Close and Return

______JICABLE '99 87.4 Experience in service of submarine power cables150 kV in Bali strait SINISUKA N.I., Bandung Institut Teknology Bandung (ITB), Bandung, Indonesia PRIYADI B. H., PT PLN (PERSERO), Jakarta, Indonesia

Résumé Abstract Le Détroit de Bali est un détroit entre l'île de Java et l'île Bali strait is a strait between Java Island and Bali Island, de Bali qui relie l'Océan Indien à la mer de Java. La which connecting Indian Ocean with Java Sea, Bali strait largeur du détroit de Bali est de 5000 m et le maximum de has a width less than 5.000 111 and a dcpth in a maximum profondeur est de 100 m. L'alimentation de Bali est of 100 m. The needs of electricity in Bali Island is assurée par des câbles sous-marins de 150kV. transmitted from Java Island by Submarine Power Cable 150 kV. PT PLN (Persero) est la compagnie d'électricité PT l'LN (Pcrsero) an Indoncsian Electricity Company Indonésienne qui a installé en 12 ans 9 câbles (5 avec 3 within 12 years has installed 9 cables (5 cables 3 core armures et 4 avec une armure) entre Java et Bali. A ce and 4 cables single core) but which up to .now only 3 jour 3 câbles seulement sont disponibles. Les autres ont cables left. The caused of damaged cables is duc to an insulation brcakdown initiated by oilleakage. été détériorés et des claquages sont intervenus suite à des fuites d'huile. This papcr will cxplain a truc story of unsuccessful subrnarinc power cable installation in Bali Strait, which could be a good experience. And remind us that unique Cette communication examine les problèmes liés aux activity or tailor made project in past experience will not conditions d'installation des câbles dans le détroit de Bali. give us a full guarantee to have succcss projcct, because in depth of the sea still remain a nature secret.

1. Introduction U. Background Indonesia comprises more than 13.000 Islands straddling Bali Island is located approximately 5 km to the East of the equator in South East Asia. Most of the population of the Java Island and brings in a lot of tourists to come 210 million are concentrated in Java Island. there, So, the supply of energy in Bali Island to be PT l'LN (Pcrsero) an Indonesian Electricity Company is needed and has been increasing year by year. the statutory authority appointed to generate, transmit, The supply of energy in Bali Island up to the midd1e of and distribute power throughout the whole archipclago. 1988 produccd by gas and oil. Thcrcforc, the encrgy cost Java, the most densely populated island, the totally of the pel' kilo watt hours especially in Bali system becomc integrated circuit of 500 kV; 150 kV and 70 kV higher than Java system and is not economical. transmission network is stretch over 1000 km from the To fulfill the increasing of energy supply caused by western of Java to Bali Island. demand with the high reliability, the alternative power Large, modem oil and coal fire, hydro power station and supplies are needed. even a geothermal power stations supply this network In accordance with the energy sales in Bali island which enables power to be produced and transrnitted bccome increase every year, 50 the management of PLN cheaply throughout the island, On the other Islands with examine the problems. On 1982, they have decided to densely populated areas, modem generation and build by interconnecting the Bali Island power systems transmission system are also developing although the to jhe integrated system on Java through submarine distances separating the load centers have meant that full cable. interconnection has not yet been possible. By comparison, the less denseIy populated Islands have to III. Experience in Service l'clayon isolated small diesel generating plants at each load center. These power station are expensive to run and The first construction of ISO kV Java-Bali Submarine maintain and do 110t all provide twenty-four hour service Cables has been signed on November 17, 1984. to consumers, The ability to transfer ofenergy is 120 MYA pel' circuit. The detailed survey of the seabed conditions in the Bali Close and Return

