Notes 12: Series Impedance of Cables

12.0 Introduction
We desire to determine methods of computing the series impedance for cable
configurations. The main steps for doing so are exactly the same as the steps for
computing series impedance of overhead lines. These steps, given in Notes 10, are:
1. Determine resistance per mile and !" of each conductor.
#. Determine distance bet$een conductors.
%. &ompute primitive impedance matrix using e's. (%),*0+ of notes ) (see belo$+ $ith
appropriate value of resistivity ,.
*. -erform .ron reduction to eliminate the presence of the neutral and thus obtain the
phase impedance matrix.

'

+ +

+
+
f r
j
f R Z
i
i ii

ln
2
1
6786 . 7
1
ln 12134 . 0
00158836 . 0

(%)+

'

+ + +

f D
j
f Z
ij
ij

ln
2
1
6786 . 7
1
ln 12134 . 0
00158836 . 0

(*0+
&onsider /ig. 1, and observe the fact that the neutral is actually comprised of several
conductors.

/ig. 1
No$ $e could treat each neutral $ire as a separate conductor and build a very large
primitive impedance matrix. This $ould be very tedious. 0 better approach is to obtain
an e'uivalent neutral.
12.1 Definitions, nomenclature, equations
/ig. # defines some distances that $e need:
1

/ig. #
These distances are:
dc: phase conductor diameter (inches+
dod: nominal diameter over the concentric neutrals of the cable (inches+
ds: diameter of a concentric neutral strand (inches+
": radius of a circle passing through the center of the concentric neutral strands (ft+
We also need these additional definitions.
!"c: geometric mean radius of the phase conductor (ft+
!"s: geometric mean radius of a single neutral strand
"c: resistance of the phase conductor (12mile+
"s: resistance of a single neutral strand (12mile+
3: the number of concentric neutral strands
0ll of the above information is obtained from the tables given in 4Notes 11,5 $ith the
exception of ". This is computed as
24 12
1
2
s od s od
d d d d
R

(1+
6ur approach is to convert the concentric neutral configuration to an e'uivalent
configuration having a single neutral, /ig. %.
2

/ig. %
"eferences 718#9 sho$ ho$ to derive relations for such an e'uivalent configuration. We
assume phase conductors are numbered first and concentric neutrals numbered last.
/or example, for % cables, each $ith concentric neutrals, $e number phase conductors
1,#,% and concentric neutrals *,:,;.
The relations are:
&oncentric neutral geometric mean radius:

k
k
si cni
kR GMR GMR
1
(#+
Note that !"i<
i
r
based on previously defined nomenclature.
Distance bet$een a concentric neutral and an ad=acent phase conductor
k
k k
nm ij
R D D (%+
$here Dnm is the center8to8center distance bet$een phase conductors
corresponding to the concentric neutrals i=. >'uation (%+ gives the geometric mean
distance bet$een all of the concentric neutral strands of one cable and the phase
conductor of the other cable.
Distance bet$een a concentric neutral and its o$n phase conductor:
R D
ij

(*+
$here " is given by e'. (1+ above.
Distance bet$een a concentric neutral of one cable and a concentric neutral of
another cable:
nm ij
D D
(:+
$here, as before, Dnm is the center8to8center distance bet$een phase conductors.
The e'uivalent resistance of the concentric neutral is
3
k
R
R
si
cni
(;+
10.2 Example
Three concentric neutral cables are buried in a trench $ith spacings as sho$n in /ig. *.
The cables are 1:3?, #:0!&! stranded all8aluminum $ith 1% strands of @1* annealed,
coated copper $ires (12% neutral+. Determine the phase impedance matrix.
1 2 3
4 5 6
/ig. *
We obtain the data for the phase conductor and neutral strands from the conductor
data table given at the end of 4Notes 11.5
dod<1.#) inches
#:0!&! 00 phase conductor:
o !"c<0.01A1 ft
o Diameter dc<0.:; inches
o "esistance "c<0.*100 12mile
@1* copper neutral strands:
o !"s<0.00#0B ft
o Diameter ds<0.0;*1 inches
o "esistance "s<1*.BA## 12mile
The radius of the circle passing through the center of the strands, e'. (1+, is:
ft
d d
R
s od
0511 . 0
24
0641 . 0 29 . 1
24

