An Experimental Investigation of Fatigue-Crack Growth in Drillstring Tubulars
- B.A. Dale (Exxon Production Research Co.)
- Document ID
- Society of Petroleum Engineers
- SPE Drilling Engineering
- Publication Date
- December 1988
- Document Type
- Journal Paper
- 356 - 362
- 1988. Society of Petroleum Engineers
- 4.2 Pipelines, Flowlines and Risers, 1.10 Drilling Equipment, 1.10.1 Drill string components and drilling tools (tubulars, jars, subs, stabilisers, reamers, etc), 4.1.5 Processing Equipment, 4.3.4 Scale, 4.1.2 Separation and Treating, 1.6.1 Drilling Operation Management, 4.2.3 Materials and Corrosion, 1.6 Drilling Operations, 1.11 Drilling Fluids and Materials
- 2 in the last 30 days
- 379 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
Summary. Drillstring failures continue to plague the oil industry, often costing millions of dollars each year. This problem is frequently intensified with the drilling of deep, deviated wellbores or "hard rock" drilling conditions. The drilling industry attempts to guard against these costly failures by performing periodic nondestructive inspections to remove damaged tubulars from service. This paper describes the results of full-scale fatigue-crack-growth tests of drill collars under rotating and bending loads. In addition, corrosion fatigue-crack-growth data are also presented for API drillpipe steels in air and in three representative water-based drilling-fluid environments. Based on this experimental investigation. the test data support the practical application of fatigue-crack-growth mechanics principles for the development of nondestructive inspection intervals to reduce drillstring failures.
Failures of drillstring tubulars are costly to the oil industry in the form of lost rig time, damaged tubular goods, and abandoned or side-tracked wells. Based on drilling records, costs associated with a downhole separation of the drillstring average about $106,000 per occurrence and have been estimated to occur on about 14% of all wells. The majority of these failures are recognized as being some form of metal fatigue. The fatigue mechanism is a progressive one. Under the action of fluctuating stress and corrosion, microscopic cracks form, become macroscopic in size, and then propagate through the wall thickness until failure. In addition, stress "raisers," such as machining scratches, slip marks, and formation cuts, have been known to accelerate fatigue-crack initiation. The oil industry currently guards against these costly failures through periodic nondestructive inspections to check for macroscopic cracks. Inspection intervals traditionally have been selected on the basis of experience in a given area or through the use of guidelines established by Hansford and Lubinski in the early 1960's. This "classical" approach is fundamentally sound for the prediction of fatigue life. This approach does not, however, provide information regarding crack growth from which nondestructive inspection intervals may be specified to avoid failure. Within the last 25 years, though, significant effort has been directed toward better understanding of fatigue-crack-growth behavior. A new philosophy known as "defect-tolerant" design--the principal concept that all structures or components possess defects, either from manufacture or from service--has evolved into engineering practice. Consequently, the fatigue life can be established based on fatigue-crack growth. Nondestructive inspection plays a very important role in defect-tolerant design in that critical flaws (i.e., defects) must be identified and either removed or repaired to avoid failure. The application of fatigue-crack-growth mechanics principles has been widely accepted and practiced by the aerospace/air-craft industries. power generation utilities, and defense contractors, in addition to the offshore structures and pipeline communities within the oil field. This paper describes the results of full-scale rotating/bending fatigue tests on drill collars and corrosion fatigue material tests of API drillpipe steels. The results of this paper support the practical application of fatigue-crack-growth mechanics principles for drill-string tubulars.
A brief review of the terminology associated with the principles of fatigue-crack-growth mechanics is warranted. Fig. 1 illustrates the basic material relationship between fatigue-crack growth rate, d a/dN,* and Mode I stress-intensity fluctuation, delta KI. Stress-intensity fluctuation is a function of fluctuating (cyclic) stress and crack depth, but is also influenced by other variables such as geometry and crack shape. As can be seen, there are three distinct regions of fatigue-crack growth. Region I is referred to as the "threshold" regime, where fatigue-crack growth rate is very slow. Region II is referred to as the "Paris" regime (after the man who first observed this behavior), where fatigue-crack growth rate is found to obey a power-law relationship with stress-intensity fluctuation. Region III is referred to as the "unstable" regime, where fatigue-crack growth rate is very rapid and brittle fracture is imminent, For practical engineering applications, Region II is of greatest importance and is described by Eqs. 1 through 4.
Full-Scale Rotating/Bending Fatigue Tests
Drillstring connections, especially those on drill collars, are particularly susceptible to fatigue failures. Therefore, they were the focal point of this investigation into the fatigue-crack-growth behavior in drillstring tubulars. Full-scale rotating/bending fatigue tests were conducted on 6 1/4 -in. [15.9-cm]-OD x 2 1/4 -in. [5.7-cm] -ID drill collars to determine the rate of crack growth at various locations within the connection. Both API NC-46 and 4 1/2-in. [11.4-cm] H-90 connections were tested.
Test Apparatus. All fatigue-crack-growth testing was conducted with a fully instrumented, servo-controlled, hydraulically actuated, closed-loop, four-point rotating/bending test frame. Fig. 2 shows an overall view of the test frame and control/data acquisition instrumentation.
|File Size||2 MB||Number of Pages||7|