The Primary Aspects of Improving the Electrical Strength of Cast

Epoxy Insulation on High-Voltage Devices

V.N. VARIVODOV, D.I. KOVALEV, D.V. GOLUBEV, E.M. VORONKOVA

Department of High Voltage Technique and Electrophysics

National Research University «Moscow Power Engineering Institute»

Krasnokazarmennaya street 14-1, Moscow

RUSSIA

Abstract: - Various technical solutions are used to meet existing requirements for insulating high-voltage

equipment, and the widespread introduction of solid insulation is one of them. Recently, there has been a

noticeable wide transition to composite materials with improved strength properties. To justify the use of such

materials, it is necessary to be guided by statistical laws of electrical strength distribution from various

parameters, particularly the size of insulation, and its volume, to analyze breakdown probabilities. When

selecting an appropriate type of material, one should also rely on the filler's type, size, and structure,

temperature coefficient difference of linear expansion for electrode and cast insulation materials, and a possible

increase in adhesion of metal elements epoxy compounds.

The article considers in detail the issues of determining the distribution of electrical strength from various

parameters, describes the theories of dielectric failure and ways to increase insulation, and also presents for the

first time the experience of high-quality adhesion of electrodes with composite materials in the absence and

pre-application of a small layer of compound on the electrode surface before the main filling with solid

insulation. The presented results cover experiments on the strength of cast epoxy insulation samples when

activating the electrode surface with alkali, potassium dichromate, and in the absence of activation. At the

same time, for a better understanding of the ongoing processes and changes in the electric field strength, the

main influencing factors and the mechanisms of the electrical breakdown development are taken into account.

Key-Words: - Adhesion, breakdown probability, composites, linear expansion coefficient, breakdown, solid

insulation, electrical strength.

Received: April 23, 2021. Revised: April 21, 2022. Accepted: May 22, 2022. Published: June 17, 2022.

1 Introduction

Currently, solid high-voltage insulation is used

in almost all areas of the electrical power

industry: in production, conversion and

distribution of electrical energy in substations,

and the transmission of this energy (within

power cables, overhead and gas-insulated lines).

Generally, high-voltage insulation means

insulation under rated voltages of several

kilovolts or higher.

Even before the middle of the XX century,

rubber, glass, and porcelain were used as solid

types of insulation. At the same time, synthetic

organic materials were used soon after that:

instead of rubber insulation in electrical cables,

instead of glass and porcelain as insulators of

overhead transmission lines. However,

widespread use of this type of insulation in the

energy industry was hampered by generally

recognized disadvantages of original polymers,

such as their relatively low (compared to steel)

mechanical strength, insufficient thermal

resistance, and also organic insulation’s

susceptibility to weathering and aging [1].

It should be noted that specific technical

requirements for solid insulation of high-

voltage devices include high electrical and

mechanical strength, thermal stability and

thermal conductivity, and resistance to defect

creation during operation. The requirement for

insulation plasticity, which is typical for high-

voltage cables, has not become the main one for

most of other high-voltage insulating structures

in substation and line equipment.

Nevertheless, the progress curve in the

development of insulation types [2] indicates a

significant increase in the use of composites and

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

103

Volume 17, 2022

polymers compared to glass and ceramics since

the 1960s.

Specific technical requirements for solid

insulation of high-voltage devices include high

electrical and mechanical strength, thermal

conductivity, thermal stability, and resistance to

defect creation. Various composite materials

with improved strength characteristics increase

electrical and mechanical strength.

It is worth noting that the requirement for

insulation plasticity, which is typical for high-

voltage cables, has not become the main one for

the most other high-voltage insulating structures

in substation and line equipment. As a result,

the use of thermoplastic compounds for high-

voltage insulators as the main insulating

material is unjustified. Since some drawbacks

characterize them, the main ones being a creep

and sharp deterioration of mechanical properties

at elevated temperatures, up to the loss of form

stability, thermosetting compounds have

become the most widely used in substation

equipment.

Among thermosetting compounds in high

voltage electrical engineering, epoxy

compounds and their modifications are the most

promising since they feature a number of the

following advantages [3]:

- sufficient dimensional stability of cured

devices;

- good resin/filler compatibility;

- excellent electrical and mechanical

properties preserved up to 130-150ºC with a

properly selected composition;

- high chemical resistance;

- good adhesion to the metal parts to be cast.

Cast epoxy insulation (CEI) has been finding

broader use in high-voltage power, plants and

switchgears, power and instrument transformers

[4]. So, for example, in the manufacture of dry

transformers with cast insulation, as a rule,

epoxy insulation with special fillers providing

low spontaneous combustion is used [5]. In the

case of switchgears structures, the surfaces of

all elements are covered with solid insulation.

Epoxy insulation is a significant material used

to insulate SIS–type switchgears and gas-

insulated switchgears (GIS) [4].

Currently, epoxy resins are the main binders

used for CEI [1, 2] – sometimes, they are also

referred to as ethoxyline resins.

It was possible to overcome the thresholds of

mechanical strength properties of polymers,

including epoxy compounds, by transitioning to

composites, mainly when composing polymers

with glass and carbon plastics. However, the

necessity to reduce the dimensions of high-

voltage devices and the creation of equipment

for higher voltage classes requires the usage of

solid insulation with even higher electrical

strength. For this purpose, the article considers

the influence of the electrical insulation strength

of composite materials on various parameters,

which are presented in Table 1 [2-3, 6-9].

