
b
Corrected to 68°F
The purpose of these tests is to c
Performed by ASTM Method D 88
evaluate the quality of new oils and
78 Insulating Oils Handbook
tension may not have as much applicability for natural higher than that normally found in mineral oil. This is to
esters as mineral oils. For example, aniline point is affected be expected, due to the chemical composition of the liquid
by the aromatics present in an oil of which natural esters causing it to be hydrophilic in nature. When used in electric
have none as refined. Because of the high hydrophilic nature apparatus with paper insulation of the same water content,
of the liquid, the interfacial tension test may not be very natural esters have a water content higher than that of
sensitive to changes in oil quality, as the highest values are mineral oil. This does not mean that the dielectric strength
in the mid-20s mN/m. is adversely affected since dielectric strength is a function of
Another distinct difference but a major advantage of the relative saturation of water in the dielectric liquid, not
natural esters is the more elevated flash and fire points. A the absolute concentration. Natural esters actually provide
distinct margin of safety is attributed to dielectric liquids a benefit in this respect. Because they are able to hold more
with more elevated flash and fire points. The fire point may water, insulating paper tends to be slightly drier in those
affect where certain liquid-filled transformers can be in- systems with natural esters as compared to mineral oils, all
stalled with or without additional fire resistance protection, other things being equal. In fact, some have considered us-
especially in indoor applications or by commercial buildings. ing natural esters as the water conducting medium during
In the United States, indoor installations of less-flammable transformer dryout processes since it can hold so much more
liquid-filled transformers are significantly less regulated water than mineral oil. For example, at 65°C the maximum
than their mineral oil filled counterparts. The NEC in the solubility of water in oil is 285 parts per million whereas for
United States specifies that the liquid in a less-flammable natural esters it is around 2100 parts per million.
transformer must have a minimum fire point of 300°C. Commonly used inhibitors such as DBPC and DBP
In general, the viscosity and pour point values of natural were not detected in the liquids tested. It is already known
esters are much more elevated than those of typical mineral that all four of the natural esters listed in the above table
oil. These physical attributes are very important and cannot contain additive packages with oxidation inhibitors, metal
be disregarded. The viscosity is important to the cooling deactivators, and pour point depressants.
functions and, therefore, to the transformer. A lower viscos- For furanic compound and benzotriazole analysis, inter-
ity liquid allows a more efficient dissipation of heat. Large ferences present in the extract of the solid phase extraction
differences in viscosity such as those attributed to natural made the chromatograms difficult to read when compared
esters can be accounted for in the design of the transformer to a typical mineral oil. This may present problems when
and, thus, can minimize any adverse effects and optimize its trying to determine the aging of the solid insulation as the
operation. However, retrofilling of a mineral oil transformer transformer ages, using this type of analysis. In one case,
with a natural ester may require special considerations and Envirotemp® FR3™, the analysis did reveal the presence of
may not be intended as a direct replacement for transformer an appreciable amount of 5-hydroxymethyl-2-furfural or an
mineral oil, although these types of retrofills have taken interfering compound detected at the exact retention time of
place (see Cooper). the standard used. It cannot be determined whether or not
The pour point is the lowest temperature at which oil this peak represented 5-hydroxymethyl-2-furfural or was an
will flow. A low pour point is important, particularly in cold interference. It is conceivable that the peak does represent
climates, to ensure that the oil will circulate and serve its 5-hydroxymethyl-2-furfural. This is because Envirotemp®
purpose as an insulating and cooling medium. Even with FR3™ is an agriculturally based product which naturally
added pour point depressants, the pour point of a natural contains fructose and glucose, both of which are used to
ester cannot match that of a typical transformer mineral oil. synthesize 5-hydroxymethyl-2-furfural (see Budavari). It
It has been documented that natural esters will solidify in may be a byproduct of the refining process. Cooper strongly
a transformer once the pour point
has been reached and, thus, must
function as a solid impregnant. Table 2 – Chemical Tests of Natural Ester Dielectric Liquids
As long as no voids exist in the
ASTM TEST Mineral Oil BIOTEMP Coconut ECO Fluid Envirotemp
solidified mass, it seems to func- Typical Values Oil FR3
tion well as no incidents have been
reported. Data Source Doble Producer Producer Doble
Corrosive Sulfur D 1275 Non-corrosive Non- Non- Non-corrosive
corrosive corrosive
Chemical Tests Water Content, D 1533, ppm ≤ 30 <80 1.0 110 <80
The results of the chemical tests Neutralization Number, D 974, mg KOH/g ≤ 0.015 0.02 0.05 0.02
performed are shown in Table 2 Oxidation Inhibitor, D 2668, % ≤ 0.30 ND 0.15 ND

and are compared to specification Furanic Compounds, D 5837, µg/L all < 5 all < 5 654 HMF

