www.netaworld.

org Winter 2009-2010 NETA WORLD 1

Niche Market Testing
by Lynn Hamrick
ESCO Energy Services
Dissolved Gas Analysis
for Transformers
T
ransformer oil sample analysis is a useful, predictive, maintenance tool
for determining transformer health. Along with the oil sample quality
tests, performing a dissolved gas analysis (DGA) of the insulating oil is
useful in evaluating transformer health. Te breakdown of electrical insulating
materials and related components inside a transformer generates gases within the
transformer. Te identity of the gases being generated can be very useful infor-
mation in any preventive maintenance program. Tere are several techniques for
detecting those gases and DGA is recognized as the most informative method.
Tis method involves sampling the oil and testing the sample to measure the con-
centration of the dissolved gases. It is recommended that DGA of the transformer
oil be performed at least on an annual basis with results compared from year to
year. Te standards associated with sampling, testing, and analyzing the results
are ASTM D3613, ASTM D3612, and ANSI/IEEE C57.104, respectively. In
this article, I will discuss why DGA is a useful, predictive, maintenance tool and
review some methodologies for evaluating test results.
Te two principal causes of gas formation within an operating transformer
are electrical disturbances and thermal decomposition. All transformers generate
gases to some extent at normal operating temperatures. Insulating mineral oils
for transformers are mixtures of many diferent hydrocarbon molecules, and the
decomposition processes for these hydrocarbons in thermal or electrical faults
are complex. Te fundamental chemical reactions involve the breaking of carbon-
hydrogen and carbon-carbon bonds. During this process, active hydrogen atoms
and hydrocarbon fragments are formed. Tese fragments can combine with each
other to form gases: hydrogen (H
2
), methane (CH
4
), acetylene (C
2
H
2
), ethylene
(C
2
H
4
), and ethane (C
2
H
6
). Further, when cellulose insulation is involved, thermal
decomposition or electric faults produce methane (CH
4
), hydrogen (H
2
), carbon
monoxide (CO), and carbon dioxide (CO
2
). Te gases listed above are generally
referred to as key gases.
Te gases listed above are considered key gases and are generally considered
combustible (note that CO
2
is not a combustible gas). Te total of all combustible
gases may indicate the existence of any one or a combination of thermal, electri-
cal, or corona faults. Te rate at which
each of these key gases are produced
depends largely on the temperature
and directly on the volume of material
at that temperature. Because of the
volume efect, a large, heated volume
of insulation at moderate temperature
will produce the same quantity of gas
as a smaller volume at a higher tem-
perature. Terefore, the concentrations
of the individual dissolved gases found
in transformer insulating oil may be
used directly and trended to evaluate
the thermal history of the transformer
internals to suggest any past or poten-
tial faults within the transformer.
After samples have been taken and
analyzed, the frst step in evaluating
DGA results is to consider the con-
centration levels (in ppm) of each key
gas. It is recommended that values for
each of the key gases be trended over
time so that the rate-of-change of
the various gas concentrations can be
evaluated. Basically, any sharp increase
in key gas concentration is indicative
of a potential problem within the
transformer. Below is a table which
has been derived from ANSI/IEEE
C57.104 information. Te suggested
action levels for key gas concentrations
are also provided:
NETA WORLD Winter 2009-2010 www.netaworld.org 2
Where DGA results include a sharp increase in key gas concentration levels
and/or normal limits have been exceeded, it is suggested that an additional
sample and analysis be performed to confrm the previous evaluation and de-
termine if the key gas concentrations are increasing. As key gas concentration
levels approach the action levels, consideration should be given to taking the
transformer out of service for further testing and inspection.
Once key gas concentrations have exceeded normal limits, other analysis
techniques should be considered for determining the potential problem within
the transformer. Te techniques involve calculating key gas ratios and comparing
these ratios to suggested limits. Some of the most commonly used techniques
include the application of Doernenburg ratios, Rogers ratios, and Duvals Tri-
angle Model.
Doernenberg ratios and Rogers ratios are recognized in the ANSI/IEEE
C57.104 and are equivalent to the Basic Gas ratios in the International Elec-
trotechnical Commission (IEC) standards. Te evaluation method applied for
Doernenberg ratios and Rogers ratios utilizes the following gas ratios: CH
4
/
H
2
,C
2
H
2
/C
2
H
4
, C
2
H
2
/CH
4
, C
2
H
6
/C
2
H
2
and C
2
H
4
/C
2
H
6
. Te use of ratios
is warranted due to the varying rates of the combustible gas generation with
temperature and energy variations for diferent fault modes. Tey also allow
for diferent rates that the gases are dissolved into the mineral oil. Diagnosis
of faults is accomplished via a simple scheme based on ranges of the ratios. A
table of these ranges is provided below.
Duvals Triangle Model is recog-
nized in the IEC Guidelines. The
Duval Triangle Model (IEC 60599)
combined feld service evidence with
laboratory experiments published in
1989 followed by enhancements in
2002. Once a potential problem has
been determined using the key gas
concentration, one calculates the total
accumulated amount of three of the
key gases, methane (CH
4
), acetylene
(C
2
H
2
), and ethylene (C
2
H
4
), and di-
vides each gas by the total of the three
gases to fnd the percentage associated
with each gas. Tese values are then
plotted on the triangle below to arrive
at a diagnosis. Sections within the tri-
angle designate: thermal fault < 300 C;
thermal fault 300-700 C; thermal fault
> 700 C; low-energy discharge; high
energy discharge; and partial discharge.
Te ratio of CO
2
/CO is sometimes
used as an indicator of the thermal
decomposition of cellulose. Te rate of
generation of CO
2
typically runs 7 to
20 times higher than CO. Terefore,
it would be considered normal if the
CO
2
/CO ratio were above 7. If the
CO
2
/CO ratio is 5 or less, there is prob-
ably a problem. If cellulose degradation
is the problem, CO, H
2
, CH
4
, and C
2
H
6

will also be increasing significantly.
At this point, it is recommended that
additional furan testing be performed.
If the CO
2
/CO ratio is 3 or under
with increased furans, severe and rapid
deterioration of cellulose is occurring
and consideration should be given for
taking the transformer out of service
for further inspection.
When cellulose insulation decom-
poses due to overheating, chemicals, in
addition to CO and CO
2
, are released
and dissolved in the oil. Tese chemi-
cal compounds are known as furanic
compounds, or furans. In healthy trans-
formers, there are no detectable furans
in the oil (<100 ppb). As the cellulose
degrades, the furan levels will increase.
Furan levels of 500 to 1000 ppb are
indicative of accelerated cellulose aging,
with furan levels >1500 ppb having a
high risk of insulation failure.
3 www.netaworld.org Winter 2009-2010 NETA WORLD
In summary, the greatest indicator of potential problems within
transformers should not be limited to the concentration levels
of the key dissolved gases. By trending the dissolved gas levels,
problems can be identifed and evaluated further before they cause
a catastrophic failure of the transformer. If key gas concentrations
increase signifcantly from one sample to the next, one
should perform another sample to verify the results
of the previous sample. Once it has been confrmed
that the concentrations have increased, use trending
and the presented diagnostic techniques to determine
what the problem may be. Based on that diagnosis,
plan an outage to investigate further and, ultimately,
to resolve the problem.
As Operations Manager of ESCO Energy Services Company,
Lynn brings over 25 years of working knowledge in design,
permitting, construction, and startup of mechanical, electrical,
and instrumentation and controls projects as well as experience
in the operation and maintenance of facilities.
Lynn is a Professional Engineer, Certifed Energy Manager
and has a BS in Nuclear Engineering from the University of
Tennessee.