A2 Power transformers

and reactors
Power system
C4 technical
performance

High-Frequency Transformer
and Reactor Models for
Network Studies
Part E: Measurements and transformer
design details
TECHNICAL BROCHURES
April 2023 - Reference 904
 TECHNICAL BROCHURE

High-Frequency Transformer
and Reactor Models for
Network Studies
Part E: Measurements and transformer
design details
JWG A2/C4.52

JWG members contributing to this Technical Brochure

B. GUSTAVSEN, TF Leader NO
R. RONCHI MX
A. PORTILLO PT

JWG Members
B. GUSTAVSEN, Convenor N A. ROCHA, Secretary BR O. STERZ, M DE
A. PORTILLO, M UY J. VEENS, M NL R. CASTRO LOPES, M PT
A. HOLDYK, CM NO J. CARLOS MENDES, CM BR R. DEGENEFF, M US
A. PALANI, CM DE J. FRANCISCO LOFRANO, CM BR R. RONCHI, M MX
B. VALECILLOS, M CK J. MONTANHA, M BR R. AZEVEDO, M BR
B. KORDI, M CA L. FERNANDO DE OLIVEIRA, CM BR S. JAMIL, M UK
B. ANDRIIENKO, M UA M.O. ROUX, M CA T. RÖHRL, CM DE
C.GONZÁLEZ-GARCÍA, M ES M. OSTRENKO, M UA G. ANDRES DIAZ FLOREZ, CM CO
C. ÁLVAREZ-MARIÑO, CM ES M. RIOUAL, M FR G. H. C. OLIVEIRA, M BR
D. MATVEEV, M RU M. FROLOV, CM RU T. NGNEGUEU, M FR
D. VUJATOVIC, M UK E. MOMBELLO, CM AR X.LÓPEZ-FERNÁNDEZ, M ES
E.RAHIMPOUR, M DE E.PORTALES, CM CA H.K. HØIDALEN, M NO
J.H. KIM, M KR F. PORTILLO, CM UY HANS DE HERDT, M BE

CM = Corresponding Member

Copyright © 2023
“All rights to this Technical Brochure are retained by CIGRE. It is strictly prohibited to reproduce or provide this publication in any
form or by any means to any third party. Only CIGRE Collective Members companies are allowed to store their copy on their
internal intranet or other company network provided access is restricted to their own employees. No part of this publication may
be reproduced or utilized without permission from CIGRE”.

Disclaimer notice
“CIGRE gives no warranty or assurance about the contents of this publication, nor does it accept any responsibility, as to the
accuracy or exhaustiveness of the information. All implied warranties and conditions are excluded to the maximum extent permitted
by law”.

WG XX.XXpany network provided access is restricted to their own employees. No part of this publication may be
reproduced or utilized without permission from CIGRE”.

Disclaimer notice
“CIGRE gives no warranty or assurance aboutISBN : the contents of this publication, nor does it accept any
978-2-85873-609-6
responsibility, as to the accuracy or exhaustiveness of the information. All implied warranties and
conditions are excluded to the maximum extent permitted by law”.
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Executive Summary
General
Lightning and electrical system generated overvoltages are some of the main causes leading to the
failure of transformers and shunt reactors (WG A2.37 - 2015).
These components are being subjected to switching overvoltages more frequently than in the past, due
to a new grid operation environment and with the introduction of intermittent (e.g renewable) energy
sources in the grid. In addition, the overvoltage waveshapes that excite the transformer/reactor's
terminals are changing with the introduction of altered grid configurations, the more extensive use of
GIS, and new technologies (e.g. vacuum circuit breakers at medium and higher voltages), often resulting
in more severe stresses on their insulation, especially the internal one.
The ability to predict the overvoltages that appear on the transformer's terminals requires the use of an
ElectroMagnetic Transient (EMT) program that is capable of representing the relevant equipment with
sufficient accuracy. It was pointed out in a previous working group (CIGRE JWG A2/C4.39 - 2014) that
the existing transformer models as available in such programs are in many situations not well-suited for
those new applications, when calculating the impinging (external) overvoltages or the resulting internal
voltage stresses.
The current JWG (A2/C4.52) was therefore established to assess the possibility of using more advanced
transformer models. The application focus is on the overvoltages on the transformer itself (externally
and internally), but models for use in general studies are also considered where the transformer is not
the center of focus. In addition to the modeling approaches themselves, the JWG has proposed file
formats for efficient model transfer and utilization in simulation programs. Extensive measurements for
model validation has also been performed on two power transformers whose designs have been
released by the manufacturer. The measurements and the designs have been utilized by the JWG for
validation and improvement purposes. This information (measurements, design) has also been placed
on an open web site, allowing for the first time validation of transformer modeling methods to be
performed by anyone, e.g. manufacturers, universities, and research institutes
The results from this JWG are documented in five Technical Brochures (TBs), covering models (white-
box, grey-box, black-box types), their inclusion in EMT programs, and validating measurements on two
different transformers.

This TB
This TB covers the part of the activity related to validating measurements. It describes measurements
that were performed on two transformers, the detailed design information of these transformers, and
examples of white-box time domain calculations based on the design data. The measurements and
design data are utilized in the other TBs written by this JWG, in particular the TB dealing with white-
box models. The measurement data and design information can be downloaded from an open web site,
thereby being a source for future WGs and for research on transformer high-frequency modeling in
general.

The TB describes details of the applied measurement setup, including a procedure for calculating
1.2/50 µs lightning impulse waveforms from frequency sweep measurements. For each transformer,
lightning impulse voltage responses were measured on external terminals and on internal points in the
regulating winding, for alternative terminal conditions: variants of open terminals and grounded
terminals, and alternative points of voltage application. The TB documents the applied terminal
conditions and provides plots of the resulting voltage waveforms. The waveforms in digital form can be
downloaded from a web-site. Additionally, terminal admittance measurements were carried out, also
documented in this TB and available in digital form on the web site. The design data of the two
transformers are described in detail, thereby permitting the reader to calculate his own white-box model,
which can be validated against the measurements: time domain voltage responses and frequency
domain admittance responses.

3
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

In Memorial
The JWG wishes to dedicate this work to Robert Degeneff who passed away during the JWG activity.
Robert was pioneering in developing high-frequency transformer models and shared generously his
knowledge and experience with the engineering community for more than thirty years.

4
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Contents
Executive Summary ............................................................................................................ 3

1. Introduction................................................................................................................ 7
1.1 Background, JWG activities, results........................................................................................................ 7
1.2 Transformer measurements and design details ................................................................................... 10
1.3 Overview of Brochure Content ............................................................................................................... 10
1.4 Acknowledgements ................................................................................................................................. 11

2. Transformer Units.................................................................................................... 12

3. Measurement approach ........................................................................................... 13

3.1 Transformer condition ............................................................................................................................ 13
3.2 Voltage transfer measurements ............................................................................................................. 14
3.2.1 General .............................................................................................................................................. 14
3.2.2 Measurement setup ........................................................................................................................... 15
3.2.3 Conversion from frequency domain to time domain........................................................................... 15
3.3 Admittance measurements ..................................................................................................................... 17
3.3.1 Measurement setup ........................................................................................................................... 17
3.3.2 Calibration ......................................................................................................................................... 18

4. Measurements on 3-ph transformer ....................................................................... 19

4.1 Internal and External connections ......................................................................................................... 19
4.2 Voltage transfer measurements ............................................................................................................. 20
4.3 Checking for erroneous connections .................................................................................................... 21
4.4 Admittance measurements ..................................................................................................................... 22

5. Measurements on 1-ph transformer ....................................................................... 24

5.1 Internal and external connections.......................................................................................................... 24
5.2 Voltage transfer measurements ............................................................................................................. 25
5.3 Admittance measurements ..................................................................................................................... 27

6. Measurement Data files ........................................................................................... 28

6.1 Voltage transfer measurements ............................................................................................................. 28
6.2 Admittance measurements ..................................................................................................................... 29
6.2.1 Measurement files ............................................................................................................................. 29
6.2.2 Measurement conditions and terminal numbering ............................................................................. 30
6.3 Download site .......................................................................................................................................... 30

7. Comparison between measurements and white-box model simulations ............ 31

7.1 time domain voltage responses ............................................................................................................. 31
7.1.1 Single-phase transformer .................................................................................................................. 32
7.1.2 Three-phase transformer ................................................................................................................... 34
7.2 Frequency domain admittance matrix elements ................................................................................... 36
7.2.1 Single-phase transformer .................................................................................................................. 36
7.2.2 Three-phase transformer ................................................................................................................... 37
7.3 Analysis of results ................................................................................................................................... 38

5
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

7.4 Time domain calculation data files ........................................................................................................ 38

7.5 Frequency domain calculation data files ............................................................................................... 41

8. Design details of 3-ph transformer ......................................................................... 43

8.1 Electrical Characteristics ........................................................................................................................ 43
8.2 Core – “Three-Legs Core-Type” ............................................................................................................. 44
8.3 TV Winding – Stabilizing Winding – “Helical Type” ............................................................................. 45
8.4 LV Winding – “Helical Type” .................................................................................................................. 46
8.5 HV Principal Winding – “Continuous Disk Type” ................................................................................. 47
8.6 RW Regulating Winding – “Multi-start Type” ........................................................................................ 48
8.7 Core & Windings Layout ......................................................................................................................... 50
8.8 Tank – “Rectangular with Radiators” .................................................................................................... 50
8.9 Tertiary Reactor ....................................................................................................................................... 51
8.10 Losses and Impedance Measured in Factory Acceptance Tests (FAT) .............................................. 52
8.11 Permittivities ............................................................................................................................................ 53
8.12 Nodes and Branches Numbering for Single Phase Modelling............................................................. 53
8.13 Nodes and Branches Numbering for Three Phase Modelling ............................................................. 57

9. Design details of 1-ph transformer ......................................................................... 65

9.1 Electrical Characteristics ........................................................................................................................ 65
9.2 Core – “1 Wound Leg with 2 Return Legs Core-Type” ......................................................................... 66
9.3 TV Winding – “Helical Type” .................................................................................................................. 67
9.4 LV Winding – “Continuous Disk Type” .................................................................................................. 68
9.5 HV Principal Winding – “Interleaved Disk Type” .................................................................................. 69
9.6 RW Regulating Winding – “Interleaved Disk Type” .............................................................................. 70
9.7 Core & Windings Layout ......................................................................................................................... 72
9.8 Tank – “Rectangular with Radiators” .................................................................................................... 72
9.9 Losses and Impedance Measured in Factory Acceptance Tests (FAT) .............................................. 73
9.10 Permittivities ............................................................................................................................................ 73
9.11 Nodes and Branches Numbering ........................................................................................................... 74

10. recommendations .................................................................................................... 77

11. References ............................................................................................................... 78

Appendix A. Impulse Plots for 3-ph transformer ............................................................ 79

Appendix B. impulse plots for 1-ph transformer ........................................................... 108

6
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

1. Introduction
1.1 Background, JWG activities, results
Background
Transient overvoltages are one of the root causes for transformer dielectric failures [1]. A previous JWG, CIGRE
JWG A2/C4.39 "Electrical transient interaction between transformers and the power system" [2], [3] gave an
overview of transient phenomena that can lead to insulation failures, such as steep-fronted waves and resonant
voltage build-up. The group further gave an overview of modeling techniques and it provided a collection of study
examples that demonstrated the use of transformer models in overvoltage studies.
It was concluded, as a feedback from the field and failure analysis, that the use of the standard lightning and
switching impulse tests may not be fully adequate for ensuring the insulation dielectric withstand against every
type of transient overvoltage that a power transformer or shunt reactor could experience in the field. In order to
ensure the withstand capability of a transformer in service, it is in general necessary to perform a two-step
procedure. In the first step, one performs an analysis of the overvoltages that excite the transformer terminals. In
the second step, one calculates the resulting internal voltages in the transformer at a set of nodes. Such analysis
could in principle result in a reinforcement or modification of the transformer design (if the transformer was not
already built), or to a modification of the transformer's environment (e.g. by use of additional overvoltage protection
or change in operational procedures).
In principle, the white box-model is eminently suited for such calculations since it can calculate both external and
internal overvoltages. Such model is established by the manufacturer, based on the transformer's detailed design
data. The use of white-box models would be of particular value in many of the new types of studies to be performed,
e.g. the introduction of vacuum circuit-breakers at higher voltage levels and their reignition effects on the
transformers. It was however found in JWG A2/C4.39 that the white-box models used by transformer manufacturers
give different results when applied to a common geometry with known material properties, although the results by
those different models would more or less lead to a similar design with respect to currently applicable standards.
Application to this so-called "fictitious transformer" showed that while the white-box models in the case of a
lightning impulse test can predict the maximum voltage stress at given positions in the winding with good accuracy,
they may be quite inaccurate in general transient studies due to inaccurate representation of resonance frequencies
and their damping. It was therefore recommended to look further into the reason for the model discrepancies.
There is also a general need for accurate terminal equivalents of the transformer, without the ability to calculate
internal voltages. One such example is the calculation of transient recovery voltage (TRV) associated with a
transformer limited fault, which may pose a severe condition for the circuit breaker. JWG A2/C4.39 accordingly
described other transformer models, known as black-box and grey-box models.
As for the black-box model type, it is often capable of reproducing quite accurately the transformer's terminal
behavior when the input data comes from frequency sweep measurements on the transformer. However, the
measurements and subsequent model extraction can at present be done only by experts with the use of customized
equipment, and there are issues with error magnification as well, depending on the vector group. Therefore, a
recommendation on measurements and model extraction procedures is needed in order for measurement-based
black-box modeling to become more widely applied. The use of standard SFRA-measurement results is not
recommended for this purpose.
As for the grey-box model type, it is probably the most widely applied model type in system studies because of its
simplicity and multiple variants, ranging from quite simple models to ladder-type models. It is however not clear
how to establish some of the required parameters, e.g. the capacitance values to be used in simple models. JWG
A2/C4.39 further presented some results for one ladder-type grey-box model, but it was not clear to what extent
its parameters could be determined with sufficient accuracy and within an acceptable computation time. In principle,
a complete ladder model could be used for determining the internal overvoltages of a transformer more accurately
than a white-box model, by optimizing the parameters of a white-box model by measurements. As the success of
such approach had not yet been demonstrated, the JWG therefore sought to look into that possibility as well.
JWG activities
In 2014, CIGRE formed a new group, JWG A2/C4.52 "High-frequency transformer and reactor models for network
studies", to follow up the work of A2/C4.39. The scope of A2/C4.52 is focused on transformer modeling techniques
in more detail, with an objective to identify the most suitable models and to make the models available to the end
users (utilities, industry) for application in general transient overvoltage studies. The JWG has considered white-
box, black-box and grey-box type models, and within each type evaluated methods and procedures for
determinating the model's parameters. The JWG has further proposed model formulations and associated file
formats to facilitate the use of the models in EMT simulations programs. The JWG has also performed
measurements on two three-winding transformers whose detailed designs have been released to the JWG. The
measurements were used internally in the JWG to evaluate the accuracy of various modeling approaches. The raw
data (measurements and detailed design information) has been released to the public on an open web site.

