1NRL Memorandum Report 1950

Anayss f h Failure of thie

SAUTEZ TOTO Il Deep Sea Moor and thePerformance of itsl Cathodic Protection System

R. L. GRoOVER

Metallurgj Diviso

November 1968

MA 2)

M1 ilc D

for 6 3 ,r,) S 'sw Ifi ; 1- chniaInfor mation -rm id Vc j 2215

NAVAL. REEARMJ LABORATORYWashingtoni, D.C.

This document ham been approvsd for public relaem and Wae;, its distfibution is uullrdltAd

CONTENTS

Abstract .................................. iiProblem Status ............................ iiiAuthorization ............................. iii

INTRODUCTION .................................... 1

FAILURE OF THE MOOR ............................. 3

SALVAGE OF THE MOOR ............................. 3

EXAMINATION OF THE WIRE ROPE ................... 5Field Observations ........................ 5Laboratory Observations .................... 6

PERFORMANCE OF THE CATHODIC PROTECTION SYSTEM ... 9

SUMMARY AND RECOMMENDATIONS................... 11

A CKNOWLEDGM. ENT ................................... 14

REFERENCES ........................... ......... 15

ABSTRACT'

This report.:contains background information on thedesign and installation of a wire rope three-point deepsea-moor which was originally coated with a bituminoussubstance and the-critical areas protected with a magne-sium galvanic anode cathodic protection system designedby the. Naval Research Laboratory. The report also de-scribes the failure of the--moor after:4 1/2years' serviceand its subsequent salvage, and presents the results of astudy of the corrosion pattern-, proposes the :cause- offailbre, and evaluates the :performance..of the -cathodic-protection system. Recommendations are presented fb theprotection of future moors and for ;possiblre researchdirected towards the understanding and prevention of-failure of wire rope structures fn sea water under con-ditions of static stress and of, fatigue.

Three of the wire rope failures were associatedwith severe corrosion and the absence of bituminouscoating. In two of the failures the bituminous coatingwas essentially absent but no severe corrosion wasobserved,. There is some circumstahtial evidence thatthe latter failures may have been caused by a shipmooring to the intermediate buoy which was not intendedto be used for this- purpose. All but one failureoccurred 1350 to 2050 ft from the nearest active anodeconnections. -One failure occurred680 ft from- anintended anode connection, but there is evidence thatthis particular anode was lost when the moor wasinstalled. Some other major components of the moorprobably received cathodic- protection for as littleas 9 to 12 months out of the approximate 54-monthslife of the moor. This was a result of a variety ofproblems associated with the replacement of theoriginal anodes.

The results of the corrosive attack and the failureof the moor indicate that where long life is demandedfor future moors the following steps should be taken:

(a) Provide a coating better than the bituminoussubstance used in the present moor.

ii

I

C C

, - -., .--.-- '--- = . . -- .e - _ _________...._______ ____,,____ _______ 2 7 .

(b) Provide cathodic protection as a secondary defenseagainst corrosion. It is essential'that cathodic protec-tion be considered early in the design phase and provideUthroughout the life of the moor if it is to be effectiVe.Scheduled inspections and replacement of consumedanodesare essential for proloiged trouble-free life- of the struc-ture. Where-adequate power is available, impressed currentsystems should be considered to obviate the ,necessity offrequent anode replacement,,iI

STATUS

This completes one phase of the program; work onother phases is continuing

AUTHORIZATION

NRL Problem 63M04-02Task No. S-4607-li894

i

INTRODUCTION

A fixed-position deep sea moor was needed by the Navyin conjunction with activities contemplated for the Tongueof the Ocean (TOTO) area.

A three-point moor, designated-AUTECTOTO II, wasdesigned by the Bureau of Ships (now Naval Ship SystemsCommand) to meet the set requirements (1). The moor con-sisted of three legs positioned 1200 apart . Ea'ch.leg was comprised of the center buoy, a main buoy, and anintermediate 'buoy. Approximately 11,000 ft of' 1 1/4-in.6'19 filler wire (WSC), extra improved plow steel' gaiva-nized, bituminous coated wire rope connected each main-buoy to an anchor chain and anchor positiohed on the seafloor. A 4,100-ft vertical riser beneath each intermediatebuoy &upported the wire rope between the main. buoy 'aid theanchor to form an upper and lower catenary. The riserswere also 1 1/4-in.-diam wire rope except the upper 105 ftof each which consisted of 30 ft of 2 1/2-in. chain and75 ft of 1 5/8-in.-diam wire rope. A schematic diagramshowing the moor and a typical anchor leg is shown inFig. 1.

After the moor design was completed and while mate-rials -were being procured, the Marine Corrosion Section

of the Metallurgy Division, Naval Research Ltboratory,was contacted concerning possible methods to extend thelife of the moor beyond one to two years. NRL made ananalysis of both the fundamental and operajtional asPecLsof providing cathodic protection to redlce corrosion: onthe AUTEC TOTO I-I moor (2,3,4). This analysisshowedthat it was not possible to protect the entire moorwithout a complete redesign (impractical) :but that itwould be p6sE'ble to protect the critical junction pqlhtsand the wire )pe for about 600 ft from these junctibnpoints. It also appeared possible to protect the wirerope to a distance of several hundred feet up from theanchors. As the best available protection withoutredesign, NRL proposed to protect the critical areasof the moor with magnesium galvanic anodes. The nec-,essary materials and equipment were procured and assem-bled at NRL and were later delivered to the vesselswhich were to lay the moor.

