{"operation":"document","citation":"PHMSA FIR, Enbridge Energy LP, 2007-01-01","title":"Failure Report-Enbridge Energy LP 1/1/07","source_type":"incident","agency":"Pipeline and Hazardous Materials Safety Administration","status":"historical","official":true,"published_on":"2013-02-11","effective_on":"2007-01-01","summary":"Enbridge Energy LP; HL; WI; failure 2007-01-01; apparent cause: Weld Seam Failure.","machine_formats":{"json":"https://regulus.evalyn.ai/document/phmsa-fir-failure-report-enbridge-energy-lp-1107-0.json","markdown":"https://regulus.evalyn.ai/document/phmsa-fir-failure-report-enbridge-energy-lp-1107-0.md"},"app_url":"https://regulus.evalyn.ai/document/phmsa-fir-failure-report-enbridge-energy-lp-1107-0","source_url":"https://www.phmsa.dot.gov/safety-reports/failure-report-enbridge-energy-lp-1107-0","body":"<<<PAGE 1>>>\n\nDOT US Department of Transportation\nPHMSA Pipeline and Hazardous Materials Safety Administration\nOPS Office of Pipeline Safety\nCentral Region\nPrincipal Investigator Karen Butler/Jim Bunn\nRegion Director David Barrett\nDate of Report 6/11/2012\nSubject Failure Investigation Report – Enbridge Line 14 Rupture Atwood\n(Owen), WI\nOperator, Location, & Consequences\nDate of Failure January 1, 2007\nCommodity Released Crude Oil\nCity/County & State Atwood/Clark County, Wisconsin\nOpID & Operator Name 11169 Enbridge Energy, Limited Partnership\nUnit # & Unit Name 1343 Fort Atkinson\nSMART Activity # 118537\nMilepost / Location MP 149.44\nType of Failure ERW Longitudinal Weld Seam Failure\nFatalities 0\nInjuries 0\nDescription of area Rural Area, Non-HCA\nimpacted\nTotal Costs $702,500\n\n<<<PAGE 2>>>\n\nFailure Investigation Report – Enbridge Line 14 Rupture Atwood (Owen), WI\nFailure Date 1/1/2007\nExecutive Summary\nOn 01/01/07 at approximately 8:49 am CST, a pipeline controller located in the Enbridge Control Center\n(ECC) detected a sudden drop in discharge pressure at Owen pump station on Line 14. The pipeline\ncontroller immediately initiated shutdown procedures for Line 14, closed station valves on either side of\nOwen and at Owen pump station, and dispatched emergency response personnel. A landowner\ncontacted the ECC to report an odor within 30 minutes of Line 14 shutdown. At approximately 12:15 pm\nCST, emergency response personnel discovered oil flowing in a ditch and a rupture in Line 14 was\nverified at MP 149.4373 near Atwood, WI (approximately one mile downstream of Owen pump station).\nCrude oil had continued to travel away from the rupture site through a natural ditch area close to a mile\nin length. An estimated 1,500 barrels of crude oil was released, 1,450 barrels were recovered, and 4,625\ncubic yards of contaminated soil were removed and taken to a disposal facility.\nThe failed line pipe was 24-inch OD, 0 .328 wall thickness, API 5L, Grade X-70, High-Frequency ERW,\ncoated with fusion bonded epoxy and manufactured by Stupp Pipe Corporation in 1998. The depth of\ncover at the failure location was 60 inches. The pipe had been hydrotested following construction. The\nMaximum Operating Pressure (MOP) was reported to be 1377 psig and the pressure at the point of\nfailure was calculated to be 1331 psig. The failed section of pipe was cut out and approximately 50 feet\nof new pipe was installed. The pipeline was returned to service on 01/03/07 with a 20 percent\nreduction in operating pressure.\nThere were no injuries, fatalities or evacuations associated with the failure. A fire was not present and\nthe failure did not occur in an HCA. The supply impact from the rupture was minimized due to a parallel\npipeline that remained in operation. The operator reported that total costs associated with the release\nwere $702,500.\nThe rupture was determined by metallurgical analysis to be the result of a lack of fusion defect located\nin the ERW longitudinal seam near the ID of the pipe and adjacent to the upstream girth weld. The\nfracture propagated across the adjacent upstream girth weld. This original defect grew to failure while\nin service (under cyclic loads) by a fatigue mechanism. The metallurgical report indicated that the\n“rupture initiated at the defect when the flaw size exceeded the critical flaw size for the material\nproperties, dimensions, and operating pressure.”\nSystem Details\nEnbridge Energy, Limited Partnership (Enbridge) at the time of this report operates approximately 5,470\nmiles of pipelines, associated pump stations and breakout tanks in the United States. Enbridge has over\n3,600 miles of right-of-way grouped under several units known as Lakehead, Ozark, CCPS (Spearhead),\nToledo and Patoka (breakout tank facility) with more in development. Several areas have pipelines in\ncommon corridors including the failure location. The majority of the system transports crude oil.\nEnbridge also operates several HVL pipelines in the US. At the time of the failure, the crude oil systems\noperated were 3,338 miles in length with 1,030 miles located in areas that could affect an HCA.