International Conference on Electrical Engineering 2002

A Study on the Reduction Methods of Sheath Circulating Current in Underground Transmission Systems
Chae-Kyun Jung Jong-Beom Lee Ji-Won Kang Tae-In Jang

Department of Electrical, Electronic and Informaton Engineering, Wonkwang University Iksan, Korea

Power System Research Lab. Korea Electric Power Research Institute Taejon, Korea

Abstract
Sheath circulating current is increased in the change of sheath mutual impedance which is caused by imbalanced cable system, and different section length between joint boxes. If excessive current flows in sheath, sheath loss which is reduced the transmission capacity is produced. And also such excessive sheath current influences severely on the operation. Recently, excessive sheath circulating current was partially measured in underground cable systems of KEPCO. Accordingly, actual schemes to reduce excessive sheath circulating current are urgently required for installed cable systems as well as newly-installing cable systems. This paper describes the analysis of sheath circulating current and various schemes to reduce excessive circulating current using EMTP/ATPDraw. And also, applicable schemes are proposed through a detailed analysis regarding cable systems by considering various electrical and environmental factors. It is evaluated that the proposed reduction schemes can be effectively applied to reduce the excessive sheath circulating current with the minimized electrical problems. And reduction effect is proved with sheath circulating current reduction equipment

effect is proved with sheath circulating current reduction equipment

2. Characteristic analysis of sheath circulating current

2.1 Analytic method of sheath circulating current The equation for sheath circulating current is expressed as eq (1) at one major cross bonding section [5].
l [Z1][Isi] + m [Z 2][Isi] + n [Z 3][Isi] + [Vs] + 2 Re

Isi = 0
i =1

(1)

where, l,m,n : Length of each cross-bond section [km] [Z1],[Z2],[Z3] : Ground return path impedance matrix of cable sheath in each section [Isi]: Cable sheath circulating current matrix[A] [Vs]: Cable sheath induced voltage matrix[V] Re : Ground resistance[]

1. Introduction
Current flowing cable causes loss of conductor resistance, dielectric and metal sheath. Especially, sheath loss is caused by circulating current between sheath and ground or sheaths[1]. A underground transmission system operated in trefoil or duct has an imbalance of length between cable joint boxes, different height, and different configurations making various imbalance of electrical phenomena. Those are reasons of occurring imbalance of mutual impedance between sheath and conductor of other phase, sheath mutual impedance. Increased sheath circulating current raises temperature of cable with sheath losses, and it reduces transmission capacity. It can harm body and cause the faults by breakdown of insulators or cable jacket, therefore sheath circulating current must be reduced. Analytic and theoretical studies, however, to solve such problems of underground power cable systems are not enough to take out the proper method [2]-[4]. This paper analyzes sheath circulating current of practical underground transmission system which has excessive sheath circulating current using EMTP/ATPDraw and calculation equations. Characteristics of sheath circulating current are analyzed in various ways with characteristics of sheath circulating current following burying types, impedance variations in formation of phase and effects of sheath circulating current in section length and ground resistance. From results, practical reduction methods are set for actual underground systems which have a sheath circulating current exceeding problem and reduction

2.2 Correlation of sheath circulating current and impedance Mutual impedances of cable sheaths, sheath and other phases conductor are affected in a cable central distance. The cable central distance is changed by cable burying type, and it has an effect directly on mutual impedance.

600 S(cable central distance)[mm] 500 400 300 200 100 0 90 85 80 75 Zp(phas e) 70 0.2 0.3 0.4 0.5 0.6 0.7

Zm (m agnitude)

Fig 1. Comparison amplitude and angle of mutual sheath impedance with variation of distance in central point of cable

To analyze the effect of cable central distance for sheath mutual impedance, fig 1 shows the characteristic of impedance varying along central cable distance. As shown in fig 1, amplitude and angle of mutual impedance have a reducing characteristic along increasing cable central distances.

