Calculation The Internal Thermal Resistance and Am Pacity of 3-Core Screened Cables With Fillers

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

IEEE Transactions on Power Delivery, Vol. 14, No.

3, July 1999

729

Calculation of the Internal Thermal Resistance and Am pacity of 3-Core Screened Cables
with Fillers
A. Napieralski Z. Kulesza
Technical University of Lodz
Lodz, Poland

G.J. Anders,
Ontario Hydro Technologies
Toronto, Ontario, Canada

Abstract. All 3-core cables require fillers to fill the


space between insulated cores and the belt insulation or a
sheath. The equations given in IEC 287 and in the NeherMcGrath paper for the internal thermal resistance of 3-core
cables were developed for paper insulated cables. For such
cables, it was assumed that the insulation and filler materials
have the same thermal resistivities. In reality, in 3-core
cables a variety of materials are used as fillers. The majority
of these will have a higher thermal resistivity than the
insulation. In the previous paper, a new formula was
developed to compute the value of the internal thermal
resistance of belted cables taking into account the thermal
resistivity of the filler. In this paper, screened cables are
considered and a new formula for the computation of the
internal thermal resistance of such cables is presented. The
effect offiller resistivity on cable ampacity is also discussed

Conductor
Conductor Screen

Copper Woven Fabrklape


or Belt Insulation

Semiconducting InsulationScreen
Metaltc Screen

1. Introduction.

Figure 1. A typical 3-core screened cable

In recent years, in addition to paper-insulated cables,


many constructions with extruded insulation have been
introduced. All modern electric power cables are constructed
with semiconducting screens around the conductor and around
the insulation. For thermal calculations, these screens are
considered to be a part of the insulation. In addition, in the
majority of modern 3-core cable constructions, a metallic
screen made of copper tapes or wires is added around each
core. An example of a typical 3-core screened cable is shown
in Figure 1. The main purpose of this screen is to provide a
uniform electric field inside the cable. However, the presence
of the screen may have a significant effect on the heat transfer
mechanism inside a cable, and hence, can influence the value
of the internal thermal resistance TI (K.m/W).

based on measurements of electrical resistance performed in a


series of tests on models comprising copper-tube electrodes
soldered to resistance-alloy sheets to represent the dielectric.
In order to take into account the effect of the metallic screen
around each core, the models were provided with an additional
copper strip soldered to the resistance alloy and surrounding
the core electrode. An obvious question to be asked is how
accurate these computations are for modern cable
constructions? The aim of these paper is to give an answer to
this question.

Calculations of the internal thermal resistance of threecore


screened cables are based on work performed by Whitehead
and Hutchings in 1930s [I].
These calculations are
PE-350-PWRPO-091998 A paper recommended and approved by
the lEEE insulated
Committee Of the lEEE Power
Engineering Society for publication in the IEEE Transactions on Power
Delivery. Manuscript submitted June 30, 1998; made available for
printing October 7,1998.

All 3-core cables require fillers to fill the space between


insulated cores and the belt insulation or a sheath. In the past,
when impregnated paper was used to insulate the conductors,
the resistivity of the filler material matched very closely that of
the paper (around 6 K . m l W ). With polyethylene insulation
having much lower thermal resistivity (3.5 K . m l W ), the
higher thermal resistivity of the filler may have a significant
influence on the overall value of
and, hence on the cable
rating. Working Group 10 of the Study Committee 20A of the
International Electrotechnical Commission (IEC) has been
Of the
given a task to review and update the
thermal resistance between the cores and the sheath for 3-core
cables. In the previous paper [ 2 ] ,we have addressed an issue

0885-8977/99/$10.00 0 1998 IEEE

730
of the internal thermal resistance of 3-core belted cables with
fillers. This paper deals with another aspect of this task,
namely the computation of I; for 3-core screened cables with
circular conductors.
In order to investigate the influence of the filler thermal
resistivity on for 3-core cables, extensive numerical studies
using the finite element method have been performed to
develop an equation for the value of I; when the filler
resistivity is different from that of the insulation. Since with
the varying thermal resistivity of the filler the values of T, will
be different from those obtained using the standards, the
influence of the new resistance values on cable rating is also
investigated.
The presentation starts with the review of the current
standard computations including the summary of the results of
the application of the finite element method to the analysis for
I; of 3-core belted cables. Next, the results for screened
cables are discussed followed by a numerical example and the
discussion closing the presentation.

