Sunteți pe pagina 1din 39

IEEE (57.~0~ 91 . . ~805702 050~351 014 . .

,
RORftized 0.1 aft IEEE Std C57.lM-1991
Ameriun Na.riona.! Standard (ANSI) <R-_nollttE 0i7.IlH-m8)

Interpretation
IEEE Guide for the
of Gases
Generated in Oil-Immersed
Transformers
Cin'uit~ and Device!>

Communicatiun!> Tf'f.'hnuluJ.~

Compulpr

Eleclrumagnelit.8 und
Radiation

IEEE Power Engineering Society

Sponsored by the
Transformers Committee

Industrial Applications

'. . .. .. 'Signo.ls and


'. . Applications

.... , . . SlanJards- .- , '.'


:,'-,'-.
;',: :!..,'"-., '. t.
"~'
Coordinating
Conunittff.s ~.
..
".

IEEE
PubJislHtd by th.lnstitut. of EllricM.r>d E ~ E n ~ It><:- 345 EMf ..7tJI SfrHt. N.... Y""k, NY IOOIZ tISA

July 22, 1992

11h~ !>y '.M IIr.iTlTurE OF HECTF1CA!.. s. ELEC~"OUCS ,.r.;t'iH"S (lEHl


"'p 01 22,3<>,~ I ......
IEEE (57. ),04 910 4B05702 0504352 TSD

ReooClIiud.1t lin IEEE


American National Standard (ANSI)

Std C57.1041991
(Rlln-ionllf
IEEE C57.104-1978l

IEEE Guide for the Interpretation of Gases


Generated in Oil-Immersed Transformers
. . -- ".,...,>..J_.U'"
","'f"'.1 1fT '\.:" \T
Sponsor
1 ,,"11'"
.}:! '.
......::.<..J-
~ ,!'~' .. ~ I_t:~
II
U..1;(
I;,..,.., K..:;slrokan _
Transformers
of theCommittee ll PT f>LN (Perscroj JCS.:l T...

IEEE Power Engineering Society

Approved June 27, 1991


IEEE Standards Board

Approved November 20, 1991


American National Standards Institute

Abstract: Detailed procedures for analyzing gas from gas spaces or gas-collecting devices as
well as gas dissolved in oil arc described. The procedures cover: (l) the calibration and use of
field instruments for detedingand estimating the amount ofoombustible gases present. in gas
blankets above oil. or in gas detector relays; (2) the use of fixed instruments for detecting and
derermining the quantity of combustible gases present in gas-blanketed equipment; (3)
obtaining samples of gas and oil from the transformer for laboratory analysis; (4) laboratory
methods for analyzing the gas blanket and the gases extracted from the oil; and (5) interpret-
ing the results in terms of transformer serviceability. The intent is to provide the operator
with positive and useful information concerning the serviceability of the equipment. An e;:ll.:ten-
sive bibliography on gM evolution, detection, and interpretation is included.

Keywords: gas analysis, oil, oil-filled transformers, transformer$.

The Inlllilule orEloctril:.I.nd Eloctronic. Enginoon, Int.


3~S Ealll ~71h Stl"Cet, New York, NY 100172394, USA

Copyrillht c: 1992 by the


InstiluteorEltliealllllcl Electronic. Engioe<>n. Inc.
All rightJ r-ese"""'d. PubLUh"d 1 ~
pr.nl"d in th" United State. orArne";".
ISBN I_S6937H7_9

NfJ p<>rl <J{ this publica/I"" moy ~ rtproduc~d in a..y {","m.


in on <lec/rank rttTi~lJ<j1 "Y""'''' ort>l.M.TWi....
wilhOl.fllh~ prior wrilt~n ~rmission fJ(lh~ pUb/i"h".

'I'";9M by the INSTITUTE OF ElECTRICAL S ELECTROU"; WG1NtERS (IEEE)


'Sep 01 22,U 04 199b
IEEE (57.104 91 . . 4!05702 0504353 997 . .

IEEE St:a.ndariU doeuments an developed within the 'Thchnical Com-


mittees of the IEEE Societies and the Standards Coordinating Commit-
tees of the IEEE Standards Board. Members of the committees serve
voluntarily and without compensation. They are not necessarily mem-
bers of the Institute. The standards developed within IEEE represent a
consensus of the broad expertise on the subject within the Institute as
well as those activities outside of IEEE that have expressed an interest
in participating in the development of the standard.
Vile of an IEEE Standard is wholly voluntary. The existence of an IEEE
Standard does not imply that there are no other ways ta produce, test,
measure, purchase, market, or provide other goods and services related
to the scope of the IEEE Slo.ndard. Furthennore, the viewpoint
expressed at the time a standard is approved and issued is subject to
change brought about through developments in the slate of the art and
comments received (rom users ofthe st.andard. Every IEEE Standard is
subjected to review at least every five years (or revision or reaffirma-
tion. When a document is more than five years old and has not been
reaffirmed, it is reasonable to conclude that its eontent&, although still
o( some value, do not wholly reflect the present state of the art. Users
are cautioned to check to determine that they have the latest edition o(
any IEEE Standard.
Comments (or revision o( IEEE Standards are welcome (rom any inter-
ested party, regBrdless of membership affiliation with IEEE. Sugges-
tions (or changes in documents should be in the form of a proposed
change of text, together with appropriate supporting etlmments.
Interpretations: Occasionally questions may arise regarding the mean-
ing of portionll of standards as they relate to specific applications. When
the need for interpretations is brought to the attention of IEEE, the
Institute will initiate action to prepl1re appropriate responses. Since
IEEE Standards represent .II consensus of all a1neemed interests, it is
important to ensure that any interpretation hu also received the con-
currence o( a balance of interests. For this reason IEEE and the mem-
bers o( its technical committees are not able to provide an instant
response to interpretation requests except in those cases where the mat-
ter has previously received (onnal consideration.
Comments on standards and requesta (or interpretations should be
addressed to:

Secretary, IEEE Slo.ndards Board


445 Hoes Lane
P.O. Box 1331
Piscataway, NJ 08855-1331
USA

IEEE Standards documents are adopted by the Institute o( Electriea


nd Electronics Enl;ineers without reJ;8rd to ..... hether !beir adoption rna
nvolve patents on articles, materials, or processes. Such adoption doe
ot assume any liability to any patent owner, nor does it assume any obli
mtion whatever to parties adopting the standards documents.

' \ bytJ>~ INSTITUTE OF Cl[CTRICo',L I. [L[CT~ICS ENGTIl[E~S (IEEEI


01 22,Jto,04 19'1to
IEEE (57.10~ 9~ .. ~a05702 050~354 823 . .

Foreword
(Thh fOJ'Cword i. Dot a put of IEEE Std C57.104.1991. IEEE Guide fDr Ille tllte",...,htioD of Cue. Genorated in
Qillrnn>el"lll;td 1'l'&.ll>Ifon:nen.l

At the time that this standard was completed, the Transformers Committee had the follow
ing officers:

d. D. Bont., Chair J. H. Harlow, VICe Chair


W. B. Binder, &cretary

At the time that this 6tandud was completed, the Insulating Fluids Subcommittee had the
following members:

B. A Pearce, Chair F. W. Heinrichs, Secretary

D. J. AJJan F. J. Grynkie....u:z R.J. Mull1


H.A:t;~ian T. J. Haupert W. Mutechler, Jr.
D. B.....""w.Jd F. W. Heinrich. E.J.Norton
J. C. DryaD.t P. J. Hoefler T.Orbeck
G. Bryant C.R.Hoe...! C. T. RaymoI'd
J. CorkraD. B. G.HuDter A.D.lWchult.e
D. w.c",n. D. L.JOh"lOOD G.J. Reitter
D.H.Dougl ... J. J. Kelly T.O.Rou.....
M.Fi~r.:ld J. P. Kio ....y L.J.Savio
R. M.}..,y J. G. Lackey G. J. Schreude...
M.Ftydma.n R.t. Lowe D. W. Sulldio.
P. Gel"lai. G.O.McR_ J. A. Thomp....
J. P. Gibeault M.M.McCee T. P.1....ub
D.A.Cilliu K. Mcl.hDllmOll R.A. Veitch
J. Goudie C.K. Miller R. M. ViDb'nt
}'. M. Gragg R. E. Millk.. itz L. Wagenaar

At the time that it balloted and approved this standard for submission to the IEEE Stan
dards Board. the Transformers Committee had the ollowing members:

E.J.Ado1ph,oD K. R. Hightoll M.MltelmJIn


L. C.A>cher P. J. Hodler H. R.Moore
D.J.AUan C.H""""I R.J.Muli1
B.AUen R. H. Holli.te-r W. H. MutllChlcr
R. AUustilU'ti C. C.Honey E. T. Norton
S.AltmM E.Howella R. A. 01'1100
J. C.Amold C. Hurty B.K. Patcl
J. Aubin Y.P.tijima W.'. Pattenon
R. Ba...,ron G. W.lIi1f H.A. PUI'C1!
D. Barnard R.G. J" .... b... n D. Pc.....
D. L. Buol D. L. JOlu:t.lIOlI L. W. PlCI"Ct!I
P. L. Bellalchi D. C. Johnlon J. M. Pollitt
S.BellDDn A.J.Jonnatti C. P. Raymond
W. B. Binder C. P. Kappeler C.A Robbio.
J. V. BonuceM R. B. Kaufman L. J. Savio
.1. n. Bo..,.t J. J. Kelly W. E. Suon
C .V. BnlWD W. N. Kennedy D. N. Sharma
O. R. Compton J. P. Kinney V.Shenoy
F. W.Cook B. Kl.pon,ki W. W. Skin
J. L. Corku.n A. D. Kline L, R. SI.ewoI&J1d
D. W.Cnlfb E.KoeDlg E.G.Str... g ...
.J_N.O"v;" J. G. Lackey D. S\lndin
D.H.Dougl .... R. E. Loo L. A. SwenllOn
J. C. Dt.>ttoD H. F. Light D. S. T.kach
J. K. E8IIle)' S. R. Lindgren A. M. Tepliuky
J.A.Ebert L. W. Lonl(" V. TheDappaD
D. J. Fallon L. A. Lowdermilk R. C. Tho.....
S. L. Foster R. J. Lc>we J. A. Thomr-o
M.Frydmsn M.L.MOl1luing T. P. Traub

>rtt>. ltlSTIlUTE OF ELECTRICAL.!. ELECIRONICS E~GWE(RS I IEEE I


22,31>.04 !'l'lt>
IEEE (51.104 91 . . 4~05102 0504355 1bT . .

n.E.Ga~1 B. D. MItIOIiI D. E. Truax


R. E_ Ceo.m.r. T. M,a-wda W.B.Uhl
D. W.~rl..:h .I. W. Matlb<twi R. E. Uptevdf, Jr
D.A.Gim- .I. McGill G. H. VaiU.-..rt
R. s. CiJori. C. J. McYille" R.A.. Veitd.
R. L. Grubb W. J.Mc:NlIu L. B. Waae--r
F. .I. cryukie...-ic& S. P.M~bt. ll:. J. Whtarty
C.RIlI C. K. Mllklr A. L. Wilb
.I. H. Hario.... C.H.MJlli... W. E. Wrt'DJI
F. W. HeillJidl. R E. Minkwil& A. C. WII rdal:l<
w. n.lI~nnl.D.g E.J.Yuud.

The Accredited Standar<u Committee on Transfonnera, Regulatora, and Reacton, C57, that.
reviewed and approved thia document, had the following members at the time of approval:

Leo J. Savio, Chair John A. Gauthier, Secntary

Electric Ualn ."dPo..... Graul' _._P . .__... .__ ..P. E. Ol1lhek


S. M. Ii.. RizYi
F.s...v.u
J. SltUi...."
J. C. T1w>a>pI<)
M. C. MiIIioI. WI.)

l t;l"t. of Electrical and lor:trotUo;:a Eng;ooort .J. D. Bont


J.Davil
.I. H. Hulow
hS.'"
H. D. Smith
R.A. V~;l.ch

Nati<loal Elec:t,-w,al Manufattu",nAaBociatio" ..____....._....._ _....H. ._........__ G. D. Collila'


P. I>.w",",
.I. D. Douilu
A.A. Char""ri...
K.R.U".koy
R. L. PI.... ter
H.RDbI"
R. E. Upt.egratr. Jr.
F. J. Hopkiraon Wr.)
J. NayWt.)

Tennea!lOO Valley "'lIt!>ority . .. H F.A. Lewll.

U"dctwritel"'l Laboratoria, Inc .. .................................................................. _ W. T. O'Grldy

US Dtpal't....e"t oru...I.owrior. Bllr...u ofRecl.o.matioo F. W. Cook, Sr.

US DeP""-,,,,"llt orthe N...,., Civil EqiDHring eor,.. .__.._... . H. P. StieJUe,.

" t by t!'>. INSTITUTE Of ElECTllIt:Al. I ElECT_ICS NGINERS 1[[([1


~ 01 20',3<>,001 19%
IEEE (57.104 91 _ 4t105702 0504356 6ib

When the IEEE Standards BOl\Td approved this standJlTd on June 27, 1991, it had the fol-
lowing membeTShip:

Marco W. Migliaro, Chair Donald C. Loughry, Vice CluJir


Andrew G. Salem, Secreta.ry

Deunt. Bod.toll Tbomn L. Hllnnan John E. M.y, Jr.


Pllul L. BorriU Donald N.lIeinnl../l L.WTenc:e V. MoC.U
Clyde Camp KeOllet.h D. Helldrix T. DOllM\(h.el'
J."",. M. Daly John W.Hon:h Still' L, Nilsson
Donllid C. Flcckenlltcin Ben C.Johnaon John L. R.nkine
J.y FOTlter Ivor N. Knight Ronald a. Reimtor
Dllvid F. Franklln JOlleph Kooplillger Cll!')' S. Robln&Oll
Ingrid Fromm Irving Kolodny TeITI../lOO R. Whilt<!""""
Mich...,l A, t.wkr

Member Emeritus

Also included were the following nonvoting IEEE Standards Board liaisons:

FeruandoAldllna
Satiolh K. AllKarwal
J."",.BeIlU
Riehud B. Engelman
Sl.llnley W.....h.w

Mary LyonI' Niel..,..


tltEE SIDnd=-<h Ihp<>rlmc"l Project Edilcr

by I"C INSTITUTE 0, ELECI~ICAl 3 ElECTRONICS ENGINEERS {IEEE)


1 22,S<> 04 lqq"
IEEE C57.10Q 9\ . . 4605702 0504357 532 . .