\nLine 14 is part of the Lakehead system and transports crude oil from the Enbridge Superior, Wisconsin to\nMokena, Illinois. This portion of the Lakehead system was originally 461 miles in length and comprised\nof 24-inch diameter, X70 pipe with wall thickness ranging from 0.328 inch to 0.500 inch. A review of\nPHMSA data indicates that Lakehead had experienced 14 reportable failures prior to this rupture\nassociated with this unit, several due to equipment or material failure.\nPage 2 of 10\n\n<<<PAGE 3>>>\n\nFailure Investigation Report – Enbridge Line 14 Rupture Atwood (Owen), WI\nFailure Date 1/1/2007\nEvents Leading up to the Failure\nRadiographic technicians inspecting each girth weld during pipeline construction in 1998 identified\ndefects in a number of ERW longitudinal seams adjacent to the girth welds. The defects were identified\nas lack of fusion. Twenty lengths of pipe that contained such defects were removed from the pipeline\nfor further analysis. Eleven of the twenty lengths of pipe removed were determined to contain\nsignificant weld defects on the inside diameter of the pipe and were sent to a metallurgical consultant\nfor further evaluation. The consultant determined that seven of the lengths of pipe contained defects\nthat would be rejected per API 5L specifications applicable to the pipe at the time of manufacture. The\nevaluation performed by the third party metallurgical specialist of these 7 defects determined that three\nof the defects would survive at least 20 years, 1 was large enough to fail during a post construction\nhydrostatic test to 1818 psi or 95% of SMYS of the 0.328 inch wall thickness, and 3 would be expected to\ngrow to failure in some time shorter than 20 years.\nEnbridge initiated an investigation with third party consultants that encompassed: a metallurgical\nanalysis to determine the characteristics of the weld defects; a review of pipe mill production records to\ndetermine if the defects had been identified during pipe mill inspection processes; an assessment of the\npipe mill’s quality assurance/quality control process. This investigation determined that all of the\ndefects were lack of fusion type, were located on the inside diameter of the pipe, and had been\nidentified during the pipe mill inspection processes. The investigation also discovered that the total\ndefect length had not been determined, completely marked or removed properly. This resulted in some\njoints having only a portion of the weld defect removed from the line pipe at the mill. Line pipe lengths\nthat contained the remaining portions of the defect were sent to the pipeline right-of-way. These\nlengths of pipe were subsequently welded into the pipeline.\nThe Stupp Pipe Corporation manufactured the pipe used to construction Line 14 at the facility in Baton\nRouge, Louisiana. The high frequency electric resistance welded (HF ERW) pipe was manufactured in\naccordance with API 5L and was grade X-70. The pipe diameter was 24-inch and wall thicknesses ranged\nfrom 0.328 to 0.500 inch. The pipe was coated with fusion bonded epoxy at the pipe mill. The pipe mill\nfacilities had been upgraded in 1997. One of the pipe mill upgrades consisted of the addition of a zip\nwelder which was used to join coils together. The zip welder allowed the pipe mill to continue to run\nwithout stopping between coils.\nEnbridge’s construction specification required that 100% of the girth welds in the pipeline be examined\nusing radiographic techniques. In an effort to find all similar defects, Enbridge had all of the radiographs\nre-evaluated to determine if any additional lack of fusion defects remained in the pipeline near a girth\nweld. Enbridge determined that there was a high statistical probability that all of the lack of fusion\ndefects had been removed from the pipeline. The pipeline was hydrostatically tested on August 31,\n1998 and the test pressure at the failure location was 1818 psig. Enbridge reported that the hydrotest\nduration was eight hours in length.\nThe Lakehead Line 14 pipeline was placed in service in January of 1999. Just prior to the time of the\nrupture on Jan. 1,2007, pumps had been started at Sheldon and Owen pump stations located upstream\nof the rupture location and at Vesper pump station located downstream of rupture location. This\nresulted in a pressure increase at Owen station as expected. Enbridge reviewed pressure data and\nreported that evidence did not exist to support any overpressure event occurrence at or near Owen\npump station at the time of the rupture. The Enbridge internal accident investigation reported that the\npressure at the Owen station reached 1347 psig with an MOP at this location of 1374 psig. Enbridge\ncalculated that a pressure at the failure location would have been 1331 psig at the time of failure and\nthat the failure location would have an MOP of 1377 psig.