7
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

The work within the JWG was conducted by five subgroups (Task Forces). Each subgroup has produced a separate
TB as follows.

Table 1.1 Technical brochures delivered from the JWG

TB subtitle Content
Part A. Transient models based on detailed design knowledge of transformer,
White-box models extensive lumped parameter models.
Part B. Measurement of terminal behavior, calculation of terminal behavior from
Black-box models white-box model, model extraction methods, application examples.
Part C. Different types of grey-box models (simplified and complex). Model typical
Grey-box models parameters . Case studies.
Part D. Model formulations, file formats for transfer of model to end user, guidelines
Model interfacing and on model requirements, guidelines on model specifications.
specifications
Part E. Measurements on two three-winding three-phase transformers,
Measurements and measurement files, transformer design details, comparison with white-box
transformer design details model simulation results.

Results
The results from the five task forces are given in the respective Technical Brochures, see Table 1.1. The main
results can be summarized as follows.
 Part A. White-box modelling. Transformer white-box (WB) models are normally proprietary models
(software) used by the manufacturers to verify the winding's ability to withstand the standard test voltages
used in the factory tests, i.e. the standard lightning impulse voltage (full and chopped waves) and
sometimes also switching impulse waves. The WB models can be categorized as lumped parameters circuit
models based on a spatial discretisation of the windings, or distributed parameters models based on
traveling wave-type approaches. The extraction of the model’s parameters is in all cases based on a
detailed description of the transformer’s geometry and material properties. The model parameters are
calculated via formulae and/or finite-elements computations. To evaluate the state of the art of WB
models, the task force used measurements on two transformers manufactured by WEG in Mexico that
were measured in the time and frequency domain (Part E). The complete design of both transformers was
shared with task force members and the calculated results with the WB member models were compared
with the measurements. In the time domain the results demonstrate a quite good agreement in the
prediction of the maximum voltages values that appears in the measured points inside the transformers
but a poor agreement in the temporal wave shapes which are dependent on the natural (resonance)
frequencies of the transformer. The task force members also calculated, using their white-box models, the
admittance matrix in the frequency domain for both transformers, to compare with the admittance
matrices measured in the frequency domain in the transformers in México, and the calculated results then
show a better agreement than in time domain. The comparison shows the importance of considering the
damping variation with frequency to improve the calculated results and the most promising way to
calculate the parameters is using frequency scan with finite elements using the principle of the complex
permeability followed by vector fitting and synthesis in a time domain of an equivalent circuit using series
Foster circuits or coupled magnetic circuits in each branch of the transformer. Another challenge is that
the manufacturers may be reluctant to give out detailed models as they may reveal proprietary design
information. It may also initiate discussions to the internal validation rules, which are considered
proprietary information. Therefore, white-box models will normally be passed to their customers in the
form of a terminal equivalent only.

 Part B. Black-box modeling. A black-box (BB) model is a terminal equivalent of the transformer, possibly
in combination with voltage transfer functions to unloaded nodes. A BB model can be calculated from
frequency domain sampled data which are subjected to a model extraction method. The sampled data
can come from frequency sweep measurements using dedicated quipment, or be calculated from a white-
box model. The TB describes details on how to perform suitable measurements while removing systematic
errors from the measurement. Once the measurement has been completed, a model is to be extracted
that satisfies certain physicality conditions that ensure a stable time domain simulation. A number of
suitable extraction methods are presented in the TB, along with methods for cross-validation of both

8
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

measurements and the extracted model. One major challenge with the BB approach is the potential
magnification of small measurement and modeling errors when the model is applied in simulation studies.
The challenges generally increase with the number of considered windings and phases, in particular when
some winding is ungrounded and when the transformer has a high voltage ratio. The TB describes ways
of alleviating the problems, by means of additional measurements and by usage of similarity
transformations. The methods are demonstrated by a number of examples, including measurements from
the Part E TB. It is also shown how a black-box modeling approach can be applied to white-box models
to obtain a pure terminal equivalent, for use in studies of the the interaction between the
transformer/reactor and the network. Currently available sFRA measurement setups are not well suited
for transformer black-box modeling.

• Part C. Grey-box modelling. A grey-box (GB) model is a topologically correct model fitted to terminal
measurements. The TB further divides GB models in simplified and ladder models. The simplified models
can be easily fitted to test-report or frequency response measurements (or even typical values), while the
ladder models require a rigorous fitting procedure. The simplified models provide only terminal response,
while the ladder models in theory can also reveal internal winding response. Some of the simplified models
are valid for low frequencies up to the first resonance frequency and all have a capacitance behaviour at
high frequencies. None of the models are accurate around the first resonance frequency where the
frequency behaviour often is complicated with several resonances. In this frequency range a black-box or
white-box should be considered. The TB defines five simplified models (M1-M5) and procedures for how
to fit these to typical values, test report and frequency response measurements. An example on how to
model the two test transformers and how their response fits to a BB based on measurements is provided.
Several examples on transient recovery voltage shows that the simplified models have limited accuracy
unless proper damping is introduced. Scaling of measured capacitance and resistive damping factor are
outlined in the TB. The detailed, fitted ladder network is still an area of research and needs more
exploration before conclusion about accuracy and application can be made. In general, the cases reported
in literature are simple, single-phase, air-core windings.

• Part D. Model interfacing and specifications. In order to properly utilize the models presented in Parts A ,
B and C, it must be possible to include them in an appropriate ElectroMagnetic Transients (EMT) program.
The simplified grey-box models can be included in such programs directly in the form of a small circuit
with known circuit parameters. The white-box and black-box models are however high-order models that
involve a large number of parameters. Fortunately, these models can be formulated in a rather standard
way, either as a pole-residue model or a state-space model. The ability of EMT programs to read in the
information and utilize it is however incomplete at the time of writing. The TB therefore defines a set of
ASCII files with a predefined format that the models should be written to. The software vendors on their
side should adapt their simulation tools to implement the proposed models when missing, and to read the
parameters from these files. That way, the model formats become independent of the applied EMT
program. The TB also provides guidelines that can assist a transformer buyer in choosing the most
appropraite model for the intended application. A set of alternative specifications are provided that can
assist in the detailed model specification. The suggested file formats and specifications do not represent
a "norm" in any kind, rather it is an attempt at facilitating the process of efficient model transfer from
manufacturer to end user, as well as the inclusion of these models in EMT simulation programs.

• Part E. Measurements and transformer design details (this TB). In order to assess the accuracy of the
various transformer modeling approaches, the JWG performed extensive frequency domain measurements
on two three-winding power transformers. The measurements included terminal admittance and voltage
transfer between terminals and from terminals to three internal points in the regulating winding. The
voltage transfer data was transformed into 1.2/50 µs time domain waveforms via rational modeling and
convolution. The manufacturer also released the detailed design information of the transformers, thereby
enabling, for the first time, white-box models to be established and checked against measurements. The
measurement data and the detailed design data can be freely downloaded from
www.sintef.no/CIGREtransformerdata.

9
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

1.2 Transformer measurements and design details

CIGRE JWG A2/C4.52 has the general objective of providing high-frequency transformer models suitable
for transient simulations in the power system. One of the main objectives is to assess the accuracy of
detailed models ("white-box") used by manufacturers. With this aim, it is necessary to perform
measurements on real transformers whose detailed internal geometry is known.
WEG Transformers offered the possibility to A2/C4.52 to perform measurements on two transformers
at its factory in Mexico City, Mexico, whose detailed geometry description can be released to A2/C4.52
members. By necessity, it was required that such measurements should be performed by an
independent third party. SINTEF Energy Research was asked to perform the actual measurements and
accepted the request. The transformers were a 1-ph transformer and a 3-ph transformer, both having
3 windings.
The measurements were performed on May 30-June 12, 2016, by Bjørn Gustavsen and Oddgeir Rokseth
(SINTEF). Alvaro Portillo (independent consultant) and Rodrigo Ronchi (WEG) had in advance set up a
detailed specification for voltage transfer measurements to be performed, between external terminals
and to internal points in the regulating winding. This specification included point of excitation, tap
position, grounding conditions and which voltages to measure. While on-site, SINTEF performed a set
of admittance measurements to give further input to the white box model validations, as well as input
to black-box and grey box activities in the group. Rodrigo Ronchi provided local support for the
measurements.
This document describes the measurement procedure, measurement conditions, and which quantities
were measured. Also is described a set of data files which contains the actual measurements, available
for download. The transformer's design details are also included. These design details were applied in
a separate Task Force ("White-box") where the members independently calculated white-box models
and then calculated transient waveforms that were compared against measurements.

1.3 Overview of Brochure Content

 Chapter 2 gives an overview of the two transformer objects being subject to measurements: a
3-ph transformer and a 1-ph transformer.
 Chapter 3 describes the applied measurement approach: Measurements setups, voltage transfer
measurements, and admittance measurements. Also is described a procedure for converting
the frequency domain voltage transfer measurements to 1.2/50 µs time domain responses.
 Chapter 4 describes the measurements performed on the 3-ph transformer: Connections,
voltage transfer measurements and terminal admittance measurements.
 Chapter 5 describes the measurements performed on the 1-ph transformer: Connections,
voltage transfer measurements and terminal admittance measurements
 Chapter 6 describes the data files containing the measurements, and information about
downloading.
 Chapter 7 shows examples of comparisons between measurements and simulations performed
by the white-box Task Force (TB Part A). Also is described the usage of (downloadable) Excel
files for facilitating the comparison
 Chapter 8 gives design details for the 3-ph transformer.
 Chapter 9 gives design details for the 1-ph transformer.
 Chapter 10 gives recommendations about future usage of the measurements and design data.
 Appendix A shows 1.2/50 µs impulse voltage plots obtained from measurements on the 3-ph
transformer.
 Appendix B shows 1.2/50 µs impulse voltage plots obtained from measurements on the 1-ph
transformer.

10
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

1.4 Acknowledgements

The measurements would not have been possible without the support from companies as described in
Table 1.1.

Table 1.2 Company contributions - measurements

WEG Transformers MX - Release of detailed design information of the two transformers

- Providing access to transformers for measurements
- Support during measurements
- Covering of local costs
SINTEF Energy NO - Planning
Research - Performing measurements
- Postprocessing of results
Ing. Álvaro Portillo - UY - Planning
Transformer - Check of measurements by white-box model
Consultant
Elvia NO Financial support, main contributor
Hitachi Energy DE Financial support
GE FR Financial support
CG Power Systems BE Financial support
EdF FR Financial support
MR DE Financial support
SMIT Trafo NL Financial support

Following the measurements, the design data was sent to the CIGRE JWG A2/C4.52 members (White-
box Task Force – TB Part A) and subjected to modeling by their company specific computer programs.
The results were systemized and compared against measurements, see Table 1.2.

Table 1.2 Company contributions – design parameter derivation and calculations

Ing. Álvaro Portillo - UY - Derive design parameters, submit to WG members

Transformer
- Collect calculation results from WG members, compare against
Consultant
measurements

11
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

2. Transformer Units
Some of the name-plate data of the two transformers are listed in Fig. 2.1.

3-ph transformer

1-ph transformer

Figure 2.1 Rated voltage and power

The three-phase unit is a 92/92/30.67 MVA three-winding transformer with rated voltages
115/34.5/13.8 kV at 60 Hz with internal connections shown in Fig. 2.2, left. The transformer is of variable
induction, as the on-load tap changer (OLTC) is in the high voltage winding, to regulate the voltage in
low voltage winding (+9/-11 Taps), with a total of 21 tap positions. The stabilization tertiary winding is
connected in series with a reactor to limit the short circuit current.
The single-phase unit is a 50/50/16.67 MVA three-winding transformer with rated voltages 230/69/13.8
kV at 60 Hz with internal connections shown in Fig. 2.2, right. The on-load tap changer (OLTC) has 11
tap positions with reversible polarity, giving a total of 21 tap positions.

Fig. 2.2. Left: three-phase transformer (one leg of); Right: single-phase transformer.

12
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

3. Measurement approach
3.1 Transformer condition
With both transformers, all measurements were performed with the transformer in tank, but without
oil, top lid and bushings (Fig. 3.1). The transformer terminals were brought to the tank rim (Fig. 3.2)
and made available for measurements. A braided wire was clamped on to the tank rim (unpainted, bare
steel) and used as ground reference, see Fig. 3.3.

Voltages at internal points were measured using unshielded connections that were brought to the
transformer rim (Fig. 3.4.).

Figure 3.1 Transformer tank with active part. External work platform for measurements

Figure 3.2 Bringing transformer terminals to tank rim

13
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

HV Short-
terminal Clamp
circuiter

Braided wire

Figure 3.3 Grounding reference at tank rim using braided wire

Figure 3.4 Measurement leads to points in tap changer, in 1-ph transformer (R1, R5)

3.2 Voltage transfer measurements

3.2.1 General
Most of the voltage transfer measurements were performed in the frequency domain. There are two
reasons for this choice.
1. Due to the large number of voltage transfer measurements that were requested, it was
considered convenient to make the responses available with identical excitations, e.g. the
1.2/50 µs wave shape. That way, the checking of manufacturer's models can be more rapidly
performed. By use of frequency domain measurements followed by convolution in the time
domain, these wave shapes can be easily created as shown in Chap. 3.2.3.
2. In the case of the 3-ph transformer, the distance between terminals was too large for standard
voltage probes to be used, due to their short lengths. Therefore, it was considered difficult to
use the four inputs of a storage oscilloscope simultaneously. However, two probes with long
leads were available. They are suitable for frequency domain measurements because only two
probes are needed in such measurement.

14
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Compared to a time domain measurement, the disadvantage of the frequency domain measurements
is the longer time consumption of the measurements.