2155

International Conference on Electrical Engineering 2002

NJ-a

300[m]

IJ-b
S12

300[m]

IJ-c
330[mm]

300[m]

NJ-d
330[mm]

S23

S13

330[mm]

330[mm]

Fig 2. Formation types of cable(duct) The central cable distance at cross-bonding section(NJ-a NJ-d) in non-grounding type with cross bonded to CCPU(Cable covering protection unit) between normal joint box(NJ) is changed to analyze the changing characteristic of sheath circulating current with varying impedance. To analyze a characteristic along central cable distance, each length of joint boxes is set to 300m and load current is set at 320A. Fig 3 shows amplitude and angle of each phase when S12 and S13 in fig 2 are changed respectively from 100mm to 500mm. The left in fig 3 shows sheath circulating current characteristic of each phase in varying S12, As shown in fig 3 sheath circulating current is reduced to minimum as distance increases until center-to-center cable is 330mm, then it is increased slowly again. The right showing characteristics of sheath circulating current for each phase in varying S13 also has a minimum at 470mm. Therefore constant center-to-center cable distance can reduce circulating current with unified formation of cable.
Fig 5. Sheath circulating current in varying S12 and S13 the amplitude of sheath current(is1) in phase A has the characteristic of variation which is reduced from 120A and then increased to maximum 140A, and angle is suddenly increased. However the amplitude of sheath current(is2) in phase B is critically increased more than variation of phase angle. The right is the case in varying S13, amplitude of sheath current in phase A is changed with angle. However amplitude of sheath current in phase C is slowly varied but angle is critically increased. From those results, balanced burying type and formation reduce sheath circulating current as reduced imbalance of impedance. Therefore the reasons that sheath circulating current rises with formation in fig 4 are section length and sheath mutual impedance by imbalanced cable formation. Then correlation of sheath circulating current and section length between joint boxes is analyzed in the next. 2.3 Correlation of sheath circulating current and cable section length The sheath circulating current is analyzed with section length. Sheath circulating current is theoretically zero with the same cable formations and equal section length. Fig 6 shows the characteristic of sheath circulating current along section length within joint boxes A S/SNJ-3 of case study installed in trefoil.
90

sheath circulating current[A]

Fig 3. Sheath circulating current in varying S12 and S13

To analyze changing characteristic for practical system with the excessive sheath circulating current problem, sheath circulating current was analyzed along varied cable central distance using underground system data between NJ-3 and NJ-6 of case study which is measured the highest sheath circulating current. Fig 4 shows the formation of cable and section length between NJ-3 and NJ-6.

80 70 60 50 40 30 20 10 100 150 200

s heath c irc ulating c urrent when "l" is c hanged s heath c irc ulating c urrent when "m " is c hanged s heath c irc ulating c urrent when "n" is c hanged

250 300 350 s ec tion length[m ]

400

450

500

NJ-3

280[m]

IJ-4
S12

324[m]

IJ-5

203[m]

NJ-6

Fig 6. Sheath circulating current with section length As shown in fig 6, sheath circulating current has minimum at equal length within 250350m. From the results of sheath circulating current at each phase along the varied length of NJ-3 IJ-4 in practical section combined with trefoil and duct of case study, sheath circulating current is reduced in inverse proportion with section length. For correlations between section length and sheath circulating current, sheath circulating current can be slightly reduced with arbitration of section length in imbalanced formation, but critically reduced with balanced formation.

S23

S13

Fig 4. Formation type of cable (duct + trefoil) Fig 5 shows that the amplitude and angle of sheath circulating currents in each phase when S12 and S13 in fig 4 are changed 100mm to 500mm. The left in fig 5 is the case in varying S12,

3. Case Study
a To reduce sheath circulating current in practical system having problem of excessive sheath circulating current, the

2156

International Conference on Electrical Engineering 2002

characteristic of sheath circulating current is analyzed in various ways and the increment reason is found with detailed simulation using EMTP/ATPDraw from analyzed, and also practical reduction method is set. 3.1 Modeling of underground transmission system Installed type of practical underground transmission system of KEPCO A S/S B S/S section used in this paper is mixed with trefoil and duct with 154kV 1200 XLPE and OF cable. Total cable length is 7.518km and normal joint boxes and insulated joint boxes are included in 26 joint boxes. Fig 7 shows cable joint type and section length of 18th joint boxes increasing sheath circulating current. As seen in fig 7, XLPE cables and OF cables are connected at NJ-16. Also sections to IJ-8 are trefoil type but 150m of IJ-2 NJ-3 and 275m of NJ-3IJ-4 are mixed with trefoil and duct type. All cables are installed in duct type after IJ-8. Direct ground resistance of sheath at terminal joint boxes is considered as 1 following guide and at normal joints are 10. 3.2 Sheath circulating current analysis Fig 8 shows the sheath circulating current simulated in each phase of practical system of case 1. Sheath circulating current is maximum at NJ-6 and rising in IJ-4NJ-9 section as seen in fig 8.
A S/S IJ-1 300m 245m IJ-2 NJ-3 150m 200m 5m 275m IJ-4 324m IJ-5 203m NJ-6 IJ-7 402m 226m IJ-8 345m NJ-9