Table 1. Dependence of the electrical

strength with solid insulation on various factors

Indicator

Literature

Dependence

Formula

in the

article

Breakdown

time τ

[6]

Life curves

1

Probability

of

breakdown

[2, 7, 9]

Weibull's Law

2

Probability

of

breakdown

time

[6, 7]

Life curves,

Weibull's Law

1 and 2

Active

volume

[7, 8]

Weibull's

Law,

empirical

formula

2 and 4

A detailed analysis of the influence of these

factors will improve existing materials and

allow combining various ways to improve

strength characteristics. In addition to the

theoretical analysis of the influencing factors on

the composite material insulation, the article

also provides practical experience in the

applicability of methods to increase electrical

strength: both scientists from different countries

and the authors of the article. Thus, the article

has beneficial interest for people studying

composite materials, employees of the current

electric power market, as well as workers in the

production of insulation and electrical

equipment. This in turn allows us to use the

article as a guide when designing.

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

104

Volume 17, 2022

It should be understood that the insulation

strength is determined by the dielectric

breakdown. The latter, in its turn, are associated

with the presence of electrodes with a small

radius of curvature and inhomogeneities (at the

molecular level, micro-and macro-levels), or

rather internal thermomechanical stresses near

these inhomogeneities. Therefore, to increase

the electrical strength, the choice of insulation

should also be based on:

1) The type, size, and structure of the

insulation filler [1, 2], which significantly

increases the insulation strength when micro

and nano-sized particles are introduced;

2) Differences in the temperature

coefficients of linear expansion for the

electrode material and cast insulation. The

difference in these coefficients is reduced by

using special damping layers at the boundary

between the electrodes and the compound or by

introducing an improved type of compound and

its manufacturing technology;

3) Increasing the adhesion of metal elements

and epoxy compounds by activating the surface

layer of the poured metal elements.

The authors of this article have done a lot of

work not only to determine the distribution of

electrical strength from various parameters,

describe the theories of dielectric failure and

ways to increase insulation, but also, for the

first time, present the experience of high-quality

adhesion of electrodes and composite materials

without a small layer of compound on the

electrode surface before the main filling with

solid insulation and with this layer. This method

has not been described anywhere in the

literature before.

The article discusses the issues of increasing

the electrical strength of the CEI. At the same

time, for a better understanding of the ongoing

processes and changes in the electric field

strength, the main influencing factors are taken

into account, as well as the mechanisms of the

development of electrical breakdown.

Thus, the work is aimed at:

1) analyzing the dependence of the strength

of composite materials on various factors based

on voltage-time characteristics (life curves),

probability theory, as well as theories of

breakdown development (theory of shock

ionization by electrons, kinetic thermal

fluctuation theory, etc.);

2) determination of ways to increase

electrical strength.

It should be understood that the considers the

main influencing factors on electrical strength

and ways to increase it.

The authors' practical experience of this

article is limited only to such a way of

increasing the insulation strength as adhesion.

The work aims to demonstrate the improvement

of adhesive properties in various ways.

2 Effect on Electrical Strength

2.1 Voltage-time Dependencies

In addition to short-term electrical strength, the

electrical strength of cast polymer insulation during

long-term exposure to voltage is also important to

ensure the reliable operation of the equipment.

Indeed, the electrical strength of insulation during

prolonged exposure to voltage can be 3 or more

times lower as compared to short-term exposure to

voltage, even in a weakly heterogeneous electrical

field, and in a sharply heterogeneous electrical field

the long-term electrical strength falls even further -

Figure 1.

Fig. 1: “Life curves” of cast epoxy insulation

when exposed to industrial frequency voltage

[6], where: 1 – average values of electrical

strength for CEI (weakly heterogeneous

electrical field);

2 – electrical strength at breakdown

probability 0.05 (weakly heterogeneous

electrical field);

3 – average values of electrical strength

(sharply heterogeneous electrical field);

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

105

Volume 17, 2022

4 – the lowest limit value of electrical

strength of insulation in a weakly

heterogeneous electrical field; 5 – permissible

electrical strength of insulation in a sharply

heterogeneous electrical field

As a rule, the voltage-temporal dependences

(or “life curves”) of electrical strength of cast

epoxy insulation in a weakly heterogeneous

electrical field when exposed to industrial

frequency voltage are well approximated for

average values by the following ratio:

(1)

where τ is the time to insulation breakdown

when exposed to the electrical field with

intensity ; is the time before insulation

breakdown when briefly exposed to the

electrical field with intensity ; m is the power

characterizing the aging rate, equal to 10-15 for

a weakly heterogeneous field, and about 12 on

average.

In a sharply heterogeneous electrical field,

when “life curves” are approximated by

круratio (4), the dependence differs from a

linear one, and power m decreases significantly

with short voltage exposures (Figure 1).

Thus, the question of improving the

electrical strength of CEI, with both short-term

and long-term exposure to voltage, remains

relevant.

Fig. 2: The functions of the CEI electrical

strength distribution in short-term exposure to

voltage (1), including partial discharges (3) and

without PD (2); =13 kV/mm

For weakly heterogeneous electrical fields of

real structures characterized by a large

dielectric volume, the electrical strength

distribution function, Figure 2 shows, is best

described by Weibull’s law [2, 7, 10]:

(2)

where E is the electrical strength at a given

breakdown probability P; is the lowest limit

of electrical strength; is the electrical

strength at breakdown probability P equal to 1-

1/e; β is a measure of dispersion.