values for transformer mineral Benzotriazole, Doble, µg/L ≤ 10a ≤ 10 ≤ 10

oil. Polychlorinated Biphenyls, D 4059, ppm <2 <2 <2 <2

The chemical tests for natural

esters, as shown in Table 2, are all ND = None Detected
very similar to typical mineral oil
a
Typical value for mineral oils from North America
except for water content, which is
Insulating Oils Handbook 79
believes that this detected peak is an interference and not 5- Conclusions
hydroxymethyl-2-furfural. Cooper has indicated it believes
The physical, chemical, and electrical tests performed in-
solid phase extraction is a viable technique for determination
dicate similarities and some significant differences between
of at least four of the five furanic compounds and is pres-
natural esters and typical mineral oil. This is not to say that
ently working on a method to remove the interference. In
mineral oil is superior to natural esters. Each liquid type has
all fairness, this analysis was not conducted on the coconut
its advantages and disadvantages and must be fitted to the
oil or the ECO Fluid, as those liquids were not available.
application. The comparison is made because mineral oil is
by far the more commonly used dielectric liquid.
Electrical Tests Many natural esters contain additive packages consisting
The results of the electrical tests performed are shown of chemicals to reduce the pour point and aid in oxygen
in Table 3. stability, and, in some cases they have an antimicrobial
agent or copper deactivators. This contrasts with mineral oil
which has either no additives or merely
Table 3 – Electrical Tests of Natural Ester Dielectric Liquids oxidation inhibitors. It is not known if
ASTM TEST Mineral Oil BIOTEMP ECO Fluid Envirotemp FR3
any adverse characteristics exist when
Typical Values natural esters are used in transformers
Data Source Doble Producer Doble over a long period.
Dielectric Breakdown, D 877, kV ≥ 30 38 45 51 Several routinely performed tests
Dielectric Breakdown, D 1816 kV, 1 mm gap ≥ 20 33 61c 36
such as water content, furanic com-
Power Factor at 20-25°C, D 924, % ≤ 0.05 0.0093 0.09 0.0610 pounds, and PCB analysis required
Power Factor at 100°C, D 924, % ≤ 0.30 0.590 0.30 1.850 modifi cations to yield accurate results
Dielectric Constant at 25°C, D 924 2.2 a
3.1 2.86 3.1 for natural esters. Because of their
Impulse Breakdown, D 3300, kV ≥ 145 134b 148 Insufficient sample chemistry, in-service natural esters will
Gassing Tendency, D 2300, µL/min negative -52.7 +22.0 -80.5 not yield the same results when com-
pared to mineral oil, especially for tests
a
Typical of a mineral oil such as water content, neutralization
b
Test performed using a negative polarity point. ABB reports that impulse break- number, interfacial tension, and power
down is slightly higher when a positive polarity point is used. factor. Instead, they must be judged
c
Testing performed using a 2mm gap instead of a 1mm gap based on their in-service criteria. Un-
fortunately, because natural esters are
The results for coconut oil are not listed as the producer relatively new, not enough in-service data has been collected
did not provide any data. Electrical test results for natural to make those informed decisions.
esters are similar to that for mineral oil. Exceptions are the
very low gassing tendency and the higher dielectric con-
stant. The low impulse breakdown strength is close to the References
minimum value for mineral oil. The extremely low gassing
tendency of some of the natural esters is most likely due 1. David Sundin, 2001, “Vegetable Seed Insulating Fluid,”
to the amount of unsaturated, nonaromatic molecules as United States Patent 6,280,659.
compared to mineral oils. This attribute may be beneficial for
applications in liquid-filled cable systems where a negative 2. Cooper Power Systems, 2000, “Recommended Retrofill
gassing tendency is a sought-after property. In addition, the Procedures for the Replacement of Oil with Enviro-
biodegradability of natural ester liquids is an added benefit temp® FR3 Fluid,” Bulletin 00046, Waukesha, WI.
if leaks should occur – especially in environmentally sensi-
tive areas. 3. Budavari, Susan, ed., 1989, “4764. 5-(Hydroxymethyl)-
The higher power factor results for some of the natural 2-furaldehyde,” Merck Index 11th Edition, Merck & Co.,
esters at 100°C are most likely due to the higher polar con- Inc., Rahway, N.J.
tent compared to that of mineral oil. The inhibitor packages
used are also partially responsible for the increased power
Lance Lewand received his Bachelor of Science degree from St.
factor. Mary’s College of Maryland in 1980. He has been employed by the
Doble Engineering Company since 1992 and is currently the Laboratory
Manager for the Doble Materials Laboratory and Product Manager for
the DOMINO®. product line. Prior to his present position at Doble, he
was Manager of the Transformer Fluid Test Laboratory and PCB and
Oil Services at MET Electrical Testing in Baltimore, MD. Mr. Lewand
is a member of ASTM Committee D 27.
80
Gassing Characteristics of
Transformer Oil Under Thermal Stress
NETA World, Fall 2005
by Lance R. Lewand and Paul J. Griffin
Doble Engineering Company
In June 2005, ASTM International adopted a new test
method for testing transformer oil. The aim of the test is to
determine the gassing pattern of an oil subjected to thermal
stress under what is considered to be low temperatures, i.e.
120 C. The method is entitled “ASTM D 7150, Standard
Test Method for the Determination of Gassing Character-
istics of Insulating Liquids Under Thermal Stress at Low
Temperature.”
Interest in the development of this test resulted from
research that Doble performed in the early 1990s and
more recently the research that the CIGRE Task Force TF
15/12-01-11 (TF11) performed in trying to discern the
differences in gassing behavior of some transformer oils at
low temperatures. CIGRE has termed the unusual gassing
behavior as “stray gassing.”
The CIGRE research was mainly looking at the gassing
behavior of new oils in which transformers were exhibiting
increasing hydrogen levels, yet no apparent cause for the ab-
normal gassing could be determined. As a result, the research
centered on the oil and found that some oils produced more
gas (in this case hydrogen) than others when they were aged
under thermal conditions in both sealed transformers (gas
blanketed, sealed or bladder/diaphragm transformers) and
free breathing transformers (open conservators). Doble had
also observed a similar pattern and as a result conducted
a massive research study into the phenomenon that has
lasted over three years and still goes on today. What was
also determined was the fact that not only could the refining
process be partially responsible for the abnormal production
of dissolved gases, but different types of contamination in
the oil, incompatible materials in the transformer, and the
addition of additives such as metal passivators could also
cause abnormal gassing to occur.
The reason that this information is so important is that
it can significantly affect the results of the dissolved gas-
in-oil (DGA) test. DGA is one of the most widely used
diagnostic tools for assessing the condition of electrical
transformers and in more recent years load tap-changers
and bulk oil circuit breakers. The test is very sensitive to a
Insulating Oils Handbook
wide range of problems and is used to characterize incipient
fault conditions. Categories of abnormal conditions such
as overheating, partial discharge, and arcing can be further
identified by the insulating materials involved (paper or
oil) or by the energy (high, moderate, or low temperatures).
DGA is ideally suited for condition-based maintenance
programs as it allows early detection of problems and can
then be used to follow most deterioration processes as they
evolve. Then, the decision can be made as to when to per-
form more definitive tests to identify the specific problem
or take remedial action. Indeed, DGA is sensitive enough
that it can be used to detect problems during factory testing
of new transformers helping manufacturers avoid shipping
defective units to the field.
It is imperative that DGA results are interpreted accu-
rately to ensure the correct action is taken concerning any
incipient faults within the transformer. It is well known that
oils can age differently, particularly where there is ample
oxygen such as may be found in older units, or free-breath-
ing transformers. A 1991 study revealed that oils exposed
to relatively low temperatures could have different gassing
behaviors and oxygen consumption rates under the same
controlled conditions. Gassing rates were different for two
oils tested if the oxygen content was fairly low at 3000 ppm
or high at 30,000 ppm and this is one of the reasons why
the new ASTM D 7150 test is performed using air and
nitrogen sparged samples. It was also shown that passing
the dissolved gas-in-oil test for a factory heat run could
depend on the oil chosen.
Early Studies on Gassing Behavior
In 1994, the Doble Oil Committee performed a study
on the low temperature gassing behavior of an oil manufac-
tured in France that was being used for the first time by a
French transformer manufacturer. The transformer did not
pass the dissolved gas-in-oil limits used by the purchaser.
The manufacturer performed some tests which showed that
the oil used (Shell Diala F) gassed much more than the oils
Insulating Oils Handbook 81
they used historically. To evaluate the gassing behavior of Nitrogen Purged, 16 hours at 120 C
the Shell Diala F oil in comparison to other products, air- Nitrogen Purged, 164 hours at 120 C
saturated samples and some samples that had been degassed
(all gases removed except for small amounts of oxygen and The aging was performed for differing times to assess
nitrogen) were aged in ground-glass matched barrel and initial gassing rate and when a plateau or equilibrium rate
plunger syringes at 95ºC for 168 hours. The following oils was reached. The samples were either air or nitrogen purged
were tested in the study: to have a range of oxygen contents representative of service
conditions. The aging times were as follows:
Shell Diala F, uninhibited, manufactured in France
16 hours – provides indication of initial generation
Esso Univolt 52, uninhibited, manufactured in France
of gases and would be similar to the time for a fac-
Exxon Univolt N61, inhibited, manufactured in US tory heat run test.
Shell Diala A, uninhibited, manufactured in US 164 hours – indicates if the gases reach a plateau or
The results are provided in Table 1 (air saturated) and a constant rate of generation
Table 2 after the oil had been vacuum processed. The oils were assigned a sequential number based on the
total amount of combustible gases formed during the experi-
TABLE 1 ment with air saturated oils aged for
Gassing Characteristics of New Oils Aged at 95ºC, ppm vol./vol. 16 hours. For example, number 1 had
Oil Hydrogen Methane Ethane Ethylene Acetylene CO CO2 TCG the least amount of total combustible
Diala F 378 6 0 1 63 2 0 215 908 764 gas (TCG) and number 30 the greatest
Univolt 52 122 1 0 1 58 10 0 113 1414 404 amount of TCG. In some cases there
Diala A 81 4 3 20 1 0 65 608 201 was more than one product from a
Univolt N61 35 0 0 0 0 46 489 81 refiner. Only the results of the 164-
hour testing are presented here as this
TABLE 2 is the time that was decided upon for
Gassing Characteristics of Vacuum-Processed New Oils Aged at 95ºC, the ASTM method.
ppm vol./vol.
Oil Hydrogen Methane Ethane Ethylene Acetylene CO CO2 TCG Air Saturated Samples
Diala F 165 133 99 2 0 116 370 515 Aged for 164 Hours
Univolt 52 0 4 2 0 0 0 42 6
at 120°C
For the samples saturated with air
and aged for 164 hours, the data shown in Figure 1 reveals
The results listed in Tables 1 and 2 show clearly that the that there is a large difference in the total combustible gas
Shell Diala F product produced larger amounts of combus- (TCG) content between some of the oils and that very sig-
tible gases than the other oils tested. Even when oxygen is nificant amounts of combustible gases are formed in some
removed or reduced in the test samples, the overall concen- cases. There were four gassing patterns that developed:
tration of gases is reduced but the same oil still produces
the largest quantity of dissolved gases. Table 2 reveals that Mostly carbon monoxide
degassing the two oils to reduce the oxygen content had a Hydrogen and carbon monoxide
much greater effect on the Univolt 52 oil than the Diala
F product. This type of experiment shows that the gassing Mostly hydrogen
characteristics of new oils under thermal stress can be drasti- Mixture of hydrogen, carbon monoxide, methane
cally different. In the case of a transformer heat run, one oil and ethane
would cause the transformer to fail the test while the other
The predominant gas formed was hydrogen. However,
oil would not be of concern.
other gases such as methane, ethane, ethylene, carbon mon-
oxide, and carbon dioxide were also formed. In general, if
Recent Studies on Gassing Behavior the oxygen content remained high, the first three gassing
Doble, with the inspiration of the CIGRE work and its patterns occurred (as listed above). If the oxygen was de-
own previous work, undertook a study to evaluate the gas- pleted, the methane and ethane were generated in greater
sing characteristics of oils commercially available in 2001 concentrations. Ethane was not generated in significant
and 2002 and in many cases still sold today. The study was amounts without methane. The oxygen consumption rate
to look at a wide range of oils to determine the amount of was quite different for the various products.
variation.
Testing was performed under the following conditions:
Air purged, 16 hours at 120 C
Air purged, 164 hours at 120 C
82
Concentration, ppm
Figure 1 — Total Combustible Gas
of Air Saturated Samples Aged for 164 Hours
Nitrogen Purged Samples Aged 164 Hours
at 120°C
Samples purged with nitrogen and then aged 164 hours
did not contain as much combustible gas as those saturated
with air before aging for the same amount of time; however,
there were five oils that still exhibited high concentrations
of hydrogen and total combustible gases. In these cases the
hydrogen made up a large percentage of the composition
of the TCG.
What is interesting in nitrogen-sparged samples is that
the methane and ethane values are consistently low but make
up a greater percentage of the TCG for more samples than
was the case when there was ample oxygen at the start of
aging. This is due to the fact that the hydrogen produced in
nitrogen-sparged samples is not as high in concentration.
Similarly, the carbon oxide gases are consistently lower for
the nitrogen-purged samples than when saturated with
air.
Conclusions
There is a new ASTM Method D 7150 that provides
the protocol for determining the gassing characteristics
of oils under air and nitrogen (sealed) conditions and at
low temperature (120 C).
It is not meant to be a routine test method in the sense
that DGA currently is. However, it can be applied in
those situations where there is a question of whether
the oil is involved in producing the gassing that is being
observed.
DGA is a very important test, but care needs to be tak-
en in the interpretation of results to understand the low
temperature gassing behavior of the oil when certain
types of problems are detected. As hydrogen is often
the predominant gas formed for the oils with the high-
est combustible gas generation rates, it is possible that
Insulating Oils Handbook
Concentration, ppm
Figure 2 — TCG and Hydrogen Content
of Nitrogen Saturated Samples Aged for 164 Hours
low temperature thermal problems could be confused
with low energy partial discharge activity. Stray gassing
may also be mistaken for excessive gassing at modest
temperatures.
Some of the “breaking in” characteristics that are ob-
served for the gassing behavior of new transformers or
newly processed transformers may be attributed to the
oil and will stabilize over time.
Stray gassing can also be due to the presence of con-
tamination in the oil, the presence of incompatible ma-
terials, and the presence of additives.
It is clearly important to know the properties and char-
acteristics of the oil in individual transformers, especial-
ly when performing DGA. Similarly, it is important to
make sure the gassing characteristics of oils used to ‘top
off ’ transformers are also well understood.
References
1. Griffin, P.J., Lewand, L. R., Heywood, R., and Lap-
worth, J. “Gassing Characteristics of Transformer Oils
at Modest Temperatures, Part 1: Transformer Experi-
ences”, Proceedings of the Seventy-First Annual Con-
ference of Doble Clients, 2004.
2. Griffin, P.J., Lewand, L. R., Heywood, R., and Lap-
worth, J. “Gassing Characteristics of Transformer Oils
at Modest Temperatures, Part 2: Laboratory Experi-
ments”, Proceedings of the Seventy-First Annual Con-
ference of Doble Clients, 2004.
3. Schmidt, J., Eitner, R., and Hartwig, R., “Gassing Be-
havior of Transformer Oils at Temperatures Above
60ºC”, 7th International Symposium on High Voltage En-
gineering, Aug. 26-30, 1991, pg. 33-36.
Insulating Oils Handbook 83
4. Griffn, P. J., “Gassing Characteristics of New Oils Used
in Factory Heat Runs”, Doble Oil Committee Minutes,
1994, pg. 18-21.

Lance Lewand received his Bachelor of Science degree from St.