Contents
SECI10N PAGE
1. Introduction 9
1.1 Scope 9
1.2 Limitations 9
1.3 References 10

2. General Theory 10
2.1 Cellulosic Decomposition 11
2.2 Oil Decomposition IL
2.3 Application to Equipment lL
2.4 Establishing Baseline Data 12
2.5 Recol:flition ofa Gassing Problem-Establishing Operating Priorities 12

3. Interpretation of Gas Analysis 13


3.1 'I'hennal Faults 13
3.2 Electrical Faults--Low Intensity Discharges 13
3.3 Electrical Fault.s--Iligb Intensity Al'eing 13

4. S~gcst.ed Operating Procedure. Utilizing the Detection and Anal)'sil of


Combustible Gases 13
4.1 Det.e-rmining Combustible Gas Generating Ratel 14
4.2 Determining the Gas Space and Dissolved Gasin-Oil Equivalents 15
4.3 Monitoring Insulation Deterioration Using Di5Solved Gas Volume 16
4.4 Evaluation ofTransfonner Condition Using Individual and TeG
Concentrations 17
4.4.1 Determining the Transformer Condition and Operating Procedure
Utilizing TeG in the Gas Space 19
4.4.2 Determining the Operating Procedure and Sampling Intervall"rom
the TOCG Levels and Generating Rates in the Oil 19
4.5 Evaluation of Poasible Fault 'I)-pe by the Key au Method 20
-4.6 Evalualion of Possible Faull 'tYpe by Analysis of the Separate Combustible
Gases Generated 22
4.6.1 Evaluation of Possible Fault Type by the Doernenburg Ratio Method 22
4.6.2 Evaluation ofPouible Fault Type by the RlJiers Ratio Method 23

5. Inslruments for Deted.ing and Determining lheAmounl of Combustible Gases


Presenl 24
5.1 Portable Instruments 24
5.2 Fixed Instrumenls 25
5.2.1 tlelhod 1 25
5.2.2 Melhod 2 26
5.2.3 tfelhod 3 26

6. Procedures fOT Obtaining Samples of Gas and Oil From the Ttansfonner for
Laboratory Analysis 27
6.1 Gas Samples for Laboratory Analysis 27
6.2 Gas Dissolved in Oil 27

7. Laboratory Methods fOT Analyzing the Gss Blankel and the Gases Extracted From
the Oil 27
7.1 Detennination ofTolal Dissolved Gas 27
7.2 Detennination of Individual Dissolved Gases 27

by the INSHTUTE CIF ElCIRICAl & ELECfRaHCS EHGlIlEERS IIEEE)


I 22.J/io.Q-t 1991>
IEEE C57.104 91 . . 4605702 0504355 479 . .

7.3 Determination of lndividual Gases Present in the Gas Blanket 27

8. Bibliography 28
8.1 Sources 28
8.2 Gas Evolution 28
8.3 Detection and Interpretation.......................................... . 30
8.3.1 Gas Detect.<Jr Relay 30
8.3.2 Gas Cushion in Sealed Transformers 31
8.3.3 Dissolved Gas in Transformer Oil 34

FIGURES

Fig 1 Halstead's Thermal Equilibrium Partial Pressures as a Function of


Temperature 12
Fig 2 Operating Procedure F10w Chart 15
Fig 3 Key Gases Evaluation 21
Fig" Doemenburg Ratio Method F10w Chart 24
Fig 5 Rogers Ratio Method Flow Chart 25

TABLES

Table 1 Dissolved Gas Concentrations 18


Table 2 Actions Based on TCG 19
Table 3 Actions Based on TDCG 20
Table" Concentrations afDissolved Gas 23
Table 5 Ratios for Key Gases-Doernenburg 23
Table 6 Rogers Ratios far Key Gases 26

"19h~ by tho INSIITUIE OF ELECTRICAl & ElECIRON,CS ENGINEHiS IIEEE)


Sop 01 22,31>,04 1'196
IEEE C57.~04 91 . . 4805702 0504359 305 . .

IEEE Guide for the Interpretation of Gases


Generated in Oil-Immersed Transformers

1. Introduction

The detection of certain gases generated in an oil-filled transformer in servi.ce is frequently


the first available indicati~n of a malfunction that may eventually lead to failure if not cor-
rec:ted. Arcing, corona disc:harge, low-energy sparking, severe overloading, pump motor faH-
ure, and overheating in the insulation system are some of the possible mechanisms. These
c:onditions oc:curring singly, or as several simultaneous events, can result in decomposition of
the insulating materials and the formation of various combustible and noncombustible gases.
Normal operation will also result in the formation of some gases. In fact, it is possible for some
transformers to operate throughout their useful life with substantial quantities of combusti-
ble gllses prescnt. Operating a transformer with large quantities of c:ombustible gas present is
not a normal occurrence but it does happen, usually after some degree of investigation and an
evaluation of the possible risk.
In a transformer, generated gases can be found dissolved in the insulating oil, in the gas
blanket above the oil, or in gas c:olleding devices. The detection of an abnormal c:ondition
requires an evaluation of the amount of generated gas present and the c:ontinuing rate ofgen-
eration. Some indication of the soun::e of the gases and the kind of insulation involved may be
gained by determining the composition of the generated gases.

1.1 Scope. This guide applies to minernl-oil-immersed transformers and addresses:

(1) The theory of combustible gas generation in a transformer


(2) The interpretation of gas analysis
(3) Suggested operating procedures
(4) Various diagnostic techniques, such as key gases, Dornenberg ratios, and Rogen ratios
(5) Instruments for detecting and determining the amount of combustible gases present
(6) A bibliography of related literature

1.2 LiDlit.ations. Many techniques for the detection and the measurement of gases have been
established. However, it must be recognil:ed that analysis of these gases and interpretation of
their significance is at this time not a science, but an art subject to vwiability. Their presence
and quantity are dependent on equipment variables such as type, location, and temperature
of the fault; solubility and degree of saturation of various gases in oil; the type of oil preserva-
tion system; the type and rate of oil circ:ulation; the kinds ofrnaterial in contact with the fault;
and finally, variables associated with the s.ampling and measuring procedures themselves.
Because of the variability of acceptable gas limits and the significance of various gases and
generation rates. a consensus is difficult to obtain. The principal obstacle in the development
of fault interpretation 8S an exact scienc:e is the lack of positive cOTTelation ofthe faultidenti
fying gases ",';th faults found in actual transformers.
The result of various ASTM testing round robins indicates that the analytical procedures
for gus analysis are difficult, have poor precision, and can be wildly inaccurate, especially
between laboratories. A replicate analysis confirming a diagnosis should be made before tak
ing any major action.

. bX t.toe INSTITUTE Of ElECTRICAL to ELECTRONlCS EN<lINEFIlS (IEEE)


)! 22,36,04 1996
IEEE (57.104 91" 4605702 0504360 027 . .

U:EI:
Std C67.1041991 IEEE GUIDE FOR THE INrERPRETATION OF GASES

This guide is, in general, an advisory document. It provides guidance on specific methods
and procedlue!> to Assist the tranf;former operator in deciding on the status and continued
operation of a transformer that exhibits combustible gas formation. However, operators must
be cautioned that, although the physical reasClnS for gas formation have.ll. finn technical basis,
interpretation of that data in tenns of the specific cause or causes is not an exact science, but
is the result of empirical evidence from whieh rules for interpretution have been derived.
Hence, u:aet causes or conditions within transfonners may not be inferred from the various
procedures. The continued application of the rules and limits in this guide. accompanied by
actual confirmation of the causes of gas fonnation, will result in continued refinement and
improvement in the correlation of the rules and limits for interpretation.
Individual experience with this guide will assist the operators in determining the best pro-
cedure, or combination of procedures, for each specific case. Some of the factors involved in the
decision of the operator are: the type of oil preservation system, the type and frequency ofthe
sampling program, and the analytical facilities available. However. whether used separately
or as Ci:lmplements to one another, the procedures disclosed in this guide all provide the oper-
ator with positive and useful information concerning the serviceability of the equipment.

1.3 Referencel. The following references lIhould be used in conjunction with this guide:

[1] ASTM D2945.90. Test Methods for Gas Content of Insulating Oils.)

[2] ASTM D3305-84 (Reaff. 89), Method for Sampling Gas (rom a Transformer.

[3] ASTM D361290, Test Methods for Analysis of Gases Dissolved in Electrical Insulating Oil
by Gas Chromatography.

[4] ASTM D361387, Methods for Sampling Electrical Insulating Oils (or Gas Analysis and
Determination of Water Content.

1.4 Definitions. The follov.ing definitions of terms are used in this guide:

key galel. Gases generated in oil-filled transfonners that can be used for qualitative deter-
mination of fault types, based on which gases are typieal or predominant at various
temperatures.

partial discharce. An electric discharge that only partially bridges the insulation between
conductors, and that mayor may not occur adjacent to the conductor.

TCG. Total combustible gas.

TOCG. Total dissolved Ci:lmbustible gas.

2. General Theory
The two principal causes of gas formation within an operating transformer are thermal and
electrical disturbances. Conductor losses due to loading produce gases from thermal decompo-
sition of the associated oil and solid insulation. Gases 8re also produced from the decomposi-
tion of oil and insulation exposed to arc temperatures. ~nerally, where decomposition gases
are formed principally by ionic bombardment, there is little or no heat associated with low
energy dischar&:es and corona.

'ASTM publi<:.lious ....veilebl.. from the eu.tom.... Service Department, American Socie~y for Telting Iud MetAl
riel", 1916 R.. <:r. St",et. Philedelphl f'A 19103, USA.

10

"19M by ,ho INSTITUT OF ElECTRICAl 8. ELECTRONICS ENGINEERS (IEEE)


Sop 01 22,3b,O<l 199"
IEEE (57.104 91" 4805702 OS043b1 Tb3 . .

IEEE
GENERATED m OIL-IMMERSED TRANSFORMERS Std C57.104-1991

2_1 Cellulosic Decomposition. The thermal decomposition of oil-impregnated cellulose


insulation produces carbon oxides (CO, COiJ and some hydrogen or methane (Hz, CH 4) due to
the oil (C0 2 is not a combustible gas). The rate at which they are produced depends exponen-
tially on the temperature and directly on the volwne of material at that temperature. Because
of the volume effect, a large. heated volume of insulation at moderate temperature will pro-
duce the same quantity of gas as a smaller volume at a higher temperature.

2.2 Oil Decoxnpo.tition. Mineral transformer oils are mixtures of many different hydrocar-
bon molecules, and the decompositicm processes for these hydrocarbons in thermal or electri-
cal faults are complex. The fundamental steps are the breaking of carbon-hydrogen and
carboncarbon bonds. Active hydrogen atoms and hydrocarbon fragments are formed. These
free radicals can combine with each other to form gases, molecular hydrogen, methane,
ethane, etc., or can recombine to form new, condensable molecules. Further decomposition and
rearrangement processes lead to the formation of products such as ethylene and acetylene
and, in the extreme, to modestly hydrogenated carbon in particulate form. These processes
are dependent on the presence of individual hydrocarbons, on the distribution of energy and
temperature in the neighborhood of the fault, and on the time during which the oil is ther-
mally or electrically strcsscd. These reactions occur stoichiometrically; therefore, the specific
degradations of the transformer oil hydrocarbon ensembles and the fault conditions cannot be
predicted reliably from chemical kinetic considerations. An alternative approach is to assume
that all hydrocarbons in the oil are decomposed into the same products and that each product
is in equilibrium with all the others. Thermodynamic models permit calculation ofthe partial
pressure of each gaseous product as a function oftemper8.lure, using known cquilibrium con-
stants for the relevant decomposition reactions. An example of the results of this approach is
shown in Fig 1 due to Halstead. The quantity of hydrogen formed is relatively high and insen-
sitive to temperature; formation of acetylene becomes appreciable only at temperatures near-
ing 1000 C. Fonnation of methane, ethane. Bnd ethylene each also have unique dependences
on temperature in the model. The thermodynamic approach has limits; it must assume an
idealized but nonexistent isothermal equilibrium in the region of a fault, and there is no pro-
vision for dealing with multiple faults in a transformer. However, the concentrations of the
individual gases actually found in a transformer enn be ul>ed directly or in ratios to estimate
the thermal history of the oil in the transformer from a model and to adduce any past or
potential faults on the unit. As the simplest example: the presence of acetylene suggests a
high temperature fault, perhaps an arc, has occurred in the oil in a transformer; the presence
of methane suggests th at-if a fault has occurred-it is a lower energy electrical or thermal
fault. Much work has been done to correlate predictions from thermodynamic models with
actual behavior of transformers.

2.3 Application to Equipment. All transformers generate gases to some extent at normal
operating temperaturu. But occasionally a gas-generating abnormality does occur within nn
operating transformer such as a local or genenl overheating, dielectric problems, or n combi-
nation of these. In electrical equipment, these abnormalities are called ~faults." Thermal,
corana, and areing faults are described in 3.1, 3.2, and 3.3. Internal faults in oil produce the
gaseous byproducts hydrogen (H2), methane (CH 4 ), acetylene (CZH2), ethylene (C ZH4 ), and
ethane (CzH s ). 'When cellulose is involved, the faults produce methane (CH 4), hydrogen (Hz),
carbon monoxide (CO), and carbon dioxide (C0 2), Each of these types offaults produce certain
gases that nre generally combustible. The total of all combustible gases may indicate the
existence of anyone, or a combination, of thermal. electrical, or corona faults. Certain combi-
nations of each of the separate gases detennined by chromatography arc unique for different
fault temperatures. Also, the mtios of certain key gases have been found to suggest fault
types. Interpretation by the individual gases can become difficult when there is more than one
fault, or when one type of fault progresses to another type, such as an electrical problem
developing from a thermal one.

11

< b <~" INSTllUTE OF ELECHHCAL & ELECTRONICS fUGJNEERS ((EEEl


01 Z2,3~,0<l Iq9b
IEEE (57.104 91 . . ~~05702 0504362 9TT . .

IEEE
Sld C57.104-1991 IEEE GUIDE FOR THE INTERPRETATION OF GASES

.3

.,

. , C,H

.,

,m 'C 1225 ~ 725 "C 22$C


TEMPERATURE

Fir 1
Halstead's Thermal Equilibrium
Partial Pressures as a Function of Temperature

Attempts to assign greater significance to gas measurements than justified by the natural
variability of the generating and measuring events themselves will lead to gross errors in
interpretation. However, in spite of these limitations, these gas-generating mechanisms are
the only existing basis for the analytical rules and procedures developed in this guide. In fact,
it is known that some transformers continue to operate for many yean in spite of above aver-
age rotes of gas generation.