strait was not carried out bcfore the tcchnical Up to 20 meier, the submarine cables were buried specification was completcd and teudered. inshore and protccted by cast iron shclls over the cable Subsequently, when the contracter carried out the survey (in Bali sidc) and cast iron pipe over the cable (in Java was found that the seabcd conditions are 10 be very sidc). ln the deeper areas (more than 20 meter depth), the onerous. submarine cables were laid directly on the seabed, The installation was coruplcted and testcd on July 1986 The submarine cables installation was completed on and energizcd al the first timo on March 26, 1987. One of Deccmber 1997 and had been operated on Decernber 26, the cables (the l", submarine cable) was found to have 1997. nd developcd a faulraud tripped out immediately. The 2 , The length of each submarine CIbles which were submarinc cable dcvclopcd a serious oil lcak within four installed as follows : wceks of energies which would also have led to electrical ~ the length of 6th submarine cable is about 4.240 .breakdown if not switchcd out of services. meier (spare of cable) U1 Based on the studv, the cause of the Iailures which were ~ the Iength of 7 submarine cable is about 4.294 occurred to the first submarine cable are due to the un­ mcter ( phase TI, appropriate route of the cable laid. seabcd conditions and ~ the length of 8t 1 submarine cable is about 4.512 the bchavior of the hard sca current which causcd the mcter (phase S~ t cable less capable 10 stand on. ~ the Iength of 9 1 submarinc cable is about 4.690 meier ( phase R) Subicct 10 the said study, the design of 150 kV Java-Bali Subrnarine Cables which was previously using the single On August 26. 1998, the 8U1 submarine cable (phase SI, t armour is now modified 10 he double annour, in addition was damaged duc to the oil lcakagc, thcn the 6 1 cement mauresses are supplernentcd to hold the cable by submarine cable (spare of cable) opcratcd t.o replace it. the distance from one cement mattrcss to the ether is th about 50 meler (bascd on the scabcd conditions). The 6 subrnarinc cable (phase S) on Octobcr 26, 1998 switch off due 10 the oil lcakagc, subsequcntly the 7'h The specification of 150 kY Java-Bali submariuc cables and 9'h submariue cablcs are switch off but opcratcd which has boen modificd by double wirc armour. without Ioading. Laying replacement of cable and cement maurcss Thcrcaûcr, on Dcccmbcr 7, 1998 the 7th submarinc cable installation al Bali strait "cre done 011 April 1988 (phase T) was damaged duc to the oil lcakage, And, the bctwecn 09.00 A.M. to 12.00 A.m. It is urgently informed rcmaining cable was again leakage on February 22, thar at Bali strait the bchavior of sea currcnt oftcn change 1999. In connection with the above damage of the cable. with the opposite current condition. Whilc the silent the PLN was carricd out to study and evaluatc what is currcnt condition norrnallv occurs on the month of the best solution. Fortuuately damage of the cables Match/April on the specifiehours only. which wcre happcncd in the guarantee pcriod. History of The said bad sea current on the timo when laying the experience in service and route of the eables is shown in north cable and the ship anchor attacked the south cable figure 1. r (the 3 t!, submarine cable) which had firstly sucecssfully laid. Bascd on site investigatiou, the south cable attacked by IV. Laying and Installation of Cable the ship anchor was damagcd and needed another IV.l. Survey und selection of route: replacement (second replacement). The north submarine 1ll cable (the 4 , submarinc cable) laid alol1g 4.810 meter The seabcd was surveyed over a \Vidc area covering the had bcen operaled sincc August 6. 1988. On Mareil 1989 proposed cable crossing. Shallow arcas down to 40 m the south cable \Vas bcing laid togcther with cement were pholographed by divers but the greatcr pari of the mallress installation and May 31, 1989 lhe SOUÙI SlraiL \vas sllrveycd with side-scan sonar. The sea bcd submarine cable (lhe Sil]. subJllarine cable) laid along was found to bc'generally firtn but undulating and very 4.260 meter lIad bccn operatcd. irregular \vith coral, rock sediment and sand formations. The two routes were selected avoiding the rocky areas. On June 1994. the norlh submarine cable was damaged Tidal currenls were observed over a period and it was callscd by shi1' anchor. PLN has already repaired t\Vicc eSlablished lhat Ihc besl period to lay the eables would but up to Deccmber 19%, unsucccss to pul in operalion. bc April to June. TI1e sea lelldcd 10 bc rough aner July To keep the energy supply contilluously with high when monsoon conditions prcvailed. The Bali Strait has rc1iability of transmission nelwork in Bali system, on always been notorious for its treacherous currents 1997 FLN has bccn installed new circuit of the submarine resuiting from the meeting of the waters of the Java Sea cabIe at Bali strait with Lhe single core and double and the Indian Ocean. 111e currents change direction annour design / incIuding the cable spare \Vith capacity twice every 24 homs and the contractor had to plan ail 240 MYA/circuit. installation work dl1ring intervals of 2 to 4 hOllrs when 111e preparation of the specifications and document, the magtùtude of the currcnt dropped bclow 2 k110ts as it negotiation of contract was carried out by FLN. TIle changed direction. . melhod of submarine cables installation are similar 10 the method previously . Close and Return

