An Analytical and Experimental Study of Stiffened Tubular Joints with Multiple Branches
- 5 May 1974
- proceedings article
- Published by Society of Petroleum Engineers (SPE)
Abstract
This paper describes a method developed by Chicago Bridge & Iron Company for analyzing a.ring-stiffened tubular joint with multiple non-intersecting branch members, each branch member being arbitrarily loaded. A computer program has been developed for analyzing a maximum of six branches, only three of which may intersect any one stiffening ring. For example, a triple K-joint may be analyzed. Model test results are presented for a three-way Y-joint and for a single K-joint. Each branch of the three-way Y-joint was loaded simultaneously with several combinations of axial, radial and tangential loads. The K-joint was loaded with axial loads only. Photoelastic coating was applied to the branch-to-chord intersections to obtain the full field distribution. Electrical resistance strain gages were placed on the models to obtain the load distribution in the branch member near its intersection to the chord member and to obtain the surface stresses on the outside of the shell and the inner fiber of the stiffening rings. Also several transducers were used for measuring radial deflections of the stiffener rings. The models representa 1:12 and 1:13 scale of a structure with a 52 foot diameter chord member. A comparison is shown of the experimental and analytical results. INTRODUCTION For platforms in deep, rough water there is a tendency to use fewer but larger members, each member supporting a larger share of the total load. As member sizes get larger, strengthening the joints by increasing chord wall thickness becomes impractical. In the above situation there is definite reduction of wave forces if the wells are drilled through one or more large legs of a platform. Also since there are no wave forces on theconductors, guides can be lighter in weight. To drill down the legs, however, requires the legs to be kept relatively open and not restricted at every joint by diaphragms and bulkheads. These two points led to a consideration of tubular joints using ring stiffeners to strengthen the chord walls. Very little analytical or experimental work on such joints was found in the literature so a research program was started to develop a method of analysis which could be used to select the number and size of stiffeners required. The primary purpose was to develop a design tool which would predict thrusts, shears, moments and deflections in the rings with sufficient accuracy for design use. The finite element method. is too cumbersome, time consuming and expensive for use as a routine design tool. A single structure requires the design of many joints, each for several loading conditions. Some finite element analysis will be done to provide an additional check on the proposed design method and hopefully to fill in any gaps between the analytical and test results. The finite element study will not be covered in this paper.Keywords
This publication has 0 references indexed in Scilit: