Transformer Insulation Upgrading

and Loading Guide Equations

Panel Session
IEEE Transformer Committee
Insulation Life Subcommittee
October 25, 2005

By
Thomas A. Prevost
EHV-Weidmann, Inc.

Hasse Nordman
ABB Transformers
Vaasa, Finland

and
T. V. Oommen
Consultant
Learning Objectives

A. Thermal Upgrading of Paper Insulation

- Benefits, method, types, application, evaluation
test results

B. Arrhenius Life Equations and Plots

- Discussion of Loading Guide
- End point criteria: tensile strength; degree of
polymerization (DP)
- Life plots in pre-1995 Loading guides
- Life equations and per unit life in 1995 Loading Guide
- Impact of changing end point criterion of life
equations and plots
- Discussion of the DP based end point and life

IEEE Transformer Committee Panel Session – October 25, 2005

 PART I

Thermally Upgraded Insulation

in Transformers

Presented by:

Thomas A. Prevost
EHV Weidmann Industries Inc.

A Member of the Group

•Introduction
•Background
•Standards Activities
•Results of Aging Tests
•Conclusion
The life of a transformer is limited to the life of its
solid insulation.

In the late 1950’s transformer manufacturers developed

Thermally Upgraded Papers (TUK).

In 1962 NEMA officially recognized TUK in standard

TR-1-1962 by establishing another temperature rise
limit of 65 °C for oil-immersed transformers using
TUK.

Today 65 °C rise transformers are the norm in N.

America.
The thermal limit of transformer windings is the insulation on the
conductor at the winding hot spot. The average winding rise is
calculated as follows:

558 C Rise 658 C Rise

Ambient 308 308
Average Wndg Rise 558 658
Hot Spot Differential 108 158
Hot Spot Temperature 958 1108 *

* Only attainable with thermally upgraded insulation.

Background:
Thermally Upgraded Insulation Processes:
General Electric Permacel

McGraw Edison Thermecel

Westinghouse Insuldur
Two types of Thermal Upgrading processes:

Modification of the cellulose chains specifically at

OH groups by cyanoethylation and acetylation.

Addition of chemicals to protect the cellulose

from oxidation: this is primarily achieved with
nitrous compounds such as urea, melamine,
dicyandiamide, and polyacrylamide.
Cellulose Molecule
Single Glucose Ring
Cyanoethylation
Ref. General Electric Company
Amine Addition - Dicyandiamide

•Chemical Additive to paper.

•Consumes water as it is produced.
•Neutralizes acids as they are produced.
•(ref Lundgaard, Lars; Hansen, Walter; Linhjell, Dag; Painter,
Terence; “Ageing of oil-immersed paper in power transformers”
IEEE Transactions on Power Delivery, Volume 19, Issue 1, Jan.
2004 Page(s):230 - 239 )

•Suppresses the self-catalyzing character of aging

process by chemical reaction.
•During this process the stabilizing agent is
consumed.
Nitrogen
•All of the various thermal upgrading processes
contain nitrogen.
•Nitrogen is not found in cellulose

Nitrogen quantity is used to determine the amount of

thermal upgrading agent added to paper.

Different thermal upgrade processes will have

different nitrogen content levels to assure sufficient
upgrading.
ASTM D-982/ TAPPI T-418 “ Organic Nitrogen in
Paper and Paperboard” (Uses Kjeldahl method)
 Test method/Lab Comparison of Nitrogen Determination Methods

3.5

3
Nitrogen Content (%)

2.5

2 Kjeldahl (Lab2)
Kjeldahl (Lab1)
Combustion
1.5

0.5

0
-0.5 0 0.5 1 1.5 2 2.5 3 3.5 4
-0.5
Definition:
Thermally Upgraded Paper (approved by IL Subcommittee,IEEE TrComm,
Spring 2003 D. Platts Chair)

