Fatigue Of Structural Steel For Offshore Platforms

Abstract

ABSTRACT Laboratory data for engineering fatigue design was obtained on cyclically strained cantilevered plates of carbon steel in air and flowing sea water. Some information on weld geometry and cathodic protection was gathered. The result were expressed as a "fatiguelife" prediction in terms of the total cyclic strain range ??t and the number of cycles N to grow a surface crack l8 eight inches long. An alternate analysis in terms of fatigue crack growth rate dl/dN was developed for a portion of the results. Most data were collected in the low-to-medium cycles life regions (103- 105 cycles), but two specimens were tested where extended lives (109 cycles) and low strains (??=200 ? inches/inch) prevailed. Precautions in corrosion fatigue tests of mild steel in sea water are highlighted. INTRODUCTON This paper presents experimental data on the fatigue properties of a representative plate of a construction steel used for the tubular joints of offshore drilling and production platforms. Since prototype testing of complex joints is expensive, some wellestablished engineering techniques for fatigue prediction are often substituted. The purpose of this program was to provide information for these techniques in two ways. Firstly, to compare the S-N curve (cyclic total strain versus fatigue life) for the present specimen to that curve provided by the American Welding Society based on data on Welded pressure vessels and tubular joints as discussed by Marshall (ref. 1). This curve provides the basic for Miner' rule linear cumulative damage calculation, which often is when a spectrum of cyclic loads is imposed on a structure during its life (ref. 2). Secondly, to test the applicability of a method of fatigue crack propagation in which the fatigue crack growth rate, dl/dN, is related to the cyclic total strain, ??t, at the region in which the crack growing according to(Mathematical Equation)(Available in full paper) The material constants C and m as well as the analytical form of f(l) are determined from experiment. A third mode of fatigue analysis, fracture mechanics, was not attempted because most of these data were obtained under extremely high plastic strain conditions, where the underlying assumptions of linear elastic fracture mechanics are invalid. Because of the environment seen by offshore structures, specimens were tested in air and sea water. Furthermore, the well known influence of cathodic protection on sea water corrosion, which in turn affects fatigue, provided the impetus to study the effect of cathodic protection under our test conditions. This paper presents test results of two programs conducted at different times: one (1968-1970) concerned with high strain, low cycle fatigue and representing the bulk of the data; and the second (1971–1972) involving two extended tests at low strain, high cycle conditions. This resulted in differences in testing procedures such as apparatus and specimen width. EXPERIMENTAL PROCEDURE Material The steel in this program came from one section of a 1–3/8 inches thick plate of modified A441 fine grain steel which corresponds closely to the present ASTM A572 specification.