IV.2. The difficult marine conditions: V.ST Fv=---- The cables which wcre laid in 1986 could not he d energizedtill carly 1987 as the substation equipment at cithcr end wcrc not rcadv but both cablcs Iailcd bcfore Where V '" tidal current vclocity, d > ovcrall diamctcr of bcing taken over and a visual survcy (video) using ROV the cable and ST is the Strouhal numbcr which is (Rcmotc.Opcratcd Yehiele) was carricd out to cstablish charactcrizcd bv the Reynolds numbcr and the form of the cause of the Iailures. Il was round that the seabed the object in the fluid.· In this instance. the range of conditions were much more severe than was revealcd bv Reynold's numbcrs calculated for the 150 kY cable for the sidc-scan sonar method. Rocks. coral formations and various combination ofconditions Iall with in the range: undulations wcrc found along the laid cable routes and the 2.7 x 10' to 7.2 x 10' cablcs wcrc caught bctwccn rocks and suspcnded in many and the corrcsponding Strouhal numbcr for this wholc places rcsulting in mcchanical damage. range is 0.2. The natural Ircqucncy Fn of a suspcndcd span is derived IV.3 ROV Surveys : [rom the weight.and physical propcrties of the cable and the length of the span. To œnsurc that synchronizcd Il was clearlv manifest that the side-scan sonar method oscillations do not take place the maximum span should (which was quito satisfactory for the Madura Strait) was be limited ta the length which satisfies the condition: tala Il)' inadequate for pre-installation or post installation Fv < 0.7 Fu survcvs in the Bali Strait. Hence the first decision to be taken was that the selection of new and botter routes, the The maximum allowable span calculated for the Java­ monitoringofthe laying of the cables and post installation Bali cable on theabove basis was 6.23 m. survcvs should ail be donc bv video with the assistance of divers and ROY. . IV.7. Stahility on the scuhcd: To ensure stability of the cable on the scabcd against the IVA. New Design and Installation Details: forces set up by tidal currents and wavc action. varions As. the ROV survcvs indicatcd that it would be difficult to Iorms of protection wcrc considcrcd and the placement select routes toiall~' dcvoid of rocks and sharp of cement Iillcd maurcsscs at suitablc intcrvals ovcr the undulations, and the scabcd was generally very rough, il cable was the mcthod ûnally choscn. Stabilitv was dccided to rcmanufacture the cable with an additional calculations wcrc carried out for the maximum tidal layer of armoring as a safe precaution. even though the current of 2.5 m/sec and an assumcd worst wavc height dcpth of the channel itself (maximum 100 m) was not too of 4 111. Wave-induced velocity amplitudes at seabcd great to cali for additional stréngthening of the cable on level vary with the depth of water, ego 1.14 rn/sec at 20 the laying barge itsclf, whcreas the ether length was m depth to O.l7 ru/sec at 90 111 depth for a wave height of transportcd on a simple turntablc barge. Aflcr laying the 4 111. Hence the combined force on the cable due to tidal Iirst cable. the second cable was transferred at site to the currents and wave action would vary with the depth of laying barge. water. The mcthod adoptcd for cstablishing the rcquircd spacing of cement mattresses was by working out the IV.5. Stability analysis and protecttou: safcty factors at different spacing. The exccptionally strong currents and the rough seabed immcrsed weight of cement mattress cornbined to present I\VO possible causes of mechanical damage ta the cable. namely, vibration of short suspcnded immersed weight required to resist fluid forces spans and movcmcnt of the cable on the seabcd. Stability For the proposed cement mauress. the iuuncrscd weight calculations wcrc made to ensurc that the installation was in sea water was 3.25 tones and the inuucrscd wcight carricd ouI correclly and add!tional proteclion \Vas requires. W 10 resist fluid forccs is given by: provided against the forces of the tidal eurrents as weIl as W = FI + L.F2 - L.IA (W2 - L2) wa"e action. Wherc: FI = lluid drag on maltress, F2 = fluid drag/Ill on cable, IV.6, Suspcndcd spans: L = fluid lift on cable/m. W2 = inunersed wt ofcable/m The s!rong tidal 'curren! can producc vortex-induccd L2 =fluid lift on cable/m oscillations on a suspended span of cable and frequency lA = coef[: offriction of such oscillations. approaches the natural frequency of the suspcnded span, mechanical fatigue of the cable will FI. F2 and L2 \Vere calculated using various force occur. The highest vortex shedding frequency Fv is coefficients and Reynolds numbers and Keulegan­ ealculated from the rclationship: Carpenter numbcrs. The safe spacings thus derived were for different depth on a level seahed and they \Vere further reducing to Close and Return another set of figures for the spacings for different IV.9. Cement mattress laying : gradients of the seabed. Finally mattress positions were A special barge was prepared for mixing concrete and plotted one by one based on the depth and the gradient of pumping it into 111e mattress on the work boat. The the seabcd along the cable route. concrete was mixed with a slow setting additive. Special As there is not a great deal of shipping activity in the lifting gear and crane were installed on 111e work boat to region, protection against the risk of anchor damage was handle the cement mattress and lower it into the sea. The not considered in view of the prohibitive cost involved. position for lowering each mattress was fixed using the 111C only way of providing this would have been to trench Trisponder system coupled with a Simrad acoustic the cable but cutting the trench in the Bali strait would not system for locating the mattress frame under water, The be a simple proposition due to the severity of the mattress frame was fitted with swivel cameras controlled conditions. The shore ends of the cable down to 20 m from the work boat. Upon positioning the mattrcsses depth were, however, to be protected against coastal accurately over 111e cable as seen by the TV cameras, 'activity by casing the cable in cast iron pipes and quick releasers were activated from the work boat. After cmbcdding them 0.3 111 below scabcd. the mattrcsscs were installèd their positions were checked by the cameras on the frame.