2.

Standard method of computation of T1

For single-core cables, the thermal resistance between


one conductor and the sheath is computed from the following
equation [3,4):

2K

G + 0.031(pf - p i ) e

diameter of conductor, mm
thickness of insulation between conductor and sheath,
mm

The computation of the internal thermal resistance of


three-core cables is more complicated than that of a singlecore cable. The general method of computation employs
geometric factors (G) in place of the logarithmic term in
equation (1); that is:

-&.

T1 -

2n

Several methods of determining such factors have been


devised as described in [ 5 ] . For a 3-core cable which is
investigated in this paper, a graphical representation of G and
the relevant equations are given in IEC Standard 287, [3].
As mentioned above, to give an indication of the effect of
the filler resistivity on the internal thermal resistance of the
belted cable, finite element studies were conducted for a range

(3)

where pf and p , are the thermal resistivities of filler and


insulation, respectively and G is the geometric factor obtained

from Figure 3 in IEC 287 [3] assuming pf = p , .


.When the values of T,, computed from equation (3) and
obtained using the finite element method for a wide range of
cable designs were plotted and regression analysis was
performed, the regression curve had a slope of 0.998 with a
standard error of 0.8% (Anders et al., 1997).
As mentioned in the introduction, a majority of modern
3-core cables are manufactured with a metallic screen, usually
made of copper tapes or wires, around 'each core. Screening
reduces the thermal resistance of a cable by providing
additional heat paths along the screening material of high
thermal conductivity, in parallel with the path through the
dielectric. IEC Standard 287 recommends that the thermal
resistance of the insulation be obtained in two steps for
screened cables. First, the cables of this type are considered as
belted cables for which t, l t =0.5 (notation as in figure 1).
Then, in order to take account of the thermal conductivity of
the metallic screens, the results are multiplied by a factor K,
called the screeningfactor. Thus, we have

2x

pi = thermal resistivity of insulation, K.m/W

t,

0.67$c

I;fi11er=

T I= K -Pi
G

where:

d, =

of values of the thermal resistivities of the insulation and the


filler. From these investigations, the following approximating
formula emerged [2].

(4)

The values of the screening factor were obtained


experimentally and are given in Figure 4 in IEC 287 [3].
The following two objectives guided the work reported
here: (1) To verify the validity of equation (4) using the finite
element method, and (2) to develop new formulas for the
for screened 3-core cables with fillers. The
computation of
results of the analysis are presented in the following sections.

3. Finite element studies


In the finite-element method, the thermal resistance of the
insulation is computed directly, assuming that the conductor
and sheath boundaries are isothermal. The thermal resistance
of the insulation with filler can be obtained from the following
relation:

I; = Tfiller+ce1t + T o r e - Tbelr

(5 )

The thermal resistance of the core can be computed from


a standard formula:

731
0.70
0 65
0 60

where:

D,,
=

diameter &&e semiconducting insulation screen, mm

d,

diameter of the conductor, mm

Er

0 50
Or05

0 40
035

The thermal resistance of the filler and belt is obtained


from the finite element analysis as:

030
0 25
020
020

0 30

0 A0

0 50

0 60

0.70

TlrkhEhne

where W, (W/m) is the heat rate generated in a single


conductor.