impedance in sheaths, conductor-sheath of other phases because sections of NJ-3NJ-9 consist of trefoil and duct. Especially, sheath circulating current is critically increased at NJ-6 by impedance variation in mixed installation type and imbalanced length of joint boxes. And sheath circulating current is getting stabilized because all cables are equally installed in duct after IJ-8. Fig 8 shows the comparison measured and simulated results of current in phase A. Each result shows similar pattern and if many environmental factors like bounded cable and incorrect phase arrangement with a drawing are taken account simulation method and results in this paper are quite accurate.
200 180 Sheath circulating current[A] 160 140 120 100 80 60 40 20 0 IJ-4 IJ-5 NJ-6 IJ-7 IJ-8 NJ-9 Cable section

measured case 1

Fig 9. The comparison of the measured and the simulated value 3.2.1 Exclusion of effect from different installation type (Case 2) To prove the reduction effect of sheath circulating current with setting mutual impedance, all sections installed duct type are changed into trefoil type in case 2. Fig 10 shows graph for the comparison of sheath circulating currents in phase A of case 1 and case 2. Sheath circulating current in case 2 is more reduced than case 1 for all sections. From results, this paper proves that sheath circulating current can be reduced by same installation type. And a section with two installed types has an effect on sheath circulating current at normal joint boxes between cross-bond sections but less effect out of the section. Therefore sheath circulating current does not increased for all cable systems, but for different installation types or section of imbalanced length between joint boxes.
200 180 160 140 120 100 80 60 40 20 0 IJ-1 IJ-2 NJ-3 IJ-4 IJ-5 NJ-6 IJ-7 IJ-8 NJ-9 IJ-10 IJ-11 NJ-12

IJ-10 297m 270m

IJ-11 208m

NJ-12 218m

IJ-13 316m

IJ-14 85m

IJ-15 350m

NJ-16 350m

IJ-17 360m

IJ-18

B S/S

: normal joint box : insulation joint box

: cross bonding : direct grounding

: OF Cable(duct) : XLPE Cable(duct) : XLPE Cable(trefoil)

Fig 7. Joint type of cable and section distance Table 1. Example of analysis for case study
Case 1 Case 2 Case 3 Case 4 Modeling of practical system Exclusion of effect from different installation type Exclusion of effect from imbalanced length Exclusion of effect from different installation and imbalanced length

Sheath circulating current[A]

Case 1 Case 2

200 180 sheath circulating current[A] 160 140 120 100 80 60 40 20 0

IJ-1 IJ-2 NJ-3 IJ-4 IJ-5 NJ-6 IJ-7 IJ-8 NJ-9 IJ-10 IJ-11 NJ-12 IJ-13 IJ-14 NJ-16 IJ-17 IJ-18 IJ-19

Sheath current(Phase A) Sheath current(Phase B) Sheath current(Phase C)

Cable section

Fig 10. The comparison of sheath circulating current with case

Cable section

Fig 8. Sheath circulating current of phase for case 1 Sheath circulating current rises along various length in joint boxes and impedance variation with imbalanced mutual

3.2.2 Exclusion of effect from imbalanced length (Case 3, 4) Fig 11 shows that the simulated results along the varied length between joint boxes of case 3 and case 4 are compared with the result of case 1. As seen in fig 11, case 3 has less effect on sheath circulating current. It proves that setting equal length of section

2157

International Conference on Electrical Engineering 2002

with various formation does not have an effect on sheath circulating current reduction. However the results of case 4 show the 97% of reduction effect with sheath circulating current less than maximum 5A. From results, it is proved that the same installation type with setting equal section length for all cable as chapter 2.2 analysis reduces the sheath circulating current.
200 180 160 Sheath circulating current[A] 140 120 100 80 60 40 20 0 IJ-1 IJ-2 NJ-3 IJ-4 IJ-5 NJ-6 IJ-7 IJ-8 NJ-9 Cable section
Case 1 Case 3 Case 4

4.1.1 The first method Table 2. Application method for installed cable systems Method 1 The changing ground resistance at NJ Method 2 The changing distance of each cable phases with trefoil type Method 3 Installation of resistor at the cross-bond lead cable The first method is the changing ground resistance at normal joint box. In this paper, correlation of ground resistance and sheath circulating current is analyzed in detail. Fig 13 shows sheath current with resistance from 1 to 100 at normal joint boxes. As shown in fig 13, it doesn't have a difference with resistance as same as case 1 shown. Those result shows that ground resistance have no effect on reduction of sheath circulating current.
200