\nPage 3 of 10\n\n<<<PAGE 4>>>\n\nFailure Investigation Report – Enbridge Line 14 Rupture Atwood (Owen), WI\nFailure Date 1/1/2007\nEmergency Response\nThe ECC detected a sharp pressure drop on the Owen station discharge at 08:49 am CST and\nimmediately initiated a shutdown of the pipeline. By 08:50 am CST, all pumps upstream of the Owen\nstation had been shut down. All remote control valves between Sheldon station and Owen station were\nclosed and all remote control valves between Owen station and Vesper station were closed shortly after\nthe shutdown. Response personnel were dispatched downstream of Owen station to search for the\nrelease. Within 30 minutes of the Line 14 shut down, a landowner who was in the vicinity of the failure\nsite called the Control Center to report a crude oil odor. The release site was located and confirmed\napproximately one mile downstream of Owen station by 12:15 pm CST. Enbridge reports that the length\nof the line that was isolated between valve locations was approximately 15 miles. The area around the\nrelease was reported to be made safe with 2.5 hours. Enbridge did not report this accident to the NRC\nuntil 15:22 EST (13:22 MST) on January 1, 2007. The initial NRC report listed an incorrect state (Indiana\nrather than Wisconsin) in the Incident Location section.\nSummary of Return-to-Service\nThe rupture site was excavated and the segment of pipe that failed was exposed. The crude oil was\ndrained from the pipe segment that contained the rupture and a fifty foot section of pipe was removed\nand replaced with pre-tested pipe. The fifty foot segment of pipe that was removed from the pipeline\nwas cut into two pieces and wrapped for protection during transport. This piece of pipe that contained\nthe rupture section was sent to a metallurgical facility for further evaluation. The other portion of the\npipeline removed from service was wrapped for storage and sent to the Enbridge yard at Griffith,\nIndiana. The portion of Line 14 impacted by the failure resumed operation on January 3, 2007 at a\nreduced operating pressure. The MOP at each pump station along the pipeline was limited to 80% of\nthe maximum discharge pressure experienced during the 15 day period immediately prior to the time of\nthe rupture. The MOP at the rupture site was limited to a maximum of 80% of the pressure at the time\nof the rupture or 1064 psig.\nPHMSA worked extensively with the operator and Enbridge agreed to complete a metallurgical failure\nanalysis; complete an in-line inspection of the entire Line 14 pipeline with an ultrasonic crack detection\ntool; complete a crack excavation and repair program; complete a fatigue testing program; and establish\nan in-line re-inspection interval based on the fatigue analysis.\nThe failure location metallurgical analysis report was finalized on March 13, 2007. The in-line inspection\nwas completed in 2007, additional metallurgical analysis of a coupon sample at MP 32.41 was\ncompleted in 2007, and the excavation and repair program was completed in 2008. The fatigue analysis\nwas completed in July of 2008 and a re-inspection interval of five years was established based on the\nsize of anomaly that could remain in the pipeline and an analysis of the pressure cycles that could occur\nbased on operating history and pipe material properties. The first in-line re-inspection was scheduled\nfor 2012. PHMSA approved the return of Line 14 to its original operating pressure in September of\n2008.\nInvestigation Details\nA PHMSA representative arrived at the site in the evening (21:00) on January 2, 2007 and witnessed the\nremoval of the failed section of pipe, the subsequent repair and the nondestructive testing of the girth\nwelds, the preparation for shipment of the failed section of pipe and the loading of the pipe to a truck\ntransport. Photographs of the failed section of pipe were taken while on site.\nPage 4 of 10\n\n<<<PAGE 5>>>\n\nFailure Investigation Report – Enbridge Line 14 Rupture Atwood (Owen), WI\nFailure Date 1/1/2007\nThis investigation involved a detailed review of metallurgical analysis on multiple sections of pipeline, a\nreview of the calculated or predicted failure pressures using CorLas software, ILI data with multiple\ntools, a fatigue assessment and analysis that involved historical pressure cycling review and dig\nverification accompanied by pipeline repair. The metallurgical analysis, fatigue analysis and ILI\ninformation will be discussed in detail.\nMETALLURGICAL ANALYSIS\nThe length of pipe that contained the rupture was sent to CC Technologies in Dublin, Ohio. CC\nTechnologies conducted a metallurgical failure analysis in accordance with the PHMSA Metallurgical\nLaboratory Failure Examination Protocol. The failure analysis included an optical examination, a\nmagnetic particle inspection, an ultrasonic inspection, a fractographic examination, a metallographic\nexamination, an examination of the fracture surface with a scanning electron microscope, a\ndetermination of the chemical composition of the pipe steel and a determination of mechanical\nproperties of the pipe steel. The following conclusions were reached at the completion of the failure\nanalysis:\n1) The fracture was at the 2 o’clock position.\n2) The rupture initiated at an ID surface breaking defect located at the ERW seam weld adjacent to\nthe upstream girth weld.\na. The lack of fusion defect penetrated the surface on the internal diameter of the pipe.\n3) There is evidence that the defect grew in service by a fatigue mechanism.\na. The fracture propagated across the adjacent upstream girth weld.