Tbelris the thermal resistance of the belt insulation. The


belt insulation was added to the construction of this cable to be
able to distinguish the temperatures of two metallic parts,
namely the insulation screen around each core and the
common sheath. Normally, screened and sheathed 3-core
cables will have a layer equivalent to the belt insulation to hold
the cores together.
The finite element studies have shown that the distribution
of temperature on the belt is strongly nonlinear. Therefore, an
application of a standard formula for the thermal resistance of
circular layers; that is:
pbeit

Tberr

= -In

2n

diam.abovebelt
diam.belowbelt

(8)

is not advisable. Instead, a procedure described in Reference 7


has been developed and used in the analysis.
In the finite element studies, we have used a broad range of
3-core screened cables with the dimensions specified in the IEC
Standard 228 [ 6 ] . Low- and medium-voltage cables have been
selected of 10, 15, 20, 30 kV, the sections of the conductor
ranged from 16 to 400 mm2. The following six values of the
thermal resistivity pf of the filler have been used: 3.5, 5.0, 6.0,
7.0, 8.0, 10.0 Km/W. Eventually, 228 various structures of the
cable were analyzed [7].
3.1 The verification of the formula (4).

Figure 2. Comparison of the values TI, calculated from


the IEC formula, and theJinite element method.
The correlation coefficient between the values TlJEC
and the finite element method (TI-FET) equals 0.989. The
regression analysis produced a straight line with the angle of
45"0'36. The regression formula (in relation to the variable
T 1 I E C ) has the form:
T1-FET = 1.000035 * TI-IEC - 0.00468;
and the error in defining Tl-FET equals 0.0169.
The difference between the results of the two methods
ranges from 0.002 to 0.039 K.m/W, and in relative values in
relation to the IEC Standard it ranges from 0.3 to 9.9%.
Therefore, with only a few (less then five out of 228)
values differing in the range of 8-9%, we can conclude that the
IEC formula (4) is an acceptable approximation for

q.

3.2 A new formula for TI as a function of the thermal


resistivity of the filler
To obtain the formula for the internal thermal resistance of the
cable from the results of the finite element studies, we used a
multidimensional correlation analysis with the selected
construction parameters of the cable. A high correlation
between the internal thermal resistance and the ratio of the
thickness of insulation to the diameter of the conductor and
also with the value of the thermal resistivity of the filler has
been found.

The aim of the first part of the study was to verify


equation (4) using the finite element analysis with the filler
resistivity equal to that of the insulation (assumed 3.5 K.m/W).

Applying optimization techniques (the simplex and


Newton methods), the following formula describing the
relationship between TI and the construction parameters of the
cable has been developed.

The graphic comparison of the resistance obtained using


equation (4) with the findings of the finite element method is
shown in Figure 2.

T1- FIT =0.89.Tl-IEC

(9)

'

732
The correlation parameter is equal to 0.00462 and the
relative error is 0.0919

thermal resistance of the insulation (see below)


=0.227K.m/W

Figure 3 shows the relationship between the values of


computed using equation (9) and those obtained from the finite
element analysis. A detailed analysis of the finite element results
leading to the development of equation (9) is given in [7].

thermal resistance of the jacket

I -5r

resistance

= 0.067 K .m/w
T4 = 0.633K . m/W

conductor resistance at 90C (50Hz) R = 0.7898.104 m m


sheath loss factor

A., = 0.018

3 610

Computing the thermal resistance of the insulation, the


thicknesses of semiconducting screens over the conductor and
over the insulation are added to the thickness of the insulation.

0 550
0 500

0 450

400

rE a0 350

h
r(

We will vary the thermal resistivity of the filler between


3.5 and 10 K.m/W. The values of T, are summarized in Table
1 and sample computations for pr = 6 K .m/W are given

0 300

.c
C 250

below.

0 200
0 150

0.150

0.250

0.450

0.350

0.550

0.650

TI-i

Figure 3. Graphical representation of equation (9).

