Fatigue Testing of Drillpipe
- G.Y. Grondin (Natl. Research Council of Canada) | G.L. Kulak (Natl. Research Council of Canada)
- Document ID
- Society of Petroleum Engineers
- SPE Drilling & Completion
- Publication Date
- June 1994
- Document Type
- Journal Paper
- 95 - 102
- 1994. Society of Petroleum Engineers
- 1.10.1 Drill string components and drilling tools (tubulars, jars, subs, stabilisers, reamers, etc), 4.3.4 Scale, 4.2.3 Materials and Corrosion, 1.10 Drilling Equipment, 1.6.1 Drilling Operation Management, 1.11 Drilling Fluids and Materials, 1.6 Drilling Operations
- 2 in the last 30 days
- 506 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 12.00|
|SPE Non-Member Price:||USD 35.00|
Twenty-nine tests in air and 27 tests in a 3.5% NaCl solution were conductedto test the fatigue strength of Grade E 4.5-in.-OD, 16.6-lbm/ft [114-mm-OD,24.7-kg/m] drillpipe. The effects of stress range, mean stress, corrosion, andupset geometry were tested on the drillpipe body alone and on specimens thatincluded the tool joint, upset, and drillpipe body. Stress range and meanstress effects were significant in noncorrosive and corrosive environments. Theeffect of upset geometry was minimal on an external upset compared with that onan internal/external upset. Among the specimens that failed, grinding marks onthe drillpipe surface, which occurred during inspection at the mill, causedfailure of about 60% of the specimens. Grinding created a notch effect andinduced a detrimental residual-stress pattern on the pipe surface, whichdecreased drillpipe fatigue life. API guidelines for cumulative fatigue ofRange 2 drillpipe were evaluated in light of these experimental test resultsand were found to be unsafe for drilling in noncorrosive conditions andpossibly too conservative for corrosive conditions.
Fatigue damage occurs when a drillpipe is subjected to sufficiently highalternating stresses, such as those created when the drillpipe rotates in thecurve of a wellbore. Drillpipe fatigue failure has been a serious concern inthe oil industry ever since sections of drillpipe were first joined to permitdrilling at depths greater than one length of drillpipe. 1-4 Thisproblem was addressed by imposing dogleg severity limits5 based ontest results published in the early 1950's.6 These tests weresatisfactory for their intended use (i.e., testing of tool-joint welds).However, these tests were not performed in a corrosive environment or underaxial tension, two factors considered important in current APIguidelines.5 During these tests, the effect of corrosion wasaddressed by use of simplifying assumptions with respect to the decrease infatigue strength in a corrosive environment. The effect of mean stress wasdealt with by use of a modified Goodman equation for the endurance limit and astandard Goodman equation for stresses above the endurancelimit.7-9
Published fatigue-test data on full-sized drillpipe are limited. Testsperformed by Hughes Tool Co. still represent a major contribution toward betterunderstanding of drillpipe fatigue strength.6 More recently, theChinese Petroleum Standardization Committee presented the results of aninvestigation on the effect of the geometry of the transition zone between thedrillpipe body and the upset.10 Full-sized drillpipes withtransition zones of various lengths and radii were tested in air as cantileverrotating beams under no axial load. Fatigue resistance improved as the lengthof the transition zone increased.
Tsukano et al. 11 shed some light on the effect ofupset/pipe-body transition-zone geometry in their investigation of theinternal-upset drillpipe geometry using finite-element analysis and tests. Theysought a combination of taper length and radius of runout that would causefatigue failure in the drillpipe body rather than in the pipebody/upsettransition zone. To verify the results of the finite-element investigation,full-sized specimens were tested in four-point rotary-bending arrangements inair at a high stress range (location of crack initiation was the only factorinvestigated). Recommendations were made for suitable internal-upset geometryto prevent drillpipe failure in the upset region.
The drilling industry recognizes corrosion as a main factor affectingfatigue life.3,8,12,13 The effect of corrosive environment ondrillpipe fatigue life has been investigated with small steel samples, but thiseffect on full-sized drillpipe has not been compared with test results fromsimilar drillpipes tested in air. Joosten et al.14 tested3-ft [0.9-m] lengths of 3 1/2-in. [89-mm]-diameter Grade G-105 drillpipe loadedin three-point bending. Tests were conducted in air and in KCl/seawatersolution. Only results of tests in a corrosive environment werepublished.14 Consequently, resistance in air and in a corrosiveenvironment cannot be compared from their published data.
In 1988; Dale15 presented the results of a test program on APIdrillpipe steels conducted to determine the influence of drilling-fluidenvironment on fatigue-crack growth rate. Although the program mainly studiedfatigue in a full-sized drill collar, it also included a series of fatiguetests on coupon specimens in drilling muds of various composition. The tests,conducted at a 5-Hz frequency, showed no significant effect of drilling mudcorrosivity on crack growth rate.
Helbig and Vogt16 presented the results of a study on drillpipefatigue life in a corrosive environment that investigated the effect of heattreatment on Grades D, E, and S-135 drillpipe. Full-sized sections of drillpipebodies were fatigue tested in two corrosive environments: tap water and 20%NaCl solution. The test results did not indicate a significant differencebetween normalized and quenched-and-tempered specimens or between the two testenvironments. From their tests on coupon specimens, Helbig and Vogtdemonstrated that the speed of testing is influential when tests are conductedin a corrosive environment. Fatigue life was reduced significantly when thetesting frequency was decreased from 1,000 to 100 rev/min; the amount ofreduction depended on the stress range at which the tests were conducted. Allthe tests of Helbig and Vogt16 on full-sized drillpipe wereperformed in a corrosive environment. Consequently, fatigue life of full-sizeddrillpipe in air and in a corrosive environment cannot be compared with theirdata. No experimental investigation to date has specifically addressed theproblem of the effect of mean stress on the fatigue life of full-sizeddrillpipe operating in air or in a corrosive environment.
As the search for petroleum moves into more-hostile environments thatrequire drilling to greater depths and in more-corrosive media, the oilindustry is again confronted with the problem of drillpipe fatigue. One failureis estimated to occur for every 6,500 ft [1980 m] drilled, including drillpipeseparations and washouts.17 Most drillpipe failures generally areagreed to result from metal fatigue. Recent publications show that drillpipefailure is still a serious concern of drilling contractors.3,14,17More research is therefore required to determine the effect of such parametersas mean stress and corrosion on drillpipe fatigue life.
A preliminary test program18 on full-sized drillpipe specimensthat incorporated the tool joint, upset ends, and drillpipe body showed thatfailures in the field could be duplicated in the laboratory. Preliminarytesting was done under cyclic tension and bending.
A full program was then designed to investigate the effects of stress range,mean stress, and corrosion on drillpipe fatigue life under rotational bending.Full-sized API Grade E 4 1/2-in.-OD, 16.6-lbm/ft [114-mm-OD, 27.4-kg/m]drillpipes with internal/ex ternal upsets were tested. Test specimens wereprepared so that the tool joint, upset region, and part of the pipe body couldbe incorporated in the test section. Testing showed that failures in allspecimens except one occurred in the drillpipe body, not the tool joint orupset region. For this reason and for reasons of economy, half the testspecimens were sections taken from only the drillpipe body.
|File Size||1 MB||Number of Pages||8|