\nb. The defect was adjacent to the upstream girth weld and the initial depth of the defect\nwas determined to be 30% of the pipe wall thickness.\nc. The defect grew in service by a cyclical fatigue mechanism to a depth of 66.7% of the\nwall thickness at which point the fracture initiated.\n4) The ERW weld seam exhibited low toughness near the fracture origin.\na. There is evidence of cleavage in the fast fracture zone between the lack of fusion (LOF)\nflaw and the OD pipe surface at the bond line, indicating low toughness of ERW seam.\n5) Away from the failure origin, the seam weld exhibited good mechanical properties and met API\nspecifications.\n6) The mechanical properties of the base metal met API 5L specifications that were in affect when\nthe pipe was produced.\n7) The chemical composition of the base metal met API 5L specifications that were in affect when\nthe pipe was produced.\nAs the result of ILI run which will be discussed in detail further in this report, a defect was found at MP\n32.41 and a second metallurgical analysis was performed on a coupon sample with similar pipe\ncharacteristics (X70, HF ERW, 0.328 w.t., etc.) located within this section of pipe. This metallurgical\nanalysis involved an optical examination, angle beam ultrasonic testing, fractographic examination,\nmetallographic examination, burst pressure analysis, examination with a scanning electron microscope\n(SEM) and chemical analysis. Analysis of the coupon resulted in the following conclusions:\nPage 5 of 10\n\n<<<PAGE 6>>>\n\nFailure Investigation Report – Enbridge Line 14 Rupture Atwood (Owen), WI\nFailure Date 1/1/2007\n1) The location and morphology of the defect was consistent with a hook crack.\n2) The ends of the weld contained a defect so the endpoints were not located. The defect was at\nleast 21.5 inches in length. The defect is longer than the angle beam UT data.\n3) The maximum depth of the defect measured from the fracture surface was 0.102 inches (31.1%\nof wall thickness) which is less than the depth category reported by the ILI tool but within the\nlimits of that reported by the UT.\n4) The UT identified two defects from the external surface but only one defect was identified\nvisually. There were some difficulties in detecting the defects from the internal surface.\n5) There was no evidence of in-service growth on the defect surface.\n6) The composition of the sample removed from the base metal met API 5L X70 pipe steel\ncomposition specifications at the time of manufacture.\nFATIGUE ANALYSIS\nThe next step in the failure investigation process was to conduct a fatigue assessment. The\nmethodology used to calculate fatigue crack growth is described in British Standard 7910, Guide to\nMethods for Assessing the Acceptability of Flaws in Metallic Structures, and American Petroleum\nInstitute Recommended Practice 579, Fitness For Service. Under this approach a flaw may be\nrepresented as a sharp-tipped crack which propagates in accordance with a formula that relates the\ncrack growth rate to the range of the stress intensity factor for the material that contains the flaw. This\nformula is known as the Paris law. The Paris law incorporates two constants, C and m, which depend on\nthe material and the applied conditions, including the environment and cyclic frequency. This equation\nrequires the understanding that the crack growth rate changes as the flaw grows. The two constants\nmay be established (determined experimentally) or may be taken from published data. The overall\nfatigue life is calculated by integrating the Paris law equation from the initial flaw size to the final critical\nflaw size. Specialized computer software is used to perform these calculations.\nThis method incorporates the following assumptions.\n1) Surface flaws are assumed to have a semi-elliptical shape.\n2) Fatigue crack growth is described by the Paris law equation, da/dN =C(Δ K)m\n.\n3) The stress intensity factor range, ΔK, is a function of the crack size and the cyclic stress range,\nand may be calculated in accordance with BS 7910 Annex M.\n4) A pressure spectrum can be converted into identifiable stress ranges using a rainflow cycle\ncounting technique (in accordance with ASTM-1049 and E-1049).\n5) A two-term Folias correction factor is incorporated to account for the bulging effect that occurs\nnear axial flaws in pipe.\n6) The effects of pipe ovality are accounted for in accordance with BS 7910 Annex D.