First, we determine the value of the screening factor. For


this cable,
tl I d , = 3.4/ 23.1 = 0.147

and the ratio

4. The effect on cable ampacity

6, . pi/(dc p,) = 0.22.3.5/(23.1.27. lo4) = 12.35

To demonstrate the effect of the filler thermal resistivity


on cable ampacity, we will consider a 10 kV, 300 mm, threecore cable with XLPE insulation. The parameters of this cable
are as follows:
conductor diameter

d, = 23.lmm

thickness of core insulation, including screening tapes plus


half the thickness of a non-metallic layer over the laid up cores
t] = 3.4 mm
diameter over the three laid up cores

4,= 64.7-

thickness of semicon tapes over the laid up cores


fbelt

From Fig. 4 in [3), we obtain K = 0.57 1.


To obtain the geometric factor G, we first assume
tl l t = 0.5 , and from Fig. 3 in [3], G = 0.7 .
The thermal resistance of the insulation as specified in
the IEC Standard is obtained from equation (4):
3.5
T, = K A G = 0.58-0.7
2n
2n

= 0.227 K . mfW

The rating of the cable is obtained from the following


equation [4,5]:

= 1.2 mm

diameter over the lead sheath

D, = 68.7 mm

diameter over the PVC jacket

De = 74.7 mm

The cable is installed underground, 1 m deep with the


soil thermal resistivity of 1 K.m/W. The ambient soil
temperature is 20C.
With the above parameters, the following values of
thermal resistances and loss factors are obtained using
standard calculation procedures:

For the cable under consideration, the values of

T, = 0,wd = 0 and 4 = 0. Hence, using the rating equation


with the internal thermal resistance given by equation (4):

90 - 20
I = 0.7898.10-4(0.22?+ 3.1.018.(0.067+0.633))
The remaining values are summarized in Table 1.

=612A

733

REFERENCES

Table 1. Ampacities using IEC method and Eq.9.


IEC

Pf

I;

(K.mlW) (K.mlW)

0.227

Eq. 9

I;

I
A

612

(K.m/W)

0.263

1. Whitehead, S., Hutchings, E.E., (1939) Current Rating of


Cables for Transmission and Distribution, ERA
Publication F/T13 1.

I
A

607

2.

Anders, G., Napieralski, A., Zamojski, W., (1997)


Internal Thermal Resistance and Ampacity Computations
for 3-Core Belted Cables with Fillers. Accepted for
publication in IEEE Trans. On Power Delivery.

3.

IEC Standard 287 (1994), Calculation of the Continuous


Current Rating of Cables (100% load factor). Part 2:
Thermal Resistance.

4.

Neher, J.H., McGrath, M.H.(1957), The Calculation of


The Temperature Rise and Load Capability of Cable
Systems, AIEE Trans., Vol. 76, Pt. 3, October 1957, pp.
752-772.

5.

Anders, G, (1997), Rating of Electric Power Cables.


Ampacity Calculations for Transmission, Distribution and
Industrial Applications, IEEE Press, New York.

The effect of filler resistivity is very small. In this


example, for the highest value of pf the ampacity is reduced
by about 1% in comparison with the value computed from
equation (4). This situation is quite different than that
encountered with the belted cables [2]. In the latter case, the
reduction of cable ampacity was quite significant as the
thermal resistivity of the filler increased. Since the presence s f
the screen helps considerably in the heat dissipation (the value
of
is reduced by about 40%), the presence of the filler is
not as noticeable as in the case of belted cables.

6. IEC Standard 228 (1978), Conductors of Insulated


Cables, Second edition. First Supplement (1982).

5. Discussion
The analysis reported above leads to the following
conclusions.
1. Equations used in the IEC Standard 287 for the internal

thermal resistance of screened 3-core cables are in a good


agreement with the results of the finite element analysis
when the thermal resistivity of the filler is assumed to be
equal to that of the insulation.
2. A simplified equation using only 3 parameters has been
proposed. This equation is accurate enough for rating
purposes considering the fact that the small variations in
the value of TI have a very small effect on the cable current
rating.
3. Since the effect of fillers on cable ampacity is very small,
the existing IEC formula could still be used to obtain the
value of T I .

4. During the analysis, it became apparent that the thermal


resistance of the belt insulation cannot be computed from a
standard formula because of a nonuniform temperature
distribution around the belt. An appropriate equation has
been developed and presented in Appendix.

