Journal of Energy Technologies and Policy www.iiste.

org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.1, No.2, 2011

Calculation of Electric Field Distribution at High Voltage

Cable Terminations
N. H. Malik, A.A. Al-Arainy. M. I. Qureshi and F. R. Pazheri*
Saudi Aramco Chair in Electrical Power, EE Department, College of Engineering, PO Box. 800,
King Saud University, Riyadh-11421, Saudi Arabia
* E-mail of the corresponding author: [email protected]

Abstract
High Voltage cables are used for transmission and distribution of electrical power. Such cables are subjected
to extensive high voltage testing for performance evaluation and quality control purposes. During such
testing, the cable ends have to be prepared carefully to make a proper end termination. Usually deionized
water terminations are used for testing XLPE cables. Alternatively conductive paint is used to prepare such a
termination. This paper presents an analytical method of calculating the voltage distribution across such a
resistive termination when subjected to AC voltage stress. The proposed method is used to determine the
effect of different design parameters on voltage and stress distribution on such cable ends. The method is
simple and can be used to understand the importance of stress control at a cable termination which constitutes
a critical part of such cables.
Keywords: AC voltage distribution, cable terminations, resistive terminations, stress control, XLPE high
voltage cable.

1. Introduction
High voltage cables are used extensively for transmission and distribution of electrical power and play an
important role in the electricity supply system. Such cables are being manufactured in many countries
including Saudi Arabia. These cables are subjected to high voltage testing for routine as well as type tests
and for other special investigative tests [1, 2]. Such testing is essential to ensure that the cable’s dielectric
properties are adequate to meet the specified performance requirements over the expected life time.
The electric field in a coaxial cable varies only in the radial direction as the field magnitude decreases with
increasing distance from the conductor center and can easily be calculated analytically. However, when a
cable end is terminated for testing and other purposes, the field at such an end region is no longer purely
radial and a tangential component is also introduced. Such a tangential field component can cause partial
and surface discharges which consequently can lead to breakdown of the cable insulation.
Due to this reason, during high voltage testing of polymeric cables, cable ends are immersed in
de-ionized water. Alternatively a conductive paint is applied at the cable end or some other form of
resistive-capacitive cable termination is formed to improve the voltage distribution at such ends. The
properties and selection of such stress control materials play an important role in the stress control
properties of such cable terminations [3-6]. The electric field calculations for cable terminations have been
reported in literature [7-10]. However, most of these methods employ numerical techniques and complex
computations for the field solutions and do not provide any analytical insight into the stress distributions.
This paper presents a simple analytical method to calculate the voltage distribution at a coaxial cable end
where a certain length of the grounded shield is removed and the cable end is enclosed in a resistive
medium. Analytical expressions are derived to determine the voltage and tangential electric field
distributions at such ends. The effects of different design parameters on the voltage and field distributions
are discussed. The proposed method can be used for understanding important concepts related to high
voltage cable accessories.

1
Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.1, No.2, 2011

2. Electric Stress in a Coaxial Cable

High voltage cables almost always have a coaxial configuration with conductors of inner and outer radii of a
and b, respectively. The capacitance C (F/m) of such a cable is given as:

2πε 0ε r
C=
ln(b a )
(1)

where ɛ0 = 8.854 x10 -12 F/m and ɛr = relative permittivity or dielectric constant of the insulation. For XLPE
insulation, ɛr = 2.3. When a test voltage VT is applied across the cable, a charge q = CVT is produced on the
cable’s conductor. By application of Gauss’s Law, the electric stress E in the coaxial cable insulation at a
distance r from the cable center is given as:

CVT VT
Er = = (2)
2πε 0ε r r r ln (b a )

This equation clearly states that the field is purely radial and is confined within the outer conductor. Its
value is maximum at r = a and is minimum at r = b.