Mary’s College of Maryland in 1980. He has been employed by the
Doble Engineering Company since 1992 and is currently the Laboratory
Manager for the Doble Materials Laboratory and Product Manager for
the DOMINO®. product line. Prior to his present position at Doble, he
was Manager of the Transformer Fluid Test Laboratory and PCB and
Oil Services at MET Electrical Testing in Baltimore, MD. Mr. Lewand
is a member of ASTM Committee D 27.

Paul J. Griffin received his BS degree at the American International

College and his MS at the University of Rhode Island. He has been em-
ployed by the Doble Engineering Company since 1978 and is currently
Vice President of Laboratory Services. He is secretary of the Doble Oil
Committee, a member of ASTM committee D 27, US Technical Advisor
to IEC TC10 for Fluids for Electrotechnical Applications, member of
the IEEE Insulating Fluid subcommittee of the Transformer committee,
and a member of the CIGRE Working Group 15.01 Fluid Impregnated
Insulating Systems.
84 Insulating Oils Handbook

Condition Assessment
of Transformers —
Analysis of Oil Data and Its Quality
NETA World, Winter 2005-2006

by Lance R. Lewand
Doble Engineering Company

Condition assessment or life assessment of a transformer really that high? Was the sample taken correctly? Did the
should use a variety of techniques. These techniques en- past test data indicate the same condition? In addition, the
compass electrical testing, a physical assessment, review of temperature of the transformer from which the sample was
historical design and test data, review of future load and use taken plays an extremely important part in understanding
requirements, oil analysis, and others. No one technique will the data that was generated in the laboratory. For example,
be able to accurately assess the condition, and they all must if a sample was taken from a transformer operating at 80°C
be used in concert to provide a good evaluation. with 1.6 percent water in the paper insulation, the water
Insulating oil analysis is one of those very powerful and content of the sample should be around 33 ppm under
invaluable tools to aid in such an investigation. The most re- equilibrium conditions. When the sample gets transported
cent data as well as historical trends are especially important to the laboratory for analysis there will be a reduction in
to provide relevant information of the transformer in ques- temperature. At room temperature, say 21°C, with 33 ppm
tion. Many analytical tests can be performed which provide of water, the relative saturation of water in the oil will be
information that can then be gathered and formulated into around 57 percent and the dielectric strength may be low
an overall diagnosis. For example, tests such as interfacial because of it. Although, there are no temperature correction
tension, color, and neutralization number provide informa- factors for dielectric strength test measurements, a seasoned
tion on the condition of the oil. Water content along with analyst must be able to recognize the importance of the
temperature data can provide information on the wetness or temperature change and its effects on the laboratory results
dryness of the liquid insulation and by calculation, the water compared to the dielectric breakdown voltage of the oil in
content of the solid insulation. Dissolved gas-in-oil analysis the transformer at operating temperatures.
(DGA) is an extremely powerful test as it is a sensitive test Another aspect of transformer condition assessment is
for detecting a wide range of incipient-fault conditions or reviewing the oxygen and nitrogen data from the DGA.
problems that may exist within the unit and may influence Take, for example, the information provided in Table 1.
operation decisions. Furanic compounds are by-products
of paper degradation and their determination provides Table 1: Oxygen and Nitrogen Concentrations
information on the mechanical strength of the paper and
localized overheating of paper insulation. Other tests such from the DGA, ppm
as corrosive sulfur, particle count, and particulate and dis- Sample Date Oxygen Nitrogen
solved metals provide information on unusual conditions
7/14/1990 1150 85,200
that may exist in a transformer and can help in providing
an understanding of the source of a problem or help clarify 9/15/1991 1640 81,400
those problems identified by other methods. 8/7/1992 1200 87,300
The quality of the data is especially important as poor 8/9/1993 1350 82,500
data can lead to a poor diagnosis or misdiagnosis. Take for 7/25/1994 2000 84,100
example a case in which the dielectric breakdown strength is 9/10/1995 1860 86,700
low (poor). A closer look at the results may indicate that the 8/27/1996 1280 85,300
water content is elevated which directly affects the dielectric 8/10/1997 1410 84,900
breakdown strength measurement. However, a series of 9/1/1998 1570 86,200
questions must then be asked such as: Is the water content
Insulating Oils Handbook
From a review of the data in Table 1, several items can
be gleaned. For instance, the sample taken on 8/7/1992 in
which a high oxygen content is present indicates that there
was probably a sampling error that occurred that allowed air
to enter the syringe prior to analysis. The nitrogen content is
too high for a sealed conservator transformer in good condi-
tion, so it is likely from a gas-blanketed unit. If the nitrogen
value were much higher it could indicate that regulator on
the nitrogen tank feeding the transformer may be set too
high causing more nitrogen to be dissolved in the oil. This
makes the oil more susceptible to being super-saturated
with nitrogen. Oils that are supersaturated with nitrogen
can release bubbles during quick cool downs which cause
a reduction in the oil dielectric breakdown voltage. Most
users of nitrogen blanketed transformers set the nitrogen
pressure at 1-3 psi to avoid this condition. Sometimes,
however, the information from the field does not match
the data. For example if the data sheet coming from the
field were marked “conservator” for the type of preservation
system, then the data in Table 1 would not match as different
nitrogen concentrations would be expected unless there was
a leak in the system. In this case, the type of preservation
system would have to be confirmed.
In many cases, the oxygen and nitrogen content data is
not so clean. The data such as that presented in Table 2 is
often seen.
Table 2: Oxygen and Nitrogen Concentrations
from the DGA, ppm
Sample Date Oxygen Nitrogen
7/14/1990 1150 85,200
9/15/1991 1640 81,400
8/7/1992 18,100 60,200
8/9/1993 1350 82,500
7/25/1994 2000 84,100
9/10/1995 4100 79,300
8/27/1996 8,200 75,100
8/10/1997 1410 84,900
9/1/1998 15,300 71,900
As shown in Table 2, there is a lot of variation in the
oxygen and nitrogen data. In the case of the sample taken
on 8/7/1992 there was probably a sampling error in which
air was allowed to enter the syringe prior to the analysis. The
syringe itself could have been compromised or the sample
was taken in another container such as a bottle in which
there is no way to preserve the oxygen/nitrogen content in
its original state. If the sample container is not gas tight,
other gases will also be affected. Examples are hydrogen
and carbon monoxide whose solubilities in oil are low and
thus can easily escape from the sample. The oxygen levels
on 9/10/1995 and 8/27/1996 are also slightly elevated and
may be due to a combination of sampling error, syringe
leakage, or sample preparation in the laboratory. The data
from the last sample point, 9/1/1998, is very interesting as it
raises many questions. Is the reason for the elevated oxygen
85
content due to sampling error, problems in the laboratory,
or an actual condition? That is a much harder question to
answer, and more data is needed before a conclusion can
be drawn. If samples taken after 9/1/1998 show that the
oxygen content reverts back to the low concentrations, in
most cases it was some error in the way the sample was
taken, stored, and/or analyzed that caused the problem. If
the oxygen content continues to increase, then there may
be a problem with the transformer in which the seals of
the unit have become compromised and oxygen is now
able to enter the system. The presence of high oxygen in a
transformer increases the rate of degradation of the oil and
solid (cellulose) insulation and is unwanted.
The lesson learned is that every test and result must be
reviewed critically in order to provide a competent diagnosis.
In addition, many tests are interrelated, and their results
must be reviewed as such. The transformer nameplate data
arriving from the field is especially important as transform-
ers of different voltage classes, preservation systems, and
other factors effects the diagnosis that is provided. Providing
comments on oil test data based on IEEE or other limits
is not enough and in most cases provides a very limited
diagnosis. The types of oil analysis tests performed must be
selected in such a way as to provide the best overall condi-
tion assessment. To obtain a good condition assessment
on transformers, more oil testing is needed than that for
routine testing. Good sampling techniques and appropriate
sample containers are especially important and can impact
all the results and the final condition assessment and have
been covered in detail in previous articles. The final lesson is
that the quality of the data from the laboratory is especially
important to obtain good information for evaluating the
condition of transformers. Poor analytical data takes time
to sort through, can result in a misdiagnosis, and can use
resources for resampling and retesting in order to obtain a
clear understanding of apparatus condition.
Lance Lewand received his Bachelor of Science degree from St.
Mary’s College of Maryland in 1980. He has been employed by the
Doble Engineering Company since 1992 and is currently the Laboratory
Manager for the Doble Materials Laboratory and Product Manager for
the DOMINO®. product line. Prior to his present position at Doble, he
was Manager of the Transformer Fluid Test Laboratory and PCB and
Oil Services at MET Electrical Testing in Baltimore, MD. Mr. Lewand
is a member of ASTM Committee D 27.
86 Insulating Oils Handbook

Recent Applications of DGA

PowerTest 2006
(NETA Annual Technical Conference)
Presenter
Karl Jakob, P.E., Weidmann-ACTI Inc.
Co-Author
Fredi Jakob, Ph.D., Weidmann-ACTI Inc.