1

OA 26-30 May 1962, the moor was laid in about 5400 ftof water in the-ThTO. The late Mr. L. J. Waldron of NRLsupervised-the attachment of approximately 4400 lb ofmagnesium anodes to the moor at the loc~tions shown inFig.. 1.

AlI anode arrays near the surface were designed to

facilitate periodic examination and replacement, which wasexpected tob6 required after one to two years.

In September 1964 (approximately 28 months after themoor was installed), one anode which had been removed fromthe ioor Was returned to NRL. Inspection showed that onlythe core remained, indicating that the magnesium had beencompletely consumed and that anode replaceiment wzs required.On 13 November 1964, NRL was authorized to procure replace-ment anod6 , and by early April 1965 they were assembledand ready for shipment to TOTO. The DavAd&y'T~ylor ModelBasin (DTMB) (now the Naval Ship ResearCh & DevelopmentCenter, Car'derock) negotiated the contract for the instal-lationof the replacement-anodes and apparently experienceddifficulties in letting the contract. This difficulty,Plus other factors, delayed the actual shipment of theanodes until 18 October 1965.

According to "Negotiated Contract" N167-173(X)FBMissued- by DTMB for instal ation of the replacement anodes

*and executed -by the contracting officer on 19 August 1965,NRL was to coordinate (or direct) the technical aspectsof the contractor's performance and was to inspect (atTOTO) the supplies and services provided by the contractorto assure conformance with the contract. Because NRL wasnot notified of the installation of the replacement anodeson 17-19 November 1965, the technical direction and in-spection by NRL during anode replacement was not accom-plished.

Fiom study of an informal written report dated29 November 1965 by the U.S. Navy Officer who workedWith the contractor and from telephone conversationswith the same oflicer, NRL concluded that it was un-likely that the replacement anodes hsd been installedaccording to anticipated replacement design.

2 i

51*1

FAILURE OF THE MOOR

In November 1966, about 4 1/2 years after the-moor wasinstalled, the 4100-ft vertical riser between the inter-mediate buoy and the ring which joined the upper and lowercatenaries on the 1500 leg parted at an estimated distanceof 1700 ft below the intermediate buoy, or about, 1700 ftfrom the nearest anode connection (5). In February 1967'a 48 1/2-ft section of the failed end of the riser wasrecovered. Parts of the recovered riser Were sent to theNaval Applied Science Laboratory (NASL), and a 1-ft section

was delivered to NRL on 8 May 1967. The failed section wasobserved to 'be severely corroded and had undergone an appre-ciable loss in breaking strength (6).

A similar riser beneath the 2700 leg intermediate buoyfailed about four months later (March 1967). A tug towedthe buoy and riser section to the Atlantic Undersea, Testand Evaluation Center (AUTEC) on the western shore of TOTOwhere the section was beached on 3 May 1967. Preliminaryfailure analysis made at AUTEC by the author disclosedthat the riser had failed about 2580 ft below the inter-mediate buoy, or about 1520 ft from the nearest anodeconnection, and it was observed that replacement anodeshad not been properly installed (7). Numerous samples.were cut from the riser for study in the laboratory; theresults of the study are reported herein.

SALVAGE OF THE MOOR

The failure of the two risers and the subsequentshifting of position of the remaining buoys led to adecision to salvage the remnants of the moor to removethe existing hazard to navigation. A contract wasawarded to the Merritt Chapman & Scott Corporation,and their ship, the MV CABLE, was dispatched from KeyWest, Florida, to TOTO on 8 July 1967. Captain HenryHalboth joined the ship at TOTO as Salvage Master.The author was also aboard the salvage, ship to obtaininformation and samples from the moor required to ana-lyze the corrosion damage and performance of thecathodic protection system.

3

W , ",

The CABLE -arrived on site at 1240 hours on 10 July 1967-aiid'moored to the main buoy/ on the 1500 leg. All buoysexcept the 0300 leg intermediate buoy were observed to liein an approximate straight line bearing to the northwes-v.The, first buoy to the northwest 'of the 1500 leg main buoywas the center buoy,, next the 0300 leg main buoy, and lastthe 2700' leg main buoy. The 0300 leg intermediate buoy wasfar from the moor ,on 'a bearing of 3520 from the 150 ° mainbuoy.- The wind was from the east at 850.

The posi-tion of the buoys indicated that the 0300anchor leg had failed, and the 0300 main buoy had driftedinto the, span wire between the center buoy and the 2700leg main buoy. Later during the salvage operation it wasdiscovered that both the upper and lower catenaries on the0300 leg had indeed failed, and the 0300 intermediate buoywas adrift but was dragging about 700 ft of wire rope onthe bottom, thus slowing its rate of drift. The failure ofthe 0300 leg apparently occurred prior to 12 May 1967 sinceon that date Naval Ship Engineering Center (NAVSEC)-per-sonnel observed that the buoys were in about the same

51 relative ,position as reported above (8).