3. Cable Termination and Stress Distribution

In modern cables, the insulation medium used typically is XLPE which has a high dielectric strength and is
capable of withstanding large value of imposed electrical stress. However, at the cable end, if a high test
voltage is applied, a flashover will take place in the air between the conductor and the shield since these are
relatively very close. To overcome this limitation, the outer conductor and semiconducting screens are
removed upto a length L from the cable’s end. Due to this modification, the field in the cable’s end region is
no longer radial and a tangential field component is introduced. This increases the stress value at the ground
shield cut back edge. As a result, if no corrective measures are taken, the cable will have electrical discharges
and breakdown in such a region. The corrective measures attempt to redistribute stress in the cable’s end
region.
One common method employed for the corrective measure during the cable testing is to insert the cable ends
in an insulting tube containing deionized water of low conductivity. Such deionized water terminators are a
common feature of high voltage cable testing facilities. The analysis of voltage distribution along the cable
end in such a cable termination is therefore of considerable practical interest and is reported rest. The
approach presented is simple and can be easily explained to students and engineers to highlight the stress
control issues in a cable termination, which is a very critical component of any high voltage cable system.

4. Method of Analysis
Fig. 1 shows such a simple deionized water termination. Let us assume that at distance x from the cable end,
the current in the deionized water is I(x) and voltage is V(x). Let C be the cable capacitance (F/m) and R
represents the resistance of the deionized water (Ω/m). Assuming that the insulation resistance of the cable
is very large (or infinity), can write the following equations for the one end region:

2
Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.1, No.2, 2011

L
dx x
I+dI I VH
V=0
V(x)
dI V1(x)

Conductor
Outer Semiconductor layer Insulation
Ground conductor Resistive Medium

Figure 1. End of coaxial cable

dV (x )
= − RI (x) (3)
dx
dI (x )
= jωRCVH − jωCV (x) (4)
dx

In eqns. 3 & 4, ω is the angular frequency of AC voltage and VH is voltage applied to the conductor. From
equations (3) and (4) one can write:

d 2 I (x ) dV ( x)
= 0 − jωC = jωRCI ( x) (5)
2 dx
dx
Equation (5) has a solution of the form:
I ( x) = A1eγx + A2 e −γx ) (6)

Where, γ = jωRC ) (7)

and A1 and A2 are constants.
From equations (4) and (6), V(x) can be expressed as:

V ( x ) = VH −
R
γ
(A eγ
1
x
− A2 e −γx ) (8)

At x=0, V(x) = VH, therefore:

V H = VH −
R
( A1 − A2 ) (9)
γ

3
Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.1, No.2, 2011

Hence A1 = A2. Also, at x = L, V(x) = 0, or

0 = VH −
γ
R
(A eγ
1
L
− A2 e −γL ) (10)

From eqns. (9) and (10)

γVH
A1 = A2 = (11)
2 R sinh γL

Upon substitution of A1, A2 into equations (6) and (8) I(x) and V(x) can be expressed as:

γVH cosh γx
I (x ) = (12)
R sinh γL

V sinh γx
V ( x ) = VH − H (13)
sinh γL

The tangential stress ET(x) along the cable insulation surface at distance x can be expressed as:

dV (x ) γVH cosh γx
ET (x ) = − = = RI (x ) (14)
dx sinh γL

Moreover, the voltage V1(x) appearing across the cable’s insulation at distance x from the cable’s end is
expressed as:
V sinh γx
V1 (x ) = VH − V (x ) = H (15)
sinh γL

6. Results and Discussions

In a typical deionized water termination rated for 300 kVrms, L=2m. In such a termination, if water
conductivity is ≈ 0.1µS/cm, then R ≈ 10 MΩ/m. The typically conductivity of deionized water used for
testing XLPE cables in such a termination is in the range of 0.01µS/cm to 1µS/cm which corresponds to R
value in the range of 1-100 MΩ/m for most cable sizes. For a 15 kV rated, 50 mm2 XLPE cables, C is about
200 pF/m. For 60 Hz voltage test, ω= 377 rad/s. Figs.2&3 show variation of absolute values of voltages V and
V1 with x for different values of R assuming that VH = 1V and L = 1m.
The results show that for R = 10 MΩ/m, the voltage distribution is linear along the length L of the cable end.
However, as the resistance of the surrounding medium is increased, this voltage distribution becomes non
uniform and for R = 10,000 MΩ/m, the distribution is extremely non uniform. Moreover in case of R larger
than 155 MΩ/m, the maximum value of V is above 1 due to capacitive effects. Fig.4 shows variations of V
with x for a few selected values of L and two values of R (one small, one large).