1. Introduction Table 1
During the past forty years Dissolved Gas Analysis, Gases Produced during Normal and Abnormal
DGA, was developed, promoted and accepted as the single Equipment Operation
most effective diagnostic tool for power transformers. There
are many additional types of electrical equipment that also Equipment Normal Gases Stress Gases
use mineral oil as a dielectric fluid and can be subjected to Transformers CO, CO2 C2H2, C2H4, C2H6,
thermal and electrical stresses that result in the production (Power) H2, CH4, CO and CO2
of fault gases. A logical question is: why has there been a LTCs and OCBs C2H2, H2 and all other Increased amounts of
“delay” in applying DGA to other electrical apparatus? There fault gases “Hot Metal” Gases
are two basic reasons for the delayed application of DGA Bushings Very Little Gas, Small All Possible Fault Gases
to an expanded array of equipment. The first reason is that Amounts of CO2 and CO
switching equipment such as load tap changers and oil cir- Voltage Regulators, Same as Power Same as Power
cuit breakers produce gases during normal operation. It was Current and Potential Transformers Transformers
thus assumed that the presence of these normally produced Transformers
fault gases would obscure the observation of additional
gases produced by thermal or electrical faults. Fortunately,
as shown below, analysis of the DGA results can reveal Further consideration of the gas formation mechanism
problems, even in the presence of “background” gases. in LTCs and OCBs led to the conclusion that the gas con-
A second reason for the delayed application of DGA was centration pattern changes when these units are subjected
the limited quantity of oil in apparatus such as bushings, to thermal or electrical stress. During normal operation
current and potential transformers. Removal of sufficient the primary arcing gases, acetylene and hydrogen, are pro-
quantities of oil for DGA analysis could lower equipment duced in abundance. Due to the cooling effect of the oil,
oil volumes below required levels. There was, and still is, temperatures at which the hot metal gases are produced
concern about possible contamination of the equipment exist further from the arcing site. Thus all of the fault gases
during the sampling process. These issues have been suc- are produced in both LTCs and OCBs. Load tap changers
cessfully addressed and, as a result, DGA has once again do not contain any cellulose-based insulation, but carbon
proven to be a powerful diagnostic tool for additional dioxide and carbon monoxide are produced from the heated
electrical equipment. oil in the presence of dissolved oxygen.
A common problem in both LTCs and OCBs is the
2. Load Tap Changers and Oil heating of contacts due to a buildup of carbon. This pro-
cess is known as coking. The deposit increases the contact
Circuit Breakers resistance which increases contact temperature and results
It was commonly assumed that DGA was not applicable in additional carbon formation from the breakdown of the
to equipment in which switching and associated arcing oc- oil. This self- accelerating process leads eventually to “ther-
curred during normal operation. This concept is illustrated mal runaway.” Abnormal heating from the contact coating
in Table 1. results in additional production of the heating gases, meth-
ane, ethane and ethylene. Temperatures are high, but not
Insulating Oils Handbook 87
high enough to produce additional quantities of acetylene.
Analysis of these observations leads one to conclude that
all of the fault gases are produced during normal operation
and that contact coking leads to the production of additional
hot metal gases.
The total gas concentration in an LTC or OCB will be
related to the number of operations and loss of fault gas
from the oil to the atmosphere. The gas concentration ratios
should be relatively independent of the operation count.
One can thus conclude that changes in gas ratios are very
important. Additional reasons to consider gas concentration
ratios as the most appropriate diagnostic data interpreta-
tion tool are:
1. Gases are lost continuously to the atmosphere. Hydrogen
is most rapidly lost and the rate of gas loss depends on tem-
perature and breathing configuration. The ratio of ethylene
(a key heating gas) / acetylene (the key arcing gas) remains
fairly constant since these gases are both lost much more Figure 1 — LTC Damage
slowly than hydrogen.
2. The ratios of heating to arcing gases have been shown
to be strongly dependent on the extent of contact coking. B. OCB Case Study
3. LTC and OCB Case Histories

A. LTC Case History

Based on the diagnostic results a “Warning” condition

The above LTC Study dramatically illustrates the im-
portance of ratios. The DGA samples from the LTC cover
a two year period. The “key” heating gas, ethylene has con-
sistently increased. The other hot metal gases, methane and
ethane also increased. The levels of the heating gases are
still quite low and may not yet be of great concern. The key
ratio of ethylene / acetylene increased dramatically during
this period. Based on this ratio the LTC was investigated.
The evidence of contact damage is quite evident as shown
in Figure 1.
code was issued. Our client was surprised with the result
because they had recently performed an IR inspection which
indicated normal operation. Based on the DGA data an in-
ternal inspection was performed. The inspection uncovered
significant contact wear and a completely destroyed resistor
contact, fragments of which were lying on the bottom of
the tank. These findings were consistent with the diagnostic
results. The resistor helps reduce the current that the contacts
interrupt. Without the resistor the contacts had to inter-
rupt more current which produced more “hot metal” and
arcing gases. The results of the IR scan make sense since the
problem is created only when the breaker operates.
88 Insulating Oils Handbook

5. Bushing Case History

115 KV Type COB Bushing Case Study

Component Serial # 25610 Serial # 25600 Serial # 15940 Serial # 19147

Gas
Hydrogen 49,945 6.5 1705 19,131
Methane 6523 <0.1 146 1256
Ethane 2185 29 71 409
Ethylene 2291 <0.1 1.8 11
Acetylene 542 TRACE <0.1 0.2

Serial # 25610 Nameplate 1979 1985 1989

C1 Power Factor .575 .857 .599 .136
C2 Power Factor .721 .453 1.56
Figure 2 — OCB Damage
Serial # 25600 Nameplate 1979 1989 1991
C1 Power Factor .30 .593 .309 .261
C2 Power Factor .45 .246 .248

4. Bushings Serial # 15940 Nameplate 1989 1993 Shop

John Stead1 has published several important papers that C1 Power Factor .31 .44 1.23 .40
show that DGA is often a more reliable indicator of bush- C2 Power Factor .21 .20 .24
ing condition than the widely used power factor test. The Serial # 19147 Nameplate 1989 1993 Shop
reluctance of U.S. manufacturers to recommend DGA for C1 Power Factor .29 .31 1.14 .77
bushings as a diagnostic tool is based on their concern with C2 Power Factor .25 .27 .27
oil contamination during the sampling process. Stead1 has
developed a sampling protocol to minimize the possibility
of bushing oil contamination. The newest DGA laboratory
method, D3612-C, uses a much smaller oil sample (5-9ml) Observations during Disassembly Serial # 25610
than the original D-3612-A (35-50ml) method. Therefore, Large region of severe carbon tracking observed lower
less oil has to be removed from the bushing in order to end of bushing core.
complete a DGA. Inner surface of lower porcelain weathershed showed
At this time interpretation of the DGA results is similar carbon tracking.
to that used for power transformers (see Table 2).
Small amount of carbon tracking had penetrated to
within 1 turn of potential tap layer.
Table 2
DGA Interpretation Considering the condition, it was extremely fortunate
that this bushing had not failed in service.
Predominant Gas(es) Interpretation Overall test (GST-Ground) at 10 kV was 8.80%
H2 Partial Discharge power factor.
C2H2, H2 Arcing Water content was measured at 6 PPM.
CH4, C2H4, C2H6 Heating
Observations during Disassembly Serial # 25600
Oil contained small flakes of discolored paper.
IEC2 recommends a DGA for bushings at a four-year
interval. Several Canadian utilities are relying more on Water content was measured at 3.5 PPM.
DGA than on power factor measurements to evaluate their Dielectric strength measured by ASTM Method D1816
bushing population. was 27 kV.
No carbon tracking was observed on either the bushing
core or the weathershed.
Insulating Oils Handbook
Observations during Disassembly Serial # 19147
Odor of burnt oil and paper was noticed approximately
half way to the core during the unwrapping process.
Paper was of a lighter color as we moved inward to the
core and did not appear to be saturated with oil as in the
outer wrappings.
X-wax was observed on the copper draw lead tube.
Power factor improved after removal from unit
Bushing serial number 15940 was not disassembled
due to time constraints.
6. Voltage Regulators
Voltage regulators provide utility and industrial com-
panies the ability to maintain constant line voltage under
varying load conditions. Voltage regulators are usually
coupled with tap changers that provide the ability to control
the voltage during peak and reduced demand to optimize
operating conditions. DGA is an important diagnostic
tool used to identify problems within voltage regulators
through the use of threshold values and ratios of heating
to arcing gases. Thresholds values are utilized to help deter-
mine whether a diagnostic ratio is applicable to a specific
regulator unit, because low levels of Ethane, Ethylene and
Acetylene are typical for voltage regulators, and the ratios
are only applicable when a minimal level of each gas used
in the ratio is present in the DGA analysis.
The gas ratios used for the voltage regulator diagnostics
are Ethylene/Acetylene, and Ethane/Methane. These ratios
can give an indication of contact coking, contact erosion,
and overheating.
89
nologies are important components of a condition-based
maintenance program. Application of DGA to every type
of oil filled equipment is rapidly becoming as significant, as
the historical application to power transformers.
References
1. John Stead, Fredi Jakob “USE OF DGA TO CON-
FIRM UNSATISFACTORY DOBLE TEST RE-
SULTS OF 115 KV BUSHINGS” Minutes of Six-
ty-Third Annual International Conference of Doble
Clients, 1996 63PAIC96
2. IEC 60599:1999-03, Mineral oil-impregnated electri-
cal equipment in service – Guide to the interpretation
of dissolved and free gases analysis.
Karl Jakob is the Director of Business Development for Weidmann-
ACTI Inc. He holds a Bachelor of Science in Electrical and Electronic
Engineering from California State University Sacramento and is a Reg-
istered Electrical Engineer in the State of California. He is a co-founder
of Analytical ChemTech International (ACTI) and has co-authored
several papers on oil diagnostic applications which have been published
in trade journals and presented at technical conferences such as EPRI,
NETA, Doble, etc. Karl is an active member of the IEEE Transformer
Committee.

7. Current and Voltage Transformers

There was never any reason to doubt that DGA would
be applicable to both voltage and current transformers. The
issue was the limited volume of oil in the apparatus. With
the recent development of ASTM 3612C the sample vol-
ume requirement has been greatly reduced minimizing this
problem. At this time DGA data interpretation is the same
as power transformers. As more DGA data is accumulated
specific interpretation may be developed.