2.4 Establishing Baseline Data. Establishing a reference point for gas concentration in new
or repaired trll.nsformers and following this with a routine monitoring program is a key ele-
ment in the application of thig guide. Monitoring the health <scrviceobility) of a transformer
must be done on a routine basis and can start anytime, not just for new units.
Generally, daily Dr weekly sampling is recommended nf'ter start-up, followed by monthly or
longer intervals. Routine sampling intervals may vary depending on application ond individ-
ual system requirements. For example, some utilities sample generator step-up (GSU) trans-
formers four to six times a year, units rated over 138 kV twice a year, and some 765 kV units
are sampled monthly.

2.5 Recognition of a Gassini' Problem-Establishini" Operatinr Priorities. Much


information has been acquired over the past 20 years on diagnosing incipient fault conditions
in transformer systems. This information is of a general nature but is often applied to veT)'

12

'19M by t~~ WSTIIUTE OF ELECTRICAl & ELECT~ONICS EU(iHl(ERS II(EE)


\.to 01 22,31>,04 1991>
IEEE
GENERATED IN OILIMMERSED TRANSFORMERS Std cti7 J 041991

specific problems or situations. One consistent finding with all schemes for interpreting gas
analysis is that the more information available concerning the history of the transformer and
test data. the greater the probability for a correct diagnosis of the health of the unit.
A number of simple schemes employing principal gases or programs using ratios of key
gases have been employed for providing a tentative diagnosis when previous information is
unavailable or indicated no fault condition existed. Principal gas or ratio methods require
detectable or minimum levels of gases to be presents or norms to be exceeded, before they can
provide a useful diagnosis.

3. Interpreta tion of Gas Analysis


3.1 Thermal Faults. Referring to Fig 1, the decomposition of mineral oil from 150C to
500 C produces relatively large quantities of the low molecular weight gases, such as hydro-
gen (H 2) and methane (CH-l), and trace quantities of the higher molecular weight {:8ses ethyl-
ene (C 2H 4) and ethane (C 21"4). As the fault temperature in mineral oil increases to modest
temperatures, the hydrogen concentration exceeds that of methane, hut now the tempera-
Lures ure accompanied by significant quantities of higher molecular weight gases, first ethane
and then ethylene. At the upper end of the thermal fault range, increasing quantities of
hydrogen und ethylene and traces of acetylene (C 2H 2) may be produced. In contrast with the
thermal decomposition of oil. the thermal decomposition of cellulose and other solid insulation
produces carbon monoxide (CO), carbon dioxide (C0 2), and watervapOl' at temperatures much
lower than that for decomposition of oil and at raus exponentially proportional to the temper-
ature. Because the paper begins to degrade at lower temperatures than the oil, its gaseous
byproductll 8re found at nonnal operating temperatures in the transformer. A GSU trans
former, for eltamplc, that operates at or near nameplate rating will normally ganeratl' several
hundred ports per million (ppm) of CO and several thousand parts per million of CO 2 without
excessive hot spots.
The ratio ofC0:IC0 is sometimes used as an indicator of the thermal decomposition of cel-
lulose. This ratio is normally more than seven. For the COiCO ratio, the respective values of
CO 2 and CO sbould exceed 5000 ppm and 500 ppm in order to improve the certainty (octor,
i.e., ratios are sensitive to minimum values. As the magnitude of CO increases, the ratio of
COliCO decreases. This may indicate an uLnQrmality that is degrading cellulosic insulation.

3.2 Electrical Faults-Low Intensity Discharges. Referring to Fig 1, low intensity dis
charges such as partial discharges and very low level intermittent arcing produce mainly
hydrogen, with decreasing quantities Q( methane and trace quantities of acetylene. As the
intensity of the discharge increases, the acetylene and ethylene coneentrations rise signifi.
cantly (see Tnble 6).

3.3 Electrical Faults-High Intensity Arcing. Referring to Fig 1, as the intensity of the
electrical discharge reaches arcing or contjnuing discharge proportions that produce tempera-
tures from 700 C W 1800 C, the quantity ofocetylene becomes pronounced.

4. Suggested Operating Procedures Utilizing the Detection and


Analysis of Combustible Gases
FrQm an op.erational point of view, it is important to establish the following priQrities;

(1) Detection. Detect the generation of any gases that exceed ~nonnal" quantities and uti~
lile appropriate guidelines so the possible abnormality may be recogniled at the carli
est possible time in order to minimize damage or avoid a failure.

13

,by the lNSTlTUl( OF HEtTI/leAL & ElECTRCl'IlCS ENGH~EE~S (IEEE I


l\22,J6,O~ 1996
IEEE (57.104 9~ . . 4605702 0504364 772 . .

IEEE
Std C57.104_1991 IEEE CUIDE FOR THE INTERPRETATION OF CASF.8

(2) Evaluation. Evaluate the impact of an obnonnality on the serviceability of the trans
former. using II l>ct of guidelines or recommendations.
(3) Action. Take the recommended action, beginning with increased surveillance and con-
firming or supplementary analysis and leading to either a detennination of load sensi
tivity. reducing the load on the transformer, or actually removing the unit from service.

The success of fault gas analysis necessitates the earliest possible detection of gases using
the following methods:

Direct measurement of the amount of combustible gas in the gas space or relay ['Ibtnl
Combustible Gas (TCO)-See 5.2.1 and 5.2.2].

Direct measurement of the amount of combustible gas dissolved in the oil (gas-in-oil
monitor~ee 5.2).

Chromatographic separation and analysis for the individual components in a gas mix-
ture extracted from a sample of the transformer oil or a sample of the transfonner gas
space (See Section 7).

An opernting procedure utilizing the gas data from the above sources is to be developed
immediately following the initial detection of combustible gases. Fig 2 is a flow chart that
traces the suggested process from the initial detection of combustible gas to the final assess
ment of the status of the transformer.

4.1 Determmine Combustible Gas GeneratiDe Rates. A given gas volume and distribu-
tion may be generated over a long time perioo. by a relatively insignificant fault or in 11 very
short time period by a more severe fault. Hence, one meaiurement does not indicate the rate
of generation and may indicate very little about the severity of the fault. Once a suspicious
gas presence is detected, it is important to be certain whether the fault that generated the gas
is active.
An evolution rate greater than (0.1) ft3 ofCQmbustible gas per day may indicate the unit has
an aelive internal fault. To calculate the rate of evolution, take the sum of the concentrations
(in ppm) of all the combustible (everything but CO 2, O 2 Nv
in the first and second samples
and use Eq 1.

(ST-SO) xVxlO-(l
R=
7.5 x 7'

(Eq 1)

where:

R ::: Rate (fl.3/day)


So = First sample (ppm)
ST = Second sample (ppm)
V = Tank oil volume (gallons)
T = Time (days)

Limits for average gas generation rates are given for gas space analysis (TCG) in -4.4.1, and
for total dissolved gas analysis rrnCGl in 4.4.2.

14

'l~M by the IN$T!1UTE OF ELECTRICAL & ELECH~ONICS ENGINEERS (IEEE)


~p 01 22,36,04 199t>
IEEE (57.104 9~ .. ~805702 050~365 609 . .

IEEE
GENERATED IN OIL-IMMERSED TRANSFORMERS Std C67.104.1991

EXAMPlES
GAS DETECTED IN
CONSERVATOR GAS SPACE
RelAY, GAS SPACE.
OR OOl STEP ,: GAS DETECTED IN GAS DETECTED IN
OIL GAS SPAce

STEP 2: DATA (PPM): Ht. TOTAL GAS. 1.S-,r.


cor"'ARE VALUES 270. cu.
w 1110, CO.
WITH TABLE 1 280. CA .31. (7H.
wI1.CIf.
TOTAL DISSOlVED
COM8USTI8LE GAS
TABLE 1 INDICATES TABLE 1 IHOICATES {TOCG).1liIS
CONOmON I: OOHOITIOfIlS 2, 3,.:
NORMAL PROBLEM MAY EXIST STEP 3: TABLE 1 INDICATES PAOCEEDTO
CONDrTlON 2 STEP.

STEP .: RESAMPLe (SEE RESAMPLE (SEE


RESUME NORMAl RESAMPLE: TO FIND '.1) INDICATES A 1) INDICATES A
SURVEILLANCE GENERATING RATE: RATE OF 20 PPM! RATE OF .025'l\J
REFER TO . I DAY AND INCREAS- OAYAND
'"0 INCREASING

I STEP 5: TABLE:) TABLE 2

GAS SPACE OR INDICATES CONDITION 2, INTERVAL C


DISSOLVED IN OIL: AND PROCEDURE 3. AO..,..,SE MANUFAC-
RELAY SAMPLE: GO TO TA8t.E 3
GO TO TABlE 2 TURER; EXTREME CAUTION: PlAN OUT-
AGE: RESAMPLE PER INTERVAl;
ANALYlE GAS SPACE AND DISSOLVED
T I GAS COMPONENTS [SEE NOTE (I)J

INVESTIGATE POSSIBLE FAULT TYPE USING STEP 15: 5 KEY GAS: H;l> CH.&-ELECTRlCAL.
METHODS DESCRIBED IN . 5.1 .&.1. ot 6.2. REO- COR""'-
Qt.1MENOED INrTIAl RESAA4PUNG INTERVAL AND .6.1 OOERNENBURG (seE Nore (')J
OPERATING PROCEDURE. FAULT TYPE: POSSlBLEARCING
.6.2 ROGERS: FAlA.TTYPE: CASE 2
I POSSiBlE ARCING.
ADJUST SAMPLING INTERVAL AND OPERATING
PROCEDURE BASED ON ACCUMUlATeD DATA
AND EXPERIENCE

NOTES: (1) A.ume equal diuol..od mmponeata in both Q.l.lDplea.


(2) Ad.ual c... "'u in.pec:l.cd when CaRl ,.,..,hed.O ppm. FouDcI arcin, belw...n inaul.ted NLTC ,ball.
pin aod <:wpl.iD,fof drive mncba..MIln..

Fie 2
Opcrntine Procedure Flow Chart

4.2 Dctcnnininr the Gal" Space aDd Di.uolved Gas-inOil Equivalents. Gas space and
oil equivalents are used to compare the results of analysis of the gas space (TeG) with Tesults
fTl,lm analysis of the gases dissolved in the oil (TDCG). Comparisons of gas Talios obtained
fTom the gns space can then be compared to similar Tatios of gases extraeted fTom the oil. It
should be nOl.e<! that the calculated equivalent values ofTeG, and experimentally measuTed
vslues of TCG pTobably do not show close agreement., since the equation fOT obtainin&, the
equivslents assumes the existence of equilibrium between the gas blanket and the oil. This
condition may not exist., particularly in the case of an actively progressing fault. However, the
equation is valuable for the determination of a limiting value for the expected total ~mbusti

15

_ by VI. INST I1UT[ OF ElEClIlleAl & E"EcTRCI-lICS ENGlIlEERS I [EEE)


II 2:2,36'(),l l'''l>
IEEE C57.10~ 91 . . ~605702 050~3bb 5~5 ..

IEEE
Std C51.lGfo-I991 IEEE GUIDE FOR THE n-.'"I"ERPRETATION OF CASES

ble gas concent.ration in the gas blanket. The dissolved gas equivalentoflhe TCG. is obtained
using the following equation:

(Eq 2)

where:

rcGe '" An estimate of the percent ofcomoostible gas in the gas space
C :: Combustible gas
G .. Each gas dissolved in oil (combustible and nonc;ombu.stible)
Fe :: The concentration expressed in parts per million (ppm) of combustible gas g
dissolved in oil
Be :: The Ostwald .olubitity coefficient of combust.ible gasg
F. :: The coneetration ofa particular gas dissolved in oil
Bit ,. The Ostwald solubility coefficient of particular EllS

OILw&ld Coefficient (8)


G.. (26Cl

R,' 0.0429

0,

co,
'""
..,00
""'"
c,H,'
OJ"

",
N, 0.0145

CO' 0.102

c,R,'
C~.o
.."
0.331

CDmlnlltiblel
Note: Owr.",.ld .....ffici""tl an for LII. oil .... ith delllily
of 0.880 ""mpt'r.u,11"e of 26 "'C, .lId p .........,.. or 1
IlmoIph"'....

4.3 Mouitorin, losulation DeteriomtioD VsinK Dissolved Ga. Volume. ODe acceptable
method for monitoring the deterioration of transfonner insulating material involves calculat-
ing tha tolal volume of gas evolved. The total volume of evolved gas is an indicator of the mag-
nitude of incipient faults. Succeeding samples indicate dlange' with time as the fault{s)
develop. Trends are readily apparent. when gas volume is plotted versus time.
1b determine the volume. in gallons, offault gas dissolved in insuiliting oil, use Eq 3.

16

lh. by u... mSlITUf( OF' EUCf"tCAl .. UCTI?OHICS (t'GINE1<S U(((I


, Of 22,)6,001 1 _
IEEE (57.104 9 1 . 4805702 0504367 l i B .

IEEE
CENERATED IN OIL-IMMERSED TnA.~SFOnMEnS Std C67.104.1991

FG(V)
1,000.000
<&I 3)
where:

FG '" Sum of H2. Cf-4. CtH,. C 2H 4, G,:H2' and CO (ppm)


V '" Volume of oil in transformer (gallons)
TCG" '" 1Ot81 disfOOlved combustible eas (gallons)

This straightforward method is useful for completely oilfilled (etlnservator-type) transform-


ers with eanditions that produce r;mall quantities of fault cas. These conditions warrant eon
tinued monitoring but ),ave not yet developed a distinct character aecording to the other
methods of fnult delemlinHtion described in this guide. This fnulL-gns volume method eontin
ues to be useful as fault conditions enlarge, with the added advantage that it permits continu-
ous monitoring of insulation deterioration in spite of any oil handling activity that includes
degassification.

4.4 Evahw.tion of Transformer Condition Using Individual and TDCG Concentra


tions. It enll be difficult to determine whether a trallsformer is behaving normally if it has no
previous dissolved gas h istoT)'. AIS(), considerable difTerem:es of opinion exist for what is con
sidered a "normal transformer" wilh acceptable concentrations of gases.
A four-level criterion has been developed to classify risks to transformers. when there is no
previous dissolved gas histoT)'. for continued operation at various combustible gas levels. The
criterion use' both concentration, for separate gases and the total concentration of all com-
bustible gases. See Table 1.

Condition 1 TDCG below this level indicates the transformer ill operating satisfactorily
(see Fig 2). Any individual combustible gas eJt:ceeding specified levels should
prompt additional investigation (see 4.5 Bnd 4.6).

Condition 2 TDca within this range indicates greater than normal combustible gas level.
Any individU!l1 combustible gas uceeding specified levels should prompt addi-
tional investigation. Proceed per l-lg 2, Step 3. Action should be taken to
establish a trend (Fig 2. SteP 4). FaulUs) may be present. Proceed to 4.4.1 or
4.4.2.