\nEnbridge contracted with CC Technologies to complete the fatigue analysis. In order to determine the\nfatigue growth rate that occurred on Line 14, CC Technologies measured the size of the initial flaw and\nmeasured the size of the flaw just prior to failure. Next they determined the hoop stress spectrum that\nthe pipeline was subjected to by performing a rainflow analysis of the pipeline pressure history since the\npipeline began operation in January of 1999. A back-calculated value for C and m constants using the\nParis law was obtained to match the results of crack growth to failure in the eight years previous to\nfailure. Established values for C and m constants were also obtained based on full scale fatigue tests\nPage 6 of 10\n\n<<<PAGE 7>>>\n\n2) 3) Failure Investigation Report – Enbridge Line 14 Rupture Atwood (Owen), WI\nFailure Date 1/1/2007\nthat were conducted on other segments of pipe that were removed from Line 14. The Paris law\nconstants were then compared. The m value in both cases was calculated to be 3, which is the same\nvalue recommended in BS 7910. The C value for the actual failure location was determined to be\n9.8x10-10 with an ovality 0% and 7.0x10-10 with an ovality of 0.5%. The C value for the full scale fatigue\ntest was determined to be 5.6x10-10 with an ovality of 0% and 3.9x10-10 with an ovality of 0.4%. The\nrecommended value of C contained in BS 7910 is 8.6x10-10\n. Until this time Enbridge had been using a C\nvalue of 2.5x10-10 and an m value of 3 to estimate fatigue growth rates. Thus, previous estimates of\nfatigue growth rates were not conservative based on established literature.\nAs a result of the investigation, CC Technologies recommended that Enbridge modify their fatigue\nanalysis procedures as follows.\n1) The Paris law constants should be in accordance with BS 7910 unless line-specific information\nwould justify the use of other values;\nThe procedure to consider ovality or weld joint misalignment should be in accordance with BS\n7910 or constants for the Paris law should be used that inherently include these effects.\nThe initial flaw size used in a fatigue analysis should be based on the likely size of remaining\nanomalies based on an in-line inspection.\nIN-LINE INSPECTION\nEnbridge contracted with GE Oil & Gas to run both a magnetic flux leakage (MFL) metal loss in-line\ninspection tool and an ultrasonic crack detection (USCD) tool through the entire length of Line 14. Both\ntools were run in the month of April 2007. The final MFL report was received in September of 2007 and\nthe final USCD report was received in July of 2007.\nThe MFL (GE Magnescan) tool did not identify any features that met the Enbridge repair criteria. Four\nvalidation digs were determined to have field assessments with two dig locations being completed in\nSeptember and December of 2007. The validation digs did confirm the presence of metal loss but the\nanomalies did not meet repair criteria.\nThe USCD tool identified 128 crack like features, 932 notch like features, 3 metal loss features and 84\ndents. Point pressure restrictions were imposed on the pipeline at each location where the anomaly\nhad a calculated failure pressure below the hydrostatic test pressure in order to achieve an immediate\nsafety factor of 1.25. Anomalies were excavated (97 field excavations required), evaluated and repaired\nas required to maintain a calculated failure pressure above 1.25 x MOP. As of February 1, 2008 there\nwere no known anomalies left in the pipeline with a calculated failure pressure below 1.25 x MOP.\nEnbridge did compare in-line inspection tool data with defect measurements made in the field with\nhand held ultrasonic inspection equipment. Based on the data comparison the tool performed in\naccordance with the tool specifications and showed a conservative bias. Enbridge did recognize that\nthere was a possibility of a defect being undersized or undetected. PHMSA believes that this possibility\nmust be taken into account when making decisions concerning the fitness for purpose of the pipeline.\nFindings and Contributing Factors\nEnbridge and their technical partners used the metallurgical data, a strength analysis to review\ncalculated failure pressures (using CorLAS software), the fatigue growth data and the in-line inspection\ntool data to evaluate the fitness for purpose of the pipeline and to establish a re-inspection interval for\nthe in-line inspection tool.\nPage 7 of 10\n\n<<<PAGE 8>>>\n\nFailure Investigation Report – Enbridge Line 14 Rupture Atwood (Owen), WI\nFailure Date 1/1/2007\nEnbridge did not report the accident to the NRC promptly.\nThe NRC report did not identify the incident location in the correct state.\nThe Operator Accident Report (30 day report) did not show the time of the accident as the time that the\nrupture occurred according to SCADA data and other sources internal to Enbridge (8:49 CST). The\nOperator Accident Report did show the time of the accident based on their confirmation of the release\nand the time at which the release reached reportable thresholds (11:15 MST).\nPHMSA provided oversight of the investigation and evaluated the future operation of Line 14 based on a\nthorough review of the data provided during the investigation. The following conclusions and findings\nwere reached as a result of a comprehensive investigation that was undertaken by Enbridge, the\ntechnical support partners and PHMSA review:\n1) During the Line 14 pipe manufacturing process, lack of fusion type defects were introduced into\nthe ERW weld seam in multiple joints of pipe. These defects were identified as cause for\nrejection during the pipe inspection process. The extent of the defects was not determined\naccurately so the defects were not completely removed from multiple pipe joints and line pipe\ncontaining defects made it to the pipeline right-of-way.