8. Conclusion
Management and stockholder expectations require
utilities to operate at increased efficiencies and generate
profit for investors. Reducing expenses is one way to meet
these demands. It is especially significant if one can simul-
taneously decrease costs and increase system reliability.
Maintenance has always been an area subject to budget
cuts because the effect of these cuts was not immediately
evident. Simply cutting maintenance costs is a short term
option. System reliability must be maintained despite the
cuts. Fortunately, condition based maintenance can both
save money and simultaneously increase system reliability.
Increased utilization of DGA and other diagnostic tech-
90 Insulating Oils Handbook

Passivators —
What They Are and How They Work
NETA World, Spring 2006

by Lance R. Lewand
Doble Engineering Company

In the past several years there have been some failures of One major concern is that there is a population of trans-
very large power transformers and shunt reactors associated formers and reactors that are in service with oils that have
with the presence of corrosive sulfur in the transformer oil. excessive corrosive sulfur or a propensity to form it. Many
This problem occurs despite the fact that the oils involved of the failures seem to have occurred in the past few years
have passed standard specification test criteria for corrosive with apparatus that has been in service from 1-7 years. Dur-
sulfur, such as ASTM D 1275. These failures occur because ing this window of time, the number of units with suspect
corrosive sulfur in the oil reacts with copper to form copper oil that have been placed in service could represent a large
sulfide, a conductive compound. The copper sulfide can form monetary value even though it is a small percentage of the
at the copper surface or with copper ions in the oil and paper. total units in service. Not only is there the cost of the ap-
Some of the copper sulfide formed at the copper surface can paratus, there is also loss of revenue and possible regulatory
also migrate to the paper insulation. The formation of copper fines if power is not delivered as specified in contracts.
sulfide is depicted in Figure 1. The conductive copper sulfide This concern about corrosive sulfur has led several trans-
causes a reduction in dielectric strength of the paper insula- former manufacturers and oil refiners to recommend the use
tion. Failure results when the dielectric breakdown strength of passivators by adding them to the oil. There are different
of the conductor insulation is exceeded by the voltage stress kinds of passivators that can be used, but one that has been
which can be influenced by transient voltages. The result recommended for the corrosive sulfur problem is added to
is arcing between two or more turns or possibly disks [1]. reach 100 ppm (mg/kg). This would mean that in a 10,000
Information about corrosive sulfur has been presented in gallon transformer, 7.5 pounds of this passivator would
this publication previously [2,3, and 4]. have to be added to the oil to make the final concentration
100 ppm.
So what exactly are passivators? Although passivators
have been around for a long time in the lubrication industry
S
Cu S S S S S
S
S
their use is relatively new in the electrical industry. Pas-
Cu
Cu Cu S S S S sivators, also know as metal deactivators, react with reactive
Cu S S
Cu S S metal surfaces and dissolved metals such as copper and silver
and reduce their rate of reaction with compounds in the oil.
This includes oxidation reactions with organic compounds
and reactions with corrosive sulfur. Passivators are composed
of two basic types, sulfur based and nitrogen based.
Cu
Cu S
Cu
Cu S Cu
The first suggested use of passivators in transformer oil of
Cu S which the author is aware was in 1967 by J.J. Melchiore and
Cu Cu Cu
I.W. Mills of the Sun Oil Company. The article that they
Cu
Cu Cu S CuCu Cu S
Cu S S
Cu Cu S CuCu Cu S
Cu S S
Cu
Cu S
wrote [5] dealt with the oxidation stability of transformer
oils in response to accelerated oxidation tests such as the
DOBLE PFVO (power factor valued oxidation) test. The
Figure 1 — Corrosive Sulfur Attack of Copper Conductor and Ions results of the testing that they performed suggested that
(Cu=Copper; S=Sulfur) the use of a metal deactivator (passivator) coated the cop-
per surface that was used as a catalyst in the experiment.
Insulating Oils Handbook 91
The passivator decreased the catalytic effect of the copper sulfide was eventually observed on copper surfaces. The
and the amount of copper that dissolved in the oil. This question is whether the passivator sufficiently retards the
reduced the oxidation of the oil and formation of deleteri- corrosive sulfur reactions so they are not of concern for the
ous by-products. Passivators have been used in some cases normal life of the apparatus. Once copper sulfide is formed
for this purpose. Passivators have also been used by some on copper surfaces and deposited in the paper insulation, it
transformer manufacturers in Japan for static suppression. is not removed by passivator or other means. The passivator
Passivators that have been used in electrical insulating undergoes a variety of reactions in the apparatus and can be
mineral oils are nitrogen based and have been predominantly degraded or consumed and might need to be added periodi-
benzotriazole (BTA) or its derivatives. BTA is a granular cally to retain sufficient binding on the copper surface to
solid at room temperature and requires heating and mixing block the corrosive sulfur reactions.
to dissolve in the oil, so it is less desirable than its derivatives.
One type of BTA derivative has a hydrocarbon molecule tail Alternative Approaches
attached. The BTA derivative is liquid at room temperature
For new oils the best approach is to use an oil that does
and, therefore, much more readily mixes with transformer
not have significant amounts of corrosive sulfur compounds
mineral oils. One passivator that has recently been suggested
as shown by passing a more rigorous modified version of the
for use for suppression of corrosive sulfur reactions is that
ASTM D 1275 corrosive sulfur test. Doble recommends
produced by CIBA named Irgamet 39. There are several
performing the test at 150ºC instead of 140ºC and for 48
other manufacturers and brand names of products that are
hours rather than 19 hours and using better nitrogen purging
commercially available.
and sealing of the test vessel. In cases where oils have been
The action of the passivator is to have the nitrogen group
used that do not meet this more stringent criteria, passivator
of the BTA molecule bind with the copper, silver and other
could be used to retard further corrosion. An alternative ap-
reactive metal surfaces. This process is a chemical bonding
proach is to use full or partial retrofills of the oil with an oil
that, given the right circumstances, can be undone (reversed).
that does not have significant amounts of corrosive sulfur
The passivator molecule is attracted to the metal surface and
compounds even when exposed to very high temperatures.
is held to the reactive site so that same site cannot be oc-
This can be tested on a case by case basis to determine the
cupied (attacked) by a corrosive sulfur molecule. Passivator
required dilution. All these methods deserve further research
molecules can also bind with free copper and silver ions or
to determine the best long-term approaches to solving the
particles present in the bulk oil (see Figure 2). In effect, the
corrosive sulfur problem.
BTA and corrosive sulfur compete for the reactive metal
sites, along with other compounds.
Conclusions
S BTA
S Passivators are metal deactivators that bind with certain
BTA BTA BTA BTA
BTA BTA
S BTA BTA BTA
BTA S BTA
BTA
BTA reactive metal surfaces to retard or delay attack by corrosive
BTA BTA BTA
BTA
sulfur. Passivators act by blocking the sites for corrosive
Step 1 sulfur compounds to attack the copper and form copper
sulfide. As shown in Figure 2 and as discussed, passivation
Cu Copper
does not completely block all metal reactions with corrosive
BTA Cu BTA Cu BTA BTA Cu
Step 2 S BTA BTA BTA BTA
BTA
BTA
BTA BTA BTA BTA
Cu S BTA Cu S
BTA BTA BTA
Conductor sulfur compounds. Further research is needed to determine
if passivation is a good long-term solution for in-service oils
S S
with excessive amounts of corrosive sulfur.
S S
S S
S S
S
Step 3 Cu Cu Cu BTA
BTA Cu Cu S BTA
BTA Cu BTA
Cu S BTA BTA
Cu S
BTA BTA BTA BTA BTA
BTA
BTACu S
BTA
BTA Cu S BTA
BTA BTA References
[1]. Lewand, Lance, “Investigating Copper Sulfide Con-
tamination in a Failed Large GSU Transformer” in Pro-
Figure 2 — Action of a Passivator (BTA = BTA or its derivatives) ceedings of the Seventy-Second Annual International Con-
ference of Doble Clients, Doble Engineering Company,
Much of the study on the use of passivator for control Watertown, MA USA, 2005
of corrosive sulfur reactions to date has been in the labora-
tory. The long-term effectiveness for retarding the action of [2]. Lewand, Lance R. “Corrosive Sulfur in Transformer
corrosive sulfur in apparatus is not known. Investigations at Systems”, NETA World, Summer, 2003.
the Doble Laboratory have shown that passivators retard [3]. Lewand, Lance R. “Sources of Sulfur in Transformer
reactions of metals with corrosive sulfur. However, even Systems”, NETA World, Fall, 2003.
at moderate temperatures (110 C) with passivator added,
under certain laboratory conditions corrosive sulfur attack [4]. Lewand, Lance R. “The Negative Effects of Corro-
still occurred. The reactions occurred over extended periods sive Sulfur on Transformer Components”, NETA World,
compared to experiments without passivator as copper Winter, 2003.
92 Insulating Oils Handbook
[5]. Melchiore, J.J., and Mills, I.W., “Factors Affecting
Stability of Transformer and Cable Oils”, in Proceed-
ings of the Seventh Electrical Insulation Conference, IEEE,
October 1967.

Lance Lewand received his Bachelor of Science degree from St.