Salvage-of the moor commenced on the afternoon of

11 July 1967. During the salvage operation, the followingevents Occurred:

(a) The 2700 lower catenary failed- about 680 ftabove the anchor chain. About 5950 ft of the upper.catenary was o6i dbck at the time, and the gage on thewinch pulling on the catenaries indicated a stress onthe rope Oi 39, 000 lb before failure of the lowercatenary which Was, reduced to 19,000 lb immediatelyafter the failure. This indicates that the stresson the lower catenary at the point of failure was20,000 lb when it failed.

0 (b) All remnants of the failed moor were sal-

vaged except two anchors and anchor chains and partof"the lower catenaries on the 030 and 2700 legs.

97.Salvage was completed at 2220 hours on 20 July1967.

, 4

EXAMINATION OF THE WIRE ROPEFi ld Observations I

The wire rope wasobserved visually as it was salvagedto determine the presence and location of broken wires andthe appearance of the corrosion products. In addition thewire rope was measured and tagged at preseiected locations. 1*Samples, were cut from these locations for study in thelaboratory. The remnants of anodes Were also recoveredfor further study. All hardware items were checked forcorrosion.

No multiple broken wires were noted near the anodeconnection points except that in a few instances somemultiple wire breaks were noted at socket connections.However, some of these wires at sockets were observedto break when the sockets came over the bow roller ofthe CABLE at an angle and under high stress.

Multiple broken wires anj a failure of the wirerope occurred 680 ft from the lower anode connection-on the 2700 lower catenary. The wire rope was severelycorroded at and above the point of failure (Fig. 2),.All other multiple broken wires and ,faiiures were atleast 1000 ft from the nearest anode connection. Fieldobservations relative to broken wires are given inTable 1.

The uper and lower catenaries on the 0309 legfailed prior to salvage. Corrosion on thp catenarieswas not severe and is not believed to have been thecause of the failures. The 25-ft -section of the uppercatenary in which failure occurred had been twistedprior to failure. Color photographs are available atNRL (Code 6325) which show in detail the twist andfailure. The lower catenary appeared to have kinkedbefore or during failure (Fig. 3). It can be surmisedthat the 0300 catenaries failed as a result of shipsimproperly mooring to the intermediate buoy ratherthan to the main buoy.

5

The 270' leg vertical riser failed about 2580 ft belowthe intermediate buoy. The appearance of the failed end isshown in Fig. 4.

The appearance of corrosion products on the wire ropecomponents was generally gray-black near anode connections.Rf-dish-brown corrosion products gradually increased withdistance from anode connections. At 1000-2000 ft from theanode connections about 100 percent of the visible wiresurfaces were coated with reddish-brown corrosion products.The bulk of the corrosion products was located betweenadjacent strands. The density of the corrosion productsappeared to increase with depth of water. This increasewas particularly evident at depths greater than 2400 ft.

Laboratory Observations

Numerous samples (3 ft or longer) were cut from the" wire ropes at the locations shown in Table 1. The tech-

nique used in the laboratory to evaluate the samples wasas follows:

(a) An outer strand about 18-in. long was removed

from each wire rope sample. The core strand was alsoremoved where failures bad occurred.

(b) The presence and. amount of bituminous coatingon the strands was determined visually.

(c) The strands were soaked in mineral spirits,then rinsed in trichloroethylene to remove the bi±uminouscoating.

(d) The cleaned strands were examined at 7Xmagnification under a stereo-microscope to estimatethe percentage of the zinc coating still present.HydL chloric acid (to cause hydrogen ebullition) wasused as a further spot check on questionable areas.The same technique was used to estimate the percentageof zinc remaining on an individual outer wire fromeach strand.

6

r

(e) Alternate outer wires (six per-strand) wereremoved from each strand, except Ihat it was necessaryto remove all (12). outer wires from core strands fromthe 0300 leg because adjacent outer wires were of dif-ferent diameters. This procedure iw in contrast tothat used for the outer wires evaluated from other corestrands, which had the same diameters.

(f) The removed outer wires and remaining wiresin strands were recleaned in trichloroethylene.Corrosion products and zinc were then removed byimmersing the wires in concentrated hydrochloric acidinhibited with 40 g/l antimony trioxide at room temper-ature. The wires were rinsed in water and coated witha rust-preventive compound.

(g) The inner wires of the strands were viewedat 7X magnification to determine the presence andextent of corrosion pits, and the remaining outer wireson the strands were checked for cracked wires andmechanical damage.

(h) The removed outer wires from the strands wereviewed at 7X magnification to determine the presence ofbroken o; cracked wires and mechanical damage. Theminimum diameter of each outer wire was then determinedto the nearest mil using a micrometer.

The results of the determination of minimum diameterof outer wires are shown in Table 1, together with thelaboratory data concerning broken wires, pitting of innerwires, and the presence of mechanical daaiage. Table 1relates these data to the approximate depth of water,distance from the nearest anode connection, total anodeconsumption, and field observations relative to brokenwires.

The data show that broken wires were observed inthe field on the 1500 leg lower catenary, but none werefound on the small laboratory samples removed from thecatenary. Field and laboratory observations of brokenwires were in good agreement in all other instances.