2.2 Breakdown Time Distribution

When selecting admissible values of the CEI

electrical strength in weakly heterogeneous

electrical fields in prolonged exposure to operating

voltage, taking into account statistical features of

dispersion of the experimental data, the optimum

approximation of the volt-temporal dependences, as

Figure 3 shows, is achieved by using the ratio (3).

This ratio accounts for the lower electrical

strength limit and compliance of the time

distribution to breakdown function to the

statistical Weibull law [7]:

(3)

where is the most frequent value of time to

breakdown in short-term exposure to the

electrical field with strength ; is the

electrical field strength in prolonged exposure

to voltage; is the lower electrical strength

limit in prolonged exposure to voltage; P is the

breakdown probability; βτ is a measure of

dispersion.

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

106

Volume 17, 2022

Fig. 3: The time distribution to CEI breakdown

function (b), particularly with partial discharges

(a) and without PD (c) at different electrical

field strengths (1 – =14±2 kV/mm, 2 –

=17±2 kV/mm, 3 – =20±2 kV/mm, 4 –

=24±2 kV/mm, 5 – =14±2 kV/mm)

Fig. 4: The CEI electrical strength in a weakly

heterogeneous electrical field in prolonged

exposure to voltage and different breakdown

probabilities (solid, dash and dash-dot lines

mean all the samples, without and with PD,

respectively; for broken samples, the

experimental data is straight horizontal lines

with constraints, unbroken samples are depicted

by points with arrows; numbers indicate the

number of samples)

Subject to [8], the calculation for ratio (3)

corresponds well to the results of experiments

shown in Figure 4.

Scientists from China [9] also conducted similar

studies of the breakdown probability distribution

versus time. The main conclusion of their work [9]

is the explanation of the effect of temperature on the

aging rate. Namely, at higher temperatures, the

breakdown time is reduced.

2.3 Active Volume

Statistical regularities can also explain the existing

dependence of the CEI electrical strength on the

insulation size.

By assuming that the electrical strength of

some single volume of dielectric does not affect

the electrical strength of another single volume,

and at the same time, the breakdown probability

of every single volume is equal to , and

where the breakdown of one single volume

entails breakdown of the entire insulation gap

with the volume times larger than the single

volume, we can determine the breakdown

probability of the entire insulation by the

formula

(4)

Using ratios (2, 4), it is easy to achieve the

dependence of insulation electrical strength on the

dielectric volume:

(5)

Paper [7] has shown the defining effect of

“active” compound volume (where the

electrical field strength is at least 85% of the

highest strength in the gap) on the electrical

strength in samples with cast metal electrodes.

However, it is precisely in this volume where

electrical and internal thermomechanical

stresses have the highest values since this

"active volume" is located near the cast

electrodes.

It should be noted that the electrical strength

value determined for the same material depends

very strongly on the method of electrical

strength testing. In Russia, for epoxy

compounds, there are standardized testing

methods to determine the electrical strength in a

homogeneous and sharply heterogeneous

electrical field [11]. In global practice, standard

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

107

Volume 17, 2022

ASTM D149 - 20 [12] is one of the main

regulatory documents for determining the

electrical strength of cast insulation.

The effect of a lower electrical strength limit

on volt-temporal dependences of electrical

strength is experienced at low breakdown

probabilities (Figure 1 - curve 2) since in this

case, the dependence becomes nonlinear, and

the results of processing of various statistical

data show that the electrical strength limit in a

weakly heterogeneous electrical field is 5

kV/mm or higher, and in a sharply

heterogeneous electrical field it is about 2

kV/mm [3].

Despite numerous theoretical and

experimental studies of electrical strength in

dielectrics, particularly polymers, there is still

no unambiguous understanding of their

breakdown mechanism when exposed to high

voltage.

3 Mechanisms of Electrical

Breakdown

When explaining the electrical breakdown

mechanism in solid dielectrics, two scientific

areas have formed in the literature. Most of the

experimental data show that the electrical

breakdown of solid dielectrics emerges due to

electron impact ionization [13-15]. However,

several papers [16, 17] deny that electron

impact ionization can develop in solid

dielectrics, particularly polymers, at least in the

initial phase: electrical breakdown of solid

dielectrics is explained in the context of

overheated electrical instability and electronic

detonation during solid dielectrics breakdown.

In particular, the kinetic thermofluctational

theory of material breakdown [18] assumes that

material breakdown is caused by the breakage

of chemical bonds due to the combined action

of energy of molecular thermal motion and

some external force (mechanical load, electrical

field, etc.).

This theory proposes to consider the

breakage of chemical bonds due to the energy

of thermal vibrational motion as a mechanism

of solid dielectric breakdown, such breakage

being facilitated by the distortion of energy

zones under the influence of the electrical field.

Here, chemical bond breakage and polymer

breakdown occur in the initial phase not under

direct electrical field effect but rather due to the

energy of atomic and molecular thermal

vibrational motion, taking into account energy

zone distortions due to the electrical field effect:

(6)

where D is the chemical bond dissociation

energy, NA is the Avogadro number, ρ is the

material density; R is the interatomic

equilibrium distance, x is the relative load

affecting the chemical bond, K is the universal

gas constant, T is temperature, and is time

to breakdown under prolonged and short-term

exposure to voltage, is the dielectric structure

heterogeneity factor, is the electrical field

heterogeneity factor, a is the constant, M is the

molecular weight, is the elastic modulus,

is the absolute permittivity, is the relative

permittivity.