Cellulose based paper which has been chemically modified to reduce the rate at
which the paper decomposes. Ageing effects are reduced either by partial
elimination of water forming agents (as in cyanoethylation) or by inhibiting the
formation of water through the use of stabilizing agents (as in amine addition,
dicyandiamide). A paper is considered as thermally upgraded if it meets the life
criteria as defined in ANSI/IEEE C57.100; 50% retention in tensile strength after
65,000 hours in a sealed tube at 110 °C or any other time/temperature combination
given by the equation:

Time (hrs)= e (15,000/ (T+273)-28.082)

 Because the thermal upgrading chemicals used today contain nitrogen,
which is not present in Kraft pulp, the degree of chemical modification is
determined by testing for the amount of nitrogen present in the treated
paper. Typical values for nitrogen content of thermally upgraded papers are
between 1 and 4 percent when measured in accordance with ASTM D-982.
(7)
Verification of 65 °C Rise Insulation

•Presently there is no clause in the standards which

state that the transformer manufacturer must verify
that Thermally Upgraded Paper is used.

•Presently no acceptance test will indicate if thermally

upgraded paper is not used.

•Currently being considered for IEEE C57.12.00

 Proposed clause for future revision of C57.12.00
DRAFT #2 Feb. 20, 2004

5.11.3 Basis for Temperature limits

The temperature limits in 5.11.1 were established to coordinate with the insulation systems in general use at the time. Transformers that
meet the Temperature and Loading Conditions in this standard, shall be built using Thermally Upgraded Paper or an alternative insulation system
that has been proven to possess minimum aging characteristics that either match or exceed those of Thermally Upgraded Paper.

The test procedures to determine the aging characteristics and the minimum life expectancy of an insulation system are provided in
C57.100.

Acceptable thermal aging performance may be assumed if the insulation system , when tested in accordance with C57.100, demonstrates a
minimum life expectancy of at least 20.5 y (180 000 h), when operated at the hottest-spot temperature, at rated load, as defined in this clause 5.
The minimum insulation life expectancy curve is displayed in Figure AA, and is determined by equation. (A).
Where
LIFE is the life in hours
T(A) is the hottest-spot temperature in ºC

Figure AA—Minimum life expectancy curve for liquid-immersed distribution, power, and
regulating transformers rated in accordance with this clause, at 65 ºC average rise, 80 ºC hottest-spot rise.

Note: A transformer’s expected life under normal loading conditions, and it’s performance under overloads, as evaluated in C57.91, have
long been based on the assumption that the insulation system would be thermally upgraded, and would meet this minimum life expectancy curve.
(The curve was formerly published in both C57.91 and C57.100.)

Task Force Definition of Thermally Upgraded Paper. February 20, 2004

 IEEE C57.100

1000000

100000
Area of acceptablitlity
10000
Life-Hours

1000

100
Area of non-acceptablitlity
10

1
90

100

110

120

130

140

150

160

170

180

190

200

210

220
Line indicates 50% loss of Tensile Strength Aging Temp

Curve A1, IEEE C57.100 ANNEX A

IEEE WORKING GROUP ON THERMAL EVALUATION OF
POWER AND DISTRIBUTION TRANSFORMERS (C57.100)

The WG suggested that there could be three possible

methods:
1) Functional Life Test (full scale models)
2) Sealed Tube Test
3) Other test that models power transformers

It was noted that we could be developing different criteria

for end of life depending on the insulation component being
measured. For example, tensile strength retention for paper and
compressibility for spacer material.
 IEEE WORKING GROUP ON THERMAL EVALUATION OF
POWER AND DISTRIBUTION TRANSFORMERS (C57.100)

6.0 Discussion IEC – 62332 – Electrical Insulation Systems (EIS) Thermal

Evaluation of Combined Liquid and Solid Components

Bill Simpson gave a summary of the work that is being done by the IEC. He presented a
sketch of the aging cell that is being proposed. This aging cell is similar to the cell
developed by DuPont which can test different materials at different temperatures
modeling insulation at the hot spot and in the bulk oil. The IEC document is currently
being balloted. At this time the working group does not have permission to circulate
this document, but this will be circulated when it is approved and available for
circulation.
Correlation of Nitrogen Content with Aging
2004 Test

Nitrogen Content TU Agent Content

Type 1 - 0% 0%

Type 2 - 1.1% 1.7%

Type 3 - 2.5% 3.8%

Type 4 - 3.4% 5.1%

Insuldur Content is equal to 1.5 times nitrogen content.