7

keqhaical damage noted in. the table referk to slight;plastic deformation of outer wires on strands. The damagewas noted on the 1500 upper and lower catenaries and the2700 upper catenary. The cause. could not be determined.

The minimum diameter of outer wire data from Table 1are plotted in Figs. 5, 7. and 9 and are related to thelocations of anodes, approximate depth of water, andlocations of wire rope failures.

Figure 5 shows t'at the average minimum diameter ofouter wires from the upper catenaries ranged from 53 to77 mil. The wires having smaller residual diameters werelocated about 1000 ft from the upper anodes, and those

- having larger residual diameters were at the lower endsof the catenaries near the 400-lb anode connections. The0300 leg upper catenary twisted and failed at the locationsnown in the figure. It is apparent that the point offailure was not associated with minimum wire diameter orthe point of highest natural stress since both would betowards the upper ejd of the catenary. The appearanceof failed ends of wires from the leg upper catenaryis shown in Fig. 6.

The average minimum diameter of the outer wires ofthe lower catenary ranged from 0 to 81 mil as shown inFig. 7. The outer wires from the 2700 lower catenaryat the point of failure were completely severed bycorrosion. Severe corrosion was also noted on the corestrand of the 2700 lower catenary at this p9int. Severekduti6fi of the diameter of' tli&h- dttei wircF was notnoted on the 0300 lower catenary which failed or on the150 lower catenary. This suggests that the 400-lb anodeat the lower end of the 2700 lower catenary was lost whenthe moor was laid and thus permitted severe corrosion tooccur. Another significant factor which contributed tothe severe corrosion of the 2700 lower catenary was therelatively poor coverage of bituminous coating on thestrands (Table 2). The appearance of failed ends ofwires from the 0300 leg lower catenary is shown inFig. 8.

8

L

The contrast in performance of the risers beneath theintermediate buoys of the 0300 and 2700 legs is shown inFig. 9. Data from the failed 1500 leg riser are not shown.because samples were not available. It is evident fromFig. 9 that the diameters of outer wires from the 030' legriser were not drastically reduced, but the opposite effectwas noted for the 2700 leg riser which failed because ofsevere corrosion. The only apparent explanation for thedivergent behavior of risers is that the 030 . leg riserwas relatively well protected by the bituminous coating(Table 2). A cross section of the 2700 leg riser approxi-mately 50 ft above the failure is shown in Fig. 10, andthe appearance of failed ends of wires from this riser isshown in Fig. 11.

Table 2 shows that the bituminous coating on the wireropes ranged from 0 to 100 percent coverage. All failuresof the wire ropes were associated with almost completeabsence of bituminous coating. The variability of coverageby the bituminous coating suggests that either the wire-ropes were not uniformly coated during manufacture, or thecoating leaked out of some sections of wire rope beforethe moor was laid. It was observed during salvage of themoor that the heat of the sun was sufficient to causemelting and leakage of the coating. Because the galvani-zing is a sacrificial coating, the presence of zinc(galvanize) on strands and wires also ranged from 0 to 100percent, with the higher percentage generally being asso-ciated with good coverage by the bituminous coating.

PERFORMANCE OF THE CUTHODIC PROTECTION SYSTEM

The magnesium anode cathodic protection system wasexpected to do the following:

(a) Protect the moor at critical junctures wherecontinuous rubbing of the surfaces might be expected tocontinuously remove corrosion products and therebypromote corrosion.

(b) Protect the wire rope to about 600 ft from thepoint of connection to the anodes provided the bituminouscoating used on the wire rope met the minimal conditionsused to specify a "very poor" coating.

9

(c) Xitend the service life of the moor beyond thatof an unprotected moor.

The-overall effectiveness (f the cathodic- protectionsystem cannot be fully determined for reasons which will'become apparent later in this report. However, we canreport that:

- - (a) No severe corrosion hardware was observed atcritical junctures, although some.broken wires wereobserved at .sockets, probably cause'by severe bendingstresses during-salvage.

(b) All but one of the wire rope failures occurred1350 to 2050 ft from the nearest anode connection. Onefailure occurred 680 ft from-an anode connection, but inthis instance there is evidence that che anode near the ,failure was probably lost when- the moor was laid.

(6) The life of the moor was 4 i/2 years.

The original anodes which were installed when themoor was laid - particularly those installed to protectthe upper ends of 'the risers and the intermediate buoys -

were consumed more rapidly than anticipated. Calculationsbased on the rate of consumption of replacement anodes atthese locations indicate that the original three 60-lbanodes beneath each intermediate buoy were consumed within9 to 12 months after the moor was laid. These and theother consumed anodes were not replaced until November1965, or 30 to 33 months too lae to insure continuouseffectiveness from the cathodic protection system. Bythis time the 400-lb anodes at the lower ends of theriser were probably consumed, and they were not replaced.From the post-mortem observations and calculations itappears that the failed risers were cathodically pro-tected for only 9 to 12 months of their 54-month life.