3.2.2 Measurement setup

The measurements were performed in the frequency domain using a Vector Network Analyzer (VNA)
together with two identical voltage probes. The probes were adjusted such that their voltage ratio was
unity in the entire frequency band. The measurements used 601 frequency samples that were
logarithmically spaced between 50 Hz and 10 MHz. Fig. 3.5 shows the connection of the two probes:
reference (right) and response (left). The signal was brought from the VNA using a coaxial cable. A few
time domain step response measurements were also performed for validation purposes, more about
this in Section 3.2.3.

Figure 3.5 Voltage transfer measurements using two identical voltage probes

3.2.3 Conversion from frequency domain to time domain

The following text is partly a verbatim copy from [4].
The frequency domain measurements were performed from 50 Hz to 10 MHz using K=601
logarithmically spaced samples. Each response h() was fitted by a pole-residue type rational model
(3.1) using the method of vector fitting [5]. The coefficients (ai, ri) are real or complex conjugate (causal
function), and all poles (ai) are in the left half-plane (stable).
N
ri
h( )  r0   (3.1)
i 1 j  ai
The impulse response (3.1) is in the time domain given as a sum of exponential functions,
N
h(t )  r0 (t )   re
i
ai t
(3.2)
i 1

The response y(t) due to an excitation u(t) can now be calculated as

y (t )  h(t )  u (t ) (3.3)

By discretizing the relation (3.3) with a fixed time step length t using the frame of recursive convolution
[6], the response at a time point n can be calculated as

15
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

xn  αxn1  βun (3.4a)

yn  cT x n   un (3.4b)

where xn is a state vector of length N.  is a diagonal matrix,  is a column vector and cT is a row-
vector. The calculation of the coefficients in (3.4a), (3.4b) is given in [7].

Fig. 3.6 shows one of the fitted frequency responses. It is observed that the many oscillations are well
captured by the model.

Figure 3.6 Measured transfer voltage and its approximation with a rational model

Fig. 3.7 shows an example where the model in Fig. 3.6 has been used with a measured voltage excitation
("Applied") as input. The output response of the model ("Convolution") is compared with the directly
measured time domain response. It is observed that a very good agreement is achieved.

Having verified the accuracy of the approach, the voltage response due to any exciting signal can be
calculated. Fig. 3.8. shows the calculated voltage response due to a 1.2/50 µs voltage excitation which
is chopped at t=20 µs.

16
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Figure 3.7 Validation of model (convolution) accuracy against a time domain measurement

Figure 3.8 Voltage response due to chopped lightning impulse wave

3.3 Admittance measurements

3.3.1 Measurement setup
A limited number of admittance measurements were performed on the two transformers using a setup
similar to the one in [8], see Fig. 3.9a and 3.9b. This setup features a VNA and a wide-band current
sensor, measuring the elements of the admittance Y as function of discrete frequency. The
measurements used 601 frequency samples that were logarithmically spaced between 50 Hz and
10 MHz

17
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Figure 3.9a Measurement setup: VNA, connection box w/ current sensor, coaxial cables

Figure 3.9b Connection of coaxial cables to transformer terminals (1-ph transformer)

3.3.2 Calibration
The measurements used 3-m long cables between the connection box and the transformer terminals.
The effect of these cable was eliminated from the measurements using the so-called "transmission line
method" [9]. Also, the insertion impedance effect and imperfect current ratio of the current sensor was
accounted for. These techniques for removal of systematic errors are described in Chap. 4.7 in TB
Part B, and in more detail in [9] and [10].

18
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

4. Measurements on 3-ph transformer

4.1 Internal and External connections
Fig. 4.1 shows the winding connections of the 3-ph transformer. The measurement conditions will be
described with reference to the labelling in the figure. The external connections for measurements can
be observed in Fig. 4.2.

Figure 4.1 Internal connections and labelling

Figure 4.2 External connections for measurement

19
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

4.2 Voltage transfer measurements

Fig. 4.3 shows the measurements that were performed. Each row defines one case by
 Tap position (2nd column)
 Applied voltage terminal (red font)
 Open/grounded terminals (black font)
 Measured quantities (blue font)

It is noted that the measurements are performed with four alternative tap positions
 Max, Nom+, Nom- and Min

and four different points of excitation,

 H1, H0, X1, X0.

It is also seen that in all cases, the voltage transfer to the two extreme ends of the regulating winding
(R1, R11), was measured.

Y1R and Y2R are internal points in the tertiary, behind the reactor.

Of the 64 cases, seven were not performed due to time restrictions (case 51, 52, 56, 59, 60, 63, 64).
Those seven are crossed with a horizontal line in Fig. 4.3.

As an example, Fig. 4.4 shows the 1.2/50 µs voltage transfer functions for Case 2.

20
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Figure 4.3 Measurements on 3-ph transformer: Excitation, grounding conditions, measured quantities

Figure 4.4 1.2/50 µs responses for Case 2

4.3 Checking for erroneous connections

The convolution approach was also used for validating the connections of the transformer, e.g. the
correct connection to internal points. This was achieved by comparing a few impulse responses with
those calculated by Portillo's white-box model. Fig. 4.5 shows one such comparison (Case 19). From
this comparison it was concluded that there were no connection errors.

21
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Figure 4.5 Comparison of measured responses with white-box calculations

4.4 Admittance measurements

The admittance matrix was measured with respect to six of the external terminals as defined in
Table 4.1, in Tap position Nom-.

Fig. 4.6 shows a plot of the measured admittance matrix elements.

Table 4.1 Terminal numbering and conditions

Terminal Node number Comment

H1 1
H2 2
H3 3
H0 Grounded, ignored
X1 4
X2 5
X3 6
X0 Grounded, ignored
Y1 Open, ignored
Y2 Open, ignored
Y3 Open, ignored

22
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Figure 4.6 Admittance matrix elements (magnitude functions)

In order to validate the accuracy and consistency of the data set, a common-pole type rational model
was calculated which is both stable and passive. Fig. 4.7 shows that a very accurate fitting has been
achieved using N=120 pole-residue terms. A time domain validation of the model's accuracy was shown
in [4].

Figure 4.7 Fitting admittance model with rational model (stable, passive)

23
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

5. Measurements on 1-ph transformer

Measurements were performed also on the 1-ph transformer, in a similar way as for the 3-ph
transformer. One notable difference is that the voltage transfer to point 5 in the regulating winding
(near mid-point) was also measured.

5.1 Internal and external connections

Fig. 5.1 shows the transformer internal points along with the labelling used for node identification. The
external connections for measurements can be observed in Fig. 5.2.

Figure 5.1 Internal connections and labelling

Figure 5.2 External connections for measurement

24
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

5.2 Voltage transfer measurements

The table in Figs. 5.3a and 5.3.b show the measurements that were performed. Each row defines one
case by
 Tap position (2nd column)
 Applied voltage terminal (red font)
 Open/grounded terminals (black font)
 Measured quantities (blue font)

The measurements were performed with four alternative tap positions

• Max, Nom+, Nom- and Min

and four different points of excitation,

• H1, H0, X1, X0.

In all cases, the voltage transfer to the two extreme ends of the regulating winding (R1, R11), was
measured as well as to a point near the regulating winding mid-point (R5). All 64 cases were done.
• Fig. 5.4 shows an example of measurements with time response extracted through
convolution.

Figure 5.3a Measurements (first part) on 1-ph transformer: Excitation, grounding conditions, measured
quantities

25
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Figure 5.3b Measurements (second part) on 1-ph transformer: Excitation, grounding conditions,
measured quantities

Figure 5.4 Example of measured responses (case #15)

26
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

5.3 Admittance measurements

The admittance matrix Y() was measured with respect to the four external terminals, with H0 and X0
grounded. The measurements were done with the transformer in Tap position Nom+, see Fig. 5.5. Due
to the presence of a floating tertiary winding, a direct measurement of terminal elements will not
properly capture the high-impedance common mode component associated with the tertiary winding.
The measurements were therefore complemented with common-mode measurements and modified as
described in [11]. The measurement data in Chap. 8 are the data that result after the processing.

Figure 5.5 Admittance matrix elements (magnitude functions). (From B. Gustavsen, A. Portillo, R. Ronchi,
A. Mjelve, IEEE Trans. Power Deliv., vol. 33, no. 2, 2018, p. 701. With permission).

In addition, the voltage transfer matrix H() from the four external terminals to the three internal nodes
(R1, R5, R11) was measured. The matrix elements are shown in Fig. 5.6 along with a calculated model.

Figure 5.6 Voltage transfer functions from external terminals to internal points. (From B. Gustavsen, A.
Portillo, R. Ronchi, A. Mjelve, IEEE Trans. Power Deliv., vol. 33, no. 2, 2018, p. 703. With permission).

27
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

6. Measurement Data files

Note 1:f
Any use of the data requires that proper referencing is made to this document and to the
web site defined in Chap. 6.3. No copying of material to another open web site is permitted.

The measurements are provided as a set of data files (Excel and ASCII) which are placed in folders

\3-ph Measurements\ (3-ph transformer)

\1-ph Measurements\ (1-ph transformer)

The folder contents are described in the following sub-chapters.

Note 2:
In some cases, a time domain response given in a file is exactly zero. This implies that the
response has not been measured.

6.1 Voltage transfer measurements

The time domain impulse responses (created from frequency-domain measurements via convolution
with 1.2/50 µs voltage excitation) are placed in folders below, provided in both Excel sheets and in
ASCII text files.

The (applied) 1.2/50 µs voltage wave shape is defined as

v(t )  vmax  A  (et  e  t ) (6.1)

where A  1.038 ,   0.015 106 ,   2.47 106 , vmax is the peak voltage, and t is time in seconds

\3-ph Measurements (3-ph transformer)

The respective Excel files are in subdirectory \H\Excel\ :
 TD3ph_Excel.xls : All 64 measurement cases with ∆t=0.02 µs, Tmax=100 µs.
 TD3ph_Excel_short.xls : All 64 measurement cases with ∆t=0.2 µs, Tmax=100 µs.
 TD3ph_Excel_long.xls : All 64 measurement cases with ∆t=0.2 µs, Tmax=1000 µs.

The respective ASCII files are in subdirectory \H\ASCII\ :

 64 measurement files with ∆t=0.02 µs, Tmax=100 µs.

\1-ph Measurements (1-ph transformer)

The respective Excel files are in subdirectory \H\Excel\ :
 TD1ph_Excel.xls : All 64 measurement cases with ∆t=0.02 µs, Tmax=100 µs.
 TD1ph_Excel_short.xls : All 64 measurement cases with ∆t=0.2 µs, Tmax=100 µs.

28
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

 TD1ph_Excel_long.xls : All 64 measurement cases with ∆t=0.2 µs, Tmax=1000 µs.

The respective ASCII files are in subdirectory \H\ASCII\ :

 64 measurement files with ∆t=0.02 µs, Tmax=100 µs.

The ASCII file contents for 3-ph and 1-ph measurements are defined in file "_TDheading.txt" as shown
in Table 6.1.

Table 6.1 File information

3-ph transformer 1-ph transformer

6.2 Admittance measurements

6.2.1 Measurement files
The frequency domain admittance measurements (with cable effects eliminated) are given both as Excel
files and ASCII text files, in folders
\3-ph Measurements\Y (3-ph transformer)
\1-ph Measurements\Y (1-ph transformer)

The respective Excel files are in subdirectory \Excel\ :

 Ymeas3Ph.xlsx
 Ymeas1Ph.xlsx

The respective ASCII files are in subdirectory \H\ :

 f.txt
 ReY.txt
 ImY.txt

File "f.txt" contains the frequency samples. File "ReY.txt" and "ImY.txt" contain respectively the real
and imaginary parts of the admittance matrix, Y.

The content in "ReY.txt" and "ImY.txt" is as follows.

 Each row k corresponds to the kth frequency sample.

29
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

 The content of each row is Y(k) stored row-wise. For instance, a 2×2 Y would have had four
elements whose real part would have been stored as follows,

[ Re{Y(1,1)} Re{Y(1,2) } | Re{Y(2,1) } Re{Y(2,2) } ]

6.2.2 Measurement conditions and terminal numbering

3-ph transformer
The admittance measurement is with respect to the HV and LV terminals, with the tertiary terminals
open, and with H0 and X0 grounded. Tap position Nom−.

The terminal numbering is as follows:

1: H1 4: X1
2: H2 5: X2
3: H3 6: X3

1-ph transformer
The admittance measurement is with respect to the HV, LV and Tertiary terminals, with the tertiary
terminals open, and with H0 and X0 grounded. Tap position Nom+. The data set has been processed
as described in [4].

The terminal numbering is as follows:

1: H1 3: Y1
2: X1 4: Y2

In the case of the 1-ph transformer, voltage transfers were also measured from the four external
terminals to three internal points (R1, R5, R11), at the same frequency points as for Y. The elements
are at each frequency stored in a matrix H of dimension 3×4. The real and imaginary parts are stored
row-wise in separate ASCII files in subdirectory \Hinternal\ :
 ReH.txt
 ImH.txt

6.3 Download site

The measurements are downloadable from this web site:
www.sintef.no/CIGREtransformerdata

Any use of the data requires that proper referencing is made to this document and to the
web site. No copying of material to another open web site is permitted.

30
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

7. Comparison between measurements and white-

box model simulations
To evaluate the accuracy of the existing white-box models, the complete design of both transformers
measured in México, and described in detail in Chapters 8 and 9 of this TB, was distributed between
the members of the working group that have software able to calculate the internal voltage
distribution when applying transient voltages on some of the transformer external terminals.
The members listed in Table 7.1 created (independently) models for the two transformers and
performed simulations in time and frequency domain corresponding to the measurements that were
made. The simulations were collected before the measurement results were released to the members.
After an initial comparison between measurements and simulations, new simulations were performed
in those (few) cases where the member had misunderstood technical details of the transformers
design.
Table 7.1 - WG members contributing with modelling and simulation results

Member Country Representing

Robert Degeneff United States Utility System Technologies Inc.

José Carlos Mendes Brazil ABB
Álvaro Portillo Uruguay Transformer Consultant
Jos Veens The Netherlands SGB-SMIT Group
Hans De Herdt Belgium CG Power Systems Belgium
Oliver Sterz Germany Maschinenfabrik Reinhausen (MR)
Xosé López-Fernández Spain Universidad de Vigo
Triomphant Ngnegueu France General Electric (ex-Alstom)
Ricardo Lopes Portugal EFACEC
Ji Hong Kim South Korea HYUNDAI
Baudilio Valecillos Switzerland Trafoexperts
Enrique Mombello Argentina Universidad de San Juan, Argentina
Daniil Matveev Russia Moscow Power Engineering Institute

The following shows comparisons between time domain and frequency domain measurements and
white-box model simulations.

