J101013 CONSIDERATION ON SEISMIC DESIGN MARGIN OF ELBOW IN PIPING

Elbow is an essential element for three dimensionally arranged piping and it is actually used in most kinds of plants. Many great researches on the flexibility and stress intensity regarding elbow elements have been performed. Moreover, we can greatly benefit from the design code where elbow elements are specified. Our research group also started a research on ultimate strength of piping systems containing elbows in 1997 and we have performed several kinds of elbow element tests and shaking table tests. All experimental results have shown that the failure loads are far higher than those described by the design criteria. The authors have confirmed that the seismic design margin is extremely conservative. In this paper, the results of shaking table tests of piping, elbow element experiments and the stress calculation for those experiments based on design code are described, their results are compared with the seismic design criteria, and the margin is discussed. The authors point out the necessity of a new design code on the basis of the detail analysis and strain criteria in order to describe more appropriate and reasonable seismic design margin of the piping. INTRODUCTION Elbow is generally used for the piping system of nuclear power plant. Regarding elbow or bending pipe, the important researches have been performed since the beginning of 20th century [1],[2],[3]. However the stress distribution of elbow is essentially complicated, the simplified and convenient method to calculate the stress is provided by design code based on previous researches. Regarding to pipe, pipe fittings and piping system, many researches were performed to verify the ultimate strength against earthquake since the latter half of 1980s [4],[5], [6],[7],[8]. And they show that most of the failures by seismic excitation were low cycle fatigue, except for only a few case of collapse failure. It is described as the seismic load is dynamic reversing load and the inertia force acts in the opposite direction to the deformation direction, then it is caused low cycle fatigue for most of the failures with dynamic reversing load. Japanese seismic design code [9] was revised in 2008. Based on the consideration that seismic load is the dynamic reversing load, primary stress evaluation for seismic load is not required, but the evaluation with other mechanical load, and fatigue evaluation is required as important evaluation for seismic load. On ASME code, primary stress evaluation is an important evaluation for seismic load. The difference of both codes is due to the difference of the number of seismic load cycles. The number of seismic load cycles from 60 to 300 is applied to the Japanese seismic design. But that of approximate 20 cycles is applied with ASME code. The difference of evaluation criteria of seismic design is caused by the variation of the number of seismic load cycles between U.S. and Japan. Elbow is an essential element for three dimensional piping and it is widely used for plants. Many researches on the flexibility and stress intensity of elbow have been performed and we can benefit from the design code. The authors started about a research about failure behavior of piping under seismic excitation from 1997 and we performed several kinds of elbow element tests and shaking table tests. Experimental results provide that the excitation level to reach the failure is so higher than limit level by design code. It is confirmed that the seismic design margin is comparatively large. This study focuses on fatigue evaluation because Japanese seismic design considers the dominant failure mode of piping to be fatigue failure. The margin of the fatigue evaluation is investigated by comparing experimental results of elbow