Jaime Buitrago

Fatigue Design and Performance Verification of Deepwater Risers

With the advent of the development of deepwater projects, ExxonMobil developed and successfully implemented a fatigue design and verification protocol for fracture-critical components, such as risers and tendons, to ensure design performance and reliability. This protocol has now become an industry practice. This paper discusses the analytical, fabrication, and testing aspects of the design process. The linkage among actual weld performance, welding procedures and inspection reliability is addressed. From the design implementation standpoint, reliability of the fabrication inspection is the key issue. Practical methodologies were developed to conduct and interpret the fatigue tests. In particular, specimen design, instrumentation, testing protocols, and postmortem examination are discussed. Data generated by testing 56 full-scale risers of various sizes and welded by different procedures are also presented. These data, including tests past 100 million cycles, show that (1) actual riser fatigue performance can be substantially better than that recommended by codes, (2) failures can occur in the long-life regime, and (3) fatigue performance varies with riser size and thickness. However, as a matter of practice, analyses, fabrication and testing are required for particular designs.Copyright

Effect of Loading Frequency on Fatigue Performance of Risers in Sour Environment

ExxonMobil requires experimental verification of fatigue performance of fracture-critical risers designed for sour environments. Interaction between the sour environment and cracks in welded risers affects the crack-growth rate and, thus, the fatigue performance of the risers. Therefore, when conducting tests on riser welds in a sour environment, the frequency at which cyclicloads are applied during testing is critical to properly capturing the physio-chemical reactions and diffusion processes at the crack tip. Unfortunately, the load frequencies required to properly capture these effects are much lower than those currently used in cost-effective, resonant fatigue testing in air. Depending on the material, sour environment composition, and loading regime, testing at too high a frequency can eliminate the potential deleterious effects of the environment acting on the riser. Yet, testing at too low a frequency may not be practical. In order to determine the most efficient but technically valid load frequency to be used in a fatigue qualification testing program, a novel experimental screening methodology has been devised and implemented. In this paper, the proposed methodology is discussed and the results of a pilot test program conducted with C-Mn steel in a mildly sour environment are presented. For the particular sour brine, C-Mn steel and loading regime, it was found that the loading frequency could be increased up to about 1Hz, thereby making the fatigue verification tests more practical and cost-effective than the 1/3Hz currently used.Copyright

High-Cycle and Low-Cycle Fatigue Resistance of Girth Welds in Sour Service

The fatigue performance of fracture-critical production lines, such as risers and flowlines, has been shown to significantly degrade in the presence of sour hydrocarbon production caused by water injection of reservoirs. To ensure the reliability of the fatigue design under such conditions, experimental verification of the degradation effect on fatigue life due the presence of H2 S is required. To that end and over the past several years, ExxonMobil has developed new testing methodologies to evaluate the riser fatigue performance for both in-air and sour conditions. This paper reviews the general elements of the fatigue qualification process and presents new sour fatigue data aimed at assessing performance at the high-cycle fatigue (HCF) and low-cycle fatigue (LCF) regimes. These new data are relevant to that seen in steel catenary riser (SCR) and flowline thermal responses, respectively. Testing methodologies for each regime are discussed and results presented. The new data are interpreted within the context of previous data in the intermediate-cycle fatigue (ICF) to provide a more robust basis for riser design. The main finding is that the new data support a constant slope S-N curve for the practical domain of fatigue lives to which offshore lines are typically designed under sour conditions.Copyright

Fatigue Assessment of Subsea Tree Connectors and Wellheads

Drilling and workover operations offshore impose fatigue loads on the subsea tree and wellhead due to the riser dynamics. In particular, the connector between the tree composite valve block and the wellhead components may be subjected to fatigue damage. This is primarily due to specific mechanical connector designs, girth welds and thickness or section transitions commonly used in trees and wellheads. The fatigue assessment methodology of subsea tree connectors and wellheads requires a holistic approach involving sequential analyses of the floating vessel, riser, subsea well stack-up, and the wellhead and downhole casing assemblies including cement and soils. In this paper, a methodology for the analysis of each of those assemblies and their interactions is discussed. The methodology reflects ExxonMobil’s experience and offers a balance among analysis efficiency compatible with project and operational constraints, availability and uncertainty of the inputs, and accuracy of the outputs. In general, the methodology is based on dynamic analysis of models to obtain global response loads of the floating unit, riser, and wellhead. Finite element analyses are then used to translate global responses to local stresses and to define SCFs for structural details of interest. The conventional cumulative damage approach is adopted, which provides fatigue lives based on local stresses and fatigue resistance defined by S-N curves. The fatigue resistance of machined components is based on a strain-life model that provides material specific S-N curves.Copyright

