Large, dry-contacting assemblies of graphite blocks are a feature of many nuclear reactors, and the seismic response of these structures is dependent on both the bulk and surface properties of the blocks. The nature of the interblock interface will in¯uence the load±de¯ection relationship of the blocks, which is important in understanding the response of the structure to seismic loading. The stiVness of this interface is dependent on surface roughness eVects.In this paper, ultrasonic re¯ection coeYcient measurements are used to investigate the stiVness distribution of this interface under various loading conditions. A low-frequency quasi-static spring model of the interface was adopted. This allowed the interfacial stiVness to be determined as a function of applied load, frequency, surface roughness and graphite composition. Cyclical loading of the graphite±graphite interface has also been carried out. This has shown that there is only a slight plastic deformation under normal loading, and also is indicative of a degree of intersurface adhesion at the interface. Direct measurement of the load±de¯ection relationship of an entire brick was also carried out which, in conjunction with the ultrasonic measurements, suggests that it is the eVect of macroscopic surface features rather than microscopic roughness that is the dominant interfacial contribution to brick-tilting stiVness. NOTATION d half-depth of the brick face (m) E Young's modulus (Pa) F force (N) h height of the brick (m) I second moment of area (m 4 ) K interfacial stiVness per area (Pa/m) l half-width of the brick face (m) p local contact pressure (Pa) p nom nominal contact pressure (Pa) R ultrasonic re¯ection coeYcient R a centre-line average roughness (m) u average interfacial separation (m) W total weight of the brick (N) x distance along the brick face from the centre of the edge (m) x 0 position along the brick face at which contact with the substrate is lost (m) z acoustic impedance (kg/m 2 s) ¬ ratio of interfacial stiVness per area to root of pressure (Pa 1=2 /m) ¡ torque applied to the brick (N m) brick de¯ection (m) angle of tilt of the brick (rad) ! ultrasonic angular frequency (rad/s) IMPORTANCE OF GRAPHITE±GRAPHITE INTERFACESThe cores of many commercial nuclear reactors contain large assemblies of graphite blocks which are used to moderate the fast neutrons in the reaction. Fuel and control rods are positioned in holes in this assembly. These blocks are typically around 1 m high and are arranged in arrays ten or more layers high. The bricks cannot be physically constrained into a ®xed matrix owing to the extreme irradiation and heat-induced deformation they undergo, but instead ®t together withThe MS was

No abstract

No abstract

Large assemblies of blocks are common in many civil engineering structures such as dams, costal defences and ancient buildings. Such assemblies are also found in graphite cored nuclear reactors. These consist of large graphite bricks, which are used to moderate the reaction and are assembled in arrays up to 10 m in height and 10 m in diameter. In this paper, the bricks considered are 200 mm  200 mm  800 mm. The seismic response of such a structure is dependent on both the properties of the bulk material and the nature of the interface between individual bricks. The bricks are in dry contact with those above and below, and it is this contacting interface that governs the rocking behaviour, which is therefore an important factor in assessing the response of the columns to vibration or a seismic event.This paper describes an investigation into the effect of microscopic and macroscopic surface features on the rocking behaviour of a single graphite brick. Typical real surfaces exhibit some form of macroscopic waviness in addition to the microscopic roughness and therefore only a small fraction of the surfaces are in contact. A number of analytical models are used to compare the effects of specific surface features with experimental data from real bricks. These models are compared with experimental results to assess their importance in explaining the rocking behaviour observed experimentally. Factors such as surface curvature, contact size and position are shown to be particularly important to the rocking stiffness of a brick. In most cases, the reduction in the rocking stiffness compared to perfectly flat surfaces is due to both the reduced moment arm of the contact forces and the increased strain at the contact region due to the high stress caused by the small area of contact.

No abstract