The !" of the e'uivalent concentric neutral is:
ft
kR GMR GMR
k
k
si cni
0486 . 0 ) 0511 . 0 ( 13 ) 00208 . 0 (
13
12
1

The resistance of the e'uivalent concentric neutral is
mile
k
R
R
si
cni
/ 1438 . 1
13
8722 . 14

No$ assume that the phase conductors are numbered 1,#, and %, and the e'uivalent
concentric neutrals are numbered *, :, and ;.
The conductor8to8conductor and concentric neutral to concentric neutral spacings are:
D1#<D#1<D*:<D:*<0.: ft.
D#%<D%#<D:;<D;:<0.: ft.
D%1<D1%<D;*<D*;<1.0 ft.
The spacings bet$een conductors and their concentric neutrals are
D1#<D#:<D%;<"<0.0:11 ft.
The distances bet$een concentric neutrals and ad=acent phase conductors are given
by e'. (%+, but because DnmCC ", $e may obtain very good accuracy by =ust assuming
this distance is =ust Dnm.
D1:<D:1<0.: ft
D#;<D;#<0.: ft
D;1<D1;<1.0 ft.
4
The self impedance for the cable in position 1 is:
mile j
j
f r
j
f R Z
/ 4564 . 1 5053 . 0
60
100
ln
2
1
6786 . 7
0171 . 0
1
ln 12134 . 0 0953 . 0 41 . 0
ln
2
1
6786 . 7
1
ln 12134 . 0
00158836 . 0

1
1 11
+

'

+ + + +

'

+ +

+
+

The self impedance for the concentric neutral for &able @1 is

mile j
j
f r
j
f R Z
/ 3296 . 1 2393 . 1
60
100
ln
2
1
6786 . 7
0486 . 0
1
ln 12134 . 0 0953 . 0 144 . 1
ln
2
1
6786 . 7
1
ln 12134 . 0
00158836 . 0

4
4 44
+

'

+ + + +

'

+ +

+
+

The mutual impedance bet$een &able @1 and &able @# is

mile j
f
j
f D
j
f Z
ij
/ 0468 . 1 0953 . 0
ln
2
1
6786 . 7
5 . 0
1
ln 12134 . 0 0953 . 0
ln
2
1
6786 . 7
1
ln 12134 . 0
00158836 . 0

12
+

'

+ + +

'

+ + +

The mutual impedance bet$een &able @1 and its concentric neutral is

mile j
f
j
f D
j
f Z
ij
/ 3236 . 1 0953 . 0
ln
2
1
6786 . 7
0511 . 0
1
ln 12134 . 0 0953 . 0
ln
2
1
6786 . 7
1
ln 12134 . 0
00158836 . 0

14
+

'

+ + +

'

+ + +

The mutual impedance bet$een

the concentric neutral of &able @1 and the concentric neutral of &able @# is:
mile j
f
j
f D
j
f Z
ij
/ 0468 . 1 0953 . 0
ln
2
1
6786 . 7
5 . 0
1
ln 12134 . 0 0953 . 0
ln
2
1
6786 . 7
1
ln 12134 . 0
00158836 . 0

45
+

'

+ + +

'

+ + +

0nd so on for the rest of the terms.

5
The resulting primitive impedance matrix in partitioned form is
[ ] [ ]
[ ] [ ]
1
1
]
1

nn np
pn pp
Z Z
Z Z
Z

$here
[ ]
1
1
1
]
1

+ + +
+ + +
+ + +

4564 . 1 5053 . 0 0468 . 1 0953 . 0 9627 . 0 0953 . 0

0468 . 1 0953 . 0 4564 . 1 5053 . 0 0468 . 1 0953 . 0
9627 . 0 0953 . 0 0468 . 1 0953 . 0 4564 . 1 5053 . 0

j j j
j j j
j j j
Z
pp
[ ]
1
1
1
]
1

+ + +
+ + +
+ + +

3236 . 1 0953 . 0 068 . 1 0953 . 0 9627 . 0 0953 . 0

068 . 1 0953 . 0 3236 . 1 0953 . 0 0468 . 1 0953 . 0
9627 . 0 0953 . 0 0468 . 1 0953 . 0 3236 . 1 0953 . 0