Condition 3 TDCG within this range indicates II high level of decompo6ition. Any individ-
ual combustible gas eJt:ceeding specified levels should prompt addition31 inves-
tigation. Pnceed per Fig 2, Step 3. Immediate action should be L'lken w
establish a trend (Fig 2, Step 4). Fault(s) ure prob3hly present. Proceed to
4.4.lor4.<l.2.

Condition 4 TDCG within this runge indicatcs cJt:cessive decomposition. Continued opera-
tion could result in failure of the transformer. Proceed immediately and with
caution per Fig 2, Step 3, and 1.4.1 or 1.4.2.

Table 1 lists the dissolved gas concentrations (or the individual gases lmd TDCG for Condi-
tions 1 through 4. This Loble is used to make thc original assessment ofa i:assing condition on
II new or recently repaired transformer or is used if there are no previous tests on the trans
former for dissolved gases or if there is no recent history. Users of this guide are advised that
the dissolved gas concentrations contained in Table 1 are consenSuS values based on the expe-
riences of mnny companies. The transformer operator may detide to use different dissolved

17

I t"~ WSllIVIE 0# HECII>IC"'L l ELECTROIlICS Ef.IGWEERS (IEEE)


22,30,04 l"~~
IEEE (57,104 9 1 " 4605702 OS043b8 318 . .

rEEE
Std C57.lC)4l\H/l IEEE GUIDE FOR THE INTERPRETATION OF GASES

gas concentrations for the individual gases (particularly acetylene) and TDCG ba~ed on enl,ol-
neering judgment. and experience with other similar t.ransformers.

Tnblc 1
Dissolved Gas Concentrations

Diuolved Key Gu ConO>entnotwll limit. (ppw)


St"tUI
II, ell, C,H, C214 C:t'I i co co, TDCGO.

no
Cnndition 1
'" 120
" '0
" 350
'''''
Cnndition 2 101_700 121-400 36-60 fil-IOO 66-100 351--li70 ,,""- ,,,.
721_
.000

Condition 3 101- 401_ ,,-"" 1 OJ -200 101-150 571_ 4001_ 192.1-


1<00
'800 '000 '0000
"'"
Conditinn 4 >1800 >1000
""
,200 "'0 >1400 "oo ,"'"
NOTES, (1) Tlbl" 1 ..Ium.... thlt liD previOUI ....It. on the tnon.form..e foe diloolve<l pi anal)"I;' hive """n
WIld... DC th"t n" receot hiltOl'Y exiotl. If" p.....mUI Inalyeis nilto, it Ihcluld be reviewed to dc.... rm.in.. if til"
.itu"li"n i t.bl.. oe unot.bl... Ror~r to Table. 2 and 3 for IpprcprilU, ldion(o) tb be tlkon.
(2) An ASTM round robiD Wdiclt.ed vlnlbility in g... ",,"Iy.i. between I,bl. Thil .hould be <XIlllideM wh... n
Illving ga. analy"i, WIlde by dirr"",nt lib"
o Th... numbere .h<lwn in Tablet .re in parte org.. pee tnillion parte o( oil (ppm) volum..tricl.!ly and ..no b....<t
on .. llrge powoe tr..,,"former Wilh ....""ml tboulllnd go,lIo,," of oU. Wilh 1 ImAller on volume, th" um.. volume
or gal wiU giv" highee 1l.A.:>.....e"t....tion. SIJl,IU dillributio" tran.rormere .,,01 vollage regul ..tore rn.y oe>D-
tai" combu8tibl.. gu.... bccflU8C ortlle operotion or internal expul.ion U .... I oe load brelk .witch..... The .tatUI
<Dd.,. in T..bl" 1 are .110 not .pplic.ble to other ,pPlU"ltlll ira which 101<1 brn.lr. Iwitch.... openou, u"der oll.
.. The TDCG VAlue dOClI "ot wclud.. CO~. which;1 DOL ... combu.tibl"gll.

The condition for a particular transfonneT is detennined by finding the highest level faT
individual gases or the TDCG in Table 1. For e:umple, if a sample contained the following gas
concentration!J (in ppm, voVvo1):

the gases that. fall into the highest. condition are H2, CH 4CzHz, CzHa, and TDCG. Therefore,
this data would indicate that the transfonner would be classified as Condition 2. This exam-
ple cnn also be used to show two other facwrs that should be considered when using this table,
that is, the age of the transformer and the type of incipient condition.
New transformers (a year or less) usually contain levels of gases that would fall well belo.....
Condition 1 and do not contain delectable levels of acetylene. Theeefore, the dcg-ree of concern
in the example would be much higher for a one-month-old transformer than a 20-year-old
transformer.
Another consideration is that acetylene may be genemled from thr~ different incipient
fault conditions, i.e., high temperature overheating of oil, partial discharge (Jow energy dis-
charge), oe arcing. In the case of overheating, acetylene will represent a small proportion of
the hydrocnrbon gnses. In the case of partial discharge, very high concentrations of hydrogen
will be generated relative to acetylene, and this would generally be a cause for concern even
though the TOea is not abnonnal1y high. The most severe condition is arcing. When high-
energy arcing OCCUT5, hydrogen and acetylene are generally of the same magnitude, liS ure the
hydrocarbon gases. When an active arcing condit"ion is found, immediate attention is
required.

18

9ht to; 'r,~ IIlS1ITU1E riO EllCTR\CAl.<: ELEC1FOtlltS ETfGWE'EF~ 11HE)


001 22,36,04 lQQ6
IEEE C57.10Q 91 . . ~S0570C 050~3b9 254 . .

GEf'ERATED rN OIL-IMMERSD 'J'Jl.o\N'SFORMERS

......1 DctcrmiDiDi' the Tr'8.ulformer Condition and Operatini' Procedure Utiliziog


TCG in the Gal Space. When sudden inuea5es in the combustible gas concentrations or
generating rates in the gas space of successfully operating transformers occur and an internal
fault is suspected, use the proceduTt recommended in Fig 2.
TlI.ble 2 indicates the recommended initial $ampling intervals and operating procedures for
various levels ofTCG (in percent).
Once the source of gassing is determined by analysis. inspection. consultation, or combina.
tions thereof and the risk hilS been assessed, then enb';neering judgment should be applied to
determine the final sampling intcrvllJ Ilnd operating procedure.

Table 2
Actions Based on TCe

So.m;>li.., Interv..l.. &lid O~..ti"'ll rr-du.-


for e... CeDeTatioln R.. tft

......
TCO

'"I
TCO
Row
(....do,.)
SampliAe
Inl.er...J ot-".. tilllll: Prooeduttell

Conditi.... > >.03 D..lly ConMder re......al r""n Mmoe.


Advi_ m....urarturer.
.0:hOl Da.iI,.
<.ot Weeki,. :oercise atrcrDe uuoo...
A"ol~ for individual gflleo.
Pian outo~.
Advt.e rnanuracturer.
Condition 3 <5 to,. .. 2 ,..03 Weekly E:oerdu C<tn!mol! cautio...
Analy... for individual I."'.
.0hOt Weekly PIon o"IIlK".
Adviae ...... nufa.rt'''fl.
<.01 Monthl,.
<210,. ...5
Coedit..... 2 :>.03

.03...0t
1.1lInlhly

Monlh.ly
Ellc!"cise eo.utl....
Analyze r"r Individual
Do:urmi_ load. dot~n<kDCI:.
,0_.
"'~'"
<.01
Condl!Jofo I d >.03 Monthly E~rriae cautic>a.
.....al,.u ror ind'vidual,.-.
Det.rmio- load .peD......

.03-.01 Quarterl)' Continuo nonnal D~rotll>C1.

<.01 Annual

E:romple: A transfonner hI'S a TCa level of 0 . % and is generating gas at a constant rate of
0.035% TCG per day. The table indicates Condition 2. It should be sampled monthly, and the
operator should follow Procedure 2 in Table 2.

".4.2 Detcrminini' the OpcmtiDC' Procedure and Samplinr Interval From the
TDCG Leveb and Generalinr Rates in the Oil. When sudden increases in the dissolved
gas conLent of the oil in successfully optratine- transformers occur and an internal fault is sus-
pected. the procedures N:commended in Fig 2 should be used. Table 3 indicates the recom
mended initial aampling intervals and operating procedures for vnrious levels of TDCe (in
ppm). An increasing gas generation rate indicates a problem of increasing severity; therefore,
II shorter urnpling interval is recommended.

19

. ~, ~n. H1S1'lTVTE ~ 1C'~Il;Al,. J E~ECTs:or<lCS Efo,]t1npS I~EEI


>l "l,~I>,04 1""'0
IEEE (57.104 9~ . . 4605702 0504370 T76 . .

U:EE
Std CIi7.104-1991 IEEE CUIDE FOR TIlE INTERPRETAnON OF' GASES

Once the source of of gassing is determined by analysis, inspection, consultation, or combi-


nations thereof and the risk, has been assessed, then eng'ineeringjudb,rment should be applied
to cietennine the final sampling interval and operating procedure.

Table g
Actions Based on meG

Sampling low""al. lod Operlting Pr<:>e(!dl,lru


{or Gn ~n"ratiol'l RAtei

....0.
TOCC
(ppm)
TDCG
Ra\.e
(ppm/doy)
Sampling
In\.olrval Operating Pl'OC'ldl,lru

Conditioo 4 >4630 >30 Dllily Conlider """",vIII (mill ..,rvi.,.,.


Adv ..... mlI,lf""tu....r.
' ....0 Daily

dO Wcckj) Eu...,u" ntnom<' c.l,ltioll.


Anll.1yze for individualgs..".
PliO outage.
Advig., manufacturer.

Condition 3 1921-4630 >30 Weekly E"",,";oe en.."ne caution.


AnalYle fot individual gnci.
' ....0 Weekly Plan outage.
Ad~ mlllluracturer.
dO Monthly

Condition 2 721-1920 >30 Monthly I::""",i." ","ution.


Anal)'le for individual ga.., .
' ....0 Monthly Ik\.ermine load dependence.

dO Qul\l1etly

Condition I ~ 720 >30 :Mo.llthly Eu...,i.., caution.


Anlllyze fot individualguc.
o..\.erminl load dotpeOdcnc1l.

llhlO QufU't.<!rly Continu.. nonnll ol"'ratim.

dO Ao.nu.1

Er.ample: If 8. tTlillsformer has a TDCG level of 1300 ppm and generates gas at a constant
rate below 10 ppm per day, it should be sampled quarterly, llnd the operator should follow Pro-
cedure 2. If the rale increases to 30 ppm per day but remains constant, the operator should
now sample monthly.

4.5 Evaluation of Possible Fault Type by the Key Gas Method. The preceding discus-
sion of the dependence on temperature of the types of oil and cellulose decomposition gases
(2.1 and 2.2) provides the basis for the qualitative determination of fault types from the gases
that are typical. or predominant, at various temperatures. These significant gases and propor-
tions nre called Mkey gases." Fig 3 indicates thesp- "key gll.ses~ and relative proportions for the
four seneral fault types.

20

it,. by t"~ IIJ~T1TU1 OF ElECf"-lCAl & ELEC1~OIJICS WGH'EE~S IIEEE)


'01 ,,",3D 04 100 ..
IEEE (57.104 9~ . . 4805702 OS0437~ 902 . .

Itt.
CENERATED IN OlL-IM:MRSED TRANSFORMERS Sl.d C67.104.1991

I. Thermal==-Oil: Deeomposi- _ '00


tion prodUClS include ethylene
and methane, together with Q 70
! '"
3
. Overheated Oil

smaller quaniries of hydrogen 6C


"
and ethane. Traces of acetylene f5
so .,
may be fanned if the fault is ~
servcn~ or involves electrical ~ 20 " , "
contracts. ::5 ", C,H,
Principal Gas - Ethylene li! H,

Gas

E .'"'"
2. Thermal--eellulose: Large l 100 I 92
Ovcrhealed Cellulose
quantities of carbon dioxide 0'1

and carbon monoxide are


evolved from overheated cellu- 2i 6C

lose. Hydrocarbon gases. such &; .,so


as methane and ethylene. will [
be fonned if the fault involves ~ "
20
an oilimpregnated structure. d ", L-'-Rr----;:,----=-c:o:--,,:o--r=--
~

Principal Gas - Carbon Mon- II: co c,ft.


I-I t CH. C,H. C~2

oxide Gas
3. Electrical----CQ[Qoa: Low- !: '00
energy electrical discharges

ane, with small quanities of


III
produce hydrogen and rneth- Q 70
~
~
.'" Corona in Oil

ethane and ethylene. Campara- ~ ."'",


ble amounts of carbon monox- Q.
ide and dioxide may result ~ 20 "
from discharges in cellulose. d ," "
Princip.al Ga.... _ Hydrogen a: co H,

4. EleClI;cal Arcing: Large t. '00


amounts of hydrogen and acet- til OJ Arcing in Oil
ylene are prodUCed, with minor o 70 '"
quantities of rneth.ane and eth ... so 60

ylcoe. Carbon dioxide and car- ~


so .,
bon monoxide may also be Q.

fanned if the fault involves eel ~ 20 "


lulosc. Oil may be carbonized. w :s ," s , 3
Principal Gas _ Acetylene II: co CH.

FirS
Key Gasu Evaluation

21

~" tt~ II'SITTUIE Of ElECT~IC"'L 3. ElECTPf't'.:S E'~I1'E'-:; 'IEEE}


I ~,3l>, o.l 1''''''
1[[[ (57.104 91" 4805702 0504372 849 . .

IEEE
Sld C57.1041991 IEEE GUIDE FOR TIrE INTI::RPRETATfON OF GASES

4.6 Evaluation of P09sible Fault 'IYpe by Analysis of the Separate Combustible


Gascs Gcncrated. The usc of gas ratios to indicate a single possible fault type is an empiri-
ttll process based upon the experience of each individual investigator in correlating the gas
analyses of many units with the fault type subsequently assigned as the cause for disturbance
or failure when the unit was examined. This process was attributed to Doernenburg and sub-
sequently confirmed by Rogers on European systems, from which the bulk of the diagnostic
correlation is obtained. US investigators have applied the European rules to units on US sys-
tems with varying degrees of success; however, a US data base ofeomparable si"l:e to !.he Euro-
pean reports does not exist.
The diagnostic theories based upon the thermal degradation principles cie~cribed in 2.1 and
2.2 employ an aITay of ratios of certain key combustible gases all the fault t.rpe indiC8tors.
These five ratios are:

Ratio 1 (Rl) = CHiHz


Ratio 2 (R2) = C2HiC 2H.
Ratio 3 (RJ) = CzHiCH.
Ratio 4 (R4) = C 2Hs!C 2H z
Ratio 5 (R5) = C ZHJC 2H c

The nrst ratio method (Doernenburg; see 4.6.1) utilizes Ratios I, 2, 3, and 4. This procedure
reQuirC8 significant levels of the gases to be present in order for the diagnosis to be valid.
The second method (Rogers; see 4.6.2) utilizes Ratios 1, 2, and 5. The ROl:"crs method does
not depend on speeific gas concentrations to exist in the transformer for the diagnosis to be
VAlid. However, it. suggests that the method be used only when the nonnallimits of the indio
vidual gases have been Cltceeded.