The specification for Nitrogen Content of Insuldur upgraded

paper is 1.3% - 2.6% .
 Correlation of Nitrogen Content with Aging
2004 Test Results

Tensile Strength
Nitrogen Content Retention (%)

Type 1 - 0% 43.0

Type 2 - 1.1% 53.8

Type 3 - 2.5% 71.1

Type 4 - 3.4% 76.3

The lower specification of 1.3% nitrogen content correlated quite closely

to the end of life criteria of 50% tensile retention
 Correlation of Nitrogen Content with Aging
Recent Tests on Paper in Market

Material Thickness Nitrogen Content

A .003” 0.32%

B .003” 0.38%

C .003” 1.83%

Note Specification for Insulur Traeted Kraft Paper is Nitrogen

Content = 1.3% - 2.6%
 Aging Test of Crepe Conductor Paper Aged at 170 C

100

90

80

70
% Tensile Retention

60
A (N2=0.32%)
50 B (N2=0.38%)
C (N2=1.83%)
40

30

20

10

0
0 100 200 300 400 500
Aging Time (Hours)
 Aging Test of Crepe Conductor Paper Aged at 170 C

1400

1200

1000
Degree of Polymerization

800
A (N2=0.32%)
B (N2=0.38%)
C (N2=1.83%)
600

400

200

0
0 50 100 150 200 250 300 350 400 450 500
Aging Time (Hours)
 Thermally Upgraded Paper in Transformers
Conclusion:
•Thermally upgraded paper has been recognized in N.
America for over forty years.
•Today, The addition of Dicyandiamide and often other amine
compounds is the principal means of thermal upgrading.
•The accepted method for quantifying the amount of the
amines in the paper is by the measurement of Nitrogen
content.
•The IEEE Transformers Committee has developed a
definition of Thermally Upgraded Paper.
•The IEC has incorporated the IEEE definition of TUK and
has included an option for higher average winding rise if TUK
is utilized.
 Thermally Upgraded Paper in Transformers
Conclusion:
•IEEE Transformers Committee is planning to add a
requirement to C57.12.00 to verify that the insulation is
Thermally upgraded.
•Test results demonstrate that measurement of Nitrogen
content is a reliable indicator of aging.
•It is critical to purchase insulation materials which are high
quality and which can be traced to the manufacturing lot
number and its respective Nitrogen content.
 Transformer Insulation
Upgrading and the
Panel Session
Loading Guide
IEEE Standards meeting
Memphis, TN. 25 10 05 Equations
© ABB Power Technology Products - 1 -

Hasse Nordman
ABB Transformers
Vaasa, Finland
10/25/2005
 Background
- The IEC transformer standards have by tradition referred to non-
thermally upgraded insulation paper.

- The IEC TC 14 (= Transformer Committee), Maintenance Team MT1

started in 2001 to prepare a new IEC Loading Guide.

- The draft for a new IEC Loading Guide is at its final stage, as a
FDIS, circulated for voting during 2005-09-02…2005-11-04.

- MT1 decided to introduce the following basic concepts from

IEEE:
* Thermally upgraded paper.
* Transformer insulation life.

This was made in 4 steps as shown on the following pages.