4
Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.1, No.2, 2011

0.8
p.u.voltage V

0.6

0.4
R=10 MΩ/m
R=100 MΩ/m
0.2 R=155 MΩ/m
R=1000 MΩ/m
R=10000 MΩ/m
0
0 0.2 0.4 0.6 0.8 1
p.u. distance x
Figure 2. Variation of V with x

1
R=10 MΩ /m
R=100 MΩ /m
0.8 R=155 MΩ /m
R=1000 MΩ /m
R=10000 MΩ /m
p.u.voltage V1

0.6

0.4

0.2

0
0 0.2 0.4 0.6 0.8 1
p.u.distance x
Figure 3. Variation of V1 with x

5
Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.1, No.2, 2011

1.4

1.2

1
p u voltage V

0.8 R=10000 MΩ /m

0.6
L =2m
0.4 L=1m
L=0.5m
0.2 L=2m
L=1m R=10 MΩ /m
L=0.5m
0
0 0.2 0.4 0.6 0.8 1
p u distance x
Figure 4. Variation of V with x for low and high values of R

It shows how the voltage distribution is affected by the length of termination in resistive medium as well as
the resistance of such a medium. Fig. 5 shows the variation of absolute value of I with p.u. distance x when
VH = 1 volt for different values of R. It shows that for larger values of R, for L=1m and VH=1V, I is
significantly increased near the ground end due to capacitive effects.
The case of R=10,000 MΩ/m means that the cable end is left in an extremely poor conductive medium. The
results for this case clearly show that the voltage drop is concentrated only on approximately a few cm near
the cut back edge of the shield. Therefore, the tangential stress is very high in this region. Moreover the stress
value is very sensitive to the value of R in the region as shown in Fig. 6.
Generally ET should be kept below 2.5 kV/cm to avoid any surface discharges. To satisfy this condition, low
value of R is required since voltage distribution is more linear for a lower R value. However, a lower R value
will generate higher power losses in the resistive medium surrounding the cable end. Thus, in the design of
such terminations a compromise between tangential stress values and the losses in the resistive medium is
required as discussed in literature [9].
The approach presented here for solving the field problem is very similar to the analysis of a long
transmission line which is covered in electrical engineering undergraduate courses and is discussed in typical
power system text books [11, 12]. The proposed approach can be used to determine the effect of different
design parameters on the voltage and stress distributions in such terminations. This method can easily be used
to explain the key concepts and challenges associated with the design and use of cable terminations, which
represents a critical part of the high voltage cables.

6
Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.1, No.2, 2011
-7
x 10
1.4

1.2

1
Current I (A)

0.8 R=10 MΩ /m
R=100 MΩ /m
0.6 R=1000 MΩ /m
R=10000 MΩ /m
0.4

0.2

0
0 0.2 0.4 0.6 0.8 1
p u distance x
Figure 5. Variation of I with x

4
x 10
3
R=10 MΩ /m
2.8
R=100 MΩ /m
2.6
R=1000 MΩ /m
2.4
R=10000 MΩ /m
2.2
Tagential Stress (V/m)

2
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
0 0.2 0.4 0.6 0.8 1
Distance x (m)
Figure 6. Variation of ET with x when VH = 1V

7
 Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.1, No.2, 2011

Generally ET should be kept below 2.5 kV/cm to avoid any surface discharges. To satisfy this condition, low
value of R is required since voltage distribution is more linear for a lower R value. However, a lower R value
will generate higher power losses in the resistive medium surrounding the cable end. Thus, in the design of
such terminations a compromise between tangential stress values and the losses in the resistive medium is
required as discussed in literature [9].
The approach presented here for solving the field problem is very similar to the analysis of a long
transmission line which is covered in electrical engineering undergraduate courses and is discussed in typical
power system text books [11, 12]. The proposed approach can be used to determine the effect of different
design parameters on the voltage and stress distributions in such terminations. This method can easily be used
to explain the key concepts and challenges associated with the design and use of cable terminations, which
represents a critical part of the high voltage cables.

2. Conclusion
Using a circuit model for high voltage cable end immersed in a resistive medium, expressions for voltage and
stress distributions are derived. The method is simple and can be used to study the importance of electric
stress control in cable terminations. The effect of different design parameters on the performance of such a
termination is briefly presented.