4.6.1 Evaluation of Possible Fault 'IYpe by the Docrnenbure- RRtio Method. The
DoenlP.uburg method suggests the existence of three general fault types as discussed in Sec-
tions 2 and 3. The method utilizes gas concentrations from which Ratios 1, 2, 3, and 4 are Cll,l
culat.ed. The step-by-step procedure (Row chort) is shown in Fig 4.
The values for these gases are first compared to special concentrations- I.I-Table 4 {see
Steps 2, 3, and 4 in Fig 4}--to oscertain whether there really is 8 problem with the unit and
then whether there is sufficient. generation of each gas for the ratio analysis to be applicable.
Then the ratios in the order Ratio I, Ratio 2, Ratio 3, o.nd &ltio 4 arc compared to limiting
values, providing II suggested fault diagnosis as given in Table 5. This table gives the limiting
values for rntios of gases dissolved in the oil and gases obtained from the transformer gas
space or gus relay.
The flow chart in Fig 4 illustrates the step.by-step application of the Docmenburg ratio
method for gases extracted from t}IC transformer oil only. Exactly the same procedure is fol-
lowed for gases obtained from the gas space or gas relays, except the limiting values for the
ratios will be those appropriate for gas space (Table 5).
Descriptions of the steps indicated in Fig 4:

Step 1 Gas concentration~ ore obtained by extrading the gases and separating them
by chromatograph (see Section 7).

Step 2 If at least one of the gas concentrations (in ppm) fOT H2, CH 4, Cj!Hz, llnd ~H4
exceeds twice the values for limit Ll (see Tnble 4) and one of the other three
gases exceeds the values for limitLl, the unit is considered faulty; proceed to
Step 3to determine validity of the ratio procedure.

Step 3 Delennining validity of ratio procedure: If at. least. one of the gases in each
ratio Rl, R2, ro, or R4 exceeds limit Ll, the ratio procedure is valid; other-

22
,

'1"" ty "H' 1I,STITl'IE Of FLEC;TF1CAl I:. HE~lPClncs EN()lIjEEI!~ ,IEEEI


... 01 ~~,31>,O~ I09~
IEEE (57.104 91" 4&05702 0504373 785 . .

lEEE
GENERATED IN OIL-IMMERSED TRANSFORMERS Std C67.1Q.41991

Table 4
Concentration or Dissolved Gas

K..,.Cu eo.......' ...tio ... L1


(ppm)

Hydro~.. (~)
""
Mctk ....e (CII.) 120

Carbon Monoxide (CO) 350

A.wtyle,..{C~H.,)

Etkylcne (C~l)
"
60

Eth..... (~~
"
"'ThC'K Vlil...... dilTcr rJ'llftl Douncnbu"ll.....d mindde'
with Cocditloa 1 or TabJ,c 1 .

wise. the ratios are not significant. and the unit should be resampled and
invcstigutcd by alternate proc:cduru.

Step 4 Assuming that the ratio analysis is valid, each successive ratio is compared to
the values obUlined from Table 5 in the order RI. R2, R3, and R4..

Step 5 If all succeeding ratios for a specific fault type fall within the values given in
Table 5, the suggested diagnosis is ....alid.

'Thble 5
Ratios fOT Key Gase.-Doernenburr

Ratio 1 (Rl) Ratio '2 (R2) Ratio 3 (ll3) Ratio 4 (R4)


CllJH, c,H,GH. c,HICH, c,H.'C,H,

EJrtrllCWd FJ'Ilm Extracl.ed Fnxo Extratt.ed From Ext..aetHF'rom


Oil Gu Space Oil Gu S,...,., Oil Gu Space Oil a.. Spaa!

1-Tbc1"Ol.&1 Decom >1.0 >0.1. <0.76 <1.' <0.3 <0.1 >0.4 >0.'2
poaitioo

2-CaJ'llna (lnw <0.1 <0.01 "'0.3 ",0.1 >0.4 >0.2


Inlenll(ty PO)

:i-Art-inl> (Hi~h >0.1 >0.01 >0.715 >1.0 >0.1 ",0.4 <0.'2


lnh.... ilyl-'J.) d.O <0.1

4.6.2 Evaluation of Pouible Pault Type by the RogeN Ratio Method. The Rogers
ratio method follows the same general pro<:edure as the Doemenburg method, except ani)'
three ratios (RI, R2, and RS) are used. This method, shown in the step-by-step flow chart (Fie
5), is also based on the thermal degradation principles described in 2.1 and 2.2. Tbe validity of
this method is based on correlation of the results of 8 much larger number offailure in....estign.
tions with the cas analysis for CIIch ease. But, as with we Doernenburg method, the Rol:ers
ratios can give ratios that do not fit into the diagnostic codes; therefure, other annlytieal meth
ods given in 4.4 and 4.5 should be considered, as well as other options oullined in Fig 2.

23

by v ... IUS'!TUI[ Of' [U"IiICAt- I; EUCTQ~:J.CS [UGltl[l"'S 1[[['


..
z,~ .. ,o. ,,,,,'"
IEEE CS7.l04 91" 4805702 0504374 611 . .

IEEE
Sld C67.1O.f-1991 IEEE CUrnE FOR THE INTERPRETATION OF CASJ,:S

PORN
(CORONAl

F'AULT NOT
IDENTIF'IABLE:
RESAM;>LE

DISCHARGE
ARCING
FAULT NOT
IDENTIFIABLE;
RESAMPLE
THEAIJAL
FAULT

Fig 4
Doernenbw;: Ratio Method Flow Chart

Tnble 6 gives the values for the three key gas ratios corresponding to suggested diagnoses
(cases). Th'ese ratios, nccording to Rogers, llre applicable to both gases taken from the gas
spaee (or relay) !lnd gases extracted from the oil. The fault types (cases) given in Tuhle 6 have
been chosen by combining some cases from the number of fault types (lriginally suggested by
Rogers.
Fig 5 is a now chart describing the step-by-step application of the Rogers ratio meLhod.

5. Instruments for Detecting and Determining the Amount of


Combustible Gases Present

5.1 Portable Instruments. Many (If the gases generated by a possible malfunction in an oil-
filled transformer arc combustible. The on-site detection and estimation of combustible gases
in the transformer in the field using a portable combustible gas meter can be the first and the
easiest indication of a possible malfunction, and it may form the basis for further testing or an
operating decision.

\Vhen II more accurate determination of the tolal amount of combustible gases or a quanti-
tative determination of the individual componen.ts is desired, n laboratory analytical method
using a gas chromatograph or mass spectrometer may be used.
Gases generated in transformers can be explosive. Strict precautions should be observed
when sampling the gases from the transformer.

2'

!!l".t by lh- II~SlITU1E Of ScUCTIIICAl S ELEC1ROllICS ENGINEERS 'IHE)


POI <:Z,3b,OJ 19%
IE (57.104 91 . . 4!05702 0504375 55~ ..

IEEE
GJ::NERATED IN OIL-IMMERSED TR.Al"'SFORMERS Std CIl1.llW.I991

INPUT
,
<. ,
y R, y CASE" 0
NO
GAS .1-1
N N FAU1..T

CAS<'
LOW TEMPERA
lORE
,,--
<,H.
C,II, THERMAL
OVERLOADING

.. -- C,It,
C,II. R'
"
y CASE 4
. . THERMAL
<700 C

-- s
,,3
N
Y CASES
THERMAL
,,700C

CAse I

~
y y PO
'"
d
N
R'
<. ,
-~-
RAOK>
INFLUENCE
VOlTAGE
lRIV)

CAS< ,
y Y HIGH-
R'
.1-1 - . . ENERGY
,,"'ING

Fig6
Rogers Rutio Method Flow Chnrt

5.2 Fixed Instru.ments. The reliability oftTansformers can be improved by either monitor-
ing the gas space or the gn5e$ dissolved in the oil using self-contained. fixedmounted instru-
ments. These eontinuous monitoring instruments indicate the presence of a certain gas or the
total combustible gases as well as sound an warm when the combustible gases exceed a prede-
termined level. Optional recorders can also be used to provide u duily record or Ole combusti-
ble gllse5 pre5ent.
Jr the amount of the individual gas components is desired, a laboratory analytical method
using a gas chromatograph or mass spectrometer should be used.
There ure three somewhnt rlllated methods or monitoring the gases, as de&cribed in the fol-
lowing subsections.
N07E: Th..... .,..;U be a "".,.L.IX')' for the ratioe R2 aD<! R$ to ;lK........ tD a .atio
above 3 . . lhe clie<:l... "i& dlrvelo", in i""""aity

5.2.1 Method 1. The nr"t type of ga" monitor continually compores the thermal conductiv-
ity of the trnnllformer gall with that or pure nitrogen and is suitable for any transformer of the
closed t)"J)e wiU1 a gas space llbove the transrormer oil.
It is calibrated v.;th hydrogen. although Ule proportions of the combustibles are not
obtained from the measurements.

25

boy ~ .... H.:.TtTUIE 01' [lh~rCAl l Eterranc::; E'lGl'lEEl><; lIEHl


l 22,3<>,Q.l lOG"
IEEE (57.104 91 . . 4~05?02 050437b 494 . .

IEEE
SLdCS7.10~.1991 TF.F:E GUIDE FOR THE INTERPRF:TATION OF GASES

Table 6
Rogers Ratios for Key Gases

'2 R> RO
C_ SuggetlWdFBult
'ifj( CHi c.)~/
DiB!fnOolo;a
H, C~II

<0.1 >0.1
d.'
<I.' Uni\ non..-l

,'" <0.1 <1.0 I..,,,,,..:<ncrgy deMity


lll"'Cing-PD(~
NOTE)

2
,.,
0.1-
..,
0.1- >3.0 Amng---.}figb .....""Ky
di..,"ugc
, ~" >0.1 1.'
,., Low lCmpc"\urc
<1.0 lh.. no.J

, <0.1 >1.0 ..... TMI'lNoI <700C

, <0.1 >1.0
'"
>3.0 'fhertt\Al >7ooC

The transformer gas is continually circulaU!d through one section of a Wheatstone bridge
and returned to the trflnsformer. The other sedion of the bridge cOntains pure nitrogen and is
balnnccd ngninst the tTllnsformer gas.
\Vhen combustible gases are produced in the transformer, they mix with the tnmdhrmer
gas and increase the thermal conductivity of the transformer gas. The increase in the thermal
conductivity of the trnnsformer gas unbalances the Wheatstone bridge, and the unbalance ig
proportional to the tolal of the combustible gases as indicated on a meter.

5.2.2 Method 2. The second type of gas monitor continuously samples the transformer gas
at fixed intervals and bums any combustible gases present to provide a measure ofthe total of
the combustible gases. This type ofmonil.or is used only on transformers with a positive pres
sure of nitrogen over the oil.
At a fixed interval (usually 24 h), a samplt! or the transformer gas is pumped from the unit,
mixed with air, and passed over 8 platinum heating sensor of a Wheatstone bridge. Any com
bustible gas in the sumple is burned. This raises the temperature of the sensor alld unbul-
ances the bridge, which was balanced against a second platinum sensor in air. The degree of
unbahmce is proportional to the amount of total combust.ible gas present in the transformer
gas as indicated on a meter.

5.2.3 Method 3. 'l'he third type orgas monitor continuously measures the amount of hydro.
gen and other combustible gases dissolved in the transfonner oil.
Hydrogen and the other combustible gases of unknown proportions diffusing through a per-
meable membrane will be oxidized on a platinum gfls-permellble electrode; oxygen from the
a mbient air will be electrochemically reduced on a second electrode. The ionic contact between
the two electrodes is provided by a gelled highconcentration sulfuric acid electrolyte. The
electric signal generated by this fuel cell is directly proportional to the total combustible gas
concentration and is sent to a conditioning electric circuit. 1'he resulting output signal is wm
perature compensated.
A relay is operated in conjunction with the percent. gus meter so that when the combustible
gases exceed a preset value the relay sounds an alarm.

26

i3h~ by the lNS1JIUTE:JF ElEC'RIC~L & ElEC1PClHCS EI1G!NEERS llEEEI


~p 01 22,3b'0': 1"9b
IEEE (57.104 91 . . 460570Z 0504377 320 . .

IEEE
CDo'ERATED IN OIL-IMMERSED 'IltANSFORMER5 Sl.d Cli7.164_1991

Alllll1it. lhat 4au~ soundfti an ClIClnn .hould lH! wmplul for complet~ 4rnJi)'si.s by a g~ chro-
maUlgraph or m(1S$ Fipectromder.
At the time of installation and each year thereafter, the equipment should be 5tandardized
to be surc the monitor is operating properly. The operator !>hould follow the instruction guide
Qf the manufacturer.

6. Procedures for Obtaining Samples of Gas nnd Oil From the


Transformer for Laboratory Analysis
6.1 Gas Sllmplcs for Labomwry Analysis. All samples of gas from the g:lS blanket above
the oil should be taken in accordance with ASTM D3305-84 (2).

6.2 Gas DislIJolved in Oil. All samples of oil from electrical apparatus being taken for the
purpose of dissolved gasin-oil analysis should be taken in accordance with ASTM D3613-87
[4].
Under certain ronditions, stratification of dissolved gases in the oil may ~ur. and complete
milling could n'lQuire many hours. In these cases, where possible, oil samples should be
obtained from more than one location on the transformer.

7. Laboratory Methods (or Analyzing the Gas Blanket and the Gases
Extracted From the Oil
Comparative tests on essentially identical samples of oil (for instance, from the same trans
former) by various laboratories have indicated a lack of precision, with the measured concen-
tration of certain key gases reported to differ by a factor of 3 or more. The principal reason
appears to be lack of uniformity in the degree, i.e., the efficiency of gas extraction. For e:13ct
and generally applicable threshold or limit values of concentrations or evolution rates of key
gases, it would be necessary to obLain uniform and high (for instance, 97%) efficiencies of
edraction for individual characterislic gases.

