Summary
The life-prediction technique of a whole-life fatigue model was applied to drillstring components. Good agreement existed between the life predictions and the experimental data of drill collars. These life predictions were then used to study the effects of material properties and commonly used thread profiles on the fatigue life of drillstring components. Results of the study indicate that the newly developed microalloying technology and thermomechanical-processing technology in steel metallurgy can significantly benefit the fatigue life of drillstring components. The sensitivity studies on the effect of thread profile on fatigue performance of drillstring components indicate that fatigue life could be enhanced by an improved thread-profile design.
Introduction
Statistics indicate that about 50 % of drillstring failures result from fatigue. Most drillstring failures occur on drill-collar and drillpipe connectors. Although periodic nondestructive inspections reduce the frequency of downhole failures, they do not improve the fatigue performance of drillstring components. Two major factors determine fatigue performance: material properties, and stress states encountered at critical locations during drilling. Because drilling loads can hardly be reduced, the alternatives to enhance fatigue life of a drillstring component are appropriate material selection and thread-profile design.he objectives of this study were to investigate whether material-property optimization can significantly improve fatigue performance of drillstring connectors and whether common thread profiles have a significant effect on fatigue life of drillstring connectors.
The drillpipes commonly used today are the seamless, Grades D and E, carbon-steel tubing. The heat-treated, high-strength/ low-alloy (HSLA) 4130–4145 steels are commonly used for tool joints and drill collars. Steel metallurgy has progressed greatly during the past 2 decades. Microalloying technology and thermomechanical-processing technology significantly affected the pipeline industry in the early 1970'S. The drilling industry, however, has not benefited greatly from these metallurgical improvements.
Conventional drillstring components have been used for decades. Few studies concern material-properties improvement as a technique for enhancing fatigue life have been published. The reasons for the lack of investigation in the past are twofold:the extremely high cost and time required for fatigue tests on drillstring components andthe lack of analytical techniques for sophisticated stress analysis and fatigue-life prediction. Only recently have the success of the finite-element method (FEM) in stress/strain analysis and the local stress/strain approach and the fracture-mechanics concept in predicting fatigue life enabled us to investigate the drillstring-fatigue problem analytically on time- and cost-efficiencases.
Whole-Life Fatigue Model
The whole-life fatigue model was successfully applied to notched specimens, welded components, and threaded pin/box connectors. The predicted fatigue-life results agree very well with the experimental results. The model nonarbitrarily defines the crack initiation size, 1, thereby nonarbitrarily joining the fatigue-crack initiation-life, N, and the fatigue-crack propagation-life, Np, estimates to give the total fatigue life, N. Fig. 1 illustrates the the-oretical background of the model. The basic concepts of the model are that the stress-concentrated area ahead of a notch root is simulated by an infinite number of microelements (Fig. 1a) and that the fatigue-crack-propagation phase begins when the damage rate resulting from crack-propagation mechanisms, R, exceeds that resulting from crack-initiation mechanisms, Ri (Fig. 1b). l, can be determined with a numerically equivalent, simpler method (Fig. 1c). The minimum of N in the plot of N vs. l also determines l . N's are the arbitrary total fatigue-life estimate found by a series of arbitrarily assumed crack-initiation sizes (l's). N at any l is the summation of the failure life of the element, N, at l and the crack-propagation life, Np, integrated from l to a critical crack size, c. The minimum of N vs. l yields the predicted l, N, N, and N .
The failure life of each element, N, can be calculated by the low-cycle fatigue approach. This requires a stress profile or stress distribution in the stress-concentrated area (Fig. 1a) and the stress/strain-vs.-life relations of the material (Fig. 2b). The crack-propagation life, Np, can be calculated by fracture-mechanics concepts. This requires the appropriate calculation of stress-intensity-factor range and crack-propagation-rate data (Fig. 2c).
The first task of this investigation was to test the life-prediction technique for drillstring applications. If successful, the results would be used to study the effect of material properties on fatigue performance of drillstring components and to explore whether the commonly used thread profiles significantly affect the fatigue life of drillstring connectors,
Verification of Life Prediction for Drillstring Applications
Few fatigue test data have been published on drillstring components. Weiner and True's drill-collar fatigue test data were used to test the accuracy of the fatigue-life predictions for drillstring components. Material-Property-Data Generation. The drill collars tested by Weiner and True were made of AISI 4145 steel with a minimum yield strength of 120 ksi. The material fatigue properties necessary for life prediction were not given, so they were generated from a similar 4145 steel of equivalent strength (4145 MOD drill-collar steel).
The small, smooth, cylindrical specimens machined from the midwall of a drill collar were used to measure the monotonic, cyclic, and fatigue properties, but not crack-propagation rate, The crack-propagation rate was measured with the single-edge-notched specimens. Fig. 2 illustrates the monotonic, cyclic, and fatigue be-haviors of this drill-collar steel. Table 1 lists the chemical compo-sitions and laboratory-measured material-property data.
FEM Stress Analysis. The stress state of a drillstring connector during drilling of a dogleg borehole is a combination of a high initial static stress (mean stress) in the connection resulting from the makeup torque, a stress concentration at the root of threads and stress-relief grooves, and a cyclic bending stress (alternating stress resulting from bending loading during rotation of the drillstring. The additional axial tension also contributes to the mean stress of the tool joint.
The FEM analyses with the ANSYS computer program were elastic. The stress profiles were provided in two terms: the assembly solution (mean stress) and the bending solution (alternating stress). Bending stresses were assumed to be superimposed on the assembly stress. The stress profiles of any makeup torque and any bending moment (or dogleg) were assumed to be linearly proportional to the stress profiles of the specified makeup torque and bending moment (or dogleg) in the FEM analysis.
The FEM analyses were performed for three drill-collar connections (5 H90, NC46, and 6 5/8 API Regular). The 5 H90 and NC46 connections are similar to the 4 1/2 H90 and API 44 connections tested by Weiner and True, respectively.
SPEDE
P. 129^