\n2) A change in the pipe manufacturing process was implemented just prior to the production of\nthe pipe for Line 14.\n3) Pipe joints that contained this lack of fusion type defect were sent to the pipeline right-of- way\nand installed in Line 14.\n4) Some of these defects were identified during construction of the pipeline by nondestructive\ninspection technicians who were examining each pipeline girth weld. When the technicians\nwere examining radiographs of the girth welds, lack of fusion in some ERW longitudinal seams\nadjacent to girth welds was identified. The extent of the defect could not be determined.\n5) The nondestructive inspection technicians made Enbridge personnel aware of their discovery\nand an investigation by Enbridge and an independent metallurgical consultant in cooperation\nwith others ensued.\n6) At the time of construction, twenty pipe segments were selected for removal from the pipeline\nas the girth weld radiographs showed that the ERW weld seam adjacent to the girth weld\ncontained a lack of fusion defect. Eleven of these pipe segments were found to have significant\nERW seam defects. After further evaluation it was determined that seven of these pipe\nsegments contained lack of fusion type defects that would be cause for rejection in accordance\nwith API 5L.\n7) 100% of the girth welds in Line 14 had been subjected to radiographic inspection. The\nradiographs of all of these girth welds were re-evaluated in an effort to determine if any\nadditional lack of fusion defects were present in the pipeline. No evidence of additional defect\nindications was discovered.\n8) At the time of construction, the evaluation performed by the third party metallurgical specialist\nof the 7 defects found in the longitudinal seam for several joints of ERW pipe determined that 3\nof the defects would survive at least 20 years, 1 was large enough to fail during a post\nconstruction hydrostatic test to 1818 psi or 95% of SMYS of the 0.328 inch wall thickness, and 3\nwould be expected to grow to failure in some time shorter than 20 years.\nPage 8 of 10\n\n<<<PAGE 9>>>\n\nFailure Investigation Report – Enbridge Line 14 Rupture Atwood (Owen), WI\nFailure Date 1/1/2007\n9) The joint investigation performed during the time of construction concluded that the number of\npipe joints that contained the lack of fusion type of defect was very limited and there was a high\nstatistical probability that all such defects had been removed from the pipeline prior to\ncommissioning.\n10) The origin of the January 1, 2007 failure was a lack of fusion type defect that remained in the\npipeline. During the metallurgical investigation the initial length of this defect was determined\nto be 3 ½ inches and the initial depth of this defect was 30% of the pipe wall thickness or 0.098\ninches.\n11) The metallurgical investigation also determined that this defect grew to a critical size by a\nfatigue growth mechanism related to normal operating pressure cycles. The defect was 3 -1/2\ninches long and 60% of the pipe wall or 0.197 inches in depth when it reached critical size.\n12) Chemical and mechanical properties of the pipe joint met specification requirements except in\nthe area of the fracture origin where the fracture toughness values were determined to be low.\n13) A fatigue study was used to establish the proper coefficients to be used in the Paris law\nequation. The default coefficients that Enbridge used prior to this failure were a C value of 2.5 x\n10-10 and an m value of 3. The fatigue study found that the m value that Enbridge had been\nusing was correct but that the C value used would result in an overestimation of fatigue life by a\nfactor in excess of 3. The study recommended that a C value of 8.6 x 10-10 and an m value of 3\nbe used in future fatigue growth analyses.\n14) The ultrasonic crack detection tool identified 1147 anomalies that had a calculated failure\npressure of less than 1.25 x MOP at the location of the defect. All of these anomalies were\nexcavated, evaluated and repaired as required. Seam defects were found at various locations\nwithin pipe joints, not just on the ends of pipe joints as originally indicated during the\ninvestigation that was performed while construction activities were ongoing.\n15) The operator, Enbridge, reported to PHMSA that all anomalies identified in the in-line inspection\nprogram that had a calculated failure pressure of less than 1.25 x MOP have been removed from\nthe pipeline.\n16) A fatigue life was calculated for the known anomalies that were left in the pipeline. The fatigue\nlife of those anomalies ranged from 10.8 years to 93.3 years.\n17) Enbridge recognizes that there is a finite possibility of a defect being undersized by an in-line\ninspection tool or not being detected by an in-line inspection tool. Enbridge recognizes that a\ndefect could be smaller than the ability of the ILI tools to detect and Enbridge recognized that\ntool results must be validated to confirm tool accuracy. In the case of Line 14, tool performance\nwas validated.\n18) Enbridge recognizes that there are additional sources of uncertainty that must be considered in\nfitness for purpose studies. As a result, the lowest calculated fatigue life in a system is divided\nby 2 to determine the re-inspection interval.