Acknowledments
The authors wish to acknowledge the financial support of
the NATO Scientific and Environmental Affairs Division,
Collaborative Research Grant No. 960050.

7.

Anders,G.J., Napieralski, A, Kulesza, Z, Insulation


thermal Resistance of 3-core Screened Cables with
Fillers, Document WGlO(GA)TR, IEC SC20A, February
1998.

Appendix
The procedure of calculating the thermal resistance
of the belt layer.
As mentioned earlier, the temperature distribution within
the belt layer is not uniform with the highest temperatures
occurring at the points where screen touches the belt. This
uneven distribution becomes even more pronounced when the
filler resistivity is much higher that of the core insulation. In
order to estimate the thermal resistance of the belt insulation, a
two stage procedure has been adopted.
1. The temperature distribution of the medium layer of the belt

was obtained from the finite element analysis. An example


of the temperature distribution inside the belt is shown in
Figure Al.
2. An equation which describes the temperature distribution
inside the belt as a function of the angle of section of the
cable rp has been developed. An equation of the following
form has been used:

where:

734
3. The mean value of the above function (Eq.Al) is then used
to compute the thermal resistance of the belt:
(A21
mean value of the temperature of the belt e,,,,,,, can be
e u l a t e d using the mean value theorem of the integral calculus:

IG

('43)
a
0

10

40

30

20

50

as long as the functionflx) is continuous in the closed interval (a,


b). The function Temp(cp)defined in Equation (Al) satisfies
such conditions.

60

Awl-

1-3CSIOK016M3.5

-a-3CS10K400I43.5

-*- 3CSlOK016MlO -*- 3CSlOK400M10]

Analyzing Figure A1 we can observe to a clear difference


of the temperature distribution for the cables with the filler of the
resistivity 3.5 and 10 K.m/W. However, the thermal resistances
of the belt are very similar in all the cases as illustrated in Table
Al.

Sample temperature distribution in a


Figure A l .
belt insulation
Ax = "'O

'

'

"'

x + px.3
+

pr.2

exp(y.x.0+ y.x.i . Radangle + y . x . 2 . T max min)


(Ala)

Table Al. ComDarison of the values of the thermal


r&stances of the belt':

A: calculations of the thermal resistance of the belt

and X denotes respectively (1) the thickness of the belt, (2)


diameter of the conductor, (3) thickness of the insulation, and
(4) thickness of the screen, while Tmin denotes the lowest
temperature within the belt, Tmaxmin is the difference of the
maximum and minimum temperatures within the belt,
Radangle is the angle in radians (in view of the occurrence of
the axis of symmetry of the section of the cable, the
examinations dealt with the segment of 1.05 rad., that is 60').

with the fder thermal resistivity equal to that of


the insulation

B: calculations with two values thermal resistivities


of the filler

The following coefficients were obtained using a nonlinear


estimation with the additional variations in the filler thermal
resistivities.

I 0.2883 I 0.6336 I 0.2287 I 0.5092 I 1.3906 13.5321 I 9.7717 I


I

10.0269 10.1418 10.4048 12.1355 12.0820 17.2453 1.7811

I 0.3035 I 0.9343 I 1.2576 I 3.8568 I 0.1538 I 3.5236 19.9658 I


0.0003 0.0007 0.4789 0.3432 8.2997 5.3010 0.5130
The correlation coefficient between analytical and
numerical results is equal to 0.99943

As we can observe that the difference of the thermal resistance


of the belt as calculated with the influence of the filler: 3.5 and
10 K.m/Wdoes not exceed 0.005 (error below 5%).

' Symbolsof cablesused in the table and figure:

3cs
- 3-core screened cable;
numberK
- voltage (kV);
number
- cross-section(mm2>;
Mnumber
- resistivity of the filler (K.m/W)
For example, 3CSlOK016M10 denotes a 3-core cable,
screened, 10 kV, 16 m2,
filler resistivity 10 K . M .

You might also like