This theory implies that the destruction of

the dielectric is associated with the sequential

destruction of molecular bonds: after the break

of one bond, other breaks follow. This

phenomenon is due to the fact that when bonds

are broken, new charges are formed,

transferring part of their energy to neutral atoms

[18].

The equations presented by the authors of

the theory show that the electrical strength of

polymers increases with increasing chemical

bond dissociation energy and decreases with

increasing temperature and voltage application

time, which does not contradict experimental

data on dielectric breakdown.

This approach also makes it possible to

justify the existing varieties of dielectric

breakdown:

- purely electric;

- electrothermal;

- electromechanical;

- electrochemical.

It has been found experimentally that

manifestation of some type of breakdown

depends on the electrical field form, dielectric

structure, properties, presence of defects,

cooling conditions, voltage exposure time, etc.,

which also does not contradict the

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

108

Volume 17, 2022

thermofluctational theory of dielectric

breakdown.

It should be noted that the thermofluctational theory

does not consider the discharge development

mechanism and the breakdown channel formation

process. After all, the low mobility of ions formed

during the breaking of chemical bonds does not

provide a short time for discharge development [18].

Therefore, in this theory, there is no criterion for

considering the discharge's development. However,

this theory describes the main directions of the

increase in electrical strength of polymers

convincingly.

4 Ways to Increase Electrical

Strength

There are various ways to increase the electrical

strength of the insulation. Recently, the most

common method is the introduction of nanofillers

into the structure of the compound [1, 2]. However,

traditional methods such as increasing the adhesion

of dissimilar materials (metal elements and

insulating compounds of high-voltage electrical

equipment) should not be excluded.

The following technological methods of

increasing the adhesion of the compound and

metal are usually adopted:

– increasing the active adhesion area

(creating roughness) of the metal;

– selection of materials with similar linear

expansion coefficients [19];

– activation of the metal surface itself by its

chemical treatment [20];

- degreasing of the connected surfaces [20];

– application of intermediate substrates

having average thermal expansion coefficients

between similar parameters of metal and

compound;

– using constructive techniques by shaping

and placing the reinforcement in a polymer

material, providing compression due to

shrinkage phenomena.

4.1 Features of Polymer Microstructure and

Its Effect on Electrical Strength

Paper [21] also notes the difficulties in calculating

the electrical strength of complex polymer

dielectrics based on existing quantum-mechanical

theories. Considering polymers structure in the

context of modern views on their structure

suggested that the so-called "free" volumes, i.e.

areas in polymer volume induced by loose packing

of macromolecular chains, play an essential role in

the electrical breakdown.

Cured epoxy resins have a

microheterogeneous structure of globular type

[22]. Sizes of globular particles (about 100

angstroms) depend on the ingredients of

composition and curing conditions (particle

sizes decrease with increasing temperature). As

globule sizes decrease, the polymer electrical

strength increases. Glass transition temperature,

compressive strength, chemical and thermal

resistance increase as distances between mesh

nodes become shorter, but this generally

increases the polymer brittleness.

When epoxy resin compositions comprise

low molecular weight compounds (such as

plasticizers) or oligomers of other types (such

as oligoethers) that contain too few or no

reactive groups, such components do not

participate in mesh formation, but rather

accumulate at globular formation interfaces,

thereby causing a significant decrease in

mechanical and electrical strength, and also

thermal and chemical resistance.

Therefore, the electrical strength of different

cured epoxy compounds can vary significantly

and manifests generally in the range of 15-35

kV/mm (for a standard electrical strength

definition in a slightly heterogeneous electrical

field) in short-term exposure to industrial

frequency voltage and 20° C temperature.

It should be noted that electrical strength is

determined not only by molecular level

heterogeneities, but also by micro-level

insulation heterogeneities (e.g., CEI filler

particles) and even by macro-level ones, namely

gas inclusions (pores, cracks, and

delaminations).

In the meantime, it is exactly the levels of

internal thermomechanical stresses near some

heterogeneity, particularly near the cast metal

elements, that form some additional external

force affecting intermolecular bonds, that

determines the electrical strength.

Thus, increasing the CEI electrical strength

can be achieved by selecting resins and

compounds with increased chemical bond

dissociation energies and low levels of local

thermomechanical stresses in the material.

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

109

Volume 17, 2022

Therefore, the selection of type, size, and

structure of epoxy insulation fillers [1, 2, 23] is

one of the most important areas for improving

the electrical strength of cast insulation.

In [24] it is shown that the electrical strength

of epoxy insulation increases by using nano-

size particle (7 to 80 nanometers) filler, and the

use of micro-size (i.e., about one micron)

particles can increase electrical strength by

improving volume dispersion (Figure 5).

Fig. 5: Breakdown voltage of 1mm CEI

samples with presence of filler, its type and

filler particle size [24]

It is generally believed that an increase in

electrical strength and other parameters of

epoxy compounds with a decrease in filler

particle sizes is associated with an increase in

the surface of interaction between filler

particles and epoxy resin [1, 2, 4].

Japanese researchers [25] have created an

epoxy compound where electrical strength

increased about 1.5 times (Figure 6). The glass

transition temperature (Tg) is 30°C higher than

conventional epoxy compounds. Breakdown

probability due to adhesion or temperature

fluctuations is decreased greatly by a very low

value of the thermal expansion factor.

The new compound uses a special type of

filler, namely spherical silica (instead of a

conventional amorphous one) and a small

number of rubber particles.