\n19) The lowest calculated fatigue life on Line 14 was 10.8 years so Enbridge determined the re-\ninspection interval for Line 14 to be 5 years. The next in-line inspection for Line 14 is scheduled\nfor 2012.\nPage 9 of 10\n\n<<<PAGE 10>>>\n\nFailure Investigation Report – Enbridge Line 14 Rupture Atwood (Owen), WI\nFailure Date 1/1/2007\nThis failure investigation identifies: the need for operators to be aware of the pipe manufacturer’s\nprocesses and any recent changes that may be implemented just before a pipe order is filled; the\nimportance of a thorough, repeatable, and understood quality assurance and quality control (QA/QC)\nprocess at the pipe mill; the value of 100% radiographs for girth welds and a thorough construction\ninspection process; the requirement that operators use properly established constants in the\ncalculations associated with the use of the Paris law for fatigue life estimates and include ovality\nconsiderations; and the need for operators to continually during the life of a pipeline asset follow up on\nthose risk elements identified during construction activities.\nAppendices\nA Maps and Photographs\nB NRC Report\nC Operator Accident Report\nD Metallurgical Analysis\nPage 10 of 10\n\n<<<PAGE 11>>>\n\nAppendix A Maps and Photographs\n-\n-\nS-'J>\n·rt\n--\n--.... ~~--...ood -\n-. --\n. ....\nPage 1 of 6\n\n<<<PAGE 12>>>\n\nAppendix A Maps and Photographs\nPage 2 of 6\n\n<<<PAGE 13>>>\n\nAppendix A Maps and Photographs\nPage 3 of 6\n\n<<<PAGE 14>>>\n\nAppendix A Maps and Photographs\nPage 4 of 6\n\n<<<PAGE 15>>>\n\nAppendix A Maps and Photographs\nPage 5 of 6\n\n<<<PAGE 16>>>\n\nAppendix A Maps and Photographs\nPage 6 of 6\n\n<<<PAGE 17>>>\n\nAppendix B NRC Report 822512\nNATIONAL RESPONSE CENTER 1-800-424-8802\n*** For Public Use ***\nInformation released to a third party shall comply with any\napplicable federal and/or state Freedom of Information and Privacy Laws\nIncident Report # 822512\nINCIDENT DESCRIPTION\n*Report taken at 15:22 on 01-JAN-07\nIncident Type: PIPELINE\nIncident Cause: EQUIPMENT FAILURE\nAffected Area:\nThe incident was discovered on 01-JAN-07 at 12:00 local time.\nAffected Medium: LAND ONTO THE GROUND\n____________________________________________________________________________\nSUSPECTED RESPONSIBLE PARTY\nOrganization: ENBRIDGE ENERGY\nGRIFFITH, IN 46319\nType of Organization: PRIVATE ENTERPRISE\n____________________________________________________________________________\nINCIDENT LOCATION\nMAIN LINE County: CLARK\nState: IN\nSection: 27 NW QUARTER Township: 28 NORTH Range: 1 WEST NEAR OWEN, WI\n____________________________________________________________________________\nRELEASED MATERIAL(S)\nCHRIS Code: OIL Official Material Name: OIL: CRUDE\nAlso Known As:\nQty Released: 25 BARREL(S)\n____________________________________________________________________________\nDESCRIPTION OF INCIDENT\nCALLER STATED THERE WAS A RELEASE OF MATERIALS FROM A 24\" STEEL PIPELINE DUE TO\nEQUIPMENT FAILURE.\n____________________________________________________________________________\nINCIDENT DETAILS\nPipeline Type: TRANSMISSION\nDOT Regulated: YES\nPipeline Above/Below Ground: BELOW\nExposed or Under Water: NO\nPipeline Covered: UNKNOWN\n____________________________________________________________________________\nDAMAGES\nFire Involved: NO Fire Extinguished: UNKNOWN\nINJURIES: NO Hospitalized: Empl/Crew: Passenger:\nFATALITIES: NO Empl/Crew: Passenger: Occupant:\nEVACUATIONS: NO Who Evacuated: Radius/Area:\nDamages: NO\nClosure Type\nAir: N\nLength of Direction of\nDescription of Closure Closure Closure\nMajor\nRoad: N Artery: N\nWaterway: N\nTrack: N\nPassengers Transferred: NO\nhttp://www.nrc.uscg.mil/reports/rwservlet?standard_web+inc_seq=822512[6/7/2012 6:07:33 PM]\nPage 1 of 2\n\n<<<PAGE 18>>>\n\nAppendix B NRC Report 822512\nEnvironmental Impact: UNKNOWN\nMedia Interest: NONE Community Impact due to Material: NO\n____________________________________________________________________________\nREMEDIAL ACTIONS\nCALLER STATED A CREW IS ON SITE WITH A VACUUM TRUCK. CALLER STATED BOOMS ARE IN\nPLACE AS WELL AND LINE HAS BEEN SHUT DOWN.\nRelease Secured: YES\nRelease Rate:\nEstimated Release Duration:\n____________________________________________________________________________\nWEATHER\nWeather: OVERCAST, 35ºF\n____________________________________________________________________________\nADDITIONAL AGENCIES NOTIFIED\nFederal: NONE\nState/Local: NONE\nState/Local On Scene:\nState Agency Number:\n____________________________________________________________________________\nNOTIFICATIONS BY NRC\nATLANTIC STRIKE TEAM (PRIMARY)\n01-JAN-07 15:32\nCG INVESTIGATIVE SVC CHICAGO (CGIS ROA CHICAGO)\n01-JAN-07 15:32\nDOT CRISIS MANAGEMENT CENTER (PRIMARY)\n01-JAN-07 15:32\nU.S. EPA V (PRIMARY)\n01-JAN-07 15:36\nU.S. EPA V (KEVIN TURNER)\n01-JAN-07 15:32\nNATIONAL INFRASTRUCTURE COORD CTR (PRIMARY)\n01-JAN-07 15:32\nNOAA RPTS FOR IN (PRIMARY)\n01-JAN-07 15:32\nORSANCO ATTN: J. SCHULTE (PRIMARY)\n01-JAN-07 15:32\nPIPELINE & HAZMAT SAFETY ADMIN (OFFICE OF PIPELINE SAFETY (AUTO))\n01-JAN-07 15:32\nIN DEPT ENV MNGMT ATTN: BEAUCHAMP (PRIMARY)\n01-JAN-07 15:32\nKY DEP/ERT ATTN: MR. DAVID LEO (PRIMARY)\n01-JAN-07 15:32\nSURFACE TRANS SECURITY INSPECT PROG (COMMAND CENTER)\n01-JAN-07 15:32\n____________________________________________________________________________\nADDITIONAL INFORMATION\nCALLER STATED STATE AGENCIES WILL BE CALLED NEXT (WI EPA).