 STEP 1
THE DEFINITION OF THERMALLY UPGRADED PAPER HAS BEEN
INTRODUCED AS THE IEEE TASK FORCE FORMULATED IT:
3.12
thermally upgraded paper
cellulose-based paper which has been chemically modified to reduce the rate at which the paper
decomposes. Ageing effects are reduced either by partial elimination of water forming agents (as in
cyanoethylation) or by inhibiting the formation of water through the use of stabilizing agents (as in amine
addition, dicyandiamide). A paper is considered as thermally upgraded if it meets the life criteria defined in
ANSI/IEEE C57.100; 50 % retention in tensile strength after 65 000 hours in a sealed tube at 110 ° C or any
other time/temperature combination given by the equation:

 15 000   15 000 15 000 

 − 28,082   − 
 ( ? h + 273 ) + ( + )
Time (h ) = e 
≈ 65 000 × e  h 
( ? 273 ) 110 273

Because the thermal upgrading chemicals used today contain nitrogen, which is not present in Kraft pulp,
the degree of chemical modification is determined by testing for the amount of nitrogen present in the
treated paper. Typical values for nitrogen content of thermally upgraded papers are between 1% and 4%
when measured in accordance with ASTM D-982.

NOTE This definition was approved by the IEEE Transformers Committee Task Force for the Definition of
Thermally Upgraded Paper on 7 October 2003.
 STEP 2
IEC HAS INTRODUCED A SUB-CLAUSE ILLUSTRATING THE BASIC
DIFFERENCE IN AGING RATE BETWEEN THERMALLY UPGRADED
AND NON-THERMALLY UPGRADED PAPER. THE SUB-CLAUSE
COMPRISES TWO ILLUSTRATIONS:
A) Sealed tube accelerated ageing in mineral oil at 150 °C:
 STEP 2. cont.
B) Insulation life measured in laboratory conditions at National Grid
Company:
Table 1 – Life of paper under various conditions

Paper type / ageing Life / years

temperature
Dry and free With air and
from air 2% moisture
Wood pulp at 80 °C 118 5.7
90 °C 38 1.9
98 °C 15 0.8
Upgraded 80 °C 72 76
wood pulp at
90 °C 34 27
98 °C 18 12
 STEP 3
IEC TOOK OVER THE RELATIVE AGEING RATE FORMULAE FOR
NON-THERMALLY AND THERMALLY UPGRADED PAPER FROM
THE EXISTING IEC AND IEEE LOADING GUIDES

- IEC: Relative ageing rate: V = 2 (? h − 98 ) / 6

Non-thermally upgraded

 15 000 15 000 
 
 110 + 273 − ? + 273 
- IEEE: Relative ageing rate: V = e h 
Thermally upgraded
 STEP 3 cont.
Table 2- Relative ageing rates due to hot-spot temperature

Hot-spot temperature Non-upgraded paper Upgraded paper

°C V V
80 0.125 0.036
86 0.25 0.073
92 0.5 0.145
98 1.0 0.282
104 2.0 0.536
110 4.0 1.0
116 8.0 1.83
122 16.0 3.29
128 32.0 5.8
134 64.0 10.1
140 128.0 17.2
 STEP 4
IEC INTRODUCED INSULATION LIFE USED IN THE PRESENT IEEE
LOADING GUIDE, FOR THERMALLY UPGRADED PAPER:
Table 3 – Normal insulation life of a well-dried, oxygen-free thermally upgraded insulation
system at the reference temperature of 110 °C

Basis Normal insulation life

Hours Years

50% retained tensile strength of insulation 65 000 7.42

25% retained tensile strength of insulation 135 000 15.41

200 retained degree of polymerization in 150 000 17.12

insulation
Interpretation of distribution transformer 180 000 20.55
functional life test data

No corresponding life figure was introduced for non-thermally

upgraded paper.
 SUMMARY
IEC INTRODUCED
- thermally upgraded paper, defined by the IEEE
Task Force,

- the equation for the relative ageing rate of

thermally upgraded paper,

- the 4 insulation life figures from the present IEEE

Loading Guide, for thermally upgraded paper.
 Part II

Transformer Insulation Upgrading

and Loading Guide Equations

T. V. Oommen
Consultant
 Transformer Coil Structure

(Medium/ Large Transformer)

IEEE Transformer Committee Panel Session – October 25, 2005

Source: ABB Power Technologies, Inc.

