Experimental evaluation of the axial capacity of cracked tubular member

Abstract

This thesis experimentally investigated the compressive axial capacity of cracked tubular members. Then compared it to NORSOK N-004 (Standard Norge 2004) formulae. Further, the test results were compared with other analytical models, such as non-linear finite element analysis from a separate thesis (Vågen 2021) and basic column formulae e.g. Perry Robertson and the Secant method.

In this study, 11 tubular columns were tested with a diameter of 70 mm, 2.9 mm thickness, and 1.5 m height, due to the testing facilities limitation. Different crack sizes were placed perpendicular to the loading direction with sizes corresponding to NORSOK N-004 reduced capacity of 25, 50 and 75%. The material properties were obtained by stub column tests. To further study the effect of cracks on tubular members, angled cracks with respect to the loading direction were introduced to the stub column. For the specimens with cracks perpendicular to the loading direction, The test results were somewhat unexpected and showed an inconsiderable reduction in the axial capacity, as the crack surfaces bear on each other. However, the angled cracks introduced to the stub columns showed a considerable capacity reduction. The reduction is assumed to result from the crack surface slipping, discontinuity in the material, and torsion due to the angled crack in the stub column.

The conclusion was that angled cracks have the most significant impact on the compressive axial capacity. On the other hand, cracks placed perpendicular to the loading direction had an insignificant effect on the capacity. Hence, the NORSOK N-004 formulae provide inaccurate formulas for capacity determination for such cases. Among the analytical methods presented in this thesis, Perry Robertson gave the most accurate capacity. This result was somewhat expected as the method was best adapted to the actual column behavior in the tests.

This thesis experimentally investigated the compressive axial capacity of cracked tubular members. Then compared it to NORSOK N-004 (Standard Norge 2004) formulae. Further, the test results were compared with other analytical models, such as non-linear finite element analysis from a separate thesis (Vågen 2021) and basic column formulae e.g. Perry Robertson and the Secant method.

In this study, 11 tubular columns were tested with a diameter of 70 mm, 2.9 mm thickness, and 1.5 m height, due to the testing facilities limitation. Different crack sizes were placed perpendicular to the loading direction with sizes corresponding to NORSOK N-004 reduced capacity of 25, 50 and 75%. The material properties were obtained by stub column tests. To further study the effect of cracks on tubular members, angled cracks with respect to the loading direction were introduced to the stub column. For the specimens with cracks perpendicular to the loading direction, The test results were somewhat unexpected and showed an inconsiderable reduction in the axial capacity, as the crack surfaces bear on each other. However, the angled cracks introduced to the stub columns showed a considerable capacity reduction. The reduction is assumed to result from the crack surface slipping, discontinuity in the material, and torsion due to the angled crack in the stub column.

The conclusion was that angled cracks have the most significant impact on the compressive axial capacity. On the other hand, cracks placed perpendicular to the loading direction had an insignificant effect on the capacity. Hence, the NORSOK N-004 formulae provide inaccurate formulas for capacity determination for such cases. Among the analytical methods presented in this thesis, Perry Robertson gave the most accurate capacity. This result was somewhat expected as the method was best adapted to the actual column behavior in the tests.