\n___________________________________________________________________________\n*** END INCIDENT REPORT # 822512 ***\nThe National Response Center is strictly an initial report taking agency and\ndoes not participate in the investigation or incident response. The NRC\nreceives initial reporting information only and notifies Federal and State\nOn-Scene Coordinators for response. The NRC does not verify nor does it take\nfollow-on incident information. Verification of data and incident response\nis the sole responsibility of Federal/State On-Scene Coordinators. Data\ncontained within the FOIA Web Database is initial information only. All\nreports provided via this server are for informational purposes only. Data\nto be used in legal proceedings must be obtained via written correspondence\nfrom the NRC.\nhttp://www.nrc.uscg.mil/reports/rwservlet?standard_web+inc_seq=822512[6/7/2012 6:07:33 PM]\nPage 2 of 2\n\n<<<PAGE 19>>>\n\nAppendix C Operator Accident Report\nNOTICE: This report is required by 49 CFR Part 195. Failure to report can result in a civil penalty not to exceed $25,000 for each violation Form Approved\nfor each day that such violation persists except that the maximum civil penalty shall not exceed $500,000 as provided in 49 USC 60122 OMB No. 2137-0047\nU.S. Department of Transportation\nResearch and Special Programs\nAdministration\nACCIDENT REPORT – HAZARDOUS LIQUID\nPIPELINE SYSTEMS\nReport Date\nNo.\n(DOT Use Only)\nINSTRUCTIONS\nImportant: Please read the separate instructions for completing this form before you begin. They clarify the\ninformation requested and provide specific examples. If you do not have a copy of the instructions,\nyou can obtain one from the Office Of Pipeline Safety Web Page at http://ops.dot.gov.\nPART A – GENERAL REPORT INFORMATION\nOriginal Report Supplemental Report Final Report\n1. a. Operator's OPS 5-digit Identification Number (if known) / ____________/\n2. b. If Operator does not own the pipeline, enter Owner’s OPS 5-digit Identification Number (if known) / ___ /\nc. Name of Operator ______________________________________________________________________________________\nd. Operator street address _______________________________________________________________________________\ne. Operator address ______________________________________________________________________________________\nCity, County, State and Zip Code\nIMPORTANT: IF THE SPILL IS SMALL, THAT IS, THE AMOUNT IS AT LEAST 5 GALLONS BUT IS LESS THAN 5 BARRELS,\nCOMPLETE THIS PAGE ONLY, UNLESS THE SPILL IS TO WATER AS DESCRIBED IN 49 CFR §195.52(A)(4) OR IS OTHERWISE\nREPORTABLE UNDER §195.50 AS REVISED IN CY 2001.\n2. Time and date of the accident\n/ / / / / / / /\nhr. month day year\n3. Location of accident\n(If offshore, do not complete a through d. See Part C.1)\na. Latitude: _____ Longitude: __________\n(if not available, see instructions for how to provide specific location)\nb. _________________________________________________\nCity, and County or Parish\nc. _________________________________________________\nState and Zip Code\nd. Mile post/valve station or survey station no.\n(whichever gives more accurate location)\n_________________________________\n4. Telephone report\n/ / / / / / / /\nNRC Report Number month day year\n5. Losses (Estimated)\nPublic/Community Losses reimbursed by operator:\nPublic/private property damage $_______________\nCost of emergency response phase $_______________\nCost of environmental remediation $_______________\nOther Costs $_______________\n(describe) _____________________________________\nOperator Losses:\nValue of product lost $_______________\nValue of operator property damage $_______________\nOther Costs $_______________\n(describe) _____________________________________\nTotal Costs $_______________\n6. Commodity Spilled Yes No\n(If Yes, complete Parts a through c where applicable)\na. Name of commodity spilled ___________________________\nb. Classification of commodity spilled:\nHVLs /other flammable or toxic fluid which is a gas at ambient conditions\nCO2 or other non-flammable, non-toxic fluid which is a gas at ambient conditions\nGasoline, diesel, fuel oil or other petroleum product which is a liquid at ambient conditions\nCrude oil\nc. Estimated amount of commodity\ninvolved :\nBarrels\nGallons (check only if spill is\nless than one barrel)\nAmounts:\nSpilled : ____________\nRecovered: ____________\nCAUSES FOR SMALL SPILLS ONLY (5 gallons to under 5 barrels) : (For large spills [5 barrels or greater] see Part H)\nCorrosion Natural Forces Excavation Damage Other Outside Force Damage\nMaterial and/or Weld Failures Equipment Incorrect Operation Other\nPART B – PREPARER AND AUTHORIZED SIGNATURE\n(type or print) Preparer's Name and Title Area Code and Telephone ","truncated":true,"body_characters":49502}