 Paper Insulation Aging
in Mineral Oil

DP DP DP DP DP DP

1000 733 549 405 309 181

End of Brittle & dark

Progressive aging with time mech str.

Effects of aging:
- darkening of color
- loss of electrical and mechanical strength; trans. failure
- shortening of cellulose chains – DP lowered
- paper becomes wetter, and acidic
- by-products contaminate the oil
Source ABB Power Technologies, Inc.
IEEE Transformer Committee Panel Session – October 25, 2005
 Source: T. V. Oommen & L. N. Arnold,15th Electrical Insulation Conf. Paper, 1981
“Cellulose Insulation Materials Evaluated by Degree of Polymerization Measurements”
Also, DEIS Videotape #12,”Cellulose Insulation Materials Evaluation” – T. V. Oommen
IEEE Transformer Committee Panel Session – October 25, 2005
Aging and decrease of mechanical strength & DP of Nonupg. Kraft paper

Relative value of indicated parameter

50% tensile strength 50% tensile str.: DP = 320

= 30% burst str.

30% tensile str. = DP 200

= 15% burst tr.

DP 200 DP = 320
DP = 0 DP = 1000
Relative ageing/DP

Ref. W. G. Lawson, M. A. Simmons, and P. S. Gale, “ Thermal Ageing of Cellulose Paper Insulation”,
IEEE Trans. Electrical Insul. Vol. EI – 12 (1977), pp. 61-66)
IEEE Transformer Committee Panel Session – October 25, 2005
 Definition of Thermal Upgrading of Paper

The slides to follow will examine this definition

IEEE Transformer Committee Panel Session – October 25, 2005

 Arrhenius Life Plots based on Life Equations

Ref. Temp. 110oC

180,000 hours
Lif
eP
65,000 hours lot
pe
Plo r1
t fo 99
r5 5L
0% oa
ten din
sil
es
gG
tre uid
gth n e Eq
en ua
dp tio
oin n
t life

The lower plot is for sealed tube aging test acceptance criterion

References: IEEE Std C57.100 – 1999

C57.12.00-1993

IEEE Transformer Committee Panel Session – October 25, 2005

Arrhenius Life Plots in Pre-1995 Loading Guides

180,000 hour life line

1995 limit
20.5 years (not based on tensile str.)

t)
tes
life
O Pre-1995 limit

on
ed
bas
.
ul .
ins

Ref Temp 110oC

ded
gra
up
ith
.w
ans
t, tr
dis
for
ot
(Pl

Note that the life at 110oC is nearly one-third of the revised life of 180,000 hours
IEEE Transformer Committee Panel Session – October 25, 2005
 Arrhenius Life Plots

P - Power trans.
D- Dist. trans.
PD- Power & Dist. Trans.

PD – 65 Upgraded paper:
new life plot (per 1995 Guide)
D - 65 Upgraded paper (pre-1995)
based on life test
P – 65 Upgraded paper (pre-1995)
old life plot based
on 50% tensile str.
D-55 Nonupgraded paper (pre-1995)
based on life test
PD - 55: Nonupgraded paper:
new life plot (per 1995 Guide)

These plots were generated from

Life equations in the Loading
Guides

Arrhenius Life Equations:

PD-65 (based on Loading

Guide 1995, exp eqn:
Log10Life (Hours)
= (6514.42/T) – 11.754
PD-55 (based on info. In
1995 Loading Guide):
Log10Life (Hours)
= (6514.42/T) – 12.447
IEEE Transformer Committee Panel Session – October 25, 2005
LIFE PLOTS COLLAGE
 DP vs. time plotting to get end point time

Example taken from Emsley et. al.

paper, “Degradation of Cellulosic
insulation in power transformers”.
Part 3: Effects of oxygen and water Raw data plotting
on ageing in oil”,
DP 200 end point line
IEE Proc. Sci. Meas. Technol.
May 2000.