7.1 Determination of Total Dissolved Gal. Determination oftolal dissolved gas should be
mode in acc:ordWlce ",ith ASTM D2945-90 fl).

7.2 Determination of Individual Dissolved Gases. Detennination of the individual dis


solved gases should be made in accordance with ASTr.l D361290 [3].

7.3 Determination of Individual Gases Present in the Gas Blanket. Analysis of the
individual cases present in the gas blanket above the oil rna)' be made by using ASTM 03612
90 (3], beginning at Section 10 of that stand:.wd. Sections 13.1 and 13.2 ofASTM D3612-90 [3J
nrc not applicable in this casco

27

~~ ",~ lll~llT1JTE OF" ElEU~tC"'" I. ElEC"Ot'IC~ [1'-:IIIEI'5 IIEH)


I :?.?I~ .. ,04 1"<>0
IEEE C57.l04 91" 48057020504378 2b7 ..

IEEE
Sid ClI7 .104-1 991 IEEE CUIDE FOR THE INTERPR.TA110N OF CASES

8. Bibliography
8.1 Sources. The sources used fire:

IE,.., and IEEE Transactions on Power Apparatus 1939-19i5


and Systems (PA)

Doble Indices of Minutes ofAnnual International 1935-19i5


COllfcrCllces ofDoble Clients

lEE Procudings 1949-1972

Electr-ical World 1949-1969

Scun~ Citation Indu 196{)-1969

Bulletin Analytique et Signaletique de France 1950-1972

Infonnation and technical papers supplied by


Memben orthe IEEE Trflnsformer Committee
(in particular a comprehensive list by T. K. Sloat>

Chemical Abstracts 1969-1971

Eb:c1r-ic;ul and Electr-onics Abstr-acts 1973-1975

CQmpendex und lnspec 197&--1987

8.2 Gas EvolutioD

(81] Clerk, F. M., '"TI>e Role of DissoJved Gases in Detennining the Behavior of Mineral Insu-
latin&" Oils,~ Journal of the Franklin Institute, vol. 215, p. 39, Jan. 1933.

(B2) Berberich, L. J. "I nnuance of Gaseous Electric Disc:har-ge of H,)rdrocarbon Oils, ~ Indus
tr-ial and EngiJU!er-ing Chemistry, vol. 30, pp. 280-228, 1938.

[B3] Murphy, E. J., "Gases Evolved by the Themlnl Decomposition ofPllpcr.~ Transactions of
Electrochemical S()(;iety, vol. 83, p. 161, 1943.

(B4) Vogel, }'. J., Peterson, C. C., nnd Mntsch, L. M., "Deterioration of Transfonner Oil and
Paper Insulation by Temperat\1re,~AlEE Transactions, vol. 78, no. I, pp. 18-21 {tithIes), 1951.

[85J Worner, T., "Behavior oflnsulating Oil Under Dielectric Stress with Respect to G::as E\o
lulion undlor Absorplion,- (in Gennan), Eltktrotech, Z. Desch, (Nuremberg), vol. 72, no. 22, pp.
656-658,1951.

[06) Oroce, C. E. Rand Whilney, W. 8., Note on thr Quantity and Constitution of Gas Liber-
ated During Arcing ill Oil CircuiJ Br-rokers. British Electrical and Allied Industries, Research
Association Technicnl Reports: G'X1'35 and ClXT66 (1951) and GlI'260 (1954).

rB7l Slelchely, V. C., ~Relation Between Cas Evolution and the Physicnl Properties of Liq
uids, ~ Applud Physics, vol. 22, p. 627, 1951.

28

lar. L, ,h 1,;"TlfUT OF nECT~ICAL g ElECfQ(llIIC<: [,lIlEr":> IIEEE


.~ 01 22,:;:',0 I~""
IEEE (57.104 91 . . 4~05702 0504379 1T3 . .

IEEE
CENERATED TN O[[,.lMJ>fERSED TRANSFORMERS Std CS7 .104_1~1

rnSI Dassechf!!o, H. and McClean, D. A., MGaning of Liquid DielectriC!. Under Electrical
Stre!os; lndustridi and Engineering Chemistry, vol. 47, no. 9, part J, pp. 1782-1794, 1955.

[89] Paul, Mlnformation on Hydrogen Generat.ion by fieat Decomposition ofPapcr," AlEE c.P.,
pp.57-21,1957.

[BIOI Meador, J. R. Bnd Dillon, K E., 'TrOonsfonner Oil Preservation,~ AlEE Transactions on
Power Apparatus and Syste.ms, vol. 33, pp. 1208-1211,1957.

[81I] HarriSOll, D., "Field Method Finds Arc-Formed Gas in Oil Filled Transformers," Eiectrj
('ai World. p. 94, August 4,1958.

[BI2} Kaufman, R. B., Pierce, J. L., and Uhlig, E. R, "The Effect ofTransformer-Oil-Preserva-
tion Methods on the. Dielectric Strength of Oil," IEEE PA, vol. 34, pp.1315-1321,1958.

1Bl3] Deenan, w. J. and Doucette, G. G., "Improved Method of Oil Preservation and Its Effect
on Gas Evolution," IEEE PA, vol. 28, pp. 657-666,1958.

(Bl"] BIlS$Khes, H. and Barnes, H. W., -oassing of Liquid Dielectric Under Electrical Stress,
Influence of Voltage and Pressure," Industrial Engineering Chemistry, vol. 50, no. 6, pp. 959-
966,1958.

18151 Krasucki, Z., Church, H. }O":, and Garton, C. G., "ANew Explanation of Gas Evolution in
Electrically Stressed Oil Impregnated Paper Insulation; Journal of the Electrochemical Soci
ety, '1.'01.107, no. 7, pp. 598-602, 1960.

(816) Saito, Y. and Hino, T., "Study of'l'hermal Deterioration of Enameled Wires by the foolass
Spectrometer Method," IEEE PA, vol. SO, pp. 653-657,1960.

(BI7] Rey, E. and Ehurt, L., "Die BeurieiluflG' vun inhibieTten und nicht inhibierten: lsolie-
rolen fUr HochspannWlgs-Transformatoren und Mcsswandler." Bull. As. Swis~ Elec. vol. 52,
110.11, p. 401, 1961.

(B181 8aguhn, A. H., Reinhard, R E., and Oake, S. 4, "Gas Generation During Interruption
under Oil," AJJ!,'E, vol. 237, 1962.

em 9] Dlodg-ett, R. B. and Bartlett,. S. C., MParnmetcrs for Predicting Gassing of Oils Under
Electric Stress," IEEE PA, vol. 55, pp. 528-536, 1961.

[820] Dnkin, T. W. and Sloat, T. K, ~Gas Generation and Its Relation to the Dielectric
Strength ofOil,~ Electrical Insulation Conference (JESEJ, 1963, pp.13O-133.

11321] Sheppard, H. R., "The Mechnnism of Gas Generntion ill Oil-Filled Transformers," Doble
Conference Index of Minutes, 1963, Slle. 6-601.

[D22) I-Iornsl>y, E. A., Irving, ft., and Patterson, E. A., "New Criterion of the Gassing l\.'Ilden
cies of Insulating Oils," Proceedings of the Institution of Electrical Engineers, vol. 112, no. 3,
pp. 590--596, March 1965.

(B23) Znky, A. S. and Hawley, R., MGas Evolut.ion from Insulating Oils," Electrical Reuicw, vol.
3, p. 828, June 1966.

(824) 81"Zusku, L. and Widmann, w., MZ ur Prufung der Gasfestigkeit von Isolierolen und Iso-
Jierun{;en, Elchtrotech. Z., pt. A, vol. 88, flO. 3, p. 69,1967.

29

V> '.!>e 1I.$TIT'J1E (IF ElECTPICIol.; [LECT"OIl;:~ ['<G't.'Ef"'S lE[[


l:C'JfI'C~ 1_
IEEE (57.\04 9l . . 4~05702 0504360 915 . .

IEEE
Std C!>7.100t_1991 IEEE GUIDE rOll TllE D\-rERPRETATtON or GASF$

{D25] Sloat, T. K, Johnson, J. L., and Sommerman. G. M. L., ~Ga, Evolution from Trans-
former Oils under High Voltage Stress," IEEE Transactions on Power ApparatllS arad System.s,
vol. PA-86, no. 3, p. 374, 1967.

[826] Pedersen, G., ~Gnssing of Insulating Oils under the lnAuence or an Electric Discharge/
Bfl)Wn &weri &u~w, vol. 55, no. 415, pp. 222-228, ApriVJI.fay 1968.

(8271 Morrison, E. L., "Evaluation of Thermal Stability of Electrical Insulating Paper," IEEE
1l'on.sactions all Electrical Insulation, vol. 1-3. pp. 76-82, August 1968.

(828) Lipsey. G. Jo-: and Ettlinger, L. T., "PermaleJ: II Insulation System." IEEE Electrical
Insulation Conference Proceeding" pp.15D-151,1969.

[8291 Varshavskij, D. S . Influence de la Valeur SpeifiQue du Dcgagcmcnt Ga.zeux Bur lo


Duree de Vie Descondensateurer au Papier Huile, Izvestia Vys$hikh Uchebnykh Zovendenii
Elekcomrkhanika, no. 7, pp. 800-803, 1970.

ra30) Gusev, N. r.. -Gas Shielding of a Transformer," En.ercetik., vol. 10, p. 24, 1970.

(831] Forster, J. A and McCrae, G. G., "Gassing Tendency Tests on Service-Aged Insulating
Oils; Dahle Conference Proceedings, 1974, See-IO-ISOl.

8.3 Detection and Interpretation

8.S.1 Gas Detector Relay

[B32) Bucholt, M., -rhe Bucholz Protective System and its Practical Applications; Elektro-
tech. Z., vol. 49, pp. 1257-1261, 1929, (In German).

[833l"Preliminary Report of Subcommittee on Gas Detector Relay,, listed in Advance Report


th~ 56th Ann.ual Meet.ing o{ Canadian Electrical Association Engineering Section. p. 39,
{Ot'
June 1946.

1834J AlEE Relay Subcommittee, "Relay Protection of Power Transfonnen.... AlEE 7l-an.soc-
tUlliS, vol. 66, pp. 9lJ-917, 1947.

[835] Gross, E. T. G., "Simplicity in Transformers Protection," ErectricaJ Enginuring, vol. 66.
pp. 564-569, June 19017.

[n36] Madill, J. T., "'Typical Transformer Faults and Gas Detector Relay Protection; AlEE
Tran.sactiolls, vol. 66. pp. 1052-1 OliO, 1947.

fB37] "Nipping Incipient Faults," editorial, Electrical World, vol. 128, p. 68, July 5, 1947.

(838j Gudmundsson, E., "Auxiliary Apparatus for Service Supervision and l>laintcnance of
Transformers," Journal of/he ASEA (Sweden), vol. 22, no. 10, PI). 159-167, 1949.

[O39} Streuli, W. R. ''Tranfoscope (Gas Actuated Transformer Protective Relay of Buchhol7.


Type," Doble Conference Index of Minute", 1949, Sec. 6-01.

(D40) Duffy, F. H . "Gas Detector Relay Experience on Saguenay System, Doble C<Jnfenncf!
Index of Afinute" 1951, See. 6-501.

30

9M bv U'e j'ISTIT'J-E or nECT~IC"'l ~ ELE(,T~crll':S ~"ultlH~~ .lHE)


~ Cl :2,?6'O~ l"?b
IEEE (57.104 91 . . 4~O;702 05043~1 851 . .

rEIU:
GE... .""ERATED ~ OIl...IMMERSED TRANSFOR.\1ERS Std C57.1D-l1991

f841 J Bean, H. L.. and Cole, H. L.., "ASudden Gas PTeuure Relay for Transformer Protection,"
AlEE 7J-ansactions, vol. 72, no. 3. pp. 480-483, 1953.

[8421 Rossier, C. "Design of Large Power Trnnsformen; Bull. SteherOlt (Suisse). no. 25f, pp.
9-25,1956.

1B431 Howe, V. H., Massey, L. nnd Wilson, A C. M., 'The Identity lind Significance of Gases
Collected in Buchholz Protectors," The Metropolitan VICkers Gazette, Manchester, England,
vol. 27, pp. 138-148, 296, 1956.

[B44] Aptov, 1. S., "Determination of the Character of Defects Inside Transformers fronl the
Composition ofGns Evolved in the Gas Relay," British Electric and Alied lndul':tries ne.~earch
Associations 7ralUaction IB1532, pp. 1-6, June 18, 1957.

[845] Hnrrison, D., 'Tnmsformer Gas Testing," Onlario Hydro Research Ntws, vol. 9, no. 4,
pp. 11-15. 1957.

[B461 Sterner, V., "Report an the Work of the Study Group on Protection and Relaying. Part II:
The Protection of Large Transformers," CIGRE C. R. 7hrnSQction$ (l8th SusionJ, 24 X 16, no.
334, pp. 8-13, Paris, 1960.

[B47] Vanlund, J. A, "General Electric Transformers," Doble Conferenct Index of .lI,{inules,


1961, Se<:. 6-301.

[B48] Sturtevant, D. D., "E:o;perience with Buchholz Gas-Operated Relay," Doble Conference
Index of Minutct, 1964, See. 6-101.

[B49) Seiler, H. "Gas Test Device for Protection of Transformer,'" Electroteeh. Z, vol. 19, no. 5,
pp. IIJ-116, 1967.

[B50J McDermid, w., -Generation of Combustible Gas in 230 kV OilFilled CutTent Trans-
formers; Doble Confenru:e lrtiUz of Minutes, 1968, See. &-301.

8.3.2 Gas Cushion in Sealed Transform.ers

[B5]j Wacner, H. H., -Method Dete<:ls Incipient. Power Transformer Faults," Electrical World,
p. 80, Dec. 1959.

[n52l Wagner, H. I-I., "Detect.ion of Incipient Faulls in Power Transformers by Gas Analysis,"
Doble Conference Index of Minutes, 1960, Sec. 6-801.

(853) PuCh, P. S. and Wagner, H. II., "Detection on Incipient. faults in 'l'rnnsformers by Gus
Analysis," AlEE c.P., pp. 60-950.1960.

[854] Wagner, H. H., "Gas Analysis Detects Incipient Faults; Electrical \\IOrld, pp. 114, June
20,1960.

{85S] Pugh, P. S. and Wagner, H. H., -Detection of Incipient. Faults in Transformers by Gas
Analysis," AlEE Transadions, vol. 80, pp. 189-195,1961.

[B56] Wagner, H. H., -Experience with Transfonner GasFault DetectorTl!Stings," Doble Con,-
ferenee Indu 0( Minutes, 1961, Sec. 6-701.