7.1 time domain voltage responses

The following subsections show a selection of comparisons between measured and simulated voltage
responses for the two transformers. The applied voltage is in all cases a 1.2/50 µs standard impulse
voltage between earth and one external terminal (7.1), where vmax is the peak value of the applied
voltage. (In the measurements, the response was obtained via frequency domain measurements,
rational function approximation, and convolution with the standard impulse voltage application as
described in Chap. 3.2.3.).
Note that the calculated results in this chapter (and in TB Part A) were provided by members without
having seen the measurement results.

v(t )  vmax  A  (e t  e  t ) ,

(7.1)
A  1.038 ,   0.015 106 ,   2.47 106

31
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

7.1.1 Single-phase transformer

Figure 7.1 shows a case where the transformer is subjected to voltage application on H1 with H0 and
X0 grounded.

Case 6
Y1 X1 Nom+ (Tap 11+)

+ 84 +149 -213 1
2
-219 3
4
5
k - 6
HV RW 7
+ 8
9
10
+217 11

+182
TV LV
+181
+216

HV RW

-214
-1 - 85 - 150

Y2 X0 H1 H0

Figure 7.1 - Voltage application on H1 with X0 grounded. Nominal tap position.

 Figures 7.2 and 7.3 show the measured and calculated response (member "i") on terminal X1
(LV terminal) and point R6 (regulating winding), respectively. It is seen that the first peak
value is captured with reasonable accuracy while substantial deviations exist regarding
damping and oscillation frequency. The shift in oscillation frequency causes a large deviation
to develop with time.
0.7

0.6

0.5

0.4
X1
0.3
X1i
0.2

0.1

0
0 10 20 30 40 50 60 70 80 90 100
-0.1

Figure 7.2 - Measured (X1) and simulated (X1i) response on X1 with voltage in p.u. and time in s.

32
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

0.06

0.04

0.02

0
0 10 20 30 40 50 60 70 80 90 100 R6
-0.02
R6i
-0.04

-0.06

-0.08

-0.1

Figure 7.3 - Measured (R6) and simulated (R6i) response on R6 with voltage in p.u. and time in s.

 Figures 7.4 and 7.5 show the same comparison, but now including the calculated results from
all members. It is observed that while there are large differences between the individual
calculation results, almost all capture the first peak value with good accuracy.

0.8
X1
0.6 X1a
X1b
0.4
X1c

0.2 X1d
X1e
0 X1f
0 10 20 30 40 50 60 70 80 90 100
X1g
-0.2
X1h

-0.4 X1i
X1k
-0.6

Figure 7.4 - Measured (X1) and simulated (X1a-X1f) response on X1 with voltage in p.u. and time in s.

0.06
R6
0.04
R6a

0.02 R6b
R6c
0
0 10 20 30 40 50 60 70 80 90 100 R6d
-0.02 R6e
R6f
-0.04
R6g
-0.06 R6h
R6i
-0.08
R6k
-0.1

Figure 7.5 - Measured (R6) and simulated (R6a-R6f) response on R6 with voltage in p.u. and time in s.

33
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

7.1.2 Three-phase transformer

Similar measurements and calculations were performed for the three-phase transformer with the
tertiary winding connected in delta. Figure 7.6 shows an example for the transformer where the
voltage is applied to H1 with H0 and X0, grounded. (Only one of the three legs are included in the
figure.)

Figure 7.6 Voltage application on H1 with X0 grounded. Maximum tap position.

 Figures 7.7 and 7.8 show the measured and calculated response (member "b") on terminal X1
(LV terminal) and point R11 (regulating winding), respectively. The voltage peak value is
captured with excellent accuracy in both locations, but the overall agreement with the
measurement waveshape is poor.
0.5

0.4

0.3

X1
0.2
X1b

0.1

0
0 10 20 30 40 50 60 70 80 90 100

-0.1

Figure 7.7 - Measured (X1) and simulated (X1b) response on X1 with voltage in p.u. and time in s.

34
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

0.1

0.05

0
0 10 20 30 40 50 60 70 80 90 100
-0.05
R11
-0.1
R11b
-0.15

-0.2

-0.25

-0.3

Figure 7.8 - Measured (R11) and simulated (R11b) response on R11 with voltage in p.u. and time in s.

 Figures 7.9 and 7.10 show the same comparison, but now including the calculated results
from all members. Substantial differences between the results can be observed. There are
also substantial differences between calculated peak values at the two locations.

0.7
X1
0.6
X1b
0.5
X1c
0.4
X1d
0.3 X1e
0.2 X1f

0.1 X1g
X1i
0
0 10 20 30 40 50 60 70 80 90 100 X1j
-0.1
X1k
-0.2
X1l
-0.3

Figure 7.9 - Measured (X1) and simulated (X1b-X1l) response on X1 with voltage in p.u. and time in s.

0.3
R11
0.2 R11b
R11c
0.1
R11d

0 R11e
0 10 20 30 40 50 60 70 80 90 100 R11f
-0.1 R11g
R11i
-0.2
R11j

-0.3 R11k
R11l
-0.4

Figure 7.10 - Measured (R6) and simulated (R11b-R11l) response on R11 with voltage in p.u. and time in
s.

35
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

7.2 Frequency domain admittance matrix elements

Measurements were also performed for the terminal admittance matrix [1]. The following subsections
show comparisons between measured and calculated admittance matrix elements for the two
transformers.
7.2.1 Single-phase transformer
The 44 admittance matrix elements were measured with respect to all terminals H1, X1, Y1 and Y2
with H0 and X0 grounded.
 Figure 7.11 shows the measured magnitude function of element Y(1,1) between 10 Hz and
10 MHz, and the calculated response by four members (a-d). Although the calculated
responses capture the main features of the measurement, there are substantial differences in
the resonances and anti-resonances. Also, there are differences among the four calculations.
Module Y11
1.E+00

1.E-01

1.E-02

mm
a
1.E-03
b
c
d

1.E-04

1.E-05

1.E-06
1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07

Figure 7.11 - Element Y(1,1). Measured (mm) and calculated (a-d) with admittance in Siemens and
frequency in Hz.

 Figure 7.12 shows the ditto result for element Y(1,2). Similar observations are made.

36
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Module Y12
1.E+00

1.E-01

1.E-02

mm
a
1.E-03
b
c
d

1.E-04

1.E-05

1.E-06
1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07

Figure 7.12 - Element Y(1,2). Measured (mm) and calculated (a-d) with admittance in Siemens and
frequency in Hz.

7.2.2 Three-phase transformer

In case of the three-phase transformer, the admittance matrix was measured in frequency domain
with respect to terminals H1, H2, H3, X1, X2, X3, with H0 and X0 grounded and Y1, Y2, Y3 open
(connected in delta). This gave a 66 matrix Y.
 Figure 7.13 shows the magnitude function of the measured element Y(1,1) between 10 Hz
and 10 MHz, and the calculated response by three members (a-c). The agreement between
measurements and calculations is less good than the corresponding result for the single-phase
transformer.
Module Y11
1.E+00

1.E-01

1.E-02

mm

1.E-03 a
b
c

1.E-04

1.E-05

1.E-06
1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07

37
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Figure 7.13 - Element Y(1,1). Measured (mm) and calculated (a-c) with admittance in Siemens and
frequency in Hz.

 Figure 7.14 shows the ditto result for element Y(1,2). The agreements are now even poorer.

Module Y12
1.E+00

1.E-01

1.E-02

1.E-03

1.E-04

1.E-05
mm

1.E-06 a
b
c
1.E-07

1.E-08

1.E-09

1.E-10

1.E-11

1.E-12
1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07

Figure 7.14 - Element Y(1,2). Measured (mm) and calculated (a-c) with admittance in Siemens and
frequency in Hz.

7.3 Analysis of results

Comparison with measurements shows that the white-box models need improvements for use in
system studies which include oscillating overvoltages on the terminals.
In the time domain, the results demonstrate a quite good agreement in the prediction of the
maximum voltage values that appears in the measured points inside the transformers but a poor
agreement in the temporal wave shapes which are dependent on the natural (resonance) frequencies
of the transformer.
The deviations in the time domain can also be understood from a comparison between measured and
calculated frequency domain admittance matrix elements. Substantial deviations exist for the location
of resonances and antiresonances in the frequency responses.
These observations apply to both the single-phase and three-phase transformer. The agreement
between calculations and measurements are in general better for the single-phase transformer.

7.4 Time domain calculation data files

Note:
Any use of the data requires that proper referencing is made to this document and to the
web site defined in Chap. 6.3. No copying of material to another open web site is permitted.
The measurements and calculations are provided as a set of Excel files which are placed in folders:
 \3-ph Calculations\Odd Cases\ (3-ph transformer)
In total 32 Excel files named:
3Ph-Case01.xlsm, 3Ph-Case03.xlsm, 3Ph-Case05.xlsm, …,3Ph-Case63.xlsm.
(These cases are with the 3-ph transformer modelled using a 1-ph representation.)

 \ 3-ph Calculations\Even Cases\ (3-ph transformer)

38
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

In total 32 Excel files named:

3Ph-Case02.xlsm, 3Ph-Case04.xlsm, 3Ph-Case06.xlsm, …,3Ph-Case64.xlsm.
(These cases are with the 3-ph transformer modelled using a 3-ph representation.)

 \ 1-ph Calculations\ (1-ph transformer)

In total 64 Excel files named:
3Ph-Case01.xlsm, 3Ph-Case02.xlsm, 3Ph-Case03.xlsm, …,3Ph-Case64.xlsm
In the first sheet of each Excel file (Figure 7.15) the measured values are shown (columns 1 to 13) and
is possible select the calculations that the user wants to see and compare with the measured values.
The members are named with letters (a,b,c, …).

Figure 7.15 – First sheet of each Excel file

The selection of the calculation is made putting 1 (“Yes”) or 0 (“No”) in the rows 12 to 31 of column
15 as is shown in Figure 7.16, where only calculations of members b and r are selected.

39
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Figures 7.16 – Selections of calculation to shown

In Figure 7.16 and enlarged in Figure 7.17 are shown in blue the measured values, in orange the
calculated values by member b and in gray the calculated values by member r, at transformer
terminal X1 in Case 6 of the single phase transformer.

Figure 7.17 – measured and calculated values at terminal X1

In that way, selecting the members and clicking in bottom “All Graphs” makes it possible to compare
the results of each member calculation with the measured values and also compare between the
calculations results of different members.
In the Excel files corresponding to cases of the single-phase transformer a total of 10 graphs are
generated, comparing measured and calculated results. In the Excel files corresponding to cases of
the three-phase transformer a total of 11 graphs are generated for the Odd Cases and 16 graphs are
generated for the Even Cases comparing measured and calculated results.
In the last sheet of each Excel file a figure is shown with the details of the connections of the Case
(Figure 7.18).

40
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Figure 7.18 – Connections corresponding to Case 6

7.5 Frequency domain calculation data files

Note:
Any use of the data requires that proper referencing is made to this document and to the
web site defined in Chap. 6.3. No copying of material to another open web site is permitted.
The measurements and calculations are provided as a set of Excel files which are placed in folders:
 \Frequency Domain\3-ph Calculations\ (3-ph transformer)
One Excel files named:
Comparatives FD 3Ph.xlsm
 \Frequency Domain\1-ph Calculations\ (1-ph transformer)
One Excel files named:
Comparatives FD 1Ph.xlsm
In the first sheet of each Excel file (Figure 7.19), it is possible to select the calculations that the user
wants to see and compare with the measured values in modulus and phase for each of the elements of
the admittance matrix. The members are named with letters (a,b,c, …).

Figure 7.19 – First sheet of each Excel file

41
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

The selection of the calculation is made putting 1 (“Yes”) or 0 (“No”) in the rows 9 to 28 of column 2
as is shown in Figure 7.20, where only calculations of members a, b, c and d are selected.

Figures 7.20 – Selections of calculation to shown

In that way, selecting the members and clicking on button “All Graphs” allows to compare the results
of each member calculation with the measured values and also compare between the calculations
results of different members.

42
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

8. Design details of 3-ph transformer

8.1 Electrical Characteristics

Three Windings, Three Phase Transformer

Power and cooling:

55/55/18.33 MVA, ONAN
73/73/24.33 MVA, ONAF1
92/92/30.67 MVA, ONAF2

Voltages, connection group and frequency:

115/34.5/13.8 kV, YNyn0d1, 60 Hz

Insulation levels:
HV Line: AC 230 kVrms / LI 550 kVpeak
HV Neutral: AC 95 kVrms / LI 250 kVpeak
LV Line: AC 70 kVrms / LI 200 kVpeak
LV Neutral: AC 50 kVrms / LI 150 kVpeak
TV: AC 34 kVrms / LI 110 kVpeak

The transformer is of variable induction, as the On-Load Tap-Changer (OLTC) is located in the high
voltage winding, to regulate the voltage in low voltage winding (+9/-11 Taps)
OLTC: In-Tank, Resistor Type, Plus/Minus (Reversing Change-Over Selector), MR VACUTAP VM III
650 Y – 72.5/C - 12 23 3 W, without tie-in resistor
Tertiary winding only for stabilization and must be operated with one corner grounded (Terminals
Y1, Y2 and Y3 only for FRA measurement purposes)

Tap Reversing Selector HV HV RW HV+RW LV LV TV TV

Rated Rated Rated BMAX (T)
Position Switch Connection Turns Turns Turns Turns Turns
Volts Volts Volts
1 + 1 115000 486 -60 426 144 38873 100 15586 1.5830
2 + 2 115000 486 -54 432 144 38333 100 15369 1.5611
3 + 3 115000 486 -48 438 144 37808 100 15159 1.5397
4 + 4 115000 486 -42 444 144 37297 100 14954 1.5189
5 + 5 115000 486 -36 450 144 36800 100 14755 1.4986
6 + 6 115000 486 -30 456 144 36316 100 14560 1.4789
7 + 7 115000 486 -24 462 144 35844 100 14371 1.4597
8 + 8 115000 486 -18 468 144 35385 100 14187 1.4410
9 + 9 115000 486 -12 474 144 34937 100 14007 1.4227
10 + 10 115000 486 -6 480 144 34500 100 13800 1.4049
11A + 11
11B +/- k 115000 486 0 486 144 34074 100 13662 1.3876
11C - 1
12 - 2 115000 486 6 492 144 33659 100 13495 1.3707
13 - 3 115000 486 12 498 144 33253 100 13332 1.3542
14 - 4 115000 486 18 504 144 32857 100 13174 1.3380
15 - 5 115000 486 24 510 144 32471 100 13019 1.3223
16 - 6 115000 486 30 516 144 32093 100 12867 1.3069
17 - 7 115000 486 36 522 144 31724 100 12719 1.2919
18 - 8 115000 486 42 528 144 31364 100 12575 1.2772
19 - 9 115000 486 48 534 144 31011 100 12434 1.2629
20 - 10 115000 486 54 540 144 30667 100 12295 1.2488
21 - 11 115000 486 60 546 144 30330 100 12160 1.2351