There are strong indications that the November 1965replacement of the consumed anodes was accomplished bydivers woh6 appzriently were not aware of the technicalrequirements of providing low resistance electricalcontacts between the anode szrings and the moor. This

10

was evident from the fact that they looped the insulatedconductors from the strings through shackles which theyattached to the moor at convenient locations. Fortunatelythe insulation on the conductors broke under the weight ofthe anodes and a degree of electrical contact was estab-.lished which permitted the anodes to function. The diverslost all 400-lb replacement anodes when they apparentlyallowed the anodes to free fall down the risers. This wasevident from the fact that the sheared eyebolts (minus the400-lb magnesium anodes) were located at the rings 115 ftbelow the intermediate buoys.

Since all available replacement anodes were salvaged,it was possible to obtain data concerning performance ofthe anodes which may be useful in future designs of cathodicprotection systems for wire rope structures. One original400-lb anode'was also salvaged from the junction of the1500 lower catenary and the anchor chain. This anode isof particular interest because it functioned continuouslyfor approximately five years at low temperature and highpressure at a depth of 5400 ft. The performance data forthe anodes are summarized in Table 3.

The estimated anode current densities ranged from161 to over550-ma/sq ft for the 120-lb anodes; thedensity was 445 ma/sq ft for the 400-lb anode. Thesecurrent densities are based on the original surfacearea of the anodes and an assumed anode efficiency of500 amp-hr/lb. The corrosion patterns developed onsome of the anodes operating at various current densi-ties are shown in Fig. 12.

SUMMARY AND RECOMENDATIONS

1. A three-point wire rope deep sea moor was laidin the Tongue of the Ocean in 5400 ft of water on 26-30May 1962. The moor was protected from corrosion by abituminous coating and critical areas were protectedwith a magnesium galvanic anode cathodic protectionsystem designed by NRL. The moor failed after about4 1/2 years. It was then salvaged and samples of thewire rope and the anodes were obtained for a study ofthe cause of failure and the performance of the cathodicprotection system.

11

+, ,

- 2. Failure of two l1/4-in.-diam wire rope risersbeneath-the:intrmediate buoys and failure of the 2700l ower catenary were associated with severe corrosion andthe absence of bituminous coating on the wire ropes.,Crrosion of the other lower catenaries was not severewhich leads us to speculate that the anode was probablylost off the lower end of the 2700 lower catenary whenthe moor was laid. Also, the coverage of the bituminouscoating was better on the riser and catenaries which didnot fail'.

3. The failure of the 1 l/4-in.-diam, wire ropeupper and lower catenaries on the 0300 leg-was associatedwith twists and kinks in the catenaries rather than with'severe corrosion. This suggests that these failures mayhave occurred as a result of a ship improperly mooringto the 0300 leg intermediate buoy rather than to the mainbbo.

4. The fractured ends of wires from the moor showeda Variety of failure appearances. These included trans-verse, longitudinal, and diagonal cracking, .and generalcorrosion. Research should be directed toward' the under-standing and prevention of the failure of wire rope insea water under conditions of static stress and offatigue.

5. A general reduction in' the diameter of outerwires of the strands (other than core strands) was noted

on all wire rope components. The diameter of outer wiresfrom core strands was reduced at or near points of failure.The amount of reduction in the diameter of outer wires did-not appear to be related to the depth of the water wherethesamples were obtained.

6. The coverage o bituminous coating on the wireropes ranged from 0 to 100 bercent indicating that thecoating was either not uniformly applied during manu-facture or leaked:-6t of some sections before the moorwas'laid. During salvage, heat from the sun was ob-served to besufficient to melt the coating. Sincecoatings should perform vital functions in protectingwire ropes, a more durable and protective coating should

12

be obtained or developed for future moors.. To assure coat-ing of outer wires, equipment should be developed as re-quired to automatically apply a final coating to the wirerope as it passes from the ship into the water.

7. The magnesium anode cathodic protection systemwas expected to protect the moor at critical junctures,protect the wire rope for approximately 600 ft from anode-connections, and extend the life of the moor. This studyof the failed moor revealed that there was no apparentsevere corrosion of hardware at critical junctures,although some broken wires were observed at sockets; allbut one wire rope failure occurred at a distance of 1350

to 2050 ft from anode connections. "There is evidencethat in one failure (approximately 680 ft from an anode)the anode was likobably lost when the moor was laid. Thelife of the moor *as 4 1/2 years.

8. The true effectiveness of the cathodic protectionsystem could not be fully determined because of its inter-mittent operation resulting from a variety of problemsassociated with the replacement of anodes after theoriginal anodes were consumed. It was estimated thatthe risers which failed had cathodic protection for onlyI. 9 to 12 months out of this 54-month life because of these~problems.

9. All available replacement anode strings weresalvaged. One original 400-lb anode was salvaged. Thisanode operated continuously for 5.12 years at low temper-ature and high pressure at a depth of about 5401 ft.Data were obtained relative to the performance.of theseanodes which should be useful if the design otQ nothergalvanic anode system for a wire rope structure isrequired.

10. Cathodic protection should be considered earlyin the design of future deep sea moors, and attachmentpoints for installing anodes should be provided atsuitable intervals (probably not over 1,000 ft apart),or sheathing materials better than the bituminouscoating must be applied to enable cathodic protectionto work over a longer distance.