Mary’s College of Maryland in 1980. He has been employed by the
Doble Engineering Company since 1992 and is currently the Laboratory
Manager for the Doble Materials Laboratory and Product Manager for
the DOMINO®. product line. Prior to his present position at Doble, he
was Manager of the Transformer Fluid Test Laboratory and PCB and
Oil Services at MET Electrical Testing in Baltimore, MD. Mr. Lewand
is a member of ASTM Committee D 27.
Insulating Oils Handbook
What is Sludge?
NETA World, Winter 2006-2007
by Lance R. Lewand
Doble Engineering Company
Have you ever been reading a laboratory report listing
the results from testing of a transformer oil sample and
wondered what the results are trying to tell you? Words
like sludge or metallic soaps, x-wax or ionic species creep
into the vernacular and leave you scratching your head and
wondering what significance that terminology really is. I
am sure you are not alone in this regard.
This article will discuss one of these terms, sludge. It
will also briefly discuss sludge formation in transformers,
what it is, and its significance. Three basic analytical tests
are useful detectors of the precursors to sludge formation
and precipitation.
Interfacial Tension (ASTM D 971, mN/m) The inter-
facial tension of an oil is the force in dynes per centimeter
(milliNewtons per meter) required to rupture the oil film
existing at an oil-water interface. When certain contami-
nants such as soaps, paints, varnishes, and oxidation by-
products are present in the oil, the film strength of the oil
is weakened, thus requiring less force to rupture. For oils
in service, a decreasing value of interfacial tension indicates
the accumulation of contaminants, oxidation by-products, or
both. This reduction is a precursor of objectionable oxidation
by-products that may attack the insulation and interfere
with the cooling of transformer windings [1].
Neutralization Number (ASTM D 974, mgKOH/g)
The neutralization number of an oil is a measure of the
amount of acidic or alkaline materials present. It is a simple
acid/base titration and some use it interchangeably with acid
content. As oils age in service, the acidity and therefore the
neutralization number increases. Used oil having a high neu-
tralization number indicates that the oil is either oxidized
or contaminated with materials such as varnish, paint, or
other foreign matter. (A basic neutralization number results
from an alkaline contaminant in the oil) [1].
93
Pentane-Insoluble Sludge (Doble) This test is a quali-
tative test in which insoluble sludge is found to be either
present or absent based on a visual inspection. The test is
conducted by mixing oil from the sample and a defined
amount of a hydrocarbon solvent, pentane. The mixture is
held for 16-24 hours in a cool, dark place and then exam-
ined by tilting the vial and visually inspecting for tiny solid
particles of gelatinous clumps [2]. This solid material forms
from soluble sludge which precipitates due to the addition
of the hydrocarbon solvent.
These tests are important as they are good indicators
of the formation of degradation by-products of both oil
and paper that are oil soluble. The value for the interfacial
tension decreases and the neutralization number increases
in response to the formation of organic acids and polar
compounds (compounds with an oxygen-bonded atom),
both of which are detrimental to the quality of the oil.
These by-products will eventually lead to sludge formation
which should be avoided. The pentane-insoluble sludge test
is a quick test that Doble uses to detect the early stages of
sludge formation.
The development of sludge in a transformer is deleterious
and presents a number of problems. Once formed in the oil,
it can layer on the winding insulation and in the oil ducts. In
these cases it acts as a thermal insulator. That is, it prevents
the transformer from being cooled properly, and thus more
overheating occurs which contributes to even more sludge
formation so it is a continuously compounding problem. The
additional heat reduces the life of both the oil and paper
insulation but mainly the paper insulation. Sludge may be
slightly conductive as well, depending on its composition,
reducing the dielectric strength of the solid/liquid insula-
tion. Because it is a polar type material it has a tendency to
attract water which increases its conductivity.
94
The formation of sludge presents maintenance issues
as well. Oils with a low interfacial tension and an elevated
neutralization number are easy to reclaim by removing
the precursors of sludge through the use of activated clay
(Fuller’s earth) or activated aluminum. However, once
sludge precipitates as a solid and bakes on the insulation or
mechanical structures of the transformer, much more ef-
fort, time, and money are needed to remove it. The primary
reason is that it tends to attach to the windings and other
surfaces so it is not readily dissolved in oil and thus not
easy to remove.
How does Sludge Form and What is it? Sludge forms
from the degradation of the oil and paper insulation due
to stressors such as oxygen and temperature and catalysts
such as copper. The early precursors of sludge are peroxides,
organic acids, alcohols, aldehydes, ketones, lacquers, and
other aromatic compounds, particularly those that have
polar functional groups. As these materials attack other
transformer components such as the paper, iron, and cop-
per, they form intermediate by-products in the oil that then
polymerize together to form a solid type material which is
called sludge. This is the terminal stage of this degradation
process [3]. Sludge tends to precipitate out in the coldest
and hottest regions of the transformer.
Figure 1 — Filters with Different Types of Sludge Deposits
Figure 2 — Filters with Varying Amounts of Carbon Deposits
Insulating Oils Handbook
Sludge and Carbon: Carbon fines and sludge fines
are often confused with each other. Sludge is formed as a
result of the process already described. Carbon is formed
from a totally separate process in which the carbon atoms
are completely separated from the hydrocarbon molecules
found in transformer oil. The process also generates hydro-
gen and hydrocarbon gases and requires a large amount of
energy such as that resulting from very high temperature
overheating or discharge conditions. Carbon is present in
transformers but most often present in load tap-changers
and oil circuit breakers as arcing occurs more frequently
in these devices. Once either sludge or carbon is filtered
from the oil it is usually very easy to distinguish between
the two. Carbon is usually very black or gray depending
on how much carbon is present. Sludge can form various
colors including tan, tannish-green, brown, beige, etc., but
not black. See Figures 1 and 2.
References
[1]. Doble Transformer Oil Purchase Specifications, publica-
tion of the Doble Oil Committee, Revised January 1,
2006, Watertown, MA 02472 USA
[2]. Oliver, F.S., “Doble Power-Factor Valued Oxidation
Test”, in the Minutes of the Twenty-Seventh Annual
Conference of Doble Clients, Sec. 10-404, Belmont,
MA USA
[3]. Myers, S.D., Kelly, J.J. and Parrish, R.H., A Guide to
Transformer Maintenance, Transformer Maintenance In-
stitute, Division of S.D. Myers, Tallmadge, OH USA
Lance Lewand received his Bachelor of Science degree from St.
Mary’s College of Maryland in 1980. He has been employed by the
Doble Engineering Company since 1992 and is currently the Laboratory
Manager for the Doble Materials Laboratory and Product Manager for
the DOMINO® product line. Prior to his present position at Doble, he
was Manager of the Transformer Fluid Test Laboratory and PCB and
Oil Services at MET Electrical Testing in Baltimore, MD. Mr. Lewand
is a member of ASTM Committee D 27.
Insulating Oils Handbook
Metals Analysis in Transformers,
Load Tap-Changers
and Oil Circuit Breakers
NETA World, Spring 2007
by Lance R. Lewand
Doble Engineering Company
Introduction
Laboratory testing is of increasing importance today as
new and better tests become available on a routine basis. This
is especially true in an environment where testing of out of
service apparatus is not possible under some conditions and
often performed less frequently. The testing, primarily per-
formed on easily-sampled electrical insulating liquids, can
help to assess the condition of the insulating materials and
more importantly serve as diagnostic procedures to detect
and identify incipient faults in apparatus. In addition, some
of these same tests can be used to verify that a failure has
occurred and then to aid in the identification of the cause.
Finally, some tests can be performed to help decide which
remedial actions might be most effective.
Metals testing in oil is not a new test but one that is being
used more frequently to diagnosis conditions in electrical
apparatus. One thing that the user should be very aware of,
however, is that there is not one test for metals in oil but
actually a variety of tests. The goal of this article is to define
the different tests and provide information on where and
when they should be used. The three tests that this article
will focus on are:
Dissolved metals-in-oil
Particulate metals-in-oil, and
SEM/EDX analysis
Analysis for metals in oil is most appropriate as a diag-
nostic or investigative tool when other symptoms indicate
an incipient-fault condition. The analysis can be performed
to identify and quantify either dissolved or particulate met-
als in the oil of which particulate metals are an order of
magnitude less. When looking for bearing wear in pumps,
a quantitative analysis for particulate copper, lead, zinc,
and iron is performed [1]. Since new oil in transformers
should not have significant quantities of any dissolved or
95
particulate metals, any appreciable increase in the oil could
be an indicator that a problem may exist. A baseline test
for dissolved metals in oil is important as the trend can be
more important than the absolute quantity in identifying the
location of an incipient fault. Usually, these tests are better
suited as investigative analyses to help aid in determining
the cause of certain incipient-fault conditions such as high
temperature overheating of the oil.
Dissolved Metals-In-Oil
This analysis is performed through the use of atomic
emission spectroscopy (AE) such as inductively-coupled
plasma (ICP). This type of metals analysis is the test most
frequently used for electrical apparatus oil as it is a quick
test to perform, can analyze for a wide variety of elements,
and can determine the concentrations of these elements
over a very large range. There are two major drawbacks. The
first one is that the dissolved metals or free particles in the
oil have to be less than about 10-20 microns in order to be
detected (depending on the analytical equipment itself ).
This is a function of the way the sample is introduced into
the analytical equipment for detection. The other draw-
back is that the lower level of detection is about 0.5 to 1.0
mg/kg (ppm) depending on the element and using older
instrumentation. Newer instrumentation may have detec-
tion limits as low as 0.01-0.10 ppm. This type of analysis
is satisfactory for certain incipient-fault conditions where
the metal is sublimed (change of a solid material into va-
por through heat) into the oil to form an organo-metallic
complex or where the particles are smaller than 10 microns.
This method, ASTM D 7151, is most commonly used to
determine if heating within the transformer has caused
metals to sublime and become entrained in or to the oil
molecules. It is also used in the analysis of LTC and OCB
oils where there is usually an overabundance of metals and
the idea is to find a quick and inexpensive way to determine
concentration of a wide variety of metals.
96
Follow-up testing for dissolved metals in oil is not usu-
ally necessary unless dissolved gas-in-oil tests indicate an
incipient-fault condition. The most common metals dis-
solved in the oil would be iron and copper or aluminum,
depending on coil construction. Lead is usually indicative of
brazes, solders, and paints, and zinc is indicative of brazes,
solders, and plated parts. Unusual metals such as titanium
in high concentrations could come from the degradation
of titanium based paints used by some manufacturers on
the interior walls of the transformer. Silicon contamina-
tion usually indicates either outside contamination (i.e.,
silicone fluids, caulks, or greases) or overheating of the iron
components such as the core metal or degradation of certain
components such as gaskets. Contamination with silicone
fluids can present a serious impediment during vacuum
processing because it can result in uncontrollable foaming
[2] of the insulating oil.
Particulate Metals-In-Oil (Bearing Wear)
For transformers cooled by pumps, the test for bearing
wear metals should be performed every few years. The most
common metals tested for are copper, lead, iron, zinc, alumi-
num, and silver. Some of these metals are produced in the
particulate state when the bearings of the transformer pump,
the pump shaft, and/or the impeller start to degrade in some
manner. The small pieces of metals are jettisoned into the
oil where, if small enough, they can remain suspended and
thus be sampled and analyzed. This type of analysis incor-
porates the use of a graphite furnace coupled with an atomic
absorption spectrometer. In addition, metal particulates are
concentrated by passing a certain volume of oil through a
0.45 micron filter to capture them. The filter is then treated
in a manner to leave only the metals in a solution of acid
and water. The analysis has the ability to detect elements at
extremely low levels, <1 µg/kg (ppb), which is necessary as
problems with pumps are usually detected where the con-
centrations of certain metals are around 20 ppb. Although
not an ASTM test method, Doble performs this analysis
on a very routine basis, and it has the capability of being
expanded to perform metals analysis on solid materials such
as paper, pressboard, and other materials.
Like ICP and AE, there are advantages and disadvan-
tages. For one, it uses a much larger volume of oil for analysis,
500 mL (0.5 quarts) whereas the ICP only requires 2-10
mL. There is also a significant amount of sample prepara-
tion involved, and the analysis time is lengthy. However,
there is really no limit on the metal particle size that can
be analyzed.
Sampling
The analysis of dissolved or particulate metals is very sen-
sitive to sampling methodology. The sample valve has to be
prepared in such a manner as to not introduce contaminant
metals or elements from that area into the sample. Depend-
ing on the type of valve, contamination from copper, tin,
and zinc (brass, bronze) as well as iron is not uncommon.
Zinc from such items as galvanized pipe nipples and drain
Insulating Oils Handbook
plugs is also a very common contaminant. In addition, copi-
ous amounts of oil should be flushed through the valve to
not only clean out the valve but to clean the area near the
valve inside the transformer. For transformers with pumps,
it is preferable that at least one bank of pumps be operat-
ing at the time of sampling so that the particulates are in
a homogenous solution inside the unit. The table below
illustrates the effect of cleaning and flushing..
Metal Results before and after
Cleaning the Drain Valve
Metal 1st Sample 2nd Sample, after cleaning and flushing
Copper 63 ppb 6 ppb
Lead 10 ppb 1 ppb
Iron 33 ppb 4 ppb
Zinc 6 ppb 14 ppb
As shown in the table, there is a significant decrease in
the concentration of most metals once the cleaning and
flushing were performed. The importance of the sampling
efforts cannot be overemphasized, as incorrect results can
lead to the wrong conclusion.
SEM/EDX Analysis
Scanning electron microscopy (SEM) and energy disper-
sive X-ray (EDX) analysis are two separate tests conducted
at the same time. SEM is an analysis in which a beam of
electrons, a few hundred angstroms in diameter, systemati-
cally sweeps over the specimen in an evacuated chamber.
The intensity of secondary electrons generated at the point
of impact on the specimen surface is measured, and the
resulting signal fed into a cathode-ray tube display which is
scanned in synchronism with the scanning of the specimen
to produce a picture. All prepared samples stubs are coated
with evaporated graphite before analysis which improves
imaging resolution and stability.
EDX is when an electron beam of the scanning electron
microscope enters the bulk of a sample producing an x-ray
emittance. The x-ray peak positions along the energy scale
identify the elements present in the sample and can provide
the percentage concentrations of each of these elements
thus providing an elemental breakdown of the material or
particles.
Obviously, this is no ordinary test. This powerful tool
can be used in investigative type analysis such as corro-
sive sulfur issues on paper or copper samples and other
contaminants from a transformer, LTC, OCB, bushing,
etc. This type of analysis is not performed on the oil itself
but on filtered materials from the oil or from components
or contamination inside the transformer that needs to be
identified. Although this analysis tool has many advantages,
like most analytical techniques there are disadvantages as
well. The foremost is cost of the instrumentation itself and
the significant operating expenses associated with it. Other
drawbacks include the amount of time necessary to prepare
the sample for analysis, the length of time for analysis and
Insulating Oils Handbook 97
the very high level of detection (0.5%, 5000 ppm) when
compared to other analysis techniques.