43
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Polarities and Connections:

Phase 1, Tap 10 (115/34.5 kV)

Y1 H1 X1

R1

Y1R
1
+ + + +
2

6
TV HV LV RW
7

10

- - - - 11

Y2R

X0 H0
R2
- +

Y2 k

8.2 Core – “Three-Legs Core-Type”

All dimensions in mm:

1415

715

2230 3630

700

3530

44
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Core Section Diameter = 725 mm

Section:14 steps (700/680/660/640/620/580/540/500/460/420/380/340/300/260)
Ducts: 1  6 mm in the middle of the central step
Stacking Height: 676.68 mm (FeSi 23ZDKH90 0.23 mm – Stacking Factor 0.95)
Core Section = 3693.35 cm2
BMAX = 1.5830 T @ Tap 1, 1.4049 T @ Tap 10 and 1.2351 T @ Tap 21

8.3 TV Winding – Stabilizing Winding – “Helical Type”

1 Circuit per Leg

Bondable CTC, Cu, proof stress Rp0.2 = 170 N/mm2, 15  (6.9 mm  1.6 mm)
Cu Section: 15  10.825 mm2 = 162.38 mm2
Enamel + B Stage Epoxy Resin: 0.15 mm
Interleaving Paper: 0.105 mm
Insulating Paper: 0.61 mm
Compression of Paper: 0.90
Theoretical dimensions, of each CTC: 14.815 mm  14.610 mm
Not compressed actual dimensions, with tolerances according to the PIRELLI criteria, of each CTC:
14.915 (+0.00/-0.15) mm  14.810 (+0.00/-0.15)

𝐻 = 2 × (6.9 + 0.15) + 0.105 + 0.61 + 0.1 = 14.915 mm

15 + 1
𝐸= × (1.6 + 0.15) + 0.61 + 0.20 = 14.810 mm
2

Compressed actual dimensions:

𝐻 = 2 × (6.9 + 0.15) + (0.105 + 0.61) × 0.90 + 0.1 = 14.844 mm

Number of Turns: 100

Axial Distance between Turns (Radial Spacers from Top to Bottom):

9  4.2 mm + 1  5.3 mm + 9  4.2 mm + 1  5.3 mm +

9  4.2 mm + 1  5.3 mm + 9  4.2 mm + 1  5.3 mm +
9  4.2 mm + 1  5.3 mm + 9  4.2 mm + 1  5.3 mm +
9  4.2 mm + 1  5.3 mm + 9  4.2 mm + 1  5.3 mm +
9  4.2 mm + 1  5.3 mm + 9  4.2 mm
90  4.2 mm + 9  5.3 mm = 425.7 mm

Axial Distance between Turns (Total Pressboard): 425.7 mm

Compression of Pressboard: 0.9540

Pressboard Radial Spacers: 28  40 mm

Non-Compressed Mechanical Height:

(100+1)14.915 + 425.7 + 4.2 = 1936.3 mm  1937 mm
Compressed Mechanical Height:
(100+1)14.844 + 425.70.954 + 4.20.954 = 1909.4 mm  1910 mm
Compressed Electrical Height:
10014.844 + 425.70.954 = 1890.5 mm

In 10 positions, uniformly distributed, they were added 3 x 1mm adjustment radial spacers, so that
the final mechanical height of the compressed coil is 1910 mm

Winding with zig-zag oil flow

45
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Radial Width: 14.8101.013  15.0 mm

Diameters (Internal/External): 765 / 795 mm

8.4 LV Winding – “Helical Type”

1 Circuit per Leg

Bondable CTC, Cu, proof stress Rp0.2 = 170 N/mm2, 45  (4.4 mm  1.5 mm)
2// Radial with Central Transposition
Cu Section: 2  45  6.385 mm2 = 574.69 mm2 (3.047 A/mm2 @ Tap 21)
Enamel + B Stage Epoxy Resin: 0.15 mm
Interleaving Paper: 0.105 mm
Insulating Paper: 0.61 mm
Compression of Paper: 0.90
Theoretical dimensions, of each CTC: 9.815 mm  38.560 mm
Not compressed actual dimensions, with tolerances according to the PIRELLI criteria, of each CTC:
9.915 (+0.00/-0.15) mm  39.260 (+0.00/-0.15) mm

𝐻 = 2 × (4.4 + 0.15) + 0.105 + 0.61 + 0.1 = 9.915 mm

45 + 1
𝐸= × (1.5 + 0.15) + 0.61 + 0.70 = 39.260 mm
2

Compressed actual dimensions:

𝐻 = 2 × (4.4 + 0.15) + (0.105 + 0.61) × 0.90 + 0.1 = 9.843 mm

Number of Turns: 144

Axial Distance between Turns (Radial Spacers from Top to Bottom):

11  4.2 mm + 1  5.3 mm + 11  4.2 mm + 1  5.3 mm +

11  4.2 mm + 1  5.3 mm + 11  4.2 mm + 1  5.3 mm +
11  4.2 mm + 1  5.3 mm + 11  4.2 mm + 1  5.3 mm +
11  4.2 mm + 1  5.3 mm + 11  4.2 mm + 1  5.3 mm +
11  4.2 mm + 1  5.3 mm + 11  4.2 mm + 1  5.3 mm +
11  4.2 mm + 1  5.3 mm + 11  4.2 mm =
11  5.3 mm + 132  4.2 mm = 612.7 mm

Axial Distance between Turns (Total Pressboard): 612.7 mm

Compression of Pressboard: 0.9540

Pressboard Radial Spacers: 32  30 mm

Non-Compressed Mechanical Height:

(145+1)9.915 + 612.7 + 24.2 = 2068.69 mm  2069 mm
Compressed Mechanical Height:
(145+1)9.843 + 612.70.954 + 24.20.954 = 2029.6 mm  2030 mm
Compressed Electrical Height:
1459.843 + 612.70.954 + 4.20.954 = 2015.8 mm

In 10 positions, uniformly distributed, they were added 3 x 1mm adjustment radial spacers, so that
the final mechanical height of the compressed coil is 2030 mm

Winding with zig-zag oil flow

Radial Width: 239.2601.006 = 78.99 mm  79.0 mm

46
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Diameters (Internal/External): 1131 / 1289 mm

8.5 HV Principal Winding – “Continuous Disk Type”

1 Circuit per Leg

Bondable CTC, Cu, proof stress Rp0.2 = 140 N/mm2, 19  (6.6 mm  1.5 mm)
Cu Section: 19  9.685 mm2 = 184.02 mm2 (2.510 A/mm2)
Enamel + B Stage Epoxy Resin: 0.15 mm
Interleaving Paper: 0.105 mm
Insulating Paper: 1.37 mm
Compression of Paper: 0.90
Theoretical dimensions, of each CTC: 14.975 mm  17.870 mm
Not compressed actual dimensions, with tolerances according to the PIRELLI criteria, of each CTC:
15.075 (+0.00/-0.15) mm  18.070 (+0.00/-0.15) mm

𝐻 = 2 × (6.6 + 0.15) + 0.105 + 1.37 + 0.1 = 15.075 mm

19 + 1
𝐸= × (1.5 + 0.15) + 1.37 + 0.20 = 18.070 mm
2

Compressed actual dimensions:

𝐻 = 2 × (6.6 + 0.15) + (0.105 + 1.37) × 0.90 + 0.1 = 14.928 mm

Number of Turns: 486

Number of Disks: 100
Turns per Disk: 5 (4.86)

Axial Distance between Disks (Radial Spacers from Top to Bottom):

1  8.6 mm + 15  6.4 mm + 4  4.2 mm +

1  5.3 mm + 6  4.2 mm + 1  5.3 mm + 8  4.2 mm +
1  5.3 mm + 6  4.2 mm + 1  5.3 mm + 6  4.2 mm +
1  5.3 mm +
6  4.2 mm + 1  5.3 mm + 6  4.2 mm + 1  5.3 mm +
6  4.2 mm + 1  5.3 mm + 6  4.2 mm + 1  5.3 mm +
4  4.2 mm + 15  6.4 mm + 1  8.6 mm =
58  4.2 mm + 9  5.3 mm + 30  6.4 mm + 2  8.6 mm = 500.5 mm

Axial Distance between Disks (Total Pressboard): 500.5 mm

Compression of Pressboard: 0.9530
Pressboard Radial Spacers: 24  35 mm

Non-Compressed Height:
10015.075 + 500.5  2008 mm
Compressed Mechanical and Electrical Height:
10014.928 + 500.50.9530 = 1969.78 mm  1970 mm

In 10 positions, uniformly distributed, they were added 3 x 1mm adjustment radial spacers, so that
the final mechanical height of the compressed coil is 1970 mm

Winding with zig-zag oil flow

Radial Width: 518.0701.018 = 91.98 mm  92.0 mm

Diameters (Internal/External): 871 / 1055 mm

47
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

8.6 RW Regulating Winding – “Multi-start Type”

1 Circuit per Leg

Cu, proof stress Rp0.2 = 140 N/mm2, 5 mm  12 mm, 2// Axial
Cu Section: 2  59.142 mm2 = 118.28 mm2 (3.905 A/mm2)
Insulating Paper: 1.22 mm
Compression of Paper: 0.90

Number of Starts or Loops = 10 (called L1, L2, L3, ……, L10)

Number of Turns per Start or Loop = 6

Total Number of Turns = 10  6 = 60

Axial Distance between Conductors (Radial Spacers from Top to Bottom):

119  6.4 mm = 761.6 mm

Axial Distance between Turns (Total Pressboard): 761.6 mm

Compression of Pressboard: 0.9510

Pressboard Radial Spacers: 32  35 mm

Non-Compressed Mechanical Height:

102(6+1)(5+1.22) + 1396.4 = 1760.4 mm  1761 mm
Compressed Mechanical Height:
102(6+1)(5+1.220.90) + 1396.40.951 = 1699.73 mm  1700 mm
Compressed Electrical Height:
1026(5+1.220.90) + 1196.40.951 = 1456 mm

In 10 positions, uniformly distributed, they were added 3 x 1mm adjustment radial spacers, so that
the final mechanical height of the compressed coil is 1700 mm

Radial Width: (12.0+1.22)1.059  14.0 mm

Diameters (Internal/External): 1349 / 1377 mm

The geometrical and electrical layout of each of the multi-start regulating windings is shown in the
figure of the next page.

48
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Top of RF Winding

L1 9 1
L2 7 +
L3 5 6 turns L5
L4 3 - 2
L5 1 START +
turn 1
L6 2 6 turns L6
L7 4 - 3
L8 6 +
L9 8 6 turns L4
L10 10 - 4
L1 +
L2 6 turns L7
L3 - 5
L4 +
L5 6 turns L3
turn 2
L6 6
L7 +
L8 6 turns L8
L9 - 7
L10 +
6 turns L2

- 8
10 L1 +
8 L2 6 turns L9
6 L3 - 9
4 L4 +
2 L5 6 turns L1
turn 6
3 L6 - 10
5 L7 +
7 L8 6 turns L10
9 L9 - 11

END 11 L10

Bottom of RF Winding

Geometrical Layout Electrical Layout

49
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

8.7 Core & Windings Layout

210 180 150 315

28 sticks x 19 mm

28 sticks x 19 mm
28 sticks x 19 mm
28 sticks x 19 mm
24 sticks x 19 mm

24 sticks x 19 mm
24 sticks x 19 mm
24 sticks x 19 mm
32 sticks x 19 mm

32 sticks x 19 mm
32 sticks x 19 mm
32 sticks x 19 mm
DFE=725
HTV=1910 HHV=1970 HLV=2030 HRW=1700
DInt TV=765

DExt TV=795

DInt HV=871

DExt HV=1055 2230

DInt LV=1131
8 8 TV 6 8 8 8 HV 6 8 8 8 LV 8 8 8 RW 10 12 10

DExt LV=1289

DInt RW=1349

DExt RW=1377
20 38 38 30 38
15 92 79 14

4 2 2 4 2 2 4 2 4 3 3
110 80 50 215

8.8 Tank – “Rectangular with Radiators”

50
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

LV Side 281
1146

347
1377 38 1377 38 1377 2050

OLTC

HV Side 392

5700

Width = 2050 mm (= 281 + 1377 + 392)

Large = 5700 mm (= 1146 + 31377 + 238 + 347)
Height = 3840 mm

8.9 Tertiary Reactor

Three phase air core reactor, with 2 half reactor coils per phase, with upper and lower silicon steel
yokes to collect reactor leakage flux.

60
50

R1 54 54 313

770 44 44 44

R2 313
488 488 488

50
60
1574
Yokes Data:

Each yoke is a rectangular prism of 490 mm wide, 60 mm high and 1574 mm long.
Yokes section: 490 mm  60 mm  0.96  10-2 = 282.24 cm2

Data of each half reactor coil (Continuous Disk Type):

Bondable CTC, Cu, proof stress Rp0.2 = 170 N/mm2, 11  (7.0 mm  1.5 mm)
Cu Section: 11  10.285 mm2 = 113.14 mm2
Enamel + B Stage Epoxy Resin: 0.15 mm
Interleaving Paper: 0.105 mm
Insulating Paper: 0.61 mm
Compression of Paper: 0.90
Theoretical dimensions, of each CTC: 15.015 mm  10.510 mm
Not compressed actual dimensions, with tolerances according to the PIRELLI criteria, of each CTC:
15.115 (+0.00/-0.15) mm  10.710 (+0.00/-0.15) mm

Number of Turns: 62
Number of Disks: 16

51
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Turns per Disk: 4 (3.875)

Axial Distance between Disks (Radial Spacers from Top to Bottom):

15  5 mm = 75 mm

Axial Distance between Disks (Total Pressboard): 75 mm

Compression of Pressboard: 0.9550

Pressboard Radial Spacers: 16  30 mm

Non-Compressed Height:
1615.115 + 75 = 316.84  317 mm
Compressed Mechanical and Electrical Height:
1615.045 + 750.9550 = 312.35 mm  313 mm

Radial Width: 410.7101.027 = 44.0 mm

Distance between phases: 54 mm
Diameters (Internal/External): 400 / 488 mm

Reactor inductance calculation:

R1 self-inductance: 1.528 mH
R2 self-inductance: 1.528 mH
Mutual-inductance between R1 and R2: 0.356 mH

8.10 Losses and Impedance Measured in Factory Acceptance Tests (FAT)

To complete the transformer data and for the purpose of verification:

 The following table contains the measured no load losses (Pfe) at nominal voltage, for
the nominal and extreme taps.