Fig. 6: Electrical strength of conventional and

optimized compounds according to [25]

Conventional material consists of a

biphenol-type epoxy resin, phthalic acid

anhydride, and amorphous silica filler. Material

of this type is used widely for high-speed

automatic injection molding. The developed

material consists of another type of biphenolic

epoxy resin and phthalic acid anhydride, and it

is filled with a large amount of spherical silica

and a small number of rubber particles.

The glass transition temperature of the

compound developed is increased due to the

changed structure of epoxy resin, thereby

improving such parameters as strength,

elasticity, electrical properties, etc.

Also, in the literature [1], there were

presented results indicating an increase in the

insulation characteristics of high-voltage

insulators with the introduction of insulation

nanofillers, such as SiO2 and h-BN. Due to

better bonding of fillers and nano- or micro-

sizes, scientists managed not only to increase

various characteristics of the component,

including increasing the resistance to corona

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

110

Volume 17, 2022

discharge, but also to slow down the aging

process of insulation.

The most common synthetic inorganic fillers

that are used to enhance the properties of

polymers are silicon dioxide (SiO2), titanium

oxide (TiO2), aluminium oxide (Al2O3), zinc

oxide (ZnO), magnesium oxide (MgO)),

nitrides (boron nitride (BN), aluminium nitride

(AlN), silicon nitride (Si4N3)) and carbides

(silicon carbide (SiC)). [1, 2, 4, 26].

4.2 Accounting for Linear Expansion

Coefficients

Where cast metal elements are used, there is a

manifestation of the effect of thermomechanical

stresses associated with the emergence of

compound deformations caused by temperature

coefficient difference of linear expansion for

electrode and cast insulation materials.

Available data [23] illustrate electrical strength

with electrode material grade. The use of metal

electrodes cast into a sample, compared to

electrodes made by metallization on a sample,

reduces the electrical strength of epoxy

insulation significantly.

During prolonged exposure to voltage, a

compound’s breakdown mechanism depends on

the accumulation of mechanical and electrical

micro-and macro-damage in insulation and on

its development.

Under the exposure to high voltage, the

material breakdown process generally begins in

a local area with the highest electrical field

strength, i.e., near electrodes with small

curvature radii or near heterogeneities contained

in the compound and distorting the electrical

field. Such heterogeneities can be, for example,

gas pores, filler accumulations, etc.

The CEI electrical strength increases slightly

with increasing temperature up to the onset of

thermal degradation [18]. This can be explained

by the positive effect of internal

thermomechanical stress decreases on electrical

breakdown processes.

The electric strength of the LEI is most

strongly reduced by gas inclusions - pores,

cracks, delaminations at the border with metal

elements, since partial discharges (PD) can

occur in these inhomogeneities.

Under the effect of PD, the following

processes can emerge in the solid insulation of

electrical conductors:

– Formation of gaseous ionization products,

namely ozone, carbon oxides, nitrogen oxides,

etc.;

– Chemical and structural breakdown of the

dielectric under exposure to ionization products,

such destruction being accompanied by

chemical bond breakages, formation of new

groups in polymer macromolecules, and also

the formation of carbon, etc;

– Direct effect on dielectric by ion and

electron bombardments, radiation generated at

PD;

– Increase in the local electrical field

strength and temperature in the PD zone.

Partial discharges generally do not induce

the rapid breakdown of insulation gaps, the

process of PD development is rather slow and

depends on partial discharge intensity, therefore

the PD effect is particularly strong under

prolonged exposure to voltage (Figure 4).

However, in weakly heterogeneous electrical

fields with high average breakdown strengths,

there is a decrease in electrical strength of cast

insulation in the presence of PD even under

short-term voltage exposures, provided that

partial discharge intensities exceed several

picocoulons (Figure 7).

Fig. 7: Electrical strength of cast epoxy

insulation in a weakly heterogeneous electrical

field, with partial discharge intensities at a

smooth rise of industrial frequency voltage [8]

Gas inclusions can appear in insulation

during manufacturing due to technological

process violations that emerge when shrinking

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

111

Volume 17, 2022

cast masses and compounds and low-quality

bonding between electrodes and dielectric.

Gas inclusions also emerge in operation due

to cracking or delamination of insulation caused

by high internal thermomechanical stresses.

Breakdown of solid insulation in electrical

conductors, when exposed to partial discharges,

is caused mainly by discharge energy release in

a gas inclusion. The absolute value of energy

dissipated in a discharge is typically low.

However, it is transmitted to quite a small area

of inclusion surface, where local temperature

rises immediately.

This exposure entails the breakdown of a

small volume of dielectric with the formation of

byproducts, sometimes chemically active

products. With multiple repeated partial

discharges, the inclusion surface breaks down

gradually, local depressions appear on it, and

they grow over time to form narrow branched

channels in the dielectric. Ultimately, the

process ends with a complete breakdown of

insulation.

High internal thermomechanical stresses

cause a negative effect on the bonding between

the compound and the metal electrodes, such

stresses emerge due to compound shrinkage

during curing, different linear expansion

coefficients of the compound, and the cast

reinforcement, and there are frequent attempts

to compensate for such negative effect by

selecting close linear expansion coefficients of

dielectric and metal using special damping

layers on the interface between the compound

and cast reinforcement, developing and

applying special compounds and technologies

of their production.

