Cavitation bubbles are generated when highly pressurized water is jetted in water. Surface residual stress is improved remarkably due to the peening effect of extremely high pressure caused by the collapse of cavitation bubbles. This technique is called water jet peening (WJP). WJP is expected to be an effective preventive maintenance technique for the stress corrosion cracking in various components of power plants. Various specimens are water jet peened to evaluate the fundamental characteristics of WJP and to select the most appropriate conditions for the residual stress improvement. The test results show that WJP markedly changes the tensile residual stress caused by welding and grinding into the high compressive residual stress and prevents the stress corrosion cracking.
The effect of water jet peening (WJP) on the improvement of stress corrosion cracking (SCC) and fatigue strength was verified by performing SCC and fatigue tests. The SCC specimens employed were sensitized type 304 stainless steel simulating the heat affected zone in welded metal. The creviced bent beam (CBB) type SCC test was performed in high temperature water using the smooth and pre-cracked specimens. WJP remarkably improved the susceptibility of SCC for the smooth specimen, and cracks were not initiated when the applied strain was less than 0.3%. WJP was also confirmed to prevent SCC for the pre-cracked specimen. In the fatigue tests both smooth and pre-cracked plate specimens were subjected to cyclic bending loads in air at room temperature. The fatigue strength of smooth and pre-cracked specimens at a fatigue life of 107 cycles was increased by 1.1 and 2.4 times, respectively, in comparison with that of non-WJP specimens. This means that WJP was much more effective for the pre-cracked specimens. CBB and fatigue tests show that WJP is an effective technique for preventing SCC in the structural components of power plants.
Fatigue tests were performed using CCT specimens having initial tensile residual stress caused by welding. The residual stress distribution on the fatigue fracture surface was measured by X-ray diffraction method. The monotonic plastic zone depth corresponds to the Kmax, true which is the sum of the Kmax related to the applied maximum stress and the Kres caused by the initial residual stress. So, if the initial residual stress distribution is known, the Kmax or the applied maximum stress can be estimated from the monotonic plastic zone depth. The residual stress on the fatigue fracture surface corresponds to the sum of the tensile residual stress related to the Kmax, true and the compressive residual stress related to the ΔKeff. So, even if the initial residual stress distribution is unknown, the ΔKeff or the stress amplitude can be estimated from the residual stress on the fatigue fracture surface and the monotonic plastic zone depth.
The surface of the bending specimens of silicon nitride was ground by the face grinding method. Face grinding was conducted with a cup-shaped diamond wheel whose mesh size was #200/230. The feed direction of face grinding was parallel and perpendicular to the longitudinal direction of the specimen. The surface roughness of the face-ground specimen was one-fifth of that of the surface-ground specimen, when a diamond wheel with the same mesh size was used for grinding. The residual stress state was compressive at the ground surface. The depth of the compressive residual stress zone was about 25μm. The compressive residual stress took the maximum of about 1GPa which was slightly smaller than that of the surface-ground specimen. When compared with the lapped specimen, the face-ground specimens had higher bending strength in the directions both parallel and perpendicular to the feed direction, in contrast to the surface-ground specimens whose strength reduced by about 200MPa in the direction perpenducular to the grinding direction. The reasons for the increment of bending strength for the face-ground specimens were as follows: (1) The main grinding flaw induced by face grinding was in the plane which was formed between the feed direction and the rotating direction of wheels, so that the grinding flaws were hardly remained in the face-ground surface. (2) The compressive residual stress contributed directly to the increment of bending strength of the face-ground specimens, because the size of grinding flaws was smaller than the depth of the compressive residual stress zone. It was concluded that the face grinding method was recommended for machining of ceramics, because of high efficient machining and a high quality of finished surfaces.
The distribution of residual stress near the surface of ground ceramics is very steep. In the X-ray stress measurements of specimens with large stress gradient, the sin2ψ diagram is nonlinear because of the change in X-ray penetration depth with X-ray tilt angle ψ. The stress measured by the X-ray method is a weighted average of the stresses over the X-ray penetration depth. However, the X-ray penetration depth has been taken rather arbitrary in the previous X-ray studies of steep stress distributions. In the present study, a steep stress gradient was generated in bending of a thin plate of silicon nitride of thickness 108μm, and this applied stress was analyzed by the cosψ and parabola methods of X-ray measurement of stress gradient. The penetration depth is concluded to be infinite, and the range of the integral of the weighted average is from the surface of infinite depth. The thickness of about six times the effective penetration depth is enough to ensure the infinite range of integration. Several sources of errors in the measurement of stress gradient were discussed on the basis of numerical simulations.
X-ray stress measurements for isotropic polycrystalline are materials are usually carried out by the sin2ψ method under the assumption of no stress gradient in X-ray penetration depth. When a steep stress gradient exists in the vicinity of surface layer, however, non-linear sin2ψ relation is observed and the sin2ψ method cannot be applied on such cases. Although several X-ray stress analyzers have been developed for materials with steep stress gradient in the surface layer, it is desirable to use diffraction data at higher incident angles of ψ0 as possible as close on 90 degrees in order to determine the both values of surface stress and stress gradient with high accuracy. In the present study, an X-ray stress analyzer based on Ω geometry was fabricated to enable X-ray incidence at higher angle of ψ0. The X-ray detector was positioned on -η side against X-ray incident beam. Both of the residual surface stress and stress gradient were determined by use of the cosψ method on shot-peened steel and silicon nitride specimens. This prototype stress analyzer was found effective to perform a biaxial or triaxial stress analysis.