Tap kV/kV Pfe(W)

1 115/38.876 42382
10 115/34.500 30362
21 115/30.330 24345

52
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

 The following table contains the measured short-circuit losses at 85ºC (Pcu) and the
measured short-circuit impedances at 85ºC (Uz), for the nominal and extreme taps.

Tap kV/kV Base(MVA) Pcu(W) Uz(%)

1 115/38.876 92 276831 13.12
10 115/34.500 92 280630 12.58
21 115/30.330 92 363256 12.43
1 115/15.586 30.67 105724 8.34
10 115/13.800 30.67 126268 10.20
21 115/12.160 30.67 163880 13.53
34.5/13.8 30.67 134639 15.43

8.11 Permittivities
All measurements were performed with the transformer in tank, but without oil, top lid and
bushings. The active part was vapour phase dried and was not impregnated with oil.

For the calculations the following permittivity values are suggested:

Permittivity
Air 1.0
Oil 2.2
Dry non-Impregnated Paper 2.7
Oil Impregnated Paper 3.5
Dry non-Impregnated Pressboard 3.8
Oil Impregnated Pressboard 4.4

For the test conditions, is recommended instead of the permittivity values for oil, impregnated
paper and impregnated pressboard as is usual, use the permittivity values for air, dry paper
and dry pressboard.

8.12 Nodes and Branches Numbering for Single Phase Modelling

In TB Part A, the transformer is modelled using 138 Nodes and 123 Branches, according to the
following description, and the figure in the next page.

 TV Winding (T1 in Figure)

25 Branches (one branch each four turns), numbering from 1 to 25
26 Nodes, numbering from 1 to 26

 HV Winding (AT1 in Figure)

50 Branches (one branch each two disks), numbering from 26 to 75
51 Nodes, numbering from 27 to 77

 LV Winding (BT1 in Figure)

36 Branches (one branch each four turns), numbering from 76 to 111

53
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

37 Nodes, numbering from 78 to 114

 RW Regulating Winding (AT2 in Figure)

10 Branches (one branch per loop, L1 to L10), numbering from 112 to 121
20 Nodes, numbering from 115 to 134

 Tertiary Reactor (R1 and R2 in Figure)

2 Branches (one branch per loop), numbering 122 and 123
4 Nodes, numbering from 135 to 138

The single phase model can be used for the calculation of the measured cases identified with odd
numbers (1, 3, 5, 7, ……., 63)

54
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Phase 1
+26 +77 +114 +116 +118 +120 +122 +124 +126 +128 +130 +132 +134
75 111
25 76
74 113
25 75 110
73
CIGRE-WEG-3Ph - Single Phase Model 24 74
112
72 109
24 73 111 +136 +138
71
23 72 108
70 110
23 71
107
69 122 123
22 70 109
68 106
22 69
67 108
21 68 105 -135 -137
66
21 67
107 R2 R1
65 104
20 66 106
64
20 65 103
63 105
19 64 102
62
19 63 104
61 101
18 62
103
60
18 61 100
59 102
17 60
58 99
17 59 101
57
98
16 58
56 100
IdTrafo: CIGRE-WEG-3Ph-01

16 57 97
55
15 56 99
54 96
15 55
98
DTDS To DDI

53
14 54 95
52 97
14 53
51 94
13 52 96 112 113 114 115 116 117 118 119 120 121
50 93 L1 L2 L3 L4 L5 L6 L7 L8 L9 L10
13 51
49 95
12 50 92
48
94
12 49
47 91
11 48 93
46
11 47 90
45 92
10 46
89
44
10 45 91
43 88
9 44
42 90
9 43 87
41
89
8 42
40 86
8 41 88
39
7 40 85
38 87
7 39 84
37
6 38 86
36 83
6 37
85
35
5 36 82
34 84
5 35
33 81
4 34 83
32
80
4 33
31 82
3 32 79
30
3 31 81
29 78
2 30
80
28
2 29 77
27 79
1 28
26 76
-1 -27 -78 -115 -117 -119 -121 -123 -125 -127 -129 -131 -133
T1 AT1 BT1 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2

55
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Y1 H1 X1

138
R1
137

Y1R
1
+124
+26 +77 +114 L5
-123 2
+126
L6
-125 3
+122
L4
-121 4
+128
L7
-127 5
+120
L3
-119 6
TV HV LV +130
L8
-129 7
+118
L2
-117 8
+132
L9
-131 9
+116
L1
-115 10
+134
L10
-1 -27 -78 -133 11

Y2R RW

136
X0 H0
R2
135 - +

Y2 k

Permanent connections:
o 1 – 136
o 26 – 137
o 123 – 126
o 125 – 122
o 121 – 128
o 127 – 120
o 119 – 130
o 129 – 118
o 117 – 132
o 131 – 116
o 115 – 134

Other connections must be made depending on the OLTC position

56
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Terminals and internal accessible points:

Terminals
and Internal Nodes
Accessible Points
H1 77
k OLTC 27
+ OLTC = 11 OLTC 133
- OLTC = 1 OLTC 124
2 OLTC 123 – 126
3 OLTC 125 – 122
4 OLTC 121 – 128
5 OLTC 127 – 120
6 OLTC 119 – 130
7 OLTC 129 – 118
8 OLTC 117 – 132
9 OLTC 131 – 116
10 OLTC 115 – 134
X1 114
X0 78
Y1 138
Y1R 26 – 137
Y2R 1 – 136
Y2 135

8.13 Nodes and Branches Numbering for Three Phase Modelling

In TB Part A, the transformer is modelled using 414 Nodes and 369 Branches, according to the
following description, and the figures in the next pages.

 TV Winding – Phase 1 (T1 in Figure)

25 Branches (one branch each four turns), numbering from 1 to 25
26 Nodes, numbering from 1 to 26

 HV Winding – Phase 1 (AT1 in Figure)

50 Branches (one branch each two disks), numbering from 26 to 75
51 Nodes, numbering from 27 to 77

 LV Winding – Phase 1 (BT1 in Figure)

36 Branches (one branch each four turns), numbering from 76 to 111
37 Nodes, numbering from 78 to 114

 RW Regulating Winding – Phase 1 (AT2 in Figure)

10 Branches (one branch per loop, L1 to L10), numbering from 112 to 121
20 Nodes, numbering from 115 to 134

 Tertiary Reactor – Phase 1 (R1 and R2 in Figure)

2 Branches (one branch per loop), numbering 364 and 365
4 Nodes, numbering from 403 to 406

 TV Winding – Phase 2 (T1 in Figure)

25 Branches (one branch each four turns), numbering from 122 to 146
26 Nodes, numbering from 135 to 160

 HV Winding – Phase 2 (AT1 in Figure)

50 Branches (one branch each two disks), numbering from 147 to 196

57
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

51 Nodes, numbering from 161 to 211

 LV Winding – Phase 2 (BT1 in Figure)

36 Branches (one branch each four turns), numbering from 197 to 232
37 Nodes, numbering from 212 to 248

 RW Regulating Winding – Phase 2 (AT2 in Figure)

10 Branches (one branch per loop, L1 to L10), numbering from 233 to 242
20 Nodes, numbering from 249 to 268

 Tertiary Reactor – Phase 2 (R1 and R2 in Figure)

2 Branches (one branch per loop), numbering 366 and 367
4 Nodes, numbering from 407 to 410

 TV Winding – Phase 3 (T1 in Figure)

25 Branches (one branch each four turns), numbering from 243 to 267
26 Nodes, numbering from 269 to 294

 HV Winding – Phase 3 (AT1 in Figure)

50 Branches (one branch each two disks), numbering from 268 to 317
51 Nodes, numbering from 295 to 345

 LV Winding – Phase 3 (BT1 in Figure)

36 Branches (one branch each four turns), numbering from 318 to 353
37 Nodes, numbering from 346 to 382

 RW Regulating Winding – Phase 3 (AT2 in Figure)

10 Branches (one branch per loop, L1 to L10), numbering from 354 to 363
20 Nodes, numbering from 383 to 402

 Tertiary Reactor – Phase 3 (R1 and R2 in Figure)

2 Branches (one branch per loop), numbering 368 and 369
4 Nodes, numbering from 411 to 414

The single phase model must be used for the calculation of the measured cases identified with even
numbers (2, 4, 6, 8, ……., 64)

58
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Phase 1
+26 +77 +114 +116 +118 +120 +122 +124 +126 +128 +130 +132 +134
75 111
25 76
74 113
25 75 110 +404 +406
73
112
CIGRE-WEG-3Ph - Three Phase Model
24 74
72 109
24 73 111
71 364 365
23 72 108
70 110
23 71
107
69
22 70 109 -403 -405
68 106
22 69
R2 R1
67 108
21 68 105
66
107
21 67
65 104
20 66 106
64
20 65 103
63 105
19 64 102
62
19 63 104
61 101
18 62
103
60
18 61 100
59 102
17 60
58 99
17 59 101
57
98
16 58
56 100
IdTrafo: CIGRE-WEG-3Ph-01

16 57 97
55
15 56 99
54 96
15 55
98
DTDS To DDI

53
14 54 95
52 97
14 53
51 94
13 52 96 112 113 114 115 116 117 118 119 120 121
50 93 L1 L2 L3 L4 L5 L6 L7 L8 L9 L10
13 51
49 95
12 50 92
48
94
12 49
47 91
11 48 93
46
11 47 90
45 92
10 46
89
44
10 45 91
43 88
9 44
42 90
9 43 87
41
89
8 42
40 86
8 41 88
39
7 40 85
38 87
7 39 84
37
6 38 86
36 83
6 37
85
35
5 36 82
34 84
5 35
33 81
4 34 83
32
80
4 33
31 82
3 32 79
30
3 31 81
29 78
2 30
80
28
2 29 77
27 79
1 28
26 76
-1 -27 -78 -115 -117 -119 -121 -123 -125 -127 -129 -131 -133
T1 AT1 BT1 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2

59
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Phase 2
+160 +211 +248 +250 +252 +254 +256 +258 +260 +262 +264 +266 +268
196 232
146 210
195 247
159 209 231 +408 +410
194
246
CIGRE-WEG-3Ph - Three Phase Model
145 208
193 230
158 207 245
192 366 367
144 206 229
191 244
157 205
228
190
143 204 243 -407 -409
189 227
156 203
R2 R1
188 242
142 202 226
187
241
155 201
186 225
141 200 240
185
154 199 224
184 239
140 198 223
183
153 197 238
182 222
139 196
237
181
152 195 221
180 236
138 194
179 220
151 193 235
178
219
137 192
177 234
IdTrafo: CIGRE-WEG-3Ph-01

150 191 218

176
136 190 233
175 217
149 189
232
DTDS To DDI

174
135 188 216
173 231
148 187
172 215
134 186 230 233 234 235 236 237 238 239 240 241 242
171 214 L1 L2 L3 L4 L5 L6 L7 L8 L9 L10
147 185
170 229
133 184 213
169
228
146 183
168 212
132 182 227
167
145 181 211
166 226
131 180
210
165
144 179 225
164 209
130 178
163 224
143 177 208
162
223
129 176
161 207
142 175 222
160
128 174 206
159 221
141 173 205
158
127 172 220
157 204
140 171
219
156
126 170 203
155 218
139 169
154 202
125 168 217
153
201
138 167
152 216
124 166 200
151
137 165 215
150 199
123 164
214
149
136 163 198
148 213
122 162
147 197
-135 -161 -212 -249 -251 -253 -255 -257 -259 -261 -263 -265 -267
T1 AT1 BT1 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2

60
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Phase 3
+294 +345 +382 +384 +386 +388 +390 +392 +394 +396 +398 +400 +402
317 353
267 344
316 381
293 343 352 +412 +414
315
380
CIGRE-WEG-3Ph - Three Phase Model
266 342
314 351
292 341 379
313 368 369
265 340 350
312 378
291 339
349
311
264 338 377 -411 -413
310 348
290 337
R2 R1
309 376
263 336 347
308
375
289 335
307 346
262 334 374
306
288 333 345
305 373
261 332 344
304
287 331 372
303 343
260 330
371
302
286 329 342
301 370
259 328
300 341
285 327 369
299
340
258 326
298 368
IdTrafo: CIGRE-WEG-3Ph-01

284 325 339

297
257 324 367
296 338
283 323
366
DTDS To DDI

295
256 322 337
294 365
282 321
293 336
255 320 364 354 355 356 357 358 359 360 361 362 363
292 L1 L2 L3 L4 L5 L6 L7 L8 L9 L10
335
281 319
291 363
254 318 334
290
362
280 317
289 333
253 316 361
288
279 315 332
287 360
252 314
331
286
278 313 359
285 330
251 312
284 358
277 311 329
283
357
250 310
282 328
276 309 356
281
249 308 327
280 355
275 307 326
279
248 306 354
278 325
274 305
353
277
247 304 324
276 352
273 303
275 323
246 302 351
274
322
272 301
273 350
245 300 321
272
271 299 349
271 320
244 298
348
270
270 297 319
269 347
243 296
268 318
-269 -295 -346 -383 -385 -387 -389 -391 -393 -395 -397 -399 -401
T1 AT1 BT1 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2 AT2

61
 Y1 H1 X1 Y2 H2 X2 Y3 H3 X3
406 410 414
R1 R1 R1
405 409 413

Y1R Y1R Y1R

1 1 1
+124 +258 +392
+26 +77 +114 L5 +160 +211 +248 L5 +294 +345 +382 L5
-123 2 -257 2 -391 2
+126 +260 +394
L6 L6 L6
-125 3 -259 3 -393 3
+122 +256 +390
L4 L4 L4
-121 4 -255 4 -389 4
+128 +262 +396
L7 L7 L7
-127 5 -261 5 -395 5
+120 +254 +388
L3 L3 L3
-119 6 -253 6 -387 6
TV HV LV +130 TV HV LV +264 TV HV LV +398
L8 L8 L8
-129 7 -263 7 -397 7