13

51,i Cathodic protection systems on future moors shouldinclude scheduled inspections, and consumed anodes shouldbe replaced in a timeIy manner so that the moor will not bewitiout protection for long intervals as was the TOTO IImoor.

12. The means for replacing consumed anodes shouldbe-provided in the initial design of future moors.

13. When consumed anodes are replaced on future moors,the installation of the replacement anodes should only beaccomplished under the technical supervision of personnelwho understand the critical requirements of the cathodicprotection system.

ACKNOWLEDGMENT

The author is indebted to Captain Henry Haiboth andMr. Hilary Crusoe (how Captain) of the Merritt Chapman &Scott Corporation who patiently and understandinglyassisted in obtaining wire rope samples during the salvageof the deep sea moor. The author is also indebted toMessrs. W. Lazier and Thomas Ford at NRL's Marine CorrosionResearch Laboratory, Key West, Florida, for their assistancein certain phases of this study.

14

(Page 16 is Blank)

REFERENCES

1. "Deep Sea Moor - TOTO II," NAVSHIPS Tech. News 16, No. 1,35-38 (1967) January.

2. M.E. Peterson, "Theoretical Considerations in theCathodic Protection of Wire Rope," NRL MemorandumReport 1301, April 1962. Naval Research Laboratory,Washington, D.C.

3. L.J. Waldron and .H. Peterson, "Cathodic Protectionof a Deep Sea Moor (AUTEC TOTO II)," NRL MemorandumReport 1338, July 1962. Naval Research Laboratory,Washington, D.(C.-

4. L.J. Waldron and M.H. Peterson, "Unique CathodicProtection System for a Deep Sea Moor," MaterialsProtection 4, No. 8, 63-67 (1965) August.

5. John A. Rusling III, "Shipley Moor," A letter reportfrom the Manager of Marine Engineering (AUTEC) to theRange Support Manager (AUTEC), 31 March 1967.

6. R.J. Wolfe, "AUTEC - TOTO II - Deep Sea Mooring WireRope Failure; Determination of Cause of Failure," Aletter report from the Commanding Officer and Directorof the U.S. Naval Applied Science Laboratory to th&Commander, Naval Ship Engineering Center (Code 6135D),23 May 1967.

7. R.E. Groover, "Preliminary Failure Analysis of the7 UTEC TOTO II Deep SeaMoor," Technical Memorandum6320-53, 14 June 1967. Naval Research Laboratory,Washinlgton, D.C.

8. H.P. Shipley, "TOTO II Mooring Buoy Array (AfterFailure)," A Memorandum from Mr. Shipley, NAVSEC(6164E), to Dr. B. F. Brown, NRL (6320), 14 August1968.

15

_____ ____

. . . ... .. . . . . =

" - : --- • -- -.; ' - :

oo,

I *A t Z -_ =2Z

Z. I. . ...

O01-Z-v

•. ... .... .

44 ,3 .. ... ..... .. .. ; - . .

It x... . .x

o; 222"22222 .n -. s" .

P -z10

HA Pq x* 2:8 8 22:: : 0 0 0 0 0 O0 ~ 0 0 0 0

E4 0

tq NIB! uu

P4 *0 a -

z IS£

fo T

17

(Page 18 is Blank)

2- 4 .

..0 ..... ............. .. x

I - .... x x.:: .. x-.... :I

2:::~ 22 2 2

4:4o

482 - - -2

~------------49940 *0049 0 ---------- 0

- - 0 .9 2 -Vg'

i Z z

* 1 1 i

z4 r0 4

0000~ ~~~~~~~~~~ v00 0 .00 00000000400 00 00*

t.-A 1 all P.. . - - c o o . n

*9904~-~~0 -17

t

E0 W 22 9

Z -

E- g

-e~f c

CII C)

C1 0'

Z P4 c

0 0 S -00 2

E-1 00 0 - If

s*

0 0 102

vo 0 N n ,

co 0 o

F-=- C-W- F F

LI. L. LI La. LI. I. I I19

(Page 20 is Vil*n)

A 2;

X A

f; -,

X9 i- -"I Z

Oi= Sir £ C1* 6: -

1.0 2 222 2 2 = 2 2

*819

- i

I ! I

2 . i 4

IILANIFAGE

DETALS ~ TABLE 3

DETAS OFPERFORMAN CE OF MAGII-E &lM ANODES

FROM THE TOTO U DEEP SEA MOORSC

M421 .0$' Losd. M.. (3 2) ( 1) Cinfeq SO CW)

O..23 -s1r n.11 C 1)2 72 43 u m 2.-2V7 13 2 ITS12

-4 13 3 3 276

-5 1"3 as 1$ m1-4 13 80 4 1"6

hi.ts MOW 3-1 (r~r) 1"2 0 1"2 sm 31 196

al o/-2 2W 2 1. "31

-4 7)2 w7 8 380_5 1I" 43 72 22

New .tb U4? m- Ccpgwr) 12 0 212 474 21"0 15.4

-. S 20 4 30272

asxml 1 3-4 --w) "45 411 348 24.