31

~1 .".. mS+HIHE OF ELECll;!CAl ;. ElECI"::fllCS Or,l.lED'S It(EEI


2';:'~.. ,o.:. 1<><>"
IEEE (57 . 04 9\ . . 4eOS702 OS043e2 79B . .

"'E
Std C67.104.1991 a: CUrD I'OR TIlE INTERPRETATION O~' CASES

113571 Horelick.A 4, "Incipient Fault Detection Method Proves Valuable for Transformer Pro
teeth'e Maintenance," Tra11Jmission and Distribution, p. 34, Jan. 1961.

[B58] Bore, Gordon llnd Reichart, "'Incipient Transformer Faults Quickly Detected," Electrkol
World, vol. 155, pp. ao-B3,April24, 1961.

[D59] Wagner, H. B., "Pennsylvania Transformers," Dobh Conference Index of Minutu, 1961,
Sec. 6-701.

1060] Duke, C. A and Tuylor, R. 0., "Detection of Combustible Gases in Nitrogen Sealed
Power Transformers; Doble Conference lndu ofMinutes, 1962, Sec. 6-101.

(B61] Gronberg, W. L., "Detection of Incipient Power Transformer faults with Combustible
Gus Jet," Doble Omference Index of Minutes, 1962, Sec. 6-301.

r062] Wagner, H. H., "Field Experiences with the TrorJsfonner Fault Gas Detector,~ Doble
Conference Index of Minutes, 1962, Sec. &-801.

[863) Taylor, R 0., "TV~5 Experiences in Analyzing Combustible Gases in Nitrogen Sealed
Power 1Tansformers," AlEE C.P., pp. 62-528, 1962.

{B64) Rickley, A L. and Clark, R. E., "Combustible Gu Detection in Transformers; Doble


Conferen~ Index ofMinuteJl, 1963, See.. 6-S01.

IB6S1 Taylor, R 0., ~C<Jmbustible Gas Tests Warn ofTrMsformer Failure,; Electrical World,
pp. 74-75, April 22, 1963.

[866] Burrett, R D. and Parker, J. C., "Experience in 'fusling NitrOl,;en in OilFilled Trans
former!i; Doble Con.ference Index of Minutes, 1963, Sec. 6-201.

(B67J Cromley, R J., MFailure and Repair of a Transformer Joint,~ Doble Con{ereru::e Index of
Minutes, 1963, Sec. 6--701.

(B6S) Van lund, J. A, "Gas Analysis for Detection of Incipient Power Transfonner Faults,~
Doble Confennce Indu of Minutu, 1963, Sec. 6--401.

[869J Vora, J. P. and Aicher, L. C., "Incipient Transformer }'ault Detection and Interpreta-
tion; IEEE G.P., pp. 63-584,1963.

IB70J Bellnschi, P. L., Mlntroduction to Symposium on Gs!t Analysis for Detection of Incipient
Faults in Power 'JTansformers,~Doble Con{uence Index of Minutes, 1963, Sec. 6-101.

IBn] Horelick, A. L., Wagner, H. H., and Lynch, J. C. , ~Method for Testing Electricallnstru-
mentalities for Incipient Faults," U.S. Patent No. 3111 388, November 19, 1963.

[B72] Wagner, H. H., ~Penllsylvania Trnnsfonner Towl Combustible Gas (TCG), Incipient
Fault Detector; Doble Conference Index ofMinutes, 1963, Sec. 6-302.

(B731 Wagner, H. H., ~Progres& Report on TeG Incipient Fault Detecl.or Practice," IEEE C.P.,
pp. 63-585, 1963.

IB741 Syrjamaki, E. M., Calibration and Comparison of Fault GllS DCtcctoT6," Doble Confer.
ence Inde:c ofMinutes, 1964, Sec. 2-101.

32

< t> .'c 1,.:.,TI1l'lE OF lE(T~H.AI. " ElEctr>OUI(;~ E!;'H',EEll$ ,IFH


~l ~;,JD'().t I.... t>
IEEE (57.104 91 . . 4A05702 0504383 624 . .

IF:n:
GENERATED IN OfL-U>L".u;RSED TRANSFOR..'oU:RS StdC57.1041991

{B75) Fer-guson, M. L., ~Combustible Gas Analysis as II. Prevenlive Maintenance Tool,~ Doble
Conference Index of Minutes, 1964, Sec 6--1801.

[D761 Duke, C. A., Goodroe, C. n., Taylor, R. 0., and Thompson, W. W., "Detection of Combus-
tiblll Gases in Nitrocen Sealed Power Transformers,~ Doble Crmfp.reru:e ltldex of Minutes,
19GJI, 6-1 00t.

{B7?] Wagner, H. H., "Development of Continuous Monitoring Total Combustible Cas (TeG)
Incipient Fault Detector,~ Doble Conference Index of Minutes, 1964, Sec. 6-1901.

[D78) Cra;vford, C. E. and Wilson, A. F., ~Incipient Transformer Fault Detected by Combusti
ble Gas Monitoring Prob'Tllm and Ultra Sound Tests," Doble Confereru:c Index of Minutes,
11164, Sec. &-1201.

(B79j Wogner, H. H., "Survcy ofTCG Tests on Transformers," IEEE G.P., pp. 6--4209.

[DBOI Parker, J. C., "Testing Nitrogen O\'cr Oil in Power Transformers," Doble Conference
Index of Minutes, 19601, Sec. 6-1501.

fDSl] Ferguson, M. L., "Combustible Gas Problem," Doble Conference Index of Minutes, 1965,
Sec. 6-401.

[B82J Graham, W. R. and Riner, R. B., Mlncipi(mt Fault Detection in Power Tra.nsfonners,~
Doble Conference Index of Minutes, 1965, Sec. 6-201.

[B83] Wllgner, H. H., ~Five YeliTS Experience with Transfontler TeG Incipient Fllult Detccwr
Thsts,~ IEEE Transactions on Power Appamtlls and Systems, vol. PAS-84, no. 8, pp. 700-706,
1965.

LB84] Yom, J. P. and Aicher, L. C., "Transformer Fault Gas Anlllysis Interpretntion," IEEE
Transactions on Powcr Apparatus and Systems, vol. PASB4, no. 2, pp. 172-176, discussion on
p. 487, 1965.

[B85] Thomas, D. J., ~Combustible Gas in Power TransfonneTS (Its DetR.ction and Signifi-
cance)," Doble CrmferelU:.e Index of Minutes, 1966, Sec. 6-1010.

lB861 Gochmauer, E. L., "TCG Analysis of Incipient Fault Forecasted," Doble Conference Index
of Minutes, 196G, Sec. 6-201.

fB871 Price, R. C., "Transformcr Failure at the Allen Steam Plants," Doble Conference Index
of Minutes, 1966, Sec. 6-301.

(E88) Rose, M., "Automatic Sampling and Measurement of Combustible Gases in TrOllsfonll-
ers," Doble Conference Index of Minutes, 1967, Sec. 6-801.

[BB9] Thompson, W. W., "Modification of Conservator Transformers to Permit Combustible


Gas Surveillance,~ Doble Conference Index of Minutes, 1967, See. 6-601.

[B90] Wagner, H. H., ~Penll5ylvania TCG Transformer Fault-Gas Continuous Monitor," Doble
Conference Index of Minutes, 1967, Sec. 6-701.

[B91] Sheppllrd, H. R., "Dangerous Conditions Are Detected by Analyzing Transfonner


Gases," Transmission and Distribution, vol. 20, no. 2, pp. 71-74, 1968.

33

~.,. the l'JSTITUTF OF FlEC1RICAI. ~ EI.ECT~OlJICS ENGIIJEEIIS tlt(()


>1 2~,3",04 lq~"
IEEE (57.\04 9L . . 48057DZ 0504384 SbO . .

IEEE
Std C51.104.1991 IEEE GUIDE FOR TIlE U.'TERPRl:TAT!ON O~' GASES

[892J Hahn. J. H., "Field Un tanking and Repair of a 2301U5 kV, 40150, 3/66.7 MVA, OAIFAJ
FA, Autotransformer!l.~ Doble Confertnce IMIlI( of Minuter. 196B, Sec. 6-101.

(893] McDermid, W., "Generation of Combustible Gas in 230 kV Oilfoll1ed Current Trans
formen,~ Doble Confeflm lrulu of Minutrs, 1968, Sec. 6-301.

[8941 Sheppard, H. R., "Why is Gas Generated in OilFilled T'ransfonners?" Transmi.uion


and Distribution, voL 20, no. I, pp. 63-65, 196B.

rB95] eillies, D. A. and Muckerhelde, L. H.. "Combustible Gas Monitorinl;--SOO kV Current


T'ransformers," [)obIt. Conferenu Index of Minutu, 1973. Sec. 6-201.

B.3..1 Dissolved Gall in Transformer Oil

(896) Clark, F M . '"The Role of Dissolved Gases in Determining the Behavior of Mineral
Insulating Oils; Journal oftM PranJdin Inuit ute. vol. 215, no. 1, p. 39, 1933.

(8971 Dornenberg, E. and Gerber, O. E., "Analysis of Dissolved and Free Gases for Monitor-
ing Performance of Oil-Filled T'ransformers," Brown BOlleri RelJiew, vol. 54, no. 213, p. 104,
1967.

[B9B1 Stamm, H. and Schul%e. w.. "Deterioration oflnsul.ators Consisting of Paper-Oil La)'ers
and its Study by Gas Chromatography,- presented at CIGRE, Paper 12-14,1968.

(8991 Sheppard, H. R.. "Why is Gas Generated in Oil-Filled Transformers?" 1rafl.llmjssion


6~5, Jan. 1968.
and Distributwn, pp.

{BIOOI Strom, R and Takahashi, S., MAnalysis of Fault Gases in Transformer Oil by Improved
Method of Gas Chromatography; Ontario Research Qunrterly, second quarter, pp. 21-25,
1968.

[BIOIl Hahn, J. H., "Field Untnnking and Repair ofa 23M15 kV, 40150, 3/66.7 MVA, OA/I'"AI
FA Autotrnnsformer: Doble Canferenot index ofMinutes, 1968, Sec. 6-101.

[BI021 Thibault, M. and Galand, J., "Modem Physical Chemistry Methods for Analysis of
InsulaUon Degradation in Service,M Review Generale de fElectricitt, vol. 78, no. 3, pp. 301-
302,1969.

[BI031 Waddington, fo-: B. and Allan, D. J., 'Transformers Fault Delel:tion by Dissolved Gas
Analyses,~ Electrical
Review, pp. 751-755, M.lI.Y 23,1969.

181G41 Coltong, R., "Combustible Gas Testing on Large OilPilled T'runsfonners," Electrical
InslJlation Conference 69C 33EJ-37, pp.1l5-118,1969.

[DI051 Muller, R., Potthoff, K, and Soldner, K., MAnnlyse des Gas Dissous dans I'Huile en
Tant que Moyen de Surveillance des T'runsrormnteurs et de Detection de Defaults Naissants, ~
Ilresented.at CIGRE, Paper 12-<12, 1970.

[BI06] t'allou, 8., Viale, F., Devaux, A., FOUTnie, R, Galand, J., Vuarthes, P., Davies, I., Rog_
ers, R R., Reynolds. E. H., and Dornenburg, E., MApplicatioll des Methodes d'Analyse Physico-
Chimiques A l'Etude des Degradations dans I'lsolation des Materiels Eledriques," presented
at CleRE, Paper 15-07,1970.

34

" ' . by t h 1'~T:I'JIE or CUCiPl('.Al I HEClROU1CS [!V,)!l;;[U'l> (!E[[l


~ 01 ~~,3<>,() 1_
IEEE CS7.10~ 91 . . ~60S702 OSO~36S ~T1 ..

lEEE
CENEllATED IN OIL-IMMERSED TRANSFORMERS Sld C57.1041991

[BI07) Moraro, D., Popesco, C. Stoic&, M., and Tanaseseo, F. I, -Detection des Degradations
Produite5par des Decharges PaTtielles pour les Types d'bolation Papier-Huile." presented at
CIGRE. Paper 15---05, 1970.

(8108) Wilputte, R., -Contribution a l'Etude de I'Evolution de Gas dans les Isolnnu LiQuides
et Solides Soumis a rAttion de Contraintes Electriques,- presented at CIGRE, Paper 15---03.

(SI09] Beer, G., Onerii, A., Bossi, A., Hjrra, C., "Contribution ezperimenLnle 8: I'etude de 1a
degradation produiLs par les delchargu partielles dans Ie papier isolant impregne a. I'huile;
presented at CIGRE, Paper 15-02, 1970.

(8110) Discussion ofPapeTS from Group 15 (Insulating Materia!), presented a1 CIGRE, 1970.

(8111 J Lindeman, K., -Analyze Transfonner Gas To Identify Faulta," Power, vol. 114, no. 4, p.
74,1970.

(81121 "Detection of Developing DaDlage in Transformer; Gorchokollo Elek. Sta., vol. 41, no.
4,p. 51, 1970.

(8113] Ishii, T. and Namba, S., "Les Gu Produits LoTS de 10 Deterioration de l'Isolation des
'ITansfonnateurs; Jour1UZ1 of the lnstitute of Electrical Ertginur. of Japan, vol. 90, no. 5, p.
910,1970.

[O1l4) Dind, J., Daoust., R, Regis, and Morgan, J . -Analysis of Gases Dissolved in 'ITans-
fonner Oil,- Doble Conferena./Il(fu: of Minutes, vol. 38, AlC71 , 1971, Sec. 6-U 01.

(BllS) Morpn-Schaffer CeJrporlltion, rra.n,fonntr Fault Ddedwn &flIice.

(BlI6] Collins, R. L.. -Hydrogen Detector," US Patent 3 559 457, 1971.

(DU7] Stoll, P. and Vuillemier, C., -Une Methode ChromaloiTaphiQue de Mesure Absolue de
C02 dans I'Huille des 1'ransfonnateun; BuU. Ass. Suisu Elee., vol. 62, no. 23, p. 1113, 1971.

(8118) Hino, T. and Suganuma, Y., -Rapid MealUTement of the Deterioration of Oil.Immersed
I'aper,~ Eketricol/n,ula1ion Conference, 71 C 38EI-41, 129,1971.

rB119) Davies, 1., MComposition oflhe Gases Formed in Electricallnsul!lting Oils: Journal of
Applied Chemi!!ry and Bio!ec;hnolO{JY, vol. 21. p. 194. 1971.

lD120) Thibault, M. and Galand, J., "Methodes Physico-Chimiques d'Etude de DCr,Tadntions


en ServiCC,R Rell. Gen. Eke., vol. 80, no. 1, p. 46, Jan. 1971.

fB121) Thibault, M. and Galand, J., "M:ethode, Physico.Chimiquel d'Etude de Degradation


en Service; ReI). Gen. Eke., vol. 81, no. 1, p. 48, 1972.

