Journal of the Society of Materials Science, Japan
Online ISSN : 1880-7488
Print ISSN : 0514-5163
ISSN-L : 0514-5163
Volume 33 , Issue 369
Showing 1-21 articles out of 21 articles from the selected issue
  • Michihiro MURATA, Kikuo WAKINO
    1984 Volume 33 Issue 369 Pages 617-627
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    Download PDF (1781K)
  • Masahiko KATOH, Shigetsune AOYAMA, Osamu FURUTA
    1984 Volume 33 Issue 369 Pages 628-633
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    A method for breaking a high carbon steel rod into billets was developed by the authors. The method consists of the three steps of forming a slight circumferential notch in the surface of the rod, developing a crack extending from the notch root toward the rod center by local heating and cooling of the surface area adjacent to the notch and breaking the rod along its notched portion by applying a mechanical load.
    The heat treatment in the second step may be either single or double heating, the latter being applicable to the breaking of small-diameter rods (less than 50mm). In the second step of this method, it is supposed that the cracks initiate and propagate owing to the axial residual stress near the notch root during the local heat treatment.
    In the present paper, after explaining the outline of the method, the hardness and residual stress distributions were measured with locally heat-treated unnotched specimens of high carbon chromium steel (SUJ2). The heat treatment conditions were selected so as to develope sufficient cracks in notched rods. And then the residual tensile stresses near the specimen surface caused by the single and double heating processes were discussed.
    The results obtained are summarized as follows.
    (1) In the single heating process, the residual tensile stress near the surface of the specimen occurred merely due to plastic shrinkage of the heated surface layer. The maximum value of tensile stress was about 40kgf/mm2.
    (2) In the case of the double heating process, the surface layer of the specimen was hardened by the first heating and cooling, and tempered to the hardness of HV 400∼550 by the second treatment. The residual tensile stress near the surface of the specimen was about 100kgf/mm2, which was never obtainable by the single heating process.
    Download PDF (1966K)
  • Koji MATSUKI, Ryoji KOBAYASHI
    1984 Volume 33 Issue 369 Pages 634-640
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    Although natural cracks such as fatigue cracks are favourable as a pre-crack for fracture toughness testing of rocks, it is very difficult to measure the exact length of natural cracks because of the opaque nature of rock and the unevenness of the crack-front.
    In this paper, by measuring the compliance of the crack opening displacement under bending, the effective length of fatigue cracks was estimated for two kinds of rocks; Ogino tuff and Tohoku marble, and the result was compared to the actual crack length on the side of the specimen. Furthermore, in order to know the fracture toughness of the rock containing natural cracks by using an artificial notch, the fracture toughness of rock specimens with a chevron-notch was measured and compared with those with another types of pre-cracks including fatigue crack and straight-through notch.
    The main results obtained are as follows:
    (1) The relationship between the observed length and the effective length of the fatigue crack depended on the texture of the rock. The effective crack length of the marble was mostly smaller than the observed one while the effective crack length of the tuff is larger than the observed one. The stable values of the fracture toughness were obtained from the effective crack length.
    (2) The fracture toughness obtained from the chevron-notch was nearly equal to that from the fatigue crack using the effective crack length.
    (3) The fracture toughness obtained from the natural cracks of chevron-notch and fatigue crack was considerably larger than that from the artificial straight-through notch because of the unevenness of the crack-front.
    Download PDF (1633K)
  • Koji NAKAGAWA
    1984 Volume 33 Issue 369 Pages 641-646
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    One of the purposes of this study is to show the relationship between the internal structural failure mechanism of concrete specimens and the confining pressure in multi-axial compression. Another one is to discuss the failure condition which may depend on the specimen confining method in loading.
    In the study, the finite element simulation method was employed and a concrete specimen was divided into finite elements. Each element represents one of the fracture patterns (cracking or sliding) when the condition (Griffith's criterion or Coulomb's criterion with tension cut-off) is satisfied in the element. The fracture development and the accumulated behavior of the elements under increasing loading were discussed.
    The numerical results gave some explanations of the fracture mechanism and failure conditions of concrete specimens under multi-axial compression.