62
+118 +252 +386
L2 L2 L2
-117 8 -251 8 -385 8
+132 +266 +400
L9 L9 L9
-131 9 -265 9 -399 9
+116 +250 +384
L1 L1 L1
-115 10 -249 10 -383 10
+134 +268 +402
L10 L10 L10
-1 -27 -78 -133 11 -135 -161 -212 -267 11 -269 -295 -346 -401 11

Y2R RW Y2R RW Y2R RW

404 408 412
X0 H0 X0 H0 X0 H0
R2 R2 R2
403 - + 407 - + 411 - +

Y2 k Y3 k Y1 k
Part E: measurements and transformer design details
TB 904 - High-frequency transformer and reactor models for network studies
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Permanent connections:

Phase 1 Phase 2 Phase 3

o 1 – 404 o 135 – 408 o 269 – 412

o 405 – 26 o 409 – 160 o 413 – 294
o 123 – 126 o 257 – 260 o 391 – 394
o 125 – 122 o 259 – 256 o 393 – 390
o 121 – 128 o 255 – 262 o 389 – 396
o 127 – 120 o 261 – 254 o 395 – 388
o 119 – 130 o 253 – 264 o 387 – 398
o 129 – 118 o 263 – 252 o 397 – 386
o 117 – 132 o 251 – 266 o 385 – 400
o 131 – 116 o 265 – 250 o 399 – 384
o 115 – 134 o 249 – 268 o 383 – 402

Between Phases

o 78 – 212 – 346
o 403 – 410
o 407 – 414
o 411 – 406

Other connections must be made depending on the OLTC position

Terminals and internal accessible points:

Terminals Nodes
H1 77
H2 211
H3 345
X1 114
X2 248
X3 382
X0 78 – 212 – 346
Y1 406
Y2 410
Y3 414

Internal Accessible
Points Nodes
Phase 1
k OLTC 27
+ OLTC = 11 OLTC 133
- OLTC = 1 OLTC 124
2 OLTC 123 – 126
3 OLTC 125 – 122

63
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

4 OLTC 121 – 128

5 OLTC 127 – 120
6 OLTC 119 – 130
7 OLTC 129 – 118
8 OLTC 117 – 132
9 OLTC 131 – 116
10 OLTC 115 – 134
Y1R 26 – 405
Y2R 1 – 404

Internal Accessible
Points Nodes
Phase 2
k OLTC 161
+ OLTC = 11 OLTC 267
- OLTC = 1 OLTC 258
2 OLTC 257 – 260
3 OLTC 259 – 256
4 OLTC 255 – 262
5 OLTC 261 – 254
6 OLTC 253 – 264
7 OLTC 263 – 252
8 OLTC 251 – 266
9 OLTC 265 – 250
10 OLTC 249 – 268
Y1R 160 – 409
Y2R 135 – 408

Internal Accessible
Points Nodes
Phase 3
k OLTC 295
+ OLTC = 11 OLTC 401
- OLTC = 1 OLTC 392
2 OLTC 391 – 394
3 OLTC 393 – 390
4 OLTC 389 – 396
5 OLTC 395 – 388
6 OLTC 387 – 398
7 OLTC 397 – 386
8 OLTC 385 – 400
9 OLTC 399 – 384
10 OLTC 383 – 402
Y1R 294 - 413
Y2R 269 – 412

64
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

9. Design details of 1-ph transformer

9.1 Electrical Characteristics

Three Windings, Single Phase Transformer

Power and cooling:

30/30/10 MVA, ONAN
40/40/13.33 MVA, ONAF1
50/50/16.67 MVA, ONAF2

Voltages, connection group and frequency:

230/3 / 69/3 / 13.8 kV, YNynd, 60 Hz

Insulation levels:
HV Line: AC 395 kVrms / SI 850 kVpeak / LI 1050 kVpeak
HV Neutral: AC 50 kVrms / LI 125 kVpeak
LV Line: AC 140 kVrms / LI 350 kVpeak
LV Neutral: AC 50 kVrms / LI 125 kVpeak
TV: AC 50 kVrms / LI 125 kVpeak

The transformer is of constant induction

101% On-Load Tap-Changer at HV Winding
OLTC: In-Tank, Resistor Type, Plus/Minus (Reversing Change-Over Selector), MR VACUTAP VM I
500 – 72.5/C - 12 23 3 W
Tie-in resistor of 198 k permanently connected from point 6 of selector to H0.

Tap Reversing Selector HV HV RW HV+RW LV LV TV TV

Rated BMAX (T)
Position Switch Connection Rated Volts Turns Turns Turns Turns Rated Volts Turns
Volts
1 + 1 253000/3 800 80 880 240 69000/3 83 13800 1.7075
2 + 2 250700/3 800 72 872 240 69000/3 83 13800 1.7075
3 + 3 248400/3 800 64 864 240 69000/3 83 13800 1.7075
4 + 4 246100/3 800 56 856 240 69000/3 83 13800 1.7075
5 + 5 243800/3 800 48 848 240 69000/3 83 13800 1.7075
6 + 6 241500/3 800 40 840 240 69000/3 83 13800 1.7075
7 + 7 239200/3 800 32 832 240 69000/3 83 13800 1.7075
8 + 8 236900/3 800 24 824 240 69000/3 83 13800 1.7075
9 + 9 234600/3 800 16 816 240 69000/3 83 13800 1.7075
10 + 10 232300/3 800 8 808 240 69000/3 83 13800 1.7075
11A + 11
11B +/- k 230000/3 800 0 800 240 69000/3 83 13800 1.7075
11C - 1
12 - 2 227700/3 800 -8 792 240 69000/3 83 13800 1.7075
13 - 3 225400/3 800 -16 784 240 69000/3 83 13800 1.7075
14 - 4 223100/3 800 -24 776 240 69000/3 83 13800 1.7075
15 - 5 220800/3 800 -32 768 240 69000/3 83 13800 1.7075
16 - 6 218500/3 800 -40 760 240 69000/3 83 13800 1.7075
17 - 7 216200/3 800 -48 752 240 69000/3 83 13800 1.7075
18 - 8 213900/3 800 -56 744 240 69000/3 83 13800 1.7075
19 - 9 211600/3 800 -64 736 240 69000/3 83 13800 1.7075
20 - 10 209300/3 800 -72 728 240 69000/3 83 13800 1.7075
21 - 11 207000/3 800 -80 720 240 69000/3 83 13800 1.7075

65
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Polarities and Connections:

Tap 1 (253/3 / 69/3 kV)

Y1 X1

+ + - 1
2
- 3
4
5
k - 6
HV RW 7
+ 8
9
10
+ 11

+
TV LV
+
+

HV RW

-
- - -

Y2 X0 H1 H0

9.2 Core – “1 Wound Leg with 2 Return Legs Core-Type”

All dimensions in mm:

1055

530
2190 2890

350 700

2460
Core Section Diameter = 720 mm
Section:14 steps (700/680/660/640/620/580/540/500/460/420/380/340/300/260)
Ducts: 2  6 mm between second and third step
Stacking Height: 670.80 mm (FeSi M4 0.27 mm – Stacking Factor 0.96)
Core Section = 3646.63 cm2

66
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

BMAX = 1.7075 T

9.3 TV Winding – “Helical Type”

1 Circuit per Leg

Bondable CTC, Cu, proof stress Rp0.2 = 90 N/mm2, 29  (3.5 mm  1.5 mm)
2// Axial, Cu Section: 2  29  5.035 mm2 = 292.05 mm2 (4.135 A/mm2)
Enamel + B Stage Epoxy Resin: 0.15 mm
Interleaving Paper: 0.105 mm
Netting Tape or Cordex: 0.45 mm
Compression of Paper: 0.90
Theoretical dimensions, of each CTC: 7.855 mm  25.200 mm
Not compressed actual dimensions, with tolerances according to the PIRELLI criteria, of each CTC:
7.928 (+0.15/-0.15) mm  26.243 (+0.30/-0.80) mm

𝐻 = 2 × 1.01 × (3.5 + 0.15) + 0.105 + 0.45 + 0.15 = 7.928 + 0.15 = 8.078 mm

29 + 1
𝐸 = 1.03 × × (1.5 + 0.15) + 0.45 + 0.30 = 26.243 mm
2

Compressed actual dimensions:

𝐻 = 2 × 1.01 × (3.5 + 0.15) + 0.105 × 0.90 + 0.15 = 7.617 mm

Number of Turns: 83

Axial Distance between Conductors (Radial Spacers from Top to Bottom):

13  4.2 mm + 1  5.3 mm + 11  4.2 mm + 1  5.3 mm +

11  4.2 mm + 1  5.3 mm + 11  4.2 mm + 1  5.3 mm +
11  4.2 mm + 1  5.3 mm + 11  4.2 mm + 1  5.3 mm +
11  4.2 mm + 1  5.3 mm + 9  4.2 mm + 1  5.3 mm +
11  4.2 mm + 1  5.3 mm + 11  4.2 mm + 1  5.3 mm +
11  4.2 mm + 1  5.3 mm + 11  4.2 mm + 1  5.3 mm +
11  4.2 mm + 1  5.3 mm + 9  4.2 mm =
152  4.2 mm + 13  5.3 mm = 707.3 mm

Axial Distance between Turns (Total Pressboard): 707.3 mm

Compression of Pressboard: 0.9500

Pressboard Radial Spacers: 28  30 mm

Non-Compressed Mechanical Height:
2(83+1)8.078 + 707.3 + 24.2 = 2072.8 mm  2073 mm
Compressed Mechanical Height:
2(83+1)7.617 + 707.30.95 + 24.20.95 = 1959.57 mm  1960 mm
Compressed Electrical Height:
2837.617 + 707.30.95 = 1936.4 mm

In 12 positions, uniformly distributed, they were added 3 x 1mm adjustment radial spacers, so that
the final mechanical height of the compressed coil is 1960 mm

Winding with zig-zag oil flow

Radial Width: 26.2431.028 = 26.98 mm  27.0 mm

Diameters (Internal/External): 756 / 810 mm

67
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

9.4 LV Winding – “Continuous Disk Type”

1 Circuit per Leg

Bondable CTC, Cu, proof stress Rp0.2 = 90 N/mm2, 21  (5.0 mm  1.5 mm)
2// Radial, Cu Section: 2  21  7.285 mm2 = 305.99 mm2 (4.102 A/mm2)
Enamel + B Stage Epoxy Resin: 0.15 mm
Interleaving Paper: 0.105 mm
Insulating Paper: 0.91 mm
Compression of Paper: 0.90
Theoretical dimensions, of each CTC: 11.315 mm  19.060 mm
Not compressed actual dimensions, with tolerances according to the PIRELLI criteria, of each CTC:
11.415 (+0.00/-0.15) mm  19.260 (+0.00/-0.15) mm

𝐻 = 2 × (5.0 + 0.15) + 0.105 + 0.91 + 0.1 = 11.415 mm

21 + 1
𝐸= × (1.5 + 0.15) + 0.91 + 0.20 = 19.260 mm
2

Compressed actual dimensions:

𝐻 = 2 × (5.0 + 0.15) + (0.105 + 0.91) × 0.90 + 0.1 = 11.314 mm

Number of Turns: 240

Number of Disks: 128
Turns per Disk: 2 (1.88)

Axial Distance between Disks (Radial Spacers from Top to Bottom):

10  4.2 mm + 1  5.3 mm + 10  4.2 mm + 1  5.3 mm

10  4.2 mm + 1  5.3 mm + 10  4.2 mm + 1  5.3 mm
9  4.2 mm + 1  5.3 mm + 9  4.2 mm + 1  5.3 mm
9  4.2 mm + 1  5.3 mm + 9  4.2 mm + 1  5.3 mm
10  4.2 mm + 1  5.3 mm + 10  4.2 mm + 1  5.3 mm
10  4.2 mm + 1  5.3 mm + 10  4.2 mm =
116  4.2 mm + 11  5.3 mm = 545.5 mm

Axial Distance between Turns (Total Pressboard): 545.5 mm

Compression of Pressboard: 0.937

Pressboard Radial Spacers: 28  30 mm

Non-Compressed Mechanical Height:

12811.415 + 545.5 = 2006.62 mm  2007 mm
Compressed Mechanical and Electrical Height:
12811.314 + 545.50.937 = 1959.33 mm  1960 mm

In 16 positions, uniformly distributed, they were added 2 x 1mm adjustment radial spacers, so that
the final mechanical height of the compressed coil is 1960 mm

Winding with zig-zag oil flow

Radial Width: 2219.2601.012 = 77.96 mm  78.0 mm

Diameters (Internal/External): 938 / 1094 mm

68
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

9.5 HV Principal Winding – “Interleaved Disk Type”

2 Circuit in Parallel per Leg - Line input in the middle of the winding
Cu, proof stress Rp0.2 = 90 N/mm2, 9.9 mm  2.7 mm, 2// Radial
Cu Section: 2  2  26.181 mm2 = 104.72 mm2 (3.995 A/mm2)
Insulating Paper: 1.52 mm
Compression of Paper: 0.90

Disk Type from Top to Bottom:

Number of Turns: 800 10 Total Interleaved Disks

Number of Disks: 42 Interleaved Disks
62 (upper half) + 62 (lower half) = 124 10 Total Interleaved Disks
Turns per Disk: 13 (12.90) 10 Total Interleaved Disks
42 Interleaved Disks
10 Total Interleaved Disks

7 7 19 19 6 6 18 18 5 5 17 17 4 4 16 16 3 3 15 15 2 2 14 14 1 1 20 7 19 6 19 6 18 5 18 5 17 4 17 4 16 3 16 3 15 2 15 2 14 1 14 1

20 20 8 8 21 21 9 9 22 22 10 10 23 23 11 11 24 24 12 12 25 25 13 13 26 26 7 20 8 21 8 21 9 22 9 22 10 23 10 23 11 24 11 24 12 25 12 25 13 26 13 26

Interleaved Disks Total Interleaved Disks

Axial Distance between Disks (Radial Spacers from Top to Bottom):