Table 2. Linear expansion coefficients of

different materials

Materials

Linear expansion

coefficient,

Epoxy resin

50 to 60

Epoxy resin + 200%

(by weight) filler in

the form of powdered

quartz

25 to 30

Aluminum

22.2

Duralumin

23.5

Brass

18.7

Copper

16.6

Cast iron

10.4 to 13.0

Porcelain

3.6 to 4.5

The table data shows that the difference

between the temperature elongation of epoxy

resin and other structural materials is extensive.

The difference is decreased by adding inert

filler such as dusty silica sand; this produces the

epoxy compound required for casting. The use

of current-carrying parts made from aluminum

in cast-insulated devices is justified since linear

expansion coefficients of the aluminum and

epoxy compound are almost equal. The method

of creating elastic liners between cast metal

poured parts and the compound has also

justified itself.

4.3 Degreasing and Chemical Activation of

Surfaces

Increased epoxy compounds adhesion to cast

metal parts can also be achieved by activating

the surface layer of cast metal reinforcement,

particularly by degreasing this surface.

Of all the theories of adhesion justification

aimed at describing the causes of adhesion of

dissimilar surfaces, the most common is the

molecular theory of adhesion [27]. According to

this theory, the adhesion of different surfaces

for bodies is associated with the action of

interatomic (chemical) bonds and (or)

intermolecular (physical) forces. The latter, in

turn, have an electrical nature of origin. Since

adhesion occurs due to the presence of

interatomic and intermolecular bonds, for a

tight adhesion of surfaces (strong) there must be

chemically active, polar, or polarized

substances on both surfaces. Fats are not such

materials, so it is necessary to perform

degreasing, i.e. the process of removing various

fatty films, etc.

5 Practical Experience for Coupling

of Heterogeneous Surfaces

It should be noted that the quality of bonding

between epoxy compound and cast metal

electrodes can be evaluated by the value of the

tensile load at failure. Where the bonding

between electrodes and the compound is poor,

tensile failure of a sample should be expected to

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

112

Volume 17, 2022

happen in a way that the tear cross-section

would accommodate almost the entire active

surface of a metal electrode, and the loads at

failure would be minimal. Conversely, the

better the bonding, the smaller the percentage of

metal electrode surface in the tear cross-section.

With perfect bonding, the tear can only happen

along the dielectric body.

As a practical experience, it was decided to

study the quality of adhesion on the metal

electrode surface with a compound in design

samples, depending on various methods of

surface activation.

The experimental work consisted of two

parts:

- production of the samples themselves;

- conducting tests of the mechanical strength

of samples.

The method of sample production was

carried out as follows:

1. manufacture of aluminum metal elements;

2. the implementation of surface treatment of

aluminum elements in various ways

(mechanical treatment, degreasing, chemical

activation (for example potassium dichromate));

3. applying a pre-coating insulation layer of

1-2 mm on the aluminum surface. This item is

present only during the second experiment;

4. filling the structure with the main volume

of the compound under vacuum (KF-1 and KE-

3 compounds).

The test procedure was as follows. The M16

thread was cut into the embedded metal

elements, where the bolts were inserted. A

smooth mechanical tensile load was applied to

the bolts at a speed of 20 kg/s until destruction

with the help of a special breaking machine.

The completed studies of the quality of

bonding between the surfaces of metal

electrodes located in electrical pathways and the

compound in structural samples (Figure 8)

depending on different methods of surface

activation and, respectively, on increasing

adhesion: without special treatment, with alkali

degreasing, and also with potassium dichromate

treatment, showed that activation of electrode

surfaces can significantly increase adhesion and

degree of bonding between electrodes and the

compound, namely with alkali treatment by

about 40% and with potassium dichromate

treatment by 88% (Figure 8).

Fig. 8: Mechanical strength of cast epoxy

insulation samples under tensile load with

different methods of electrode surfaces

activation

The main reason for such adhesion is the

force field of molecules located in the solid

metal surface, such field pulling molecules

located in close proximity to this surface of the

liquid polymer during its curing. Actually, the

effect of the field of molecules located in the

compound extends to the depth of a single

molecule. The contact layer is formed by

surface polymer molecules, and they, in

contrast to chaotic and disorderly arrangement

in the thickness, acquire an oriented and

ordered structure.

There were also studies to relieve internal

mechanical stresses near electrodes by applying

an intermediate layer of the compound to the

surfaces of electrodes before the main casting

(Figure 9).

The results obtained (Figure 9) show that

this technological technique is one of the most

effective ways of improving the bonding

between electrodes and the compound since

even if untreated electrodes are covered with a

compound layer it gives a 54% increase in

strength and a 126% increase with additional

potassium dichromate treatment of electrodes

(by 183% for the lowest values).

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

113

Volume 17, 2022

Fig. 9: Effect on the mechanical strength of cast

epoxy insulation samples under the tensile load

of compound layer precoat on electrode

surfaces

The noted positive effect of applying a

compound layer precoat and potassium

dichromate treatment of electrodes is caused by

two reasons: oxide film removal from electrode

surfaces and decrease in internal

thermomechanical stresses in the compound

near metal electrodes, provided that they are

covered with a thin compound layer.

From the results presented in Figures 8 and

9, the following conclusions can be drawn.

Firstly, degreasing the surface with precoat

increases the mechanical strength of the

samples by about 1.5 times, and with chroma

peak by about 2 times. Secondly, the

application of an additional layer of insulation

1-2 mm thick on aluminum elements before the

main filling with the compound contributes to

the greater mechanical strength of the samples,

i.e., more adhesion. However, the emergence of

precoat requires solving the issues of compound

autohesion in the technological process of

manufacturing, i.e., ensuring high physical and

chemical properties of the bonding between

individual layers.