Sheath circulating current[A]

180 160 140 120 100 80 60 40 20 0 IJ-1 IJ-2 NJ-3 IJ-4 IJ-5 NJ-6 IJ-7 IJ-8 NJ-9 IJ-10

Fig 11. The comparison of sheath circulating current with case

Case 1 1 ohm 10 ohm 20 hom 50 ohm 100 ohm

4. Sheath circulating current reduction methods and application of reduction equipment

It is the fact that power cables are not installed on basis of planned research but taken account of geographical or environmental factors in Korea. The increment of sheath circulating current has problems in the simulated system as well as other practical systems, which causes insulation breakdown by rising temperature of cable, reducing power transmission capacities and electric shock to personnel. Therefore setting a reduction method for increasing sheath circulating current is urgent. This chapter proposes the reduction methods of sheath circulating current for newly installing cable systems. Practically, changing the whole power system which is already constructed is impossible. So this reduction method will be a basic data for planned system. Fig 12 shows the order of reduction for each case. Sheath circulating current of case 1 which had maximum value is considered to be 100%, and then compared with the other cases. The order of reduced sheath circulating current is case 1, case 3, case 2, case 4. Case 2 shows 66% of reduction effects and case 4 shown 97% has highest. Equal length between joint boxes shows low reduction method with 8.5%. As sheath circulating current reduction method based on results of the paper for newly installing cable systems, it is recommended that cable formation in all sections is set as duct or trefoil type with equal length.

IJ-11 NJ-12

Cable section

Fig 13. The comparison of sheath circulating current at ground resistances with case 1 4.1.2 The second method The second method is the changing distance of each cable phases with trefoil type to IJ-8. This method proves the reduction effect of sheath circulating current when varying the central cable distance with 50mm, 100mm, 200mm, 300mm to minimize the increasing factor of sheath circulating current caused by imbalanced sheath mutual impedances in trefoil.

Fig 14. The application model of the second method From the results of the second method, sheath circulating current is reduced with central cables distance near 330mm in duct type. Fig 15 shows the order of sheath circulating current reduction compared proportion with maximum value at NJ-6. As seen in fig 15, sheath circulating current has 63% of case 1 at 50mm central cable distance, and imbalanced mutual sheath impedance is reduced until central cable distance is 300mm. That distance has 63.1% of the highest reduction effect. And maximum sheath induced voltage is 21V, it is a stable voltage as sheath induced voltage side because the limit voltage is 50V. In this paper, therefore the third method varying the central cable distance in trefoil type can be considered as one of suggested reduction method of sheath circulating current for installed cable systems.

Case 1 (100 %)

Case 3 (91.5 %)

Case 2 (54.1 %)

Case 4 (2.73%)

Fig 12. The order of sheath circulating current reduction method for each case. 4.1 Reduction method of sheath circulating current for installed systems This paper suggests that the practical reduction method for installed systems can be applied. Sheath circulating current reduction effects will be proved in comparing simulation results of three methods in table 2 with case 1. Three methods are possible to apply for the installed cable systems.

2158

International Conference on Electrical Engineering 2002

Method 2

and installed on the practical system which has a problem with excessive sheath circulating current to test the third method.
300mm

Case 1

50mm

100mm

200mm

(100%)

(63%)

(57.6%)

(43.4%)

(36.9%)

4.2.1 Development of sheath circulating current reduction equipment 1 of resistor to reduce sheath circulating current is developed and tested in practical system. Fig 18 shows the reduction equipment of sheath circulating current.

Fig. 15 The reduction order of sheath circulating current with the second method 4.1.3. The third method The third method is a installation of resistor at the cross-bond lead cable of insulated joint box in the section where sheath circulating current rises. Although sheath induced voltage with resistor is increased within limit value, sheath circulating current is reduced. Fig 16 shows the application model of the third method.
N J
C C PU

C C PU

N J
R

R C C PU

C C PU

R
C C PU

R
C C PU

Fig. 16 The application model of the third method. Fig 17 shows the sheath circulating current with 1 of resistor at cross-bond lead cable in a section IJ-5 and IJ-7 near NJ-6. Totally, it has a good reduction effect and shows 89% of the reduction effect with 20A at NJ-6 compared with case 1. Resistors are set on IJ-4, IJ-5 section and IJ-7, IJ-8 before and after NJ-6 respectively, NJ-6 show the 94% of high reduction effect with 11A of sheath circulating current. Sheath induced voltage also shows the stable value under 50V.
200 180 Case 1 1 ohm