Some authors such as Schroff & Semi-log plotting

Stannet used this method, but
could fit only 80% of the time DP 200 line
scale into linear plots.
See, IEE Paper Nov. 1985, “A review
of paper ageing in power
transformers”

The most reliable linear plots DP

are obtained by plotting 1/DP
vs. time. See also:
A. M. Emsley & G. C. Stevens,
“Degradation of cellulosic electrical
paper insulation in oil-filled transformers”
IEEE Proc-Sci. Meas. Technol., Sep 1994.

IEEE Transformer Committee Panel Session – October 25, 2005

 87,400 111,600

IEEE Transformer Committee Panel Session – October 25, 2005

 Some Experimental Data

Paper Ref. Temp 110C Time to reach Time to reach

50% tensile str. * DP 0f 200

A 140,000 hours 250,000 hours

B 93,000 168,000

C 90,000 160,000

* approx. DP of 300

All the papers met the 150,000 hours for a DP value of 200, but only one
met the 180,000 hour requirement. All the papers exceeded the 65,000
hour minimum life based on the tensile strength limit.

A & B are upgraded papers, but A is superior to B. C is non-upgraded

European paper almost matching B.

It obvious that papers that just met the minimum life based on the tensile
limit would have much lower life based on DP end point.

IEEE Transformer Committee Panel Session – October 25, 2005

Short term Test to check of Upgrading agent Effectiveness
1. Conduct a one month aging study at 150 oC (‘sealed tube’)
Get DP and plot 1/DP vs time and extrapolate to 1/DP = 0.005 equiv. to
DP of 200. Starting point should correspond to DP of 1000 (1/DP = 0.001)
Check time against life shown in Arrhenius plot. If there is reasonable
agreement, the upgrading may be considered satisfactory.

1/DP DP

0.005 200 (End)

0.0018 X 560 (measured)

0.001 1000 (new)
720 hours 4,431 hours
(1 month) (End point)
Time, Hours of Aging at 150o C

2. Conduct a nitrogen content of the paper; the value should be in the1.5 to 4.0% range,
preferably 2.0 – 4.0% range.

Use apparatus such as Perkin - Elmer PE 2410 model. Nitrogen compounds

are combusted to produce nitrogen which is measured by GC. Time needed: one hour or less.

IEEE Transformer Committee Panel Session – October 25, 2005

 Conclusions

1. Thermal upgrading of paper increases insulation life and allows

transformer loading class to be elevated from 55oC Rise to 65oC
Rise.

2. The amount of upgrading agent should be ensured by measuring

the nitrogen content of paper or by sealed tube tests based on a
DP end point of 200, and not based on 50% tensile str. end point.

3. To meet the present requirements for life of 180,000 hours or

150,000 hours at 110oC, the insulation paper must be adequately
upgraded. Meeting tensile strength requirement of 65,000 hours at
110oC does not ensure the prescribed life.

4. Papers should be evaluated by long term aging study using a DP

end point criterion, eg., 200. For routine checks, nitrogen content
measurement would suffice.

5. Appropriate changes must be made in the Standards and in the

definition of upgraded paper.

6. The Arrhenius life plots and life would be useful in the Guide.

IEEE Transformer Committee Panel Session – October 25, 2005

 Future Plans
• Today, the transformers committee has 3
different documents that contain references
to the minimum expected aging criteria.
• The definition to be included in C57.12.80;
• C57.100 “IEEE Standard Test Procedure for
Thermal Evaluation of Liquid-Immersed
Distribution and Power Transformers”;
• C57.91, The loading guide.
 Future Plans
• Our Objective:
• The requirements for the minimum life expectancy of the
insulation will appear in C57.12.00,
• The test procedures to determine the aging characteristics
and the minimum life expectancy of a proposed insulation
system will be provided in C57.100,
• The loading guide which is a tool to provide guidance for
utilizing the transformer in a prudent manner that
recognizes the effects of various loading scenarios on the
minimum life expectancy. As a guide, it can not contain
any requirements, and will merely reflect the information
and limits established in the other 2 documents.