j j j
j j j
j j j
Z
pn
[ ] [ ]
T
pn np
Z Z


[ ]
1
1
1
]
1

+ + +
+ + +
+ + +

3296 . 1 . 2391 . 1 0468 . 1 0953 . 0 9627 . 0 0953 . 0

0468 . 1 0953 . 0 3296 . 1 2391 . 1 0468 . 1 0953 . 0
9627 . 0 0953 . 0 0468 . 1 0953 . 0 3296 . 1 2391 . 1

j j j
j j j
j j j
Z
nn
0pplying .ron reduction results in:
[ ] [ ][ ] [ ]
1
1
1
]
1

+ +
+ + +
+ +

,
_

4463 . 0 7981 . 0 0328 . 0 3191 . 0 0143 . 0 2849 . 0
0328 . 0 3191 . 0 4041 . 0 7891 . 0 0328 . 0 3191 . 0
0143 . 0 2849 . 0 0328 . 0 3191 . 0 4463 . 0 7981 . 0

1
j j j
j j j
j j j
Z Z Z Z Z
np nn pn pp
10. !ape s"ielded cables
The tape shielded cable uses bare copper tape helically applied around the insulation
screen. 0n insulation =ac3et encircles the tape shield. /ig : illustrates.

/ig. :
6
/or a given ampacity, a tape shielded conductor is less expensive than a concentric
neutral conductor because of the simpler construction and manufacturing cost of the
tape shield relative to the concentric neutral. Tape shielded cables are often called
po$er cables because they are more fre'uently used in higher current applications
such as three8phase mainlines, since such applications typically see less unbalance
and the neutral conductor need not be designed to carry full load.
-arameters of the taped shielded cables are:
dc: phase conductor diameter (inches+
dod: outside diameter over the =ac3et (inches+
ds: outside diameter of the tape shield (inches+
T: thic3ness of copped tape shield (mils+
,: resistivity of tape shield (18meters+ at :0D &.
With this information $e may compute the resistance of the tape shield as:
mile
T d
r
s
shield
/ 10 9385 . 7
8


(A+
$here the constant A.)%B:>B converts from meters per in8mil to per mile.
Eince the shield is in fact a hollo$ thin8$alled conductor, $e need not correct for the
internal flux. Therefore, the number to use as the shieldFs geometric mean radius is
exactly its radius, given as half the difference bet$een its outside diameter ds and its
thic3ness T21000 ($here $e divide by 1000 to convert from mils to inches+:
ft
T d T d
GMR
s s
shield
24
1000 /
12
1
2
1000 /

(B+
0s in the concentric neutral, $e number phase conductors 1,#,% and tape shields *,:,;.
Then spacings bet$een a tape shield and the conductors and other tape shields are:
Distance bet$een a tape shield and an ad=acent phase conductor
Di=<Dnm
$here Dnm is the center8to8center distance bet$een phase conductors
corresponding to tape shields i and = (ft+. Note that $ith the concentric neutrals, this
distance $as given as
k
k k
nm ij
R D D
but $e concluded Dnm $as usually a very good approximation.
Distance bet$een a tape shield and its o$n phase conductor:
Di=<!"shield
$hich is the radius to midpoint of the shield (ft+. Gn the concentric neutral case, this
$as ", the radius of a circle passing through the center of the concentric neutral
strands.
Tape shield to a tape shield of another conductor:
Di=<Dnm (ft+
This is the same as in the concentric neutral case.
"eferences:
719 W. He$is and . 0llen, 4Eymmetrical component circuit constants and neutral
circulating currents for concentric neutral underground distribution cables,5 G>>> Trans.
on -0E, ?ol. -0E8)A, No. 1, pp 1)181)), Ian2/eb 1)AB.
7#9 W. He$is, . 0llen, and I. Wang, 4&ircuit constants for concentric neutral
underground distribution cables on a phase basis,5 G>>> Trans. on -0E, ?ol. -0E8)A,
no. 1, pp #008#0A, Ian2/eb 1)AB.
7
8