IV.S. Installation of cables: Three pre-laid anchor positions were used so that the V. Conclusion cable laying eould take place in stages during slack tides. For interconnecting the Bali Island power system to the A four point mooring system was used at the anchor integrated system on Java could be considered the positions. The forward motion of the barge was achieved establishment of 150 kV Ol-IL. If the submarine cables by towing on a pre-laid steel hauser. The position and would he used, so the more detail survcy must be carried Iying aspect of the hasher was checked by ROV-TV out and the submarine cable design should be followed bcfore cable laying commenced to ensure that tlle route to the survey result. was clear. The barge was fitted with two 220 HP thrusters and was restraincd al the stern by a second hauser to provide bctter control during cable laying. Vi Reference During the pre-installation ROV-TV survey it bccame 1. A. DURAI, Sofyan Saibir, Hasim Surotaruno; clear that it would be practically impossible to lay the "Special Technical Features Associated with the cable without any suspensions whatever exceeding 6.23 Installation of Submarine Power Cable in m. Il actual fact a Humber off small suspensions over 6.23 Indonesian Waters". m in span occurred and these were rectified later : 2. PT PLN (Persero) P3B; "Report on cause of (a) By casing the cable clear of minor outcrops by incident" Java-Bali Project, Ref. No. )E-3-1655, picking il with a special grapncl anchor Iastcncd 10 June 24, 1988. the mattress handling frame which was fitted with a 3. PT PLN (persero) P3B; "Suspension Analysis for pair of rotating TV cameras, or 150 kV 3x300 mm2 Submarine Cable (DF)" Java­ Bali Project, Ref. No. OE-3-1717-B. (b) By building up a support using concrete mattresses alongside the cable and then moving the cable onto 4. PT PLN (persero) P3B Sektor Denpasar; "Frivole the new support. Communication" 5. D. Mc. Allister; "Electric Cables handbook"; Granada, 1983. Close and Return Jt.WA· BALI 150 KV SUBMARINECAB LES ROUTE