-3 212 so To-4 M) 5 as -80-5 12 53 6$ 0

-4 1"2 54 1"3

3.2.8tb 12W Yl Salvsgwd -- -

Iss .. di~te 3.a'7

zsemtb 270* 1 (CtWO 120 Go so 306 315 22.21sterwdite Door iz 120 65 55 339I.3 120 74 44 242

4 1)2 73 47 247-1 120 71 49 258

At the Joa01100 of 60 400 12 386 3"6 445 4.3

Cxt.2ry sod tb6

I All Cri Ze-- 6-23... t@121 cm-2 All b-- CM (s tal.) of tbe otjlw.l 100-lb ~so~ W8 t01say cea s~t q~w, thoe e th,-ft0-x 07 %L- OW3 sz2127:P I-- W24-210s elth tbe *-a= cb.2. (taw. -20dts - am02.. All .d...1 *o1. are 1=1815 12 tue b,,,

t.1.1 .- LTO-lb awes sxe cat e lah I w ,. 11.7 Wm5 zw22 so2 te. se of r1Jtot. Camos-im@@ o-. Cs "s to XII.A.21&12.

2. 4;0*0~ o0.g~w12e.±.2~.17-3 -. :65 21-v l. cm- 7196

(It60 4-yo 12.65 jeers I nopc..) n. %=-s b*5,th. 270* llf-e. v b=7.1 -a woelC cm. 3 367-1967 (.i- 5P1 12j lS je

%b am.5 w. -2 ti. -- =o of te 1w0 Lw cwtv.aod. tke .0sz eba Wo -I- L. sf-e 5.12 jeers *2;0m.

3. =6. U.2b5r &om~£ odes vW2.1z dh~l2e -se 36 x 7 x I 12., *21 the 0061b &..)dts W 60 x.10 x 10 So.

0.,caet.i1 of -wae cV-re- do-.Ity: (us/sq ft)

V - Velg#4 1--Is 17 tUe &000 (D)8,;0, - Ymr of b~ s a2 jeer

7 * 1e4t of tene (T). ":'#eels .%$A for .oel s b.ae52h the 2W0 12em.21.t* ba~y,1.63 f=r .11 other .12I s., .o 5.,2 for the 002.13 awl.

A V.Ortlel mftce eree of the oew. (eq ft~l. W e*wl 7.5 for 12041b .001*. . 9.71 for the

5Cslc~alof .1e cvTem qT~lcl to ths . (Amp)

3 ~ ~ rozt - 5w0~T~2

t= !O rLlzl =-ilbTrwced y t. aodewt -T2-A "tt Io lythe w~wj) -Obs

F,6 * .r se~ldi yes.%

LOWER CATENARY LEG 00LEMC23-

MAWE BUOY (B)

CNTERME T BUOY (A)

UPPER CATENARY LEG NEM0AEBO C

1 - 3574 IFT -L--G 9Ft--..-IOFT- CENTER BUJOY

I INTERMEDIATE MAIMN - SURFFACE~E

I~ 3 -ANODE I - SPAN D. W IREE

KAPH ROPESTIN

j~ 3e fT. -t/eD.WIRE ROPE BOTTOM

CHAIN (1) 400 LB ANODE AT END OF ANCHOR CHAIN6000 LB LWT (2) 400 LB ANODE AT GROUND RING UNDER INTERMEDIATE BUOY

ANCHOR (3) SIX-60 LB ANODE STRING AT GROUND RING UNDER MAIN BUOY(4) SIX-60 LB ANODE STRING AT FLOUNDER PLATE UNDER CENTER BUOY(5) THREE-6O0LB ANODE STRING ATTACHED TO INTERMEDIATE BUOY

()17-19 NOV. 19651REPLACED By SIX- 120 LB MAG14ESIUM AN~ODES.400 LB AN.ODES WERE LOST DURiNG REPLACEMENT,

Fig. 1 - Plan view of the moor in the upper sketch and a typicalmooring leg elevation in the lower sketch (not to scale).

22

~---- ~ ~ -- ~ -

-'o

Ig

-. 0 0

c .0

E00

C4 0 0

0, 0I jj.~j~V;,..7CD 0-

..--- i: : -- K-~ iRILIdtoC

23

(L

0 a

0 -0

$4 0

'.a 0

00

0 . 0 0

Eqq

00

244F ________o

Ie

A PC

co 0

4 P4

25

-44

I~ 4.

* I 0

26O

~. __ _ _ _ _ _ _ _ _O

izr-

CD Cd

Wit~

rr Cd

0 I-

a~Po

I.~E)4 -

P4 A

ID'~o

0 0 00

WNWVO DY3ly "HI 3N MOS SVf~ XI SO30V3AY

Win) ~ ~ UNMHMS 31nI nI~

27~

1 71

,k 1,

4) 0

on 428

.94

10 I

01 >~

w& X

oj Co "

0a: 0 0

aV- 1-

w i C3o wa

z ~~ z 3

4 05, ) 60

I - k b

I I

0 04~4

I-az 0 00

Iz ~ x 4.) b~J~j.J

"wwz 3: U.

C.

0 0 0 02 OD D va

(33NVISI(I HOV3 iv a~~~~vIlS3 WU MI XSJO DUAV(SI ) S I 3n q 33~ nII

I29

-12 FAILED

IKt

-, -

- -I

222

Fi. - Fale enso iefo Ch 00 oe aeay

. a I- N

Fig. 8 - Failed ends of wire from the 0300 lower catenary.