The Effect of Reeling on the Fatigue of Welded Risers

ExxonMobil Upstream Research Company (URC) conducted a pilot testing program to systematically assess the effect of reeling on the fatigue performance of welded risers. Given the criticality of deepwater risers, a robust reeling simulation method was developed. The method properly addresses the fundamental weld sampling issue arising from the randomness of the plane in which risers are reeled. To address that issue the nominal reeling strain is accumulated around the entire circumference of girth weld, thereby capturing the weak fatigue bending plane. The reeling simulation is implemented by using multi-plane bending. The number and orientation of bending planes needed to simulate a total cumulative reeling strain are based on (1) the strain experienced during the process being simulated (2) weld fabrication and inspection records and (3) residual stresses left after reeling calculated via finite element analyses. Thirty production-quality welds were made on 323mm OD × 20.5-mm WT (12.75-in OD × 0.81-in WT) seamless X60 pipe, yielding a total of 14 full-scale specimens. Two different welding procedures and two different cumulative nominal strain levels were used. The effect of reeling on fatigue was assessed by comparing the fatigue lives obtained with companion reeled and non-reeled full-scale specimens. Results indicate that, although there was a distinct deleterious effect of reeling on fatigue life, the performance may still be adequate for most riser applications when high quality welds are used and qualified.© 2010 ASME

Effect of Reeling on Small Umbilical Tubing Fatigue

Umbilicals use steel tubing of different sizes to carry various pressurized fluids. During manufacturing, transportation and installation of the umbilicals, the tubing is subjected to reeling and unreeling operations, resulting in a cumulative amount of plastic strain. The amount of strain varies with different cross-section designs and umbilical manufacturers. This plastic strain may limit the fatigue performance of the tubing in dynamic umbilicals, thereby limiting the number of reeling operations of umbilicals in deepwater. A previous paper presented a novel experimental methodology to simulate reeling and its effect on the fatigue of 57.6-mm ID × 3.4-mm WT made of super duplex steel. Results indicated that reeling can significantly degrade the fatigue performance of the welded tubing as the cumulative reeling strain increases up to 20%. This paper discusses an experimental program aimed at assessing the size effect on fatigue of reeled tubing. Ten 10 additional fatigue tests were conducted with 12.6-mm ID × 1.46-mm WT tubing reeled to 20% strain. Results indicate that the smaller tubing, once reeled to 20%, its fatigue performance degrades to the same level as that of the larger tubing. However, the combined fatigue data do not support current design criteria DnV-RP-C203. Therefore, given the uncertainty of all of the variables involved, fatigue qualification for specific applications is considered necessary.Copyright

A Reliability Assessment of Collapse Safety Factors for Tendon Design

Since its first edition in 1987, the tendon system design section of The Recommended Practice for Planning, Designing, and Constructing of Tension Leg Platforms, API RP 2T, has not been materially revised. An entirely new draft of that section, including both strength and fatigue recommended practices, has now been completed. One key aspect for future designs is the maximum depth in which tendons can be installed while still maintaining adequate and consistent reliability. The new strength guidance recommends the axial tension-hydrostatic collapse interaction equation currently used by API RP 2A LRFD, coupled with a working stress design format with explicit tension and collapse safety factors. The latter factor controls the water depth applicability. To establish a basis for a hydrostatic collapse safety factor, ExxonMobil performed an independent reliability study to arrive at a safety factor consistent with the new interaction equation accuracy and current fabrication and design tendon practices. This paper presents the data gathering, analysis procedure, and interpretation of results leading to a safety factor that is consistent with standard practices. The analyses were conducted via Monte Carlo simulations for a number of design cases across a range of safety factors. Results indicate that a collapse safety factor of 1.67 for the Category A operational load case that controls tendon design in deep water yields levels of reliability consistent with those historically sought by design codes such as AISC and API.© 2006 ASME

