Heki Shibata

Impact bending fatigue and impact response behavior of a nuclear-grade graphite beam

Abstract Repeated impact bending tests have been carried out on a fine-grained Isotropic graphite for HTTR under constant energy condition. Each specimen whose length differed was centrally impacted by an instrumented pendulum. The simple one-dimensional model taking the Hertzian contact stiffnesss into consideration has been applied to evaluate the relationship between the maximum stress of specimen generated by impact, the impact force, and the impact energy. It has been confirmed that the endurance curves in terms of impact energy are dependent on the length of the specimen, but the one in terms of the evaluated maximum stress is independent of that, and the strength of graphite is lower in impact fatigue than in nonimpact fatigue.

On estimated modes of failure of nuclear power plants by potential earthquakes

Abstract This paper deals with the estimated ‘modes of failure’ of nuclear power plants during future violent earthquakes. The authors have been surveying the damage to industrial plants caused by several violent earthquakes since 1960. Some of them have already been reported in English, but here the authors try to rearrange them from the viewpoint of ‘modes of failures’ of nuclear power plant buildings, equipment, vessels and piping. The authors categorize the mechanisms of failure as follows: (i) damaged by the dynamic effect of acceleration waves, (ii) by resonance in displacement waves, (iii) by the static effect of seismic force, (iv) by external force from attached piping and others, or forced deformation, and (v) by liquefaction of soil. The authors try to determine the modes of failure of the following items in a matrix form of the mechanisms: (i) the reactor building, (ii) steel containment vessel, (iii) auxiliary building, (iv) reactor vessel, (v) core internals, (vi) primary and secondary coolant system, (vii) emergency power supply system, (viii) emergency gas treatment system and stack, (ix) fuel cooling pond and fuel rack, (x) refuel machine crane, (xi) auxiliary system and component, (xii) turbine and its pedestal, and (xiii) main power system and control instrumentation. They also examine them from another point of view, i.e. in ‘the classification of the important factor’ of items for their aseismic design.

Development of Seismic Design Code for High Pressure Gas Facilities in Japan

The Seismic Design Code for High Pressure Gas Facilities was established in 1982 in advance of those in other industrial fields, the only exception being that for nuclear power plants. In 1995, Hyogoken Nanbu earthquake caused approximately 6000 deaths and more than

Bending fatigue behavior of nuclear-grade graphite under impact loading

1 billion (US) loss of property in the Kobe area, Japan. This unexpected disaster underlined the idea that industrial facilities should pay special consideration to damages including ground failure due to the liquefaction. Strong ground motions caused serious damage to urban structures in the area. Thus, the Seismic Design Code of the High Pressure Gas Facilities were improved to include two-step design assessments, that is, for Level I earthquakes (operating basis earthquake: a probable strong earthquake during the service life of the facilities), and Level 2 earthquakes (safety shutdown earthquake: a possible strongest earthquake with extremely low probability of occurrence). For Level 2 earthquakes, ground failure by possible liquefaction will be taken into account. For a Level 1 earthquake, the required seismic performance is that the system must remain safe without critical damage after the earthquake, including no gas leakage. For a Level 2 earthquake, the required seismic performance is that the system must remain safe without gas leakage. This means a certain non-elastic deformation without gas leakage may be allowed. The High Pressure Gas Safety Institute of Japan set up the Seismic Safety Promotion Committee to modify their code, in advance of other industries, and has continued to investigate more effective seismic design practices for more than 5 years. The final version of the guidelines has established design practices for the both Level I and Level 2 earthquakes. In this paper, the activities of the committee, their new design concepts and scope of applications are explained.

Philosophy and practice of the aseismic design of nuclear power plants — Summary of the guidelines in Japan

Abstract The graphite components in an HTR (high-temperature gas-cooled reactor) are subjected to impact forces due to earthquake. It is important from the viewpoint of seismic safety design to investigate the difference of strength under impact loading and nonimpact loading. Both bending strength and bending fatigue strength tests, therefore, were carried out under impact and nonimpact loading on two kinds of graphite materials: isotropic and near-isotropic. The impact response analyses, which used a beam model taking account of the contact behavior between specimen and tup, were performed to evaluate the relationships between impact energy, impact force and stress. The main conclusions obtained are summarized as follows: 1. (1) A beam model taking account of the contact behavior through the Hertzian theory is applicable to describe the impact behavior. 2. (2) The bending strength of graphite is independent of strain rate in the range from 10 −6 to 5/s. 3. (3) The bending fatigue strength of graphite is lower under impact loading than under non-impact, independent of the specimen volume and type of graphite.

Development of aseismic design of piping, vessels and equipment in nuclear facilities