References
[1] R. Bartnikas & K.D. Srivastava (2000), Power and Communications Cables: Theory and Applications., New York:
Wiley.
[2] G. Lupo, et al. (1996), “Field Distribution in Cable Terminations from a Quasi-static Approximation of Maxwell
Equations”, IEEE Trans. on DEI, 3(3), 399-409.
[3] P. N. Nelson & H. C. Herving (1984), “High Dielectric Constant for Primary Voltage Cable Terminations”, IEEE
Transaction on PAS, 103 (11), 3211-3216.
[4] L.Bayon, et al. (2004), “Field Distribution Measurement and Simulation of Stress Control Materials for Cable
Accessories”, Proceedings of the IEEE International Conference on Solid Dielectrics, pp.534 – 537.
[5] R. Strobl, et al. (2001), “Evolution of Stress Control Systems in Medium Voltage Cable Accessories”, Proceedings
of the IEEE T&D Conference, pp. 843-848.
[6] S. Nikolajevic, et a. (1994), “Development of High Dielectric Constant Materials for Cable Accessories and Design
of XLPE MV Cable Terminations”, Proceedings of the IEEE International Symposium on Electrical Insulation, pp.
570 – 573.
[7] J. Rhyner & M.G. Bou-Diab (1997), “One-dimensional Model for Non Liner Stress Control in Cable
Terminations”, IEEE Trans on DEI, 4(6), 785-791.
[8] V. Tucci & J. Rhyner (1999), “Comment on 1-Dimensional Model for Nonlinear Stress Control in Cable
Terminations”, IEEE Transactions on DEI, 6, 267-270.
[9] R. Gleyvod & P. Mohaupt (1993), “Operation of Water Terminations for Testing Power Cables”, Paper # 697, ISH
Yokohama, Japan..
[10] S.V.Nikolajevic, N.M.Pekaric-Nadj and R.M. Imitrijevic (1997), “The Influence of Some Construction Parameters
on Electrical Stress Grading in XLPE Cable Terminations”, Proceedings of the International Conference and
Exhibition on Electricity Distribution, pp. 3.15.1.-3.15.4.
[11] J.J. Grainger & W.D. Stevenson Jr. (1994), Power System Analysis, New York: McGraw- Hill.
[12] A.A.Al-Arainy, N.H.Malik & S.M. Al-Ghuwainem (2007), Fundamentals of Electrical Power Engineering, Saudi
Arabia: King Saud University

8
Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.1, No.2, 2011

N. H. Malik graduated with B.Sc.in E.E from UET, Lahore in 1973, MASc in
Electrical Engineering from University of Windsor, Canada in 1977 and received
Ph.D. from the University of Windsor, Canada in 1979. He has authored over 140
research papers and four books. Presently he is Chair Professor of “Saudi Aramco
Chair in Electrical Power”, Electrical Engineering Department, King Saud
University, Saudi Arabia.
(e-mail: [email protected]).

A. A. Al-Arainy graduated with B.Sc.in E.E from KSU, Saudi Arabia in 1974,
MASc in Electrical Engineering from University of Torondo, Canada in 1977 and
received Ph.D. from University of Torondo, Canada in 1982. He has authored over
100 research papers and four books. Presently he is Supervisor of “Saudi Aramco
Chair in Electrical Power”, Electrical Engineering Department, King Saud
University, Saudi Arabia.
(e-mail: [email protected]).

M. I. Qureshi received his B.Sc. (Electrical Engg.) from University of Peshawar,

Pakistan in 1969 and Ph.D. (Electrical Engg.) from University of Salford, United
Kingdom in 1992. He has authored over 120 research papers and two books. Currently
he is with College of Engineering, King Saud University. Saudi Arabia.
(e-mail:[email protected])

F. R. Pazheri received his B.Tech. degree in Electrical & Electronics Engineering

from the Calicut University, India in 2001 and M. Tech. degree in Electrical
Engineering from Kerala University, India in 2004. He was with MES College of
Engineering, India from 2005 to 2006 and with College of Engineering, Thalassery,
India from 2006 to 2009. Currently he is Lecturer at “Saudi Aramco Chair in
Electrical Power”, King Saud University,Saudi Arabia. He has authored over 18
technical papers.
(e-mail: [email protected]).