Internal thermomechanical stresses can also

be decreased by increasing the compound

plasticity near an electrode since the emergence

of thermomechanical stresses in the compound

is associated with shortening interatomic

distances. Dielectric’s elastic state is caused by

an increase in the range of interatomic distance

fluctuations.

The application of a semiconductive layer

around the conductor is one of the design

solutions aimed at preventing partial discharges

as a result of insulation delamination from the

conductor. However, even here the issues of

good bonding between this layer and the

compound are still relevant.

The value of the obtained experimental

results in the study of various ways to improve

the adhesion of metal elements with epoxy

compounds is the fact that they were obtained

on samples as close as possible to the design of

real insulators (namely, support insulators of

gas-filled high-voltage devices), characterized

by large volumes of dielectric and metal

elements, which is associated with significant

internal thermomechanical stresses in this

dielectric.

6 Conclusion

The results given in the paper confirm that the issue

of improving the electrical strength of cast epoxy

insulation is an important one. Also, they indicate

the following points regarding this type of

insulation:

- Volt-temporal dependences of electrical

insulation strength can be described using the

proposed ratios for average electrical field

strength values, and also in small breakdown

probabilities;

- Statistical specifics of electrical insulation

strength must be accounted when calculating

the latter based on Weibull’s limit statistical

law;

- Internal local thermomechanical stresses in

dielectric have a crucial effect on the electrical

strength of cast epoxy insulation, and this effect

can be substantiated based on the

thermofluctational theory of dielectrics

breakdown;

- It is expedient to reduce internal local

thermomechanical stresses and, consequently,

increase the electrical strength via a proper

selection of insulation structure and

manufacturing technology: by optimizing

fillers, selecting linear expansion coefficients of

individual insulation components, activating the

surfaces of cast metal parts, applying special

damping layers during manufacturing.

- Gas inclusions must be prevented within

insulation, both during manufacturing and in

operation, since they decrease the electrical

strength most dramatically.

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

114

Volume 17, 2022

The correctness of conclusions made is

confirmed not only by the analysis of available

data, but also by own experimental studies

performed by the authors and given in the

article: taking into account the activation for

electrode surfaces as a result of their treatment

with solutions, and also application of damping

layers to electrodes during their manufacturing.

The authors would like to thank the

managers and specialists of the V.I. Lenin All-

Russian Electrotechnical Institute. They

allowed using the benches for experimental

studies of cast epoxy insulation strength.

References:

[1] Saleem M.Z., Akbar M., Review of the

Performance of High-Voltage Composite

Insulators, Polymers, Vol.14, No.3 (431), 2022,

pp. 1-17.

https://doi.org/10.3390/polym14030431.

[2] Pleşa I.; Noţingher P.V.; Schlögl S.; Sumereder

C. Muhr M., Properties of Polymer Composites

Used in High-Voltage Applications, Polymers,

Vol. 8, 2016, pp.1-173.

https://doi.org/10.3390/polym8050173.

[3] Varivodov V.N., Kovalev D.I., Zhulikov S.S.,

Golubev D.V., Romanov V.A., Mirzabekyan

G.Z., Technological Aspects of the Use of Cast

Polymer Insulation for High-Voltage

Switchgear and Busbars, Power Technology

and Engineering, Vol. 54, 2021, pp. 915–922.

doi: 10.1177/1050651999013002002.

[4] Hari R., Mohana Rao. M, Characterization of

nano-additive filled epoxy resin composites

(ERC) for high voltage gas insulated

switchgear (GIS) applications, International

Journal of Emerging Electric Power Systems,

Vol.23, No.1, 2022, pp. 47 - 57.

[5] Jalakas T., Janson K., Molder H., Roasto I.,

High‐voltage pulse transformer for IOT

modulators, IET Electric Power Applications,

Vol. 14, No.12, 2020, pp.2348-2354.

[6] Varivodov V.N., Kovalev D.I., Krupenin N.V.,

Khrenov S.I., Permissible Electrical-Field

Intensities in the Epoxy Cast Insulation of a 6

to 110 kV switchgear, Russian Electrical

Engineering, Vol. 89, No. 5, 2018, pp. 294–

297.

[7] Varivodov V.N., Electrical strength of cast

epoxy insulation in quasi-homogeneous fields,

Applied Physics, Vol. 5, 2001, pp. 34–40.

[8] Varivodov V.N., Internal insulation in gas-

filled high-voltage and ultrahigh-voltage

equipment, PhD dissertation, Dept. High

voltage technique, Moscow, 2000.

[9] C. Liu et al., Combined Electrical and Thermal

Aging of Alumina Filled Epoxy Solid

Insulators for GIS, 2019 IEEE International

Conference on Environment and Electrical

Engineering and 2019 IEEE Industrial and

Commercial Power Systems Europe (EEEIC /

I&CPS Europe), 2019, pp. 1-4, doi:

10.1109/EEEIC.2019.8783950.

[10] Andersen A., Dennison J., Mixed Weibull

distribution model of DC dielectric breakdowns

with dual defect modes, 2015 IEEE Conference

on Electrical Insulation and Dielectric

Phenomena (CEIDP), 2015, pp. 570-573, doi:

10.1109/CEIDP.2015.7352017.

[11] GOST6433.3-71. Solid electrical insulating

materials. Methods for evaluation of electrical

strength at а.с. voltage at power (50 Hz)

frequency and d. c. voltage. [Online].