Aseismic design is considered to be one of the most important factors for the safety of the nuclear power plants built in zones of high seismicity such as Japan. All structures, equipment and piping are classified in accordance with the importance of their radioactive safety to the plant, and the dynamic analysis and/or factored seismic coefficient analysis are applied accordingly. Site and ground conditions, as well as seismicity, should be studied thoroughly in order to estimate the intensities of the design earthquake and the safety margin check earthquake. Dynamic analyses of buildings and structures are performed using the multi-lumped-mass-system supported by soil springs with time history analysis conceptions. This idea is also applied to the design of equipment and piping by coupled system to the major structure or by the floor response spectra criteria. Tolerances are applied to damping factors although some experiments show more realistic results. Allowable stresses of ferrous metals for equipment and piping during earthquakes are more scientifically precise. This report summarizes a guideline for aseismic design of nuclear power plants. The guideline was prepared by the Japan Electric Association in April, 1970, after three years laborious work. In sect. 1, the philosophy and criteria are described. All components of a plant should be classified into three classes in accordance with their contributions to reactor safeties. Design to earthquake loadings should be based on “design basis earthquake” which is decided in consideration of local seismicity. In sect 2, site selection and review for ground are described in the sense of seismic aspects. In sect 3, deciding the earthquake motion for design is discussed. In Japan, semi-statistical approaches are used in normal practice. In sect. 4, design philosophy and practice of building structures and containment vessels are described. They are designed under statical seismic forces, and the design of the class “A” structures should be checked by a dynamic response technique. In sect. 5, design philosophy and practice of piping, vesels and equipment are described. Those which belong to class “A” items should be designed in a dynamic sense. Several programs for dynamic analyses of these items are prepared. Allowable stress under earthquake conditions is discussed in relation to other codes, for example, ASME Section III. The greater part of the philosophy and design criteria have been adopted to all nuclear power plants which have been and are currently being built in Japan.

Proving test of earthquake-resistant pipings, equipment and active components☆

Abstract This report involves the development of aseismic design procedures of piping, vessels and equipment in Japan. These mechanical structures show their various characteristics of vibration. Pressure boundaries, a containment vessel and safety systems belong to such structures. The vital components of nuclear power plants are classified to “A” class according to the classification for the aseismic design in Japan. All components in “A” class are required to be based on dynamic earthquake-resistant design, of which level is decided in consideration of local seismisity. For dynamic design purposes, the following processes are the most important: 1. estimating eigenfrequencies and modes of the system; 2. estimating its damping characteristics; 3. estimating the behavior of the system during strong earthquakes; 4. deciding the design criteria, especially the allowable stresses to earthquake loadings.

Bell-ring vibration response of nuclear containment vessel with attached masses under earthquake motion

Abstract In Japan we initiated the project of a shaking table to prove the earthquake-resistant properties of key items in nuclear power stations. This two-axial shaking table will be able to shake a 1000 ton object on a table of 15 × 15 m by 2600 tonG in horizontal force and by 1300 tonG in vertical force. In this paper, the philosophy of such projects as well as various experiences on such proven tests done in Japan will be described. The main purposes of the proven tests are to understand: 1. (1)The behavior of nuclear structures and equipment under strong earthquakes both from the viewpoint of structural dynamics and process dynamics. 2. (2)The endurance limits of structures and equipment to destructive earthquakes both from the viewpoint of structural integrity and function. 3. (3)The behavior and availability of active components under deformations and accelerations induced by destructive earthquakes. 4. (4)The margin of safety of structures and equipment under assumed destructive earthquake conditions both for society at large and related engineers. Although we have almost eleven centuries of historical data on earthquake damage, we still learn new facts from each new destructive earthquake. We have almost no experience of earthquake damage to nuclear structures and equipment. We have endeavoured to estimate the ‘modes of failure’ of various structures and items in nuclear power stations. Therefore, we should be afraid of lack of knowledge on that, because earthquakes are natural phenomena and have a somewhat unpredictable nature. However, we can understand the behavior of structures and equipment in their ultimate condition through the shaking test. One of our uncertainties on earthquake effects is the effect of vertical ground motions. The three-dimensional response seems to cause no particular problems; however, overturning and unstable moving of a solid structure are highly non-linear problems. The behavior of free-surface water in a containment under three-dimensional excitation is not known completely also. These things can be clarified only through shaking experiments using a two- or three-dimensional shaking table, including vertical component motions. The availability of active components, such as control rod driving mechanisms, valves and pumps can be checked only through shaking and/or forced deformation tests, because troubles in such active components may be induced by mechanical friction or contact of moving parts. The malfunction of electrical components is also complicated, for example chattering of the contact of relays. To evaluate the behavior of such active components and electrical components, the so-called mathematical model is inadequate. It is almost impossible to establish an adequate model of those components including all elemental factors related to its function. In Japan, we have many experiences of shaking tests of various size models and for various purposes, not only for nuclear power stations, but also for other areas. Their philosophy, methodologies, results and remarks will be briefly described.

Developments of Seismic Design Code for High Pressure Gas Facilities in Japan

Abstract There are two types of vibrations, designated as ‘beam-type’ and ‘bell-ring type’ occurring with axisymmetric thin shell nuclear containment vessel. Up to this time, the seismic analysis for such thin axisymmetric shells has mostly been carried out only for the ‘beam-type vibration’ because the response participation factor for the ‘bell-ring type vibration’ under seismic motion is zero when the shell structure is perfectly axisymmetric. However, as with nuclear containment vessels, when the thin axisymmetric shell has several attached heavy masses such as the equipment hatch or the manholes, the resulting seismic response of bell-ring type vibration is unexpectedly large and becomes remarkably more important than the beam-type vibration. For the seismic analysis of bell-ring type vibration an approximate uncoupled analysis using the natural mode shapes of unweighted perfect axisymmetric shell has been advocated on the assumption that the effect of the attached mass on their natural modes might be very small. However, application of this method to some models showed that the response of bellring type vibration calculated was noticeably smaller than the experimental results. In this paper we show the seismic response analysis of the bell-ring type vibration coupled with the beam-type vibration through the attached masses with the new consideration. These results show good agreement between the theoretical calculation and the experiment.

Human response in gazing at a moving figure

The Seismic Design Code for High Pressure Gas Facilities was established in advance of other industrial fields in 1982. Only exception was that for nuclear power plants. In 1995, Hyogoken Nanbu earthquake brought approximately 6,000 deaths and more than 100,000 M