Three dimensional residual stress distributions in a 4 inch diameter carbon steel pipe welded joint were measured by neutron diffraction technique. The results showed that the residual stress distributed near the weld metal, namely within about 30mm. The major tensile stresses occurred in the hoop direction in the fusion and heat affected zones of the weldment, and they attained a level greater than 200MPa throughout the pipe wall thickness. While the axial residual stress at the inside surface was 40MPa, the stress at the outside surface was -100MPa. These residual stress distributions were compared with those measured by the X-ray diffraction techinique and strain gauge method, and they agreed with each other.
Incremental torsional impact experiments based on the stress bar method revealed that the over stress is uniquely related to the instantaneous plastic strain rate throughout the dynamic deformation in the strain rate jump tests for annealed pure copper. This unique relationship is formulated using an empirical equation of the power law. The precise constitutive equation is evaluated from experimental-numerical analyses of the plastic wave propagation under the incremental torsional impacts. The behavior of the incremental wave propagation is successfully simulated by the empirical equation. This equation is rewritten by a constitutive law based on the dislocation motion.
Impact tension and compression tests on ferritic ductile cast iron were performed by means of the split Hopkinson bar technique. Tensile and compressive stress-strain data for the ductile cast iron were determined at the strain rates of over 103/s and compared with those obtained at quasi-static strain rates. The test results indicate that ductile cast iron has higher stress-strain characteristics in compression than in tension at low and high rates of strain. In an attempt to examine the different strength characteristics in tension and compression, microscopic examinations of the post-test specimens were conducted with an optical microscope and a scanning electron microscope. Furthermore, the microhardness tests were carried out to estimate the strength of spheroidal graphites in a ferritic matrix. It is found that this different mechanical behavior in tension and compression is mainly attributed to the presence of spheroidal graphites in the ferritic matrix of ductile cast iron.
The relation between damage and absorbed energy in four kinds of carbon/epoxy composite laminates was studied using an instrumented falling weight impact machine. Orthotropic and quasi-isortropic laminates, which were fabricated from unidirectional or plain woven prepreg, were used to investigate the effect of lay-up sequence. The thermal deply technique was applied to determine the extent of delamination and fiber breakage at various impact energy levels. In each kind of laminate, the absorbed impact energy showed a linear relation with the total delamination area and crack area of fiber breakage. No major difference was observed between orthotropic and quasi-isotropic laminates. Thus, the effect of lay-up sequence on the relation between damage and absorbed energy was negligible.
Bending strength of a continuous carbon fiber-reinforced silicon nitride was experimentally investigated at high rate of strain by a split Hopkinson pressure bar method. Ramped incident waves were applied to the specimen to obtain a smooth load-deflection curve without high frequency oscillations. Acoustic emission during impact deformation was successfully detected as well as quasi-static deformation and the evidence of micro-cracking in the relatively early stage of high velocity deformation was confirmed. The dependences of strength and fracture morphology on fiber orientation and specimen size were clarified and their rate dependences were also identified. Tensile strength was found to be almost independent of deformation rate, while shear strength along fiber direction was significantly increased with rise of strain rate. Fiber bridging over a fracture surface affected shear strength and provided a reasonable explanation of experimental results.
In this paper, through theoretical discussion and by obtaining numerical results, the values of critical impact velocity of various materials, such as metals of various crystal structures and non-metals, have been estimated at different temperatures and strain rates, when the specimens or bars of these materials are subjected to a tension or a torsion impact with sufficiently high velocities. The theoretical approach is made by the method of characteristic curve, provided that the conditions are the same as those in the case of Karman-Duwez solution where simple waves are propagated. Then, the values of the critical impact velocities for various materials at different temperatures and strain rates are evaluated. In addition, some experimental investigations for processing thin plates are made by high speed perforation. Some of the most significant experimental results are presented.
In order to develop an impact punching technique in brittle materials, slow impact punching tests were carried out in a soda lime glass and some engineering ceramics by using a specially designed clamping device made of a frame and a compact powerful jack. Punching works were performed by using 3 types of punches; the works by conical and spherical punches are successfully carried out without serious radial cracks around the punched hole, while by flat punch, not only the radial cracks are often produced in thick plates, but the load to form a hole is quite large. Therefore, the conical and spherical punches are adopted to investigate a possible range of punching. Although hole formation load depends upon the top radius R of the punch in thick plate, the effect of R is quite small in thin plate. This trend is explained very well by the formation of a Hertzian ring crack and its growth. The load to form a hole is very high in impact punching and the rate effect clearly appeared, but no characteristic feature of impact effect has been found in the observation of punched plate.
This paper extensively studies the effects of discharging condition on properties of sprayed coating by wire explosion. Charging energy and condenser capacity are ranged from 185J to 415J and 144μF to 280μF, respectively; small condensor capacity corresponds to large circuit frequency. Large charging energy as well as small condensor capacity result in small size and larger velocity of flying particles. It is shown that the particle size distribution is mainly dominated by charging energy. Flying particles with small diameter and large velocity give fine sprayed coating surface. The optimal combination of size distribution and velocity of flying particles, however, is necessary to increase mass of deposits and to obtain strong adhesion strength with the substrate. It is found that adhesion strength between substrate and sprayed coating is mainly dominated by mechanical bonding, i.e., anchoring effect.
A simple measuring method for impact tensile strength of brittle materials is proposed and examined for plaster and ceramic specimens. The experiment is based on the propagation of the tensile stress waves reflected from the free end of the specimen bars by utilizing a modified Hopkinson bar method. The stress waves transmitting into the specimens were detected by a set of two strain gages pasted diametrically at two positions on the input bar. The measuring method and fracture behavior of the specimens are presented and discussed by means of a simple theory of one-dimensional elastic waves in a bar. The impact fracture tests are performed under various loading conditions. The statistical analysis of the impact tensile strength of the specimens by using the Weibull distribution is presented. It is found that the impact tensile strength of both materials is not influenced so much by loading rate.