Available:

https://docs.cntd.ru/document/1200011888.

[12] Standard ASTM D149 – 20. Standard Test

Method for Dielectric Breakdown Voltage and

Dielectric Strength of Solid Electrical

Insulating Materials at Commercial Power

Frequencies. [Online]. Available:

https://webstore.ansi.org/standards/astm/astmd

14920.

[13] Zhao L., A formula to calculate solid dielectric

breakdown strength based on a model of

electron impact ionization and multiplication,

AIP Advances, Vol. 10, No. 2, 2020, pp.

025003-1 – 025003-9.

[14] Cao L., Yuan Y., Li E., Zhang S. Relaxor

regulation and improvement of energy storage

properties of Sr2NaNb5O15-based tungsten

bronze ceramics through B-site substitution.

Chemical Engineering Journal, Vol. 421, 2021,

pp. 1-12.

[15] Gockenbach E., High Voltage Engineering. In:

Papailiou, K.O. (eds) Springer Handbook of

Power Systems. Springer Handbooks, 2021.

[16] Raju G. G. Dielectrics in electric fields, second

edition, Dielectrics in electric fields, second

edition, 2016, pp. 1-752.

[17] Li B., Salcedo-Galan F., Xidas P. I., Manias,

E., Improving electrical breakdown strength of

polymer nanocomposites by tailoring hybrid-

filler structure for high-voltage dielectric

applications, ACS Applied Nano Materials,

Vol. 1, No. 9, 2018, pp. 4401-4407.

doi:10.1021/acsanm.8b01127.

[18] Kartashov E. M., Thermal destruction of

polymeric fibers in the theory of temporary

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

115

Volume 17, 2022

dependence of strength, Tonkie Khimicheskie

Tekhnologii, Vol.16, No.6, 2021, pp. 526-540.

doi:10.32362/2410-6593-2021-16-6-526-540.

[19] Walker R. C., Hamedi H., Woodward W. H.

H., Rajagopalan R., Lanagan M. Impacts of

crosslinking and degassing on the conductivity,

dielectric loss, and morphology of low-density

polyethylene and crosslinked polyethylene,

2021, doi:10.1021/bk-2021-1375.ch011.

[20] Edward M. Petrie, Passive processing methods,

The World of Electroplating magazine, Vol. 3,

No. 03, 2007, pp. 183-272.

[21] Großmann S., High currents and contact

technology, 2021. doi:10.1007/978-981-32-

9938-2_4.

[22] Serra A., Ramis, X., Fernandez‐Francos X.,

Epoxy Sol-Gel Hybrid Thermosets, THE

Coatings, Vol. 6, No. 8, 2016, pp. 1-19.

[23] Varivodov V.N., Kovalev D.I., Golubev D.V.,

Mirzabekyan G.Z., Development of insulation

systems for high-voltage busbars with solid

insulation, Russian Electrical Engineering, Vol.

92, 2021, pp. 185–192.

[24] Liang M., Wong K.L., Electrical Performance

of Epoxy Resin Filled with Micro Particles and

Nanoparticles, Energy Procedia, Vol. 110,

2017, pp. 162–167. doi:

0.1016/j.egypro.2017.03.122.

[25] Sato J., Susumu K., Osamu S., Masaru M.,

Shimizu T., Homma M., Solid Insulated

Switchgear and Investigation of its Mechanical

and Electrical Reliability, Electrical

Engineering in Japan, Vol. 174, no. 4, 2011,

pp. 28–36.

[26] Rajkonwar L., Ram Prasath R.A., D J., Roy

N.K., Studies on Epoxy Based TiO2Nano-filler

for High Voltage Application, 2018

International Conference on Power, Energy,

Control and Transmission Systems (ICPECTS),

2018, pp. 128-132.

[27] Bukhari M.D., Gohar G.A., Akhtar A., Ullah

S., Akram M.B., Abid J., Raza H., Adhesion

Theories and Effect of Surface Roughness on

Energy Estimation and Wettability of

Polymeric Composites Bonded Joints: A-

Review. VW Applied Sciences, Vol. 2, No. 1,

2019, pp. 74-86.

Contribution of Individual Authors to the

Creation of a Scientific Article (Ghostwriting

Policy)

-Vladimir Varivodov carried out methodology

development for article, participated in writing and

conducting the research in Section 4.

-Dmitry Kovalev provisioned laboratory samples

and carried out project administration.

-Dmitry Golubev has organized oversight for the

research activity planning and execution and

conducted the research in Section 4.

-Ekaterina Voronkova was responsible for data

visualization, writing article, carried out conducting

the research in Section 4.

Sources of Funding for Research Presented in a

Scientific Article or Scientific Article Itself

Financial support for the research at National

Research University, "Moscow Power

Engineering Institute" for developing

innovative bus ducts for a nominal voltage of

6–110 kV was provided by the Ministry of

Science and Higher Education of the Russian

Federation

Creative Commons Attribution License 4.0

(Attribution 4.0 International, CC BY 4.0)

This article is published under the terms of the

Creative Commons Attribution License 4.0

https://creativecommons.org/licenses/by/4.0/deed.en

_US

WSEAS TRANSACTIONS on POWER SYSTEMS

10.37394/232016.2022.17.12

V. N. Varivodov, D. I. Kovalev,

D. V. Golubev, E. M. Voronkova

E-ISSN: 2224-350X

116

Volume 17, 2022