Tension and Collapse Tests of Fabricated Steel Cylinders

Clarence D. Miller, R.K. Kinra, Roy S. Marlow

Offshore Technology Conference · 1982 · 13 citations · 0 references

Concepts

Abstract

Abstract This paper summarizes the results of collapse tests on forty-two stiffened and unstiffened fabricated steel cylinders under combined axial tension and external pressure loading. The test program was initiated because the existing data base consisted only of tests on small diameter manufactured tubes with a maximum diameter/thickness (D/t) ratio of 27. The objective of this test series was to develop a comprehensive data base for fabricated cylinders land to determine the adequacy of current API RP 2A design rules for tension and collapse loading. The tests were conducted in two groups and the results of the first group were used to select the test specimens for the second group. The first group of twenty-four tests, completed during 1980, was sponsored by a joint industry group. The second group of eighteen tests, completed during 1981, was sponsored by the American Petroleum Institute (API). The test specimen parameters were selected to represent typical offshore platform member sizes (D/t from 36 to 96), commonly used platform steels with yield strengths of 36 and 50 ksi (A36, A633 and A572) and routine platform fabrication procedures. Eighteen and twenty-four inch diameter test specimens were designed to fail at different stress levels, ranging from 0.3 to 0.9 times the material yield strength, when loaded by radial pressure only. Prior to testing, extensive out-of-roundness measurements were made of each test specimen. The test results are compared with the API design rules1 and with two interaction equations developed by Miller2 based upon a study of available tests performed on small diameter tubes. Good agreement is found between the test results and the method of Ref. 2. The API rules are found to be conservative for the geometries tested. The amount of conservatism varies with the geometry and with the axial stress. The maximum ratio of the hoop stress at failure to the corresponding API predicted stress is 2.0. Introduction Studies have been made of the theoretical methods for predicting the collapse strength of cylinders under combined axial tension and hoop compression. These methods are limited to modes of failure which are either elastic or in the yield range. No theoretical equations have been found for the transition range between these limits. Studies have also been made of the available test data for this combination of loads. When the present test program was initiated the only available test data were on small diameter tubing and a3few plastic cylinders. In 1981 Kyogoku, et. al.,3 presented the results of tests performed on oil well casing. The present test program was conducted to develop data on fabricated cylinders. The results of the tests on tubing were used to develop interaction equations2 applicable to all modes of failure. The current API design rules for fabricated cylinders are based upon a more conservative approach than that predicted from the small tube test results to account for unforeseen effects resulting from fabrication. The present test results, however, are found to closely follow those of the tests on small tubing and oil well casing.