The fracture appearance on individual wire s varied andwire diameters were not severely reduced.

30

0 0

xU

8 p =14o- #

ifcc

4- -- 0

I' u;MI'_ _jO _j

P.4.

CC

311

I

}. ,I

Fig. 10 - Cross section of the 270* leg intermediate buoy riserapproximately 50 ft above the failure. Note the general reductionin the diameter of the outer wires on all strands. White areas onthe circumference of some individual wires are residual galva-nizing (zinc). Etched with 1% Nital.

32

L i I

IM 0 IovIt P

1 4 '

-icdC

~~14;Ol P4 Pi

I3

.C 0

=1 0 4

to 0 0

V 344

_____ ~~~~~ ~ ~ ~ 4 __ _ __ _ __ _ _ ;4__ _ __ _ _ _ __ _

o-

0

35P

[~~~~~J CS__ ______ ____

E-4r

:1 C,;0

0 *d

o0 1-

a36 t

o____

(Page 38 is'Blank)

fo~ -

m 10

0 M

0 "

4.) C :

-< ci

UNCLASSIFIEDS"urit Classification

= -DOCUMENT CONTROL DATA - R & D,Secuwir tlasssfrat a .1 Zul. boar of abstract aid nd Acostn.. .x..t be t, e when she overall report is ¢taslied)

I. OIIINJAT ING ACTIVITY (COM 2ate AU&SCe) Ta. REPORT SECURITY CLASS.FICAT:o

Naval Research Laboratory UNCLASSIFIEDWasbington, D.C.20390 2. GROUP

3. XCPONT TITLE -

ANALYSIS OF THE FAILZURE OF THE AUTEC TOTO II DEEP SEA MOOR AND THEPERFORNICE, OF ITS CATHODIC PROTECTION SYSTEM

;. OESCRIPTIVE NOTES (7pc ot revot ad lnclu ivc dates)

This:completes one phase pf the program; work on other phases is continuing.S. AUTHORIS) (FiJrl nam.e jide initial, last name)

II. E. Groover

REPORT DATE 7a. OTAL HO. OF PAG. 1 NO. OF gEFSNoverber 1968 44

S.. CONTRACT UR GRANT 14O. 2. ORIGINATOR'S REPORT NUUSER9S)

b. PROJECT HO. NRL Memorandum Report 1950S-4307-11894

C. 9b. OTHER RCPORT ,NOtS) (Any other nw.Iber$ that way be assigned

* ~~~~d. _ _ _ _ _ _ _ _ _

10. DISTRISUTION STATEuENT

This document has been approved for public release and sale; its distribution is unlimited.

I1. SUPPLEMENTARY NOTES 112. SPONSORING ,AILITARY ACTIVITY

Deep Submergence Systems ProjectI 6900 Wisconsin Avenue

13 T ._r Chevy Chase, Maryland

.)k This report contains background information on the design and installation of a wirerope three-point deep sea moor %iich was originally coated with a bituminous substance andthe critical-areas protected with a magnesium galvanic anode cathodic protection systemdesigned by the Naval Research Laboratory. The report also describes the failure of the moorafter 4 1/2 years service and its subsequent salvage, and presents the results of a study of

the corrosion pattern, proposes the cause of failure, and evaluates the performance of the Icathodicprotection system. Recommendations are presented for the protection of future moorsand for possible research directed towards the understanding and prevention of failure of wirerope structures in sea water under conditions of static ktress and of fatigue. ( ¢.._

Three of the wire rope failures were associated with severe corrosion and the absence of

bituminous coating. In two of the failures the bituminous coating was essentially absent but no

severe corrosion was observed. There is some circumstantial evidence that the latter failuresmay have been caused by a ship mooring to the intermediate buoy which was not intended to beused for this purpose. All but- ond failure occurred 1350 to 2050 ft from the nearest activeanode connections. One failure occurred 680 ft from an intended anode connection, but thereis evidence that this particular anode was lost when the moor was installed. Some othermajor components of the moor probably received cathodic protection for as little as 9 to 12months out of the approximate 54-months life of the moor. This was a result of a variety ofproblems associated with the replacement of the original anodes. (Continued)

DD IFOVMJ 4 7 3 39 UNCLASSIFIEDS/N 0101807-6801 Security Classification

'UNCLASS FIEDSecuity Classification

14 I I A LINK 8 LINK CKEYT wORD$

OL E WT OL E W T O LE W TV

Deep sea moorAUTEC TOTO II moorNre ropeCorrosion

I,

The results of the corrosive attack and the failure of the moor indicate that where ,ilong life is demanded for future moors the following steps should be taken:

(a) lProvide a coating better than the bituminous substance used in the present moor.(b) Provide cathodic protection as a secondary defense against corrosion. It is

essential that cathodic protection be considered early in the design phase and providedthroughout the life of the moor if it is to be effective. Scheduled inspections and replace- ,ment of consumed anodes are essential for prolonged trouble-free life of the structure.Where adequate power is available, impressed current systems should be considered toobviate the necessity of frequent anode replacement.

DD, NOVss1473 (A 40 UNCLASSIFIED(PAGE- 2) Security Classification

if