Conclusion
The three main types of metals analysis that can be
performed are dissolved metals-in-oil, particulate metals-
in-oil, and SEM/EDX analysis. The goal of the article is to
explain the differences between each test and to list certain
advantages and disadvantages to help the user in determin-
ing which test to request.

References:
1. Oommen, T.V. and Petrie, E.M. “Particle Contamina-
tion Levels in Oil-Filled Large Power Transformers”,
IEEE Transactions on Power Apparatus and Systems,
May 1983, Vol. PAS-102, No. 5, pp. 1459-1465.
2. Griffin, P.J. “A Cup Full Of Trouble”, Report on the
Minutes of the Doble Oil Committee Meeting, April
1989, pp. 8-14.

Lance Lewand received his Bachelor of Science degree from St.

Mary’s College of Maryland in 1980. He has been employed by the
Doble Engineering Company since 1992 and is currently the Laboratory
Manager for the Doble Materials Laboratory and Product Manager for
the DOMINO® product line. Prior to his present position at Doble, he
was Manager of the Transformer Fluid Test Laboratory and PCB and
Oil Services at MET Electrical Testing in Baltimore, MD. Mr. Lewand
is a member of ASTM Committee D 27.
98
Testing for Corrosive Sulfur Effects
NETA World, Summer 2007
by Lance R. Lewand
Doble Engineering Company
Over the past several years there have been a number
of failures of power transformers and reactors where there
are indicators of copper corrosion and formation of cop-
per sulfides on conductors and insulating materials. The
conductive copper sulfide reduces the dielectric breakdown
voltage of the solid insulation. This can result in a dielectric
puncture through the paper insulation. The failures have
occurred after the apparatus have been in service for a year
or more, often several years. At normal operating tempera-
tures the corrosion process appears to take time to form
critical amounts of conductive sulfides. To date the failures
have occurred without prior evidence of abnormal gassing
behavior; therefore, it appears that this problem is difficult
to detect and manage.
This problem is fairly wide known in the large power
utilities but less so with industrial and commercial users
with oil-filled electrical equipment, but the same issues
can arise. Failures due to corrosive sulfur attack have also
been documented in load tap-changers. One of the most
frequently asked questions is “Which apparatus is most
susceptible?” In response to this, Doble has developed the
following list:
1) Units that have the following characteristics
a) Manufactured from 1999 to present
b) Operate at high temperatures over long periods of
time such as generator step-up transformers and re-
actors and continuously fully-loaded transformers
c) Are either gas blanketed or have a sealed conserva-
tor system (a few failures have occurred with free-
breathing conservators)
d) Have oils that fail the corrosive sulfur tests (ASTM
D 1275B and CCD tests)
2) Those unit that do not have completely enameled or
varnished conductors
3) Corrosive sulfur problems have been known to occur in
units older than 1999, but to a much lesser extent. Test
the ones that are most critical to the system first. The
Insulating Oils Handbook
lowest voltage class transformer known to have failed
due to corrosive sulfur is 35 kV, but most failures have
been above distribution voltages.
In the past several years, in response to this problem, there
has been a great deal of research activity in developing new
tests to help identify oils with the potential to have corrosive
sulfur problems or transformers that have already developed
copper sulfide contamination. Some of the tests are already
ASTM tests and some are in various stages of development
and are broken down into two groups.
Oil Tests (discussed in this article)
Transformer Electrical Tests (discussed in a future ar-
ticle)
Oil Tests
These tests aid in determining if corrosive sulfur com-
pounds are present in the oil or identifying sulfur com-
pounds that are presently stable but may become corrosive
under conditions that exist in the transformer (temperature,
time, oxygen, etc.). It must be remembered that some of
these tests are accelerated aging tests and do not represent
a transformer environment. The purpose of the accelerated
aging test is to get a reasonable expectation of what may oc-
cur without testing the oil for years instead of a few days. A
listing of some of the tests used with descriptions follows:
ASTM D 1275B, Corrosive Sulfur in Oil
This test ages 220 mLs of oil in contact with a copper
strip in a sealed vessel for 48 hours at 150 C. The primary
purpose of the test is to determine if any corrosive sulfur
compounds in the oil will react with the copper strip to
turn it gray or black. It is a subjective test in that there is a
comparison of colors of the copper strip with some colored
standards and a table of descriptions listing what is corrosive
and noncorrosive (Figure 1A and 1B). The test was recently
modified from just D 1275 in April of 2006 to D 1275A
Insulating Oils Handbook
and D 1275B. The B method is more rigorous then the old
method D 1275 or D 1275A and was developed in response
to the problem with corrosive sulfur. The issue was that some
of these oils met the requirement of the oil specifications
using the old D 1275 (D 1275A) test yet became corrosive
while in service and eventually caused failures of transform-
ers, LTCs, and reactors. ASTM D 27 committee recognized
this flaw and modified the test method to include D 1275B.
This is a very routine test.
Figure 1A — D 1275B Setup (Copper Strip in Bottle)
ASTM D 5623, Sulfur Speciation
The ASTM D 5623 analysis is performed by gas chro-
matography with sulfur selective detection and covers the
detection of volatile sulfur-containing compounds. The test
method will not identify all individual sulfur components.
Detector response to sulfur is linear and essentially equi-
molar for all sulfur compounds; thus, both unidentified and
known individual compounds are determined. However,
many sulfur compounds, for example, hydrogen sulfide and
mercaptans, are reactive, and their concentration in samples
may change during sampling and analysis. This test can be
used to isolate specific sulfur compounds, some of which
may or may not be responsible for corrosive sulfur attack.
One of the issues is that depending on the lab, the database
of sulfur compounds that can be analyzed can be small or
large but usually not more than about 70 compounds. There
99
Figure 1B — Corrosive (Top) and Noncorrosive Strips (Bottom)
are thousands of sulfur compounds, and this test does not
have the ability to cover them all. Certain types of sulfur
compounds cannot be fully isolated and identified. It is
usually not considered a routine test and is mostly used for
research purposes.
ASTM D 4294, Total Sulfur in Oil
There are actually several ASTM tests that can be used
for the detection of total sulfur in oil. This just happens to be
the one that is used by our laboratory. This test determines
the total amount of sulfur in the oil but does not determine
whether or not the compounds being detected are corro-
sive. Transformer oil is made from a petroleum crude that
has naturally occurring sulfur. Depending on where in the
world the oil is from dictates the amount of sulfur in the
crude. The process of refining the crude to transformer oil
and other products attempts to remove reactive (corrosive)
sulfur compounds by converting them to hydrogen sulfide
gas which is easily removed from the process. The refin-
ing process can also convert some of the less stable sulfur
compounds into more stable unreactive sulfur compounds
that can provide benefits in the final refined product. The
amount of conversion and removal depends on the crude
and the refining process itself, as each refiner has its own
distinctive process. Most modern transformer oils have a
final sulfur content of less than 1500 ppm, and some are even
less than 10 ppm. This is a routine test, but it only provides
information on how much sulfur is in the oil, not whether
it is deleterious or not.
100
ASTM D 3227, Mercaptans in Oil
Mercaptans are sulfur compounds that can be very reac-
tive. This is a potentiometric titration test that is used to
determine the concentration of mercaptans as a class of
compounds in the oil, but it will not be able to identify
specific compounds. This test is used in the fuel and lubrica-
tion industries often and is starting to be used in the trans-
former oil industry. One of the problems with mercaptans
is that some of them are highly volatile and thus not easily
sampled for. It is presently not a routine test, but it is being
used more and more.
Doble Covered Conductor Deposition
(CCD) Test, Doble Test
There are two other variations, one developed by ABB and
the other by Siemens, which preceded the Doble method. A
variation of the Siemens method is being developed into an
IEC test. This new test is very important in that the purpose
of the test not only determines if the corrosive sulfur will
attack the copper and form copper sulfide, but also if cop-
per sulfide formations will develop in the paper insulation.
The test is conducted by taking an abraded copper rod and
wrapping new Kraft paper insulation around the rod. Two
Figure 2A — CCD Vial Setup
Insulating Oils Handbook
Figure 2B — Paper with No Deposits (Left),
Paper with Deposits (Right)
rods are prepared for each sample, and they are placed in
20 mL headspace vials. Oil is added and then the vials are
sealed. A stainless steel needle of a particular diameter is
then pierced through the septum of one vial and left there
to allow air ingress during aging (see Figure 2A and 2B).
The vials are then aged at 140 C for four days. At the end
of the aging cycle, the vials are removed and the copper rods
with paper retrieved. The paper is then removed from the
copper, and both are washed in a solvent to remove the oil.
The rod is inspected to determine if it has been tarnished
by corrosive sulfur, and the paper is inspected to determine
if deposition has occurred. The presence of deposition in
the paper is important as most of the recent transformer
failures have been due to copper sulfide deposits in the
paper causing a severe reduction in dielectric strength. The
resulting "deposition," whether heavy, moderate, or light,
means that the interaction of the copper and oil with the
paper insulation over the duration of the test resulted in
an obvious deposit of copper sulfide, copper, or oil/paper
degradation by-products onto or in the paper insulation.
This is becoming a very routine test.
Doble Covered Conductor Deposition
(CCD+DT), Doble Test.
A CCD test with dielectric breakdown voltage is per-
formed on the paper after aging. During the research con-
ducted for the development of the CCD test, it was noticed
that many oils, under the conditions of the test, actually
formed deposits on and in the paper. It was determined
that if a dielectric breakdown voltage test was performed on
the paper from the copper rod, it could be used to qualify
the deposits. Basically there are two categories of deposits:
those having no impact on the dielectric strength of the
paper and those that negatively affect the dielectric strength
of the paper. Obviously the deposits affecting the dielectric
strength by lowering it are of concern. This is also becoming
a very routine test.
Insulating Oils Handbook 101