18122) Daoust, R., MAnalysis of GasCll Dissolved in Transfonner Oils,R Doble Confuenu Index
of Minutu, vol. 39, AlC71,1972, Sec. 1()....601.

(BI23) McCrae. G. G_, "Experience with Incipient Faults Detection by Gasin-Oil Analysis:
Doble COIlfu'cnoe /ndero( Minu!u, vol. 39, lAC72, 1972, Sec. 6-101.

(8124) Hardy, Hurworth, and Robinson, -Diagnostic Testing of 1'Tens(ormer Insulation by


Gas-in-Oil Analysis: presented a1 SUPEC 72 Conference, University of Bradford, 1972.

35

by ~"e WSlllUIE OF ElECIIlICAL & ELEC1R(J-llCS F.tlGltlEEIl$ (IEEEI


12,J40,~ lq<jl>
IEEE (57.~04 9~ . . 4605102 05043~b 333"

IEEE
Sld C57 J04-l991 1 GUID FOR nm INTt:RPRTAnON OF GASES

(8125) ~Cas Analysis as an Aid to Transfonner Fault Location; EI~ctric 7Tmu, vol. 161, no. 5,
p. 46. 1972.

181261 Cameron, D. I.. "Gas-in-OilAnalysis: A'Thol for Early Warning of Equipment Trouble,~
presented at CEA, Spring 1972.

[B1271 Barraclough, B., Bayley, E., Davies, I., Robinson. K., Rogers, R R., nnd Shanks, E.
KCEGB Experience of the Amdysis of Dissolved Gas in Transfonner Oil for the Detection of
Incipient Faults," I n TEE Conference Publication, no. 94, Sec. 1, Part 2. March 1973.

[8128) 8allaschi. P. L.. "Application of Fault-Gas Detection Methods on Transfonner Test at


the Factory," Doble Conference Twx 0{ Minutrs. 1973, Sec. 10-301.

(8129] Pugh. D. R., "Combustible G8I Analysis, Doble COTIferrna Iru;ux 0{ Minutes, 1973.
Sec. 10-401.

{BISOJ Martin. W. 8. Jr. and 1'I.ylor, R. 0., -Experience with Chromatcgraphic Analysis of
Fault Gases in Power Transfonners; Doble Conference Index 0{ Minutu, 1973, Sec. 11)....201.

[8131) Booker, J. R., "Ellperience with Incipient Fault Detedion Ulili:ting Gas Chromatogra-
phy Analysis,- Doble Confuen~ Index ofMinwtJ, 1973, Sec. 10-.501.

[B132J Raab. E. L., "Symposium on Gas Chromatography: Doble Confereme Index of Min-
ules, 1973, Sec. 10-501 A.

fBI33l Yamaok, M. et al., "GJO Analyser (Auto.Analyser for Gases in Oil),~ Fuji Elretr. Rev.
(Japan), vol. 19, no. 2, pp. 83-86, 1973.

18134J Smimov, M. A. et al.. "Gas-Content of the Oil in Working Transformers," Elektr. Ston-
tsii (RusJ, no. 8, pp. 57-60, 1973.

(BI35] Pugh. D. R., "Advances in Fault Diagnosis by Combustible Gas Analysis: Doble Con
ferenu Indu of Mi.nut~$, 1974, Sec. 10-1201.

rB136) Beeler. G., "Cambustible Gas Analysis on Peach Bottom No.1 Transformer: Doblt
COIIferrnce 1nOOo( Minutes, 1974, Sec. 10-1401.

[BI37) Keene. D. J., -rransfonner Problem Delected by F.o.uItrGosAnalysis," Doble Confer-


tmt Tndex of Minutes, 1974, Sec. 10-1301.

[8138] Duval, M., "Fault Gasu lo'ormed in Oil-Filled Breathing E.H.V. Transformers; Inter
prel.Rlion of Gas Analysia D.o.ta, IEEE C.P" no. C 74, pp. 476-478,1974.

[lU39) Domenburg, E. and Strittmatter, w., -Monitoring Oil Cooled Transformers by Gas
Analysis.- Brown Bovui Revitw, vol. 61, nO. 5, pp. 238-247. May 197<1..

[8140) Muller. R, Schliesing. H., and Soldner, K., "Testing and Supervision of Transformers
by Analysis oftbe Cases Dissolved in the Oil,- Ehlt.trizitoetswirlsch4tt (Germany), vol. 73, no.
239.pp.683-687.1974.

[8141]" Discussion From Group 15 ((nsulating Materials): CIGRE Conference Proc:rrdings,


vol. 1.1974.

3.

'.I,IM try ~h .. WSTITUTE Of' ELECUlIUol. !. [lEClllONICS EH3WHRS IIHE!


s..p 01 22,:>t>,().ol tQ<Jt>
IEEE (57.11J4 91 40305702 QS043B7 27T

IF.F.E
GENERATED IN OIL1MMERSED TRANSFORMERS 9t.d C5? .1Q.f-19111

[BI42} Rogers, R. R., "U.K Experience in the Interpretation of Incipient Faults in Power
Transformers by Dissolved Gas-In-Oil Chromatographic Analysis,~ DobIe Conference Index of
Minutes, 1975, Sec. 10-201.

IB143] Armstrong, G. W. Jr., and B::aker, A. E., "Transformer Problems Detected or Confirmed
by Use ofCombustibJe Gas Analysis," Doble Conference Index of Minutes, 1975, Sec. 1()-301.

[B144J Dvoracek, E. and Walla, J., "Analysis of Gases Dissolved in Transformer Oils." Enerae-
/ika, vol. 15, no. 1, pp. 8-11,Jan.1975.

[D145J Pejovic, V. and Bojkovic, A., "Control of Transformer Insulation Systems by Gas Anal-
ysis," Thknika (Yugoslavia). vol. ao, no. 5, pp.~, 1975.

[B146] Thibault, M. and Rabaud, J., "Application of the Analysis of Dissolved Gases to the
Maintenance of Transformers," Rev. Gen. Elec., vol. 84, no. 2, p. 21, Feb. 1975.

IB147J Dind, J. E. and Regis, J., "How HydroQuebec Diagnoses Incipient Transformer Faults
by Using Gas-inOil Analysis," Pu.lp Paper Can., vol. 76, no. 9:T264, Sept. 1975.

[B148] Davies, I. A., "'Review of Methods for Examining the Analysis of Gases in Oil Filled
Electrical Equipment, With ObselVations on the Concentrations Found and Their Depen-
dence an Time," G.E.G.B., Paper No. SSD/SElRR/68J75.

[B149] Hubacher, E. J., "Analysis of Dissolved Gas in Transformer Oil to Evaluate Equip-
ment Condition," P.C.E.A. Engineering and Operation Section Spring Conference, March
1976.

(BI50) Tsukioka, H., Sugawara, K, Mori, E., Hukumori, S., and Sakai, S., "New Apparatus
fur Detecting H 2, CO, and CHi Dissolved in 'lTansformer Oil," IEEE 'I'ransactions on Electri-
cal Insulation, vol. EI-18, no. 4, pp. 409-419, Aug. 1983.

[BI5Il Lampe, W., Spicar, E., and CarTender. K.., "Gas Analysis as a Means of Monitoring
Powcr Trallsfonners,~ASEA Journal (Allmaenna Svenska Elektrisha Aktiebol!llget), vol. 52,
no. 2, pp. 39-42, 1979.

[BI52l DiGiorgio, J. B., Jakob, F., and Haupert, T., 'Transformer Fault Gas Analysis," min-
utes of the meeting of Pennsylvania Electr. Assoc. Eng. Sect. Electr. Equip. Cerom., Bedford.
PA. Oct. 18-19,1977. Published by Pennsylvania Electr. Assoc. Eng. Sect., HafTisburg, p. 22,
1977.

(B153] Aiyar, D. and Nanda, J. R., "Identification and Estimation orGnses DissO'lved in Elec-
tric!!l Insulating Oil by Gas Chromatography," Electric India, vol. 16, no. 9, pp. 17-21, May
15,1976.

lDI5-1] Tsukioka, H., Sugawara, K, More, E., Hukumori, S., and Sakai, S. , "New Apparatus
for Detc<:ting H2, CO, and CH 4 Dissolved in Transfonner Oil," IEEE Transformer Ekctrical
IMulation (USA), vol. EL-18, no. 4, pp. 409-419, Aug. 1983.

[B15sJ Surakka, M. L., "Gas AnRlysis of Oil Monitors Transformer Condition," Modern Power
Systems {USA}, vol. 3, no. 4, pp. 60-61, April 1983.

[B1561 Surakka, M. L. and ,Jaakkola, V. K.., "Gassing Properties ofTransfonner Oils," Pre-
sented at Nordisk Isolation Symposium, 1982, Odense, Denmark. Published: ASEA, Stock-
holm, Sweden.

37

'y th. INSTlTUT OF ElfCTRIClll & ELECTRONICS ENGINEERS II((EI


22,36,04 n~6
1E[E (57.104 91 . . 4605702 05043&8 lOb . .

IEEE
Std. CS7.llM.lelll 1EE GUIDE FOR THE D-."TERPRETATJON OF CASES

(DIS?] Kelly. J. J. and Myers. S. D., "The Value of Gas-in-Oil Analysis as II Guide to the
Operational Conditions Existing in a Power Transformer," Conference Record of the 1972
Thirty-Fourth Annual Conference of Electrical Engineering Problems in the Rubber and Plas-
t.ics Industries.

rB158J Yamada. H., Nomura, Y., Kat.ayma, Y., Ishii, '1':, Imamura, 0 . and buchiyama, K..,
'"Automatic: Field Monitoring of Dissolved Gas in Transformer Oil," IEEE TrallSformer Power
Apparatus and Systems (USA), vol. PAS-lOO, no. 4, pp. 1538-1544,AprilI981.

[13159] Brasel. E. and Langec:ker, K., '"E.traction of the Gas Dissolved in the Transformer
Oil," EleJctrie (Germany), vol. 34, no. 12, pp. 637-641,1980.

(B160] Kan, B., '"Automatic Dissolved-Gas Analyzers for Transformer Monitoring," Mit:tub
ilfhl" Electric Adt> (Japan), vol. 12, pp. 8-10, June 1980.

[B161] Groves, B. C. and Brinkworth, D. C. "On-Line Gasin-Oil Monitor," CEGB nchnical


Disdosun Bulletin (GB), no. 349, pp. 1-2. June 1980.

rB162J Kelly. J. J . 'Transformer Fault Diagnosis by Dissolved-GuAnalysis: presented at


IEEE 26th Annual Petroleum and Chemical Industry Conference, pp. 145-151, 1979. San
Diego, CA, Sept.10-l2, 1979.

(8163J Rogers, R. R, "IEEE and IEC Code. to Interpret Incipient Faults in Tranlformers
Using Gas-in-Oil Analysis: IEEE Transforme,. Eleetricallruulation (USA), vol. EI-13, no. 5.
pp. 349-354, Oct. 1978

[B164] Belanger, G. and Duval, H.. "Monitor for Hydrogen Dissolved in Transformer Oil,
IEEE Transfonner Elmricollnsulation (USA), vol. EI-12, no. 5, pp. 334-340, Oct. 1977.

{DIG51 Kauffman, G. H., "Gas Bubbles in Distribution Transformers; IEEE Transformer


Power Apparatus and Systems (USA). vol. PAS96. no. 5, pp.1596-1601, Sept.-Oct. 1977.

[BI661 Shirai, H., Shimoji. 5., and Ishii, T., 'Thermodynamic Study on the Thermal Decompo-
5ition oflnsulating Oil," IEEE '1hJruformer ElectrkallnsuJation (USA), vol. EI-12, no. 4, pp.
272-280, Aug. 1971_

(B167J Muller, R., Schliesing, "., and Soldner, K, Assessing the Operational Stale ofTrana-
formers by Gas Analyai5," EleJctriziJ.oetw;irtschoft (Gennany), vol. 76, no. 11, pp. 345-349,
May 23, 1977.

[B1GB] Glass, R. H. "Dissolved GasAnlllysis ofTrnnsfonner Oil," 'n-ansmisswn and Distribu-


tion (USA), vol. 29, no. 1, pp. 22-24, 26, 28-29, Jan. 1977.

[B169] Domenburg, E. and Hutzel. 0., "Operational Supervision by Testing the Insulating
Oil; Ele.lttrotech. Z. A (Germany), vol. 98, no. 3, pp. 211-215, March 1977.

(B170] Heinrichs, F. W, "'I'he Impact of Fault-Detection Methods and Analysis on the Trans-
former Operating Decision," presented at IEEFJPES Transmission and Distribution Confer-
ence, Paper 86 T&D 559-9, Anaheim, California, Sept. 1986.

(DI7l) Haupert, T. J., Jakob, F., DiGiorgio, J . and Sperry, M., "Use ofTransfonner Oil Anal
ysis for Reducing Equipment Failures," Electrical World, vol. 192, no. 7, 1979.

38

I'''t by 0.... INSTI1UIE OF (LEctPIC"," " HEClROIllCS ENGTIlEEPS !lEUI


~p 01 22,36,(),( \qq6
IEEE (57.104 91 4B05702 0504369 042

lEEE
GENERATED m OILIMMERSED TRANSFORMERS SldC57JQ.I,I991

[BI72) Jokob, F., Haupert, T. J., ond DiGiorgio, J., Cut Failures by Analy>;ing Trnnsfonner
Oil," Electrical World, pp. 52-54, Oct. 1, 1979.

[BI73] Jakob, P., Haupert, T. J., and DiGiorgio, J., "New Technique Pinpoints Fault Loco.
tions," Ekctrical World, pp. 119-121, May 1981.

[B1741 Church, J. 0., Haupert, T. 0., Jakob, F., "Analy:t.e Incipient Faults with Dissolved Gas
NomOTaph,~ Electrical World, Oct. 1987.

[B17S] Haupert, T. O. and Jakob, F., "A Review of the Operating and Practice of Dissolved Gas
Analysis," Ekctricallnsulation Oils, STP 998, pp. 108-115, 1928.

[B176) Haupert, T. 0., Jakob, F., and Hubo.cher, E.J., "Application of a New 'Thchnique for the
1nterprelation of Dissolved Gas Analysis Data," Il1h Annual Thchnical Conference of the
International Electrical Thsting Association, PP. 43-51, 1989.

39

y the INSTITUTE OF ElECTRICAl! ElECTRONICS ENGINEERS (IEEEI


22.36,04 1"1"16

S-ar putea să vă placă și