Fatigue, Creep and Electrical Performance of Subsea Power Cable

Subsea processing is a key aspect of ExxonMobil’s subsea technology qualification efforts. Consequently, the reliable delivery of high voltage power to subsea equipment offshore, such as multi-phase pumps, is essential. The base case involves power delivery up to 30 kV via copper cables embedded in umbilicals to a water depth of 1,500m over a design life of 30 years. Since umbilicals are often suspended from a floating facility in deepwater, the power cables are subjected to static load due to the umbilical weight and to cyclic loads in response to the floating facility motions. Therefore, sufficient mechanical resistance to preclude electrical failure is needed to achieve safe and cost-effective designs.To that end, an experimental program was completed aimed at developing a qualification methodology and collecting data on the effect of static and cyclic loads on the mechanical and electrical performance of copper power cables. This paper presents the experimental approaches and results obtained from creep, fatigue, and electrical tests performed on 18/30kV, 95mm2 copper conductor cable specimens.Specifically, the program included (a) five monotonic tension tests to failure to characterize the cable, (b) 11 creep tests to develop a load-time performance curve, and (c) 59 fatigue tests under pulsating tension, of which 50 used short-cable samples to evaluate the effect of mean load and nine used long samples for electrical testing following fatigue. A fatigue curve is proposed that includes the effect of mean load and temperature. However, fatigue results can be sensitive to the cable material and mechanical characteristics of a particular cable.Copyright

Cryogenic Structural Performance of Corrugated Pipe

One alternative to developing offshore gas reserves is to use a floating LNG plant (FLNG) on site and export the LNG using tankers. This alternative requires the use of a reliable LNG transfer system between the FLNG and the tanker under offshore conditions. One such system involves a cryogenic hose, whose main body is a vacuum insulated, pipe-in-pipe hose made of corrugated stainless steel pipe (c-pipe) and flanged terminations. Given the novelty of the transfer system, ExxonMobil conducted an experimental program to understand the structural performance of the basic c-pipe under static and cyclic loading at room and cryogenic temperatures. This paper discusses overall qualification issues and presents the experimental methodology and results of structural performance tests of the full-scale c-pipe at both ambient and cryogenic temperatures. Fourteen full-scale, c-pipe static tests are reported, including tension, compression, bending, torsion, and internal pressure. In addition, 11 axial and three pressure fatigue tests are presented. One key result is that, overall, cryogenic temperature improves structural performance for the limit states tested, indicating that future qualification at room temperature would be sufficient. Moreover, the fatigue performance at both ambient and cryogenic temperatures surpassed the design curve reported in the literature for c-pipe.Copyright

Verification of Fracture and Fatigue Performance of Titanium Gr. 29 Welds in Tapered Stress Joints

Design of a Steel Catenary Riser (SCR) requires the use of connection hardware to accommodate the bending moment that arises from the abrupt change in stiffness at the floater hang-off. Reliability of this connection hardware is of paramount importance in ultra deepwater applications (up to 3000m), especially those involving high pressure and temperature fluids. One type of such connection hardware is a metallic tapered stress joint. Because of its inherent density, strength, and stiffness, steel is not well suited for these applications due to the length and weight constraints. Titanium Gr. 29 (Ti), which is as strong as steel but lighter and more flexible, has been identified as a good material candidate for a tapered stress joint.The required length (∼40ft) and thickness (∼3.5in.) of the Titanium Stress Joint (TSJ) cannot be fabricated as a single piece due to forging size limitations. Thus, an intermediate girth weld becomes necessary. The fracture and fatigue performances in the presence of the external seawater and cathodic protection (CP) and the internal sour production with galvanic effects between the Ti and steel must be assessed to verify the service life of the stress joint. ExxonMobil has developed and initiated a Joint Industry Project to fully address the fracture and fatigue qualification of titanium welds. In particular, the project plans to establish a robust methodology for the future qualification of TSJs that parallels, to the extent possible, the qualification process currently used for SCRs.This paper discusses the primary aspects of the titanium weld qualification: (1) selection of test specimens, (2) load frequency effects on initiation and propagation lives, (3) environmental assisted cracking due to hydride formation under cathodic and galvanic conditions, (4) full-thickness small-scale fatigue, (5) size effect on fatigue, and (6) weld inspection.Copyright