Dibenzyl disulfide (DBDS in Oil),

Doble Test
DBDS is a sulfur compound found in certain transformer
oils. DBDS is not thermally stable at higher temperatures
and breaks down into benzyl mercaptan which is very
corrosive and attacks the copper quite quickly. It should
be remembered that DBDS is not the only compound to
cause severe copper sulfide deposition on the copper. There
are other compounds that cause corrosive sulfur attack
but as of yet have not been identified. This is now a very
routine test.

Passivator in Oil, Doble Test

Passivators are not part of the corrosive sulfur problem
but are used as a remediation technique to guard the copper
against corrosive sulfur attack. What passivators are and how
they work was described in a previous NETA World article
and will not be discussed here. With the addition of the
passivator, it then becomes important to be able to monitor
the concentration. Benzotriazole (BTA) and Irgamet® 39
along with similar type products can be detected.

Conclusion
Corrosive sulfur attack causing failure in oil-filled elec-
trical apparatus is occurring more frequently although the
number of failures are still a low percentage of the popula-
tion. However, the loss of revenues from these units as well
as capital equipment is significant. As a result, numerous
researchers are engaged in activities to determine and
eliminate the source(s) of the corrosive sulfur. Because of
this, many tests are being used in concert with each other
to determine the corrosive sulfur aspects of the oils used in
the electrical apparatus.
These tests will be used to screen out or identify oils that
may not be suitable.

Lance Lewand received his Bachelor of Science degree from St.

Mary’s College of Maryland in 1980. He has been employed by the
Doble Engineering Company since 1992 and is currently the Laboratory
Manager for the Doble Materials Laboratory and Product Manager for
the DOMINO® product line. Prior to his present position at Doble, he
was Manager of the Transformer Fluid Test Laboratory and PCB and
Oil Services at MET Electrical Testing in Baltimore, MD. Mr. Lewand
is a member of ASTM Committee D 27.
 NETA Accredited Companies
The following is a listing of all NETA Accredited Companies as of August 2011.
Please visit the NETA website at www.netaworld.org for the most current list.

A&F Electrical Testing., Inc...................................................................................Kevin Chilton

Advanced Testing Systems ............................................................................Patrick MacCarthy
American Electrical Testing Co., Inc. ......................................................................Scott Blizard
Apparatus Testing and Engineering ....................................................................... James Lawler
Applied Engineering Concepts .................................................................... Michel Castonguay
Burlington Electrical Testing Company, Inc. ........................................................... Walter Cleary
C.E. Testing, Inc. ........................................................................................... Mark Chapman
CE Power Solutions of Wisconsin, LLC............................................................. James VanHandel
DYMAX Holdings, Inc. ....................................................................................... Gene Philipp
Eastern High Voltage ....................................................................................... Joseph Wilson
ELECT, P.C. .................................................................................................Barry W. Tyndall
Electric Power Systems, Inc. .................................................................................. Steve Reed
Electrical and Electronic Controls ..................................................................... Michael Hughes
Electrical Energy Experts, Inc............................................................................... William Styer
Electrical Equipment Upgrading, Inc. .......................................................................Kevin Miller
Electrical Maintenance & Testing, Inc........................................................................ Brian Borst
Electrical Reliability Services ..................................................................................Lee Bigham
Electrical Testing, Inc. ................................................................................. Steve C. Dodd Sr.
Elemco Services, Inc. ...................................................................................... Robert J. White
Hampton Tedder Technical Services ....................................................................... Matt Tedder
Harford Electrical Testing Co., Inc. ................................................................... Vincent Biondino
High Energy Electrical Testing, Inc..................................................................... James P. Ratshin
High Voltage Maintenance Corp. ........................................................................... Eric Nation
HMT, Inc. .........................................................................................................John Pertgen
Industrial Electric Testing, Inc. ........................................................................ Gary Benzenberg
Industrial Electronics Group ................................................................................. Butch E. Teal
Industrial Tests, Inc. .............................................................................................. Greg Poole
Infra-Red Building and Power Service ............................................................ Thomas McDonald
M&L Power Systems, Inc. .................................................................................. Darshan Arora
Magna Electric Corporation ................................................................................... Kerry Heid
Magna IV Engineering – Edmonton ...................................................................Jereme Wentzell
Magna IV Engineering (BC), Ltd. ........................................................................ Cameron Hite

Setting the Standard

MET Electrical Testing, LLC .......................................................................... William McKenzie
National Field Services...................................................................................... Eric Beckman
Nationwide Electrical Testing, Inc. ...............................................................Shashikant B. Bagle
North Central Electric, Inc. ...............................................................................Robert Messina
Northern Electrical Testing, Inc. .......................................................................... Lyle Detterman
Orbis Engineering Field Service, Ltd. ....................................................................... Lorne Gara
Pacific Power Testing, Inc. ...................................................................................Steve Emmert
Phasor Engineering ........................................................................................... Rafael Castro
Potomac Testing, Inc. ........................................................................................... Ken Bassett
Power & Generation Testing, Inc.......................................................................... Mose Ramieh
Power Engineering Services, Inc. ..................................................................... Miles R. Engelke
POWER PLUS Engineering, Inc. ...................................................................Salvatore Mancuso
Power Products & Solutions, Inc. ........................................................................ Ralph Patterson
Power Services, LLC ........................................................................................ Gerald Bydash
Power Solutions Group, Ltd ...........................................................................Barry Willoughby
Power Systems Testing Co. ............................................................................... David Huffman
Power Test, Inc. ..............................................................................................Richard Walker
POWER Testing and Energization, Inc. ............................................................... Chris Zavadlov
Powertech Services, Inc. ................................................................................... Jean A. Brown
Precision Testing Group .................................................................................... Glenn Stuckey
PRIT Service, Inc. ........................................................................................ Roderic Hageman
Reuter & Hanney, Inc....................................................................................... Michael Reuter
REV Engineering, LTD ................................................................................ Roland Davidson IV
Scott Testing, Inc................................................................................................Russ Sorbello
Shermco Industries ............................................................................................... Ron Widup
Sigma Six Solutions, Inc. ....................................................................................... John White
Southern New England Electrical Testing, LLC ................................................. David Asplund, Sr.
Southwest Energy Systems, LLC .......................................................................Robert Sheppard
Taurus Power & Controls, Inc. ............................................................................... Rob Bulfinch
Three-C Electrical Co., Inc.................................................................................James Cialdea
Tidal Power Services, LLC ....................................................................................Monty Janak
Tony Demaria Electric, Inc. ............................................................................ Anthony Demaria
Trace Electrical Services & Testing, LLC ...................................................................Joseph Vasta
Utilities Instrumentation Service, Inc. ........................................................................Gary Walls
Utility Service Corporation.................................................................................. Alan Peterson
Western Electrical Services ......................................................................................Dan Hook

Setting the Standard

 About NETA
NETA (InterNational Electrical Testing Association) is an association of leading electrical testing companies;
visionaries, committed to advancing the industry’s standards for power system installation and maintenance
to ensure the highest level of reliability and safety.
NETA is an accredited standards developer for the American National Standards Institute (ANSI) and defines
the standards by which electrical equipment is deemed safe and reliable.
NETA is also the leading source of specifications, procedures, testing, and requirements, not only for
commissioning new equipment but for testing the reliability and performance of existing equipment.

QUALIFICATIONS OF THE TESTING ORGANIZATION

An independent overview is the only method of determining the long-term usage of electrical apparatus and
its suitability for the intended purpose. NETA Accredited Companies best support the interest of the owner, as
the objectivity and competency of the testing firm is as important as the competency of the individual technician.
NETA Accredited Companies are part of an independent, third-party electrical testing association dedicated to
setting world standards in electrical maintenance and acceptance testing.

Hiring a NETA Accredited Company assures the customer that:

• The NETA Technician has broad-based knowledge — this person is trained to inspect, test, maintain, and
calibrate all types of electrical equipment in all types of industries.
• NETA Technicians meet stringent educational and experience requirements in accordance with ANSI/NETA
Standard for Certification of Electrical Testing Technicians, (ANSI/NETA ETT).
• A registered Professional Engineer will review all engineering reports.
• All tests will be performed objectively, according to NETA specifications, using calibrated instruments traceable
to the National Institute of Science and Technology (NIST).
• The firm is a well-established, full-service electrical testing business.

CERTIFICATION
NETA Certified Technicians conduct the tests that ensure that electrical power equipment meets the ANSI/NETA
standards’ stringent specifications.
Certification of competency is particularly important in the electrical testing industry. Inherent in the
determination of the equipment’s serviceability is the prerequisite that individuals performing the tests be
capable of conducting the tests in a safe manner and with complete knowledge of the hazards involved. They
must also evaluate the test data and make an informed judgment on the continued serviceability, deterioration,
or nonserviceability of the specific equipment. NETA, a nationally-recognized certification agency, provides
recognition of four levels of competency within the electrical testing industry in accordance with ANSI/NETA
Standard for Certification of Electrical Testing Technicians, (ANSI/NETA ETT).

Setting the Standard