4  (4.2 mm + 6.4 mm) + 1  5.3 mm + 2  (6.4 mm + 4.2 mm) +

4  4.2 mm + 1  5.3 mm + 6  4.2 mm +1  5.3 mm + 8  4.2 mm +
1  5.3 mm + 8  4.2 mm +1  5.3 mm + 4  4.2 mm +
7  (6.4 mm + 4.2 mm) +
1  6.4 mm (center of the winding)
7  (4.2 mm + 6.4 mm) +
4  4.2 mm + 1  5.3 mm + 8  4.2 mm + 1  5.3 mm +
8  4.2 mm + 1  5.3 mm + 6  4.2 mm + 1  5.3 mm + 4  4.2 mm +
2  (4.2 mm + 6.4 mm) + 1  5.3 mm + 4  (6.4 mm + 4.2 mm) =
86  4.2 mm + 10  5.3 mm + 27  6.4 mm = 587 mm

Axial Distance between Disks (Total Pressboard): 587 mm

Compression of Pressboard: 0.9580

Pressboard Radial Spacers: 32  35 mm

Non-Compressed Height:
124(9.9+1.52) + 587  2003 mm
Compressed Mechanical and Electrical Height:
124(9.9+1.520.90) + 5870.958 = 1959.58 mm  1960 mm

In 12 positions, uniformly distributed, they were added 2 x 1mm adjustment radial spacers, so that
the final mechanical height of the compressed coil is 1960 mm

Winding with zig-zag oil flow (with 1x8.0 mm axial oil duct)

Radial Width: 132(2.7+1.52)1.027 +8.0= 120.68 mm  121.0 mm

Diameters (Internal/External): 1242 / 1484 mm

69
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

9.6 RW Regulating Winding – “Interleaved Disk Type”

2 Circuit in Parallel per Leg

Cu, proof stress Rp0.2 = 90 N/mm2, 14.2 mm  2.2 mm, 2// Radial
Cu Section: 2  2  30.877 mm2 = 123.51 mm2 (3.387 A/mm2)
Insulating Paper: 1.52 mm
Compression of Paper: 0.90

Number of Turns: 10 taps  8 turns = 80

Number of Disks: 20 (upper half) + 20 (lower half) = 40
Turns per Disk: 4 (4.00)
Total Interleaved Disks – Interleaved each 10 Disks

Central Gap (between upper and lower part of RW): 600 mm

Axial Distance between Disks (Radial Spacers from Top to Bottom):

upper half: 19  5.3 mm = 100.7 mm

lower half: 19  5.3 mm = 100.7 mm

Axial Distance between Disks (Total Pressboard): 100.7 mm + 100.7 mm

Compression of Pressboard: 0.9490

Pressboard Radial Spacers: 32  40 mm

Non-Compressed Mechanical Height:

20(14.2+1.52) + 100.7 = 415.1 mm  416 mm
416 + 600 + 416 = 1432 mm
Compressed Mechanical and Electrical Height:
20(14.2+1.520.90) + 100.70.949 = 406.92 mm  407 mm
407 + 600 + 407 = 1414 mm

In 6 positions, uniformly distributed in each winding half, they were added 2 x 1mm adjustment
radial spacers, so that the final mechanical height of the compressed coil is 416 mm

Radial Width: 42(2.2+1.52)1.039 = 30.92 mm  31.0 mm

Diameters (Internal/External): 1612 / 1674 mm

The geometrical and electrical layout of the regulating windings is shown in the figure of the next
page.

70
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

RW – Upper Half RW – Lower Half

71
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

9.7 Core & Windings Layout

170 170 170 443 Electrostatic Shield

28 sticks x 19 mm

28 sticks x 19 mm
28 sticks x 19 mm
28 sticks x 19 mm
28 sticks x 19 mm
28 sticks x 19 mm
28 sticks x 19 mm

28 sticks x 19 mm
28 sticks x 19 mm
28 sticks x 19 mm
28 sticks x 19 mm
28 sticks x 19 mm
28 sticks x 19 mm
28 sticks x 19 mm

32 sticks x 19 mm
32 sticks x 19 mm
32 sticks x 19 mm
32 sticks x 19 mm
32 sticks x 19 mm
32 sticks x 19 mm
DFE=720
HTV=1960 HLV=1960 HHV=1960 HRW=407
DInt TV=756

DExt TV=810

DInt LV=938
RW
DExt LV=1094

Retrun Leg
DInt HV=1242
6 8 TV 8 8 10 10 8 8 LV 6 8 8 10 10 10 8 HV 8 8 10 10 8 8 GAP=600 8 9 8 8 2190
ç

DExt HV=1484

DInt RW=1612

DExt RW=1674 RW
18 64 74 64 43
27 78 121 31

HRW=407

4 2 2 2 2 4 2 2 2 2 2 4 2 2 2 2 4 3 3 4
60 60 60 333

Electrostatic Shield:
2 mm Pressboard + 1 mm Copper + 1 mm Pressboard = 4 mm

9.8 Tank – “Rectangular with Radiators”

Width = 2334 mm = 155 (LV Side) + 1674 (HV Ext Diameter) + 505 (HV Side)
Large = 2800 mm = 170 + 2460 (Total Core Large) + 170
Height = 3105 mm

72
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

9.9 Losses and Impedance Measured in Factory Acceptance Tests (FAT)

To complete the transformer data and for the purpose of verification:

 The following table contains the measured no load losses (Pfe) at nominal voltage, for
the nominal and extreme taps.

Tap kV/kV Pfe(W)

1 253/69 33184
11 230/69 33184
21 207/69 33148

 The following table contains the measured short-circuit losses at 75ºC (Pcu) and the
measured short-circuit impedances at 75ºC (Uz), for the nominal and extreme taps.

Tap kV/kV Base(MVA) Pcu(W) Uz(%)

1 253/69 50 211346 22.48
11 230/69 50 213925 21.42
21 207/69 50 248126 21.04
1 253/13.8 16.67 38065 12.38
11 230/13.8 16.67 38512 12.05
21 207/13.8 16.67 42380 11.96
69/13.8 16.67 34206 3.95

9.10 Permittivities

All measurements were performed with the transformer in tank, but without oil, top lid and
bushings. The active part was vapour phase dried and was not impregnated with oil.

For the calculations the following permittivity values are suggested:

Permittivity
Air 1.0
Oil 2.2
Dry non-Impregnated Paper 2.7
Oil Impregnated Paper 3.5
Dry non-Impregnated Pressboard 3.8
Oil Impregnated Pressboard 4.4

For the test conditions, is recommended instead of the permittivity values for oil, impregnated
paper and impregnated pressboard as is usual, use the permittivity values for air, dry paper
and dry pressboard.

73
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

9.11 Nodes and Branches Numbering

The transformer is modelling using 219 Nodes and 213 Branches, according to the following
description, and the figure in the next page.

 TV Winding (T1 in Figure)

83 Branches (one branch per turn), numbering from 1 to 83
84 Nodes, numbering from 1 to 84

 LV Winding (BT1 in Figure)

64 Branches (one branch each two disks), numbering from 84 to 147
65 Nodes, numbering from 85 to 149

 HV Winding – Lower Part (AT1 in Figure)

31 Branches (one branch each two disks), numbering from 148 to 178
32 Nodes, numbering from 150 to 181

 HV Winding – Upper Part (AT1 in Figure)

31 Branches (one branch each two disks), numbering from 179 to 209
32 Nodes, numbering from 182 to 213

 RW Regulating Winding – Lower Part (AT2 in Figure)

2 Branches (one branch each ten disks), numbering from 210 to 211
3 Nodes, numbering from 214 to 216

 RW Regulating Winding – Upper Part (AT2 in Figure)

2 Branches (one branch each ten disks), numbering from 212 to 213
3 Nodes, numbering from 217 to 219

74
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

F1 +84 +149 -213 -219

83 147 209
83
148 212
82
82 146 208
81 147 211
81
145 207
80
80 146 210
79 144 206
79

CIGRE-WEG-1Ph-01c.txt
145 209
78
78 143 205
77 144 208
77 142 204
76
143 207
76
141 213
75 203
75 142 206
74
140 202
74
73 141 205
73 139 201
72 140
72 204
71 138 200
71 139
203
70 137
70 199
138
69 202
69 136
198
68 137
68 201
135
67 197 218
67 136
200
66 134
196
66 135
65 199
133

DTDS To DDI - 11/06/2016 12:20:55

65 195
64 134
198

IdTrafo: CIGRE-WEG-1Ph-01
64 132
63 194
133
63 197
62 131
193
62 132
61 196
130
61 192
60 131
195
60 129
59 191
130
59 194 212
128
58 190
58 129
57 193
127
57 189
128
56 192
56 126
55 188
127
55 125 191
54 187
54 126
53 124 190
53 125 186
52 189
123
52
124 185
51
51 122 188
+217
50 184
123
50
121 187
49
49 122 183
48 120 186
48
121 182
47
47 119 185
46 120 181
46
118 184
45
45 119 180
44 117 183
44 179
118
43
43 116 +182
42 117
42 115 +181
41
116 178
41
40 114 180
40 115 177
39
113 179
39
38 114 176
38 112 178
37 113 175
37
36 111 177
36 112 174
35 110 176
35
111 173
34
34 109 +216
175
33 110 172
33
108 174
32
32 109 171
31 107 173
31 108
30 170
30 106 172
29 107
169
29 105
28 171
106
28 168
27 104
170
27 105
26 167
103
26 169 211
25 104
166
25 102
168
24 103
24 165
23 101
167
23 102
164
22 100
22 166
101
21 163
21 99
165
20 100
20 162
98
19 164
19 99
161
18 97
18 163
98
17 160 215
96
17 162
16 97
16 159
95
15 161
96
15 158
14 94
14 160
95
13 93 157
13 159
12 94
12 92 156
11 93 158
11 155
91
10
92 157
10
9 90 154
210
9 156
91
8
89 153
8
7 90 155
7 88 152
6
89 154
6
5 87 151
5 88 153
4
86 150
4
3 87 152
3 85 149
2 151
86
2
1 84 148
-1 -85 -150 -214
T1 BT1 AT1 AT2

75
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Y1 X1

+84 +149 -213 1

2
-219 3
4
5
k - 6
HV RW 7
+ 8
9
10
+217 11

+182
TV LV
+181
+216

HV RW

-214
-1 - 85 - 150

Y2 X0 H1 H0

Permanent connections:
o 181 – 182
o 150 – 213
o 216 – 217
o 215 – 218
o 214 – 219

Other connections must be made depending on the OLTC position

Terminals and internal accessible points:

Terminals
and Internal Nodes
Accessible Points
H1 181 – 182
k OLTC 150 – 213
+ OLTC = 11 OLTC 216 – 217
- OLTC = 1 OLTC 214 – 219
6 OLTC 215 – 218
X1 149
X0 85
Y1 84
Y2 1

76
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

10. recommendations
This TB has presented extensive (downloadable) measurements on two transformers, along with a
detailed description of the transformer's geometry and internal connections. The Task Force members
consider that the data offers many uses.

Research on transformer modeling

Universities, research institutes and transformer manufacturers should utilize the provided geometry
data and measurements to develop improved methods for transformer modeling.

Benchmarking
Transformer manufacturers should use the data to validate their in-house transformer models that they
use for calculating the 1.2/50 µs lightning impulse voltages.

Model export to customers

If a transformer manufacturer intends to provide transformer models to its customers (see TB Part D),
he could use the data to check that his modeling approach will result in models with sufficient accuracy.

77
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

11. References
[1] Technical Brochure 642 (2015), "Transformer reliability survey", WG A2.37, www.e-cigre.org.
[2] Technical Brochure 577A (2014), "Electrical transient interaction between transformers and the
power system. Part 1 – Expertise", JWG A2/C4.39, www.e-cigre.org.
[3] Technical Brochure 577B (2014), "Electrical transient interaction between transformers and the
power system. Part 2 – Case studies ", JWG A2/C4.39, www.e-cigre.org.
[4] B. Gustavsen, A. Portillo, R. Ronchi, A. Mjelve, "Measurements for validation of manufacturer's
white-box transformer models", 4th International Colloquium Transformer Research and Asset
Management, May 10-12, 2017, Pula, Croatia. Procedia Engineering 202 (2017), pp. 240-250.
[5] B. Gustavsen, A. Semlyen, “Rational approximation of frequency domain responses by vector
fitting”, IEEE Trans. Power Delivery, vol. 14, no. 3, pp. 1052-1061, July 1999.
[6] A. Semlyen, A. Dabuleanu, “Fast and accurate switching transient calculations on transmission
lines with ground return using recursive convolutions”, IEEE Trans. Power Apparatus and
Systems, vol. 94, pp. 561-575, March/April 1975.
[7] B. Gustavsen, H.M.J. De Silva, "Inclusion of rational models in an electromagnetic transients
program – Y-parameters, Z-parameters, S-parameters, transfer functions", IEEE Trans. Power
Delivery, vol. 28, no. 2, pp. 1164-1174, April 2013.
[8] B. Gustavsen, “Wide band modeling of power transformers”,
IEEE Trans. Power Delivery, vol. 19, no. 1, pp. 414-422, Jan. 2004.
[9] B. Gustavsen, "Eliminating measurement cable effects from transformer admittance
measurements", IEEE Trans. Power Delivery, vol. 31, no. 4, pp. 1609-1617, August 2016.
[10] B. Gustavsen, “Removing insertion impedance effects from transformer admittance
measurements”, IEEE Trans. Power Delivery (Letters), vol. 27, no. 2, pp.1027-1029, April 2012.
[11] B. Gustavsen, A. Portillo, R. Ronchi, A. Mjelve, "High-frequency resonant overvoltages in
transformer regulating winding caused by ground fault initiation on feeding cable", IEEE Trans.
Power Delivery, vol.33, no. 2, pp. 699-708, April 2018.

78
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Appendix A. Impulse Plots for 3-ph transformer

79
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

80
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

81
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

82
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

83
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

84
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

85
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

86
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

87
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

88
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

89
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

90
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

91
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

92
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

93
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

94
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

95
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

96
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

97
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

98
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

99
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

100
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

101
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

102
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

103
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

104
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

105
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

106
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

107
 TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

Appendix B. impulse plots for 1-ph transformer

108
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

109
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

110
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

111
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

112
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

113
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

114
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

115
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

116
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

117
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

118
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

119
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

120
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

121
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

122
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

123
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

124
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

125
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

126
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

127
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

128
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

129
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

130
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

131
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

132
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

133
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

134
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

135
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

136
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

137
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

138
TB 904 - High-frequency transformer and reactor models for network studies
Part E: measurements and transformer design details

139
 ISBN : 978-2-85873-609-6

TECHNICAL BROCHURES
©2023 - CIGRE
Reference 904 - April 2023