Jacket Transportation: Effect of Ballast Distribution on Jacket Stresses and Fatigue Lives

ABSTRACT This paper describes selected results from the transportati0n analyses performed for two jackets scheduled to be transported by barge from Japan to the west coast of California in 1986. For both jackets, the distribution of the ballast water in the barge was found to have a significant effect on the distribution of load transferred between the jacket and barge. The main conclusion drawn from these results was that, by controlling the distribution of ballast water in a jacket transportation/ launch barge, the designer can limit the amount of member resizing required for the transportation conditi0n. It was also found that, for the case studied, fatigue results were relatively insensitive to ballast distribution. INTRODUCTION For many offshore oil and gas production projects, ecomonics dictate the use of low cost fabrication facilities located hundreds, if not thousands, of miles from the eventual sites. More emphasis is therefore placed on designing the structures to withstand the transportation induced stresses. As an example of the significance that a long duration tow can have on the design of a conventional fixed jacket, a study for the Exxon Harmony and Heritage structures (1) shows that the tow of these deep water, single piece jackets from the Orient resulted in:A 40% increase in steel weight over the weights required for minimum sizing. (Minimum sizing is based on a diameter to thickness ratio of 53 and a slenderness ratio of 80.)A 30% increase over the minimum joint can weight required. (Minimum joint can weight corresponds to a can thickness equal to the nominal chord member size.) The same study shows that the transporation loading condition of a single piece jacket has a greater effect on the steel weight requirement than other loading condit1ons such as in-place and seismic. While significant advances in transportation analysis techniques have been made, many problems remain. Among these are the uncertainty in the sea state probability results from forecasting or hindcasting techniques, the lack of correlation between analytical results and actual motion measurements, the cost benefit trade-off between simplified model1ing and elaborate detailed models, the determination of the appropriate spectral density function to use, the selection of spectral versus time domain analysis, etc. This paper's contribution to the topic of jacket transportation is to examine the barge/jacket interaction and to high1ight the advantages that can be achieved by controlling the distribution of ballast water in a jacket transportation/launch barge. This approach may lead to reduction in the additional stee1 required to withstand the transportation loads. DESCRIPTION OF JACKETS AND BARGES Numerical results presented in this paper are from analyses performed for two West Coast jackets currently being fabricated in Japan. Both jackets are conventional steel template structures with eight main legs and additional skirt pile sleeves. All main legs and skirt pile sleeves will be piled and grouted following installation. Design water depths for the platforms are 430 ft. and 740 ft.