Reinforced Joints and Fabricated Trunnions 
for Marine and Offshore Structures

There are thousands of offshore platforms operating in oil and gas fields around the world. Many components in offshore structures, such as boat landings, risers and sacrificial anodes, are welded to the structure (T-joint) by the provision of a reinforcement plate. During the construction and lift installation of large offshore structures, fabricated trunnions are cost effective components and have distinct advantages over cast padears (which are expensive) and padeyes (which are constrained by the capacity of shackles). However, the design approach for reinforced connections and fabricated trunnions is very rudimentary and not based on test evidence. It is therefore important to carry out systematic studies, through well-instrumented tests and parametric finite element analyses, to study the complex behaviour and to derive practical design recommendations. 

This article describes the large-scale tests and corresponding nonlinear finite element studies conducted by a research team in the Centre for Construction Materials and Technology, NUS. The fabricated trunnion investigations are part of the overall R&D project "Development of a Lift Dynamics and Decision Support System for Lift Installation of Structures" funded by National Science & Technology Board, and hosted by Sembawang Marine & Offshore Engineering Pte Ltd and Asian Lift Pte Ltd. The reinforced joint research is sponsored by Exxon Production Research Company (Houston, U.S.A.), with the specimens contributed by Sembawang Marine & Offshore Engineering Pte Ltd. 

The fabricated trunnion research was carried out in two phases. In the first phase, a total of 16 specimens (consisting of 8 plate trunnions and 8 pipe trunnions) with different geometric arrangements and dimensions were tested to failure to observe the failure loads and mechanisms. The results provided a sound basis for the project team to extend the testing to 17 large-scale specimens using the 1000 tonne test rig in the Structural Engineering Laboratory of the Department of Civil Engineering. Figure 1 shows an overall view of the test set-up for one of the tests. A white-washed specimen can be seen in the foreground. 

Figure 1: Testing of large-scale trunnion specimens in the 1000 tonne test rig.

The 17 specimens are thought to be the largest tubular DT joints tested under shear and in-plane moment loads by researchers, with the observed strength reaching up to 7,000 kN. The specimens have been designed with a variety of structural schemes and geometric ratios to provide good reference results for calibration of computational and theoretical models. Figure 2 shows the failure modes of two specimens (CT2 and CT6) which have different structural arrangement. Specimen CT6, with a through pipe arrangement, has a higher strength due to more efficient load transfer mechanism. The 17 large-scale tests have yielded valuable results highlighting the importance of proper design to account for consistent load path, and possible design improvements over current industry practice.

 
Figure 2: Deformed Shapes of Trunnion Specimens CT2 and CT6.

The experimental program for reinforced joints consisted of fifteen large-scale T-joint test specimens subjected to brace axial load (either compression or tension). The chords were API 16" (406.4 mm) Schedule 20 and 40 pipes, while the braces were API 4" and 8" (114 mm and 219 mm) pipes. The geometry and dimensions of the chord, brace, doubler or collar plate, were selected to correspond to typical values for T-joints with doubler plates in offshore platforms. Figure 3 shows Specimen EX-05 after completion of the test. The specimen is pin-supported at the chord ends and the brace end is bolted to the computer-controlled Instron 200kN actuator mounted on top of the four-column test frame. Specimen EX-05 was subjected to brace compression and the significant deformation at the reinforced joint can be readily observed. The test results indicated that significant strength enhancement for a T-joint was possible, if reinforced by a doubler plate or a collar plate. The strength enhancement was around 40% for brace compression and 15% for brace tension. 

Figure 3: Reinforced T-joint Specimen EX-05 after test.

In order to observe the deformed shape of each of the un-reinforced and reinforced joints, the test specimens were flame-cut across the transverse axis of the chord and subsequently ground smooth. In Figure 4, the deformed shapes for three specimens with identical geometric ratios: EX-01 (un-reinforced), EX-03 (with collar reinforcement) and EX-07 (with doubler reinforcement) highlight the different load transfer and failure mechanisms of the joints. Each of the circular pipe chords (in the un-deformed geometry) has undergone significant deformation and there is a large gap between the chord and doubler plate of EX-07. 

   
Figure 4: Deformed Cross-Sections of T-Joints with Identical Chord Dimensions: EX-01 (Unreinforced), EX-03 (Collar Reinforcement) and EX-07 (Doubler Reinforcement).

Selected experimental results were presented in the paper "Static Strength of T-Joints Reinforced with Doubler or Collar Plates" in the Eighth International Symposium on Tubular Structures (which was hosted by the research team in 1998 in Singapore). This paper is referred in the new International Code, "Petroleum and Natural Gas Industries - Offshore Structures - Part 2: Fixed Steel Structures". Nonlinear finite element analyses that account for the geometric, material and contact non-linearities have been performed and very good correlation with the experimental results have been obtained. Parametric studies have also been carried out on the strength of reinforced T-joints subjected to moment loads. The research results will provide an appropriate basis for the formulation of design recommendations to the industry. The collaborators for this project are Prof NE Shanmugam, Assoc Prof RJY Liew, Mr CK Quah and Dr GJ van der Vetgte (of Delft University of Technology, The Netherlands).