Fig. 18 The sheath circulating current reduction equipment (resistor) Total cable length to be tested with the sheath circulating current reducing equipment is 6km and consists of OF cable in trefoil and duct type. And normal and insulated joint boxes are included in 16 of total joint boxes. A resistor is set at lead cable of NJ-15 and sheath circulating current is measured at NJ-12 NJ-15. Table 3 shows results from the comparison of sheath circuit in distance with and without resistor. Table 3. The results of sheath circulating current comparison Joint Phase Unit without With resistor resistor A-B Arms 43.7 11.1 IJ-13 B-C Arms 63.4 10.3 C-A Arms 56.7 10.2 A-B Arms 58.0 12.1 IJ-14 B-C Arms 56.8 12.6 C-A Arms 40.3 12.9 A Arms 80.1 1.00 NJ-15 B Arms 78.2 1.10 C Arms 58.1 1.40 From the comparison results of sheath circulating current in table 3, it is proved that resistor setting method has a good effect on reducing sheath circulating current. For resistors work as a thermal source, continuous studies are needed.

Sheath circulating current[A]

160 140 120 100 80 60 40 20 0 IJ-1 IJ-2 NJ-3 IJ-4 IJ-5 NJ-6 IJ-7 IJ-8 NJ-9 Cable section

Fig.17 The comparison of sheath circulating current method 3 with case 1. In this paper, setting resistor or reactor at cross-bond lead cable can be one of the reduction method of sheath circulating current for installed cable systems from the results of the third method which shows high reduction effect of sheath circulating current in fig 17 as well as sheath induced voltage under limit. 4.2 Reduction equipment for sheath circulating current The third method from results of reduction method for sheath circulating current is proved as most effective. Therefore reduction equipment for sheath circulating current is developed

5. Conclusion
This paper proves the accuracies of simulation with comparing measured sheath circulating current and simulated results from modeling a practical system having a problem of raising sheath circulating current. And then it proposes the reduction methods of sheath circulating current. And the sheath circulating current reduction equipment is also developed and applied for field test in practical system. This paper can be summarized as following

2159

International Conference on Electrical Engineering 2002

(1) Combined installation types(trefoil+duct) work as factor causing a imbalanced mutual impedance of sheaths and sheath-conductors of other phase (2) For the case of unified cable installation type, sheath circulating current is zero with equal length between joint boxes. Mixed formations with various joint boxes length don't have a good reduction effect on sheath circulating current. (3) A ground resistance has no effects on reducing the sheath circulating current. (4) The Proposed reduction methods for sheath circulation current on installed cable systems -For the case of combined installation types(trefoil+duct), it is effective to give the center-to-center cable interval of trefoil type. -It is effective to install a resistor or a reactor on lead cable of cross-bond (5) Sheath circulating current reduction equipment(resistor) is developed and the results of field test shows that sheath circulating current is critically reduced. The proposed reduction methods of sheath circulating current also will be a basis to contribute for establishment of reducing method for newly installing or installed cable systems having a problem of sheath circulating current, and the authors will keep studying to establish novel reduction method. References
[1] KEPCO, Practical business of underground transmission power cables, 1996 [2] KEPRI. A Study on the Establishment of Grounding Technology for Under ground Power Transmission System, 1998 [3] IEEE, "IEEE Guide for the application of sheath bonding Methods for single conductor cables and calculation of induced voltages and currents in cable sheaths, ANSI/IEEE Std 575-1988 [4] J.S.Barrett, G.J.Anders, "Circulating current and hysterisis losses in screens, sheaths, and armour of lectric power cables - mathematical models and comparison with IEC Standard 287 [5] M.G.Dwek, M.N.Eggleton, "Electric parameters of 400kV and 275kV cables used in England and Wales" [6] KEPRI, A Study on the Theory and Applications of the Electromagnetic Transients Program(EMTP) I, 2, 3, 1991 [7] Chae-kyun jung, Jong-Beom Lee, Ji-Won Kang, Tae-In Jang, Analysis and Reduction Methods of Sheath Circulating Current in Underground Transmission System KIEE, Vol. 50A. No. 11, Nov. 2001

2160