EAST JAVA l 't l,. 1 )Oll K t nur t . lll ·I~ -lï' D1.E~[ J. - 1. . / 10000N '" ..

#3

1'" " ]-. 1011111 SOUlft 1000 mater iï il ·ir- · ·'::l~ir.n·iii1~_: I! :~-t! l ].!.-.r. .~ (" l' 1 t l.~ L_ _

#u ! 189m

1000 moter

/~ ..:t. ) :'(\111 11 ; 11 ).1 ~ 1 l , t . J,. .. ,;,

W W w <> <> W <> ,-; <> g <> <> <> <>o -s - -N -.., CABLE' 1 : 3 Core 1Single Armoured (Installed May - Augus-t 1986. enargize March 1987. cir ect insulation breakdown. no oil leakage) CABLE' 2 I lE![,Jl:œiji]l : 3 Core / Single Armoured ( Insl alled May · August 1986, energize March 1987. swttcb ort May 1987, oU leakeçe ) CABLE' 3 ~ : 3 Core 1Double Armou red ( InstaU.d April· August 1988, enargize August 19ô8, dam;l ge by ship anchor on June 1984. oil leakac'" CABLE' 4 : 3 Core 1Double Armoured ( Installed April 1988, damage br ship lJnehor in the laying operat ion of cable 1:3 on April 198a . oil leaklilal " CAB LE tt 5 : 3 Core J DOûble Armoured ( Installed March · May 1989, . 5 replacement based on msu rance esause for cable U , energtze May 1St 9 ) CABLE 16 : 1 Core l Double Armoured ( lnstallt)d Nov - Dec 1997, &5 spare cable. energize 2~08/199S , switch off 2611011998 , 011 leakaQe ) CABLE' 7 ...,.:ç....., : 1 Core 1 D ou~le l',~moufed' Ptu..se T ( tnstaûed Nov . Dec 1997 , energize 26112/1997, ~witch off 2611011998, damage 07/ 1211SD8,o il Ict.biQ e ) CA8L E ' 6 ~ : 1 Core ' Double Armoured 1Phase S ( Installed Nov - Oee 1997, energize 2611Z!1997. switch off 261081 199S, cil leakage, repleee by spere cab le #6) CA8LE ' 9 ~~·r..zl : 1 Core l Double Armoured' PhioS& R( Instalied Nov. Cee 1997. energize 2611211&97, swüc h oft 25.'1011998, Oil lt'akage stzrting on 2:.:. Pcbrven Î .. .:..;.

- Cable # 1 end cebta :: 2 W2S eut on shore end positron due to smooth instalation for cable #3, #4, #5. - Cab re #- 5 the remain ing scbrnarlne cables ccnnectinç Jawa - Bali system, wi th transfer c apacity 120 MVA ~