    Download PDF (967K)
  • Chuji KAGAYA, Masanori KATO, Noboru EGAMI
    1984 Volume 33 Issue 369 Pages 647-651
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    The tufftride process is widely applied to increase the fatigue-resistance and the wearresistance of machine parts. However, their toughness is conversely reduced by the process. Therefore, the tufftrided steel is often fractured when subjected to impact load. The purpose of this paper was to make an experimental investigation with relation to the effectiveness of inorganic coating to restrain the embrittelement. The relation between the charpy impact value and the tufftriding time of the anti-tufftrided steel was examined, and the impact fracture surface was observed through SEM.
    The main results obtained are as follows;
    (1) The charpy impact value of the anti-tufftrided steel was reduced in accordance with tufftriding time. But its degree was remarkably small when compared with the tufftrided steel.
    (2) The impact fracture surface of the anti-tufftrided steel showed the ductile fracture surface and was not seriously changed by tufftriding time.
    (3) The inorganic coating can be applied most effectively to prevent the embittlment or decrease in impact value, as well as to partially prevent the tufftride process.
    Download PDF (2699K)
  • Katsumi MIZUTANI, Morihiko TANISAKI, Yoshio TANAKA, Mamoru IDO
    1984 Volume 33 Issue 369 Pages 652-658
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    It was found previously that material removal in orthogonal cutting of ceramics was characterized by the trajectory of a macro crack: at the beginning it unstably extends downwards by some distance from the region near the tool tip, then re-extends after a pause, and finally, being accompanied with material crushing ahead of the tool rake face, turns upwards to emerge out at the free surface.
    In this study the ultrasonic vibration cutting of mullite and glass ceramics (Macor) was carried out orthogonally to hasten the material crushing which helps to stop an undesired extension of the macro crack. The main results obtained are as follows.
    (1) The material crushing was accelerated by repeated impacts of the ultrasonically vibrating tool on the work being cut. A large part of the material removal was done by this crushing, and craters on the cut surface made by macro crack extension become shallow.
    (2) The material crushing was facilitated more by shorter time repeated impacts, which can be attained by adopting the proper values of frequency and amplitude of the tool vibration and the feeding velocity of the work.
    Download PDF (3609K)
  • Tsuneshichi TANAKA, Masahiko KOSUGI
    1984 Volume 33 Issue 369 Pages 659-665
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    Electro-polished plate specimens of fully annealed commercial pure iron were tested under completely reversed bending stress at 60Hz, until fine cracks were found on the specimen surface, and the orientation analysis of all the grains, more than 200, in the test area of the specimen surface was made by the etch pit method developed in the authors' previous study. The purpose was to find out the crystallographic and mechanical conditions satisfied at crack nucleation sites. All the cracks were found to nucleate at grain boundaries, and it was concluded that at least eight conditions were satisfied at these sites, with respect to grain size and geometrical configuration of preferential slip systems with grain boundary.
    Download PDF (2371K)
  • Hironobu NISITANI, Tooru TAKASE
    1984 Volume 33 Issue 369 Pages 666-672
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    Previous studies of fatigue under two-step loading have shown that the value of cumulative cycle ratio is usually smaller than unity when the first stress level is higher than the second, but in some cases it becomes larger than unity. This phenomenon may be related to the existence of work hardened layer in the specimen.
    In the present study, rotating bending fatigue tests were performed by using the annealed or the pre-strained 0.21%C steel specimens under two-step loading. The behavior of fatigue cracks was observed successively by the replication method and the effect of the work hardened layer on the cumulative fatigue damage was examined. The main results obtaind are as follows.
    (1) The Miner's rule holds good in the crack initiation process under double repeated stress in two stress levels.
    (2) On the annealed specimens:
    (a) When σ12, the value, of the cumulative cycle ratio based on fracture, n1/Nf1+n2/Nf2, is nearly equal to unity in general.
    (b) When σ12, the value of n1/Nf1+n2/Nf2 is nearly equal to unity excepting the case where the cycle ratio of the first stress level, n1/Nf1, is small. In that case, the value of n1/Nf1+n2/Nf2 is often larger than unity. This may be due to the work hardening under the first stress level.
    (3) On the pre-strained specimens:
    (a) When σ12, the value of n1/Nf1+n2/Nf2 is larger than unity.
    (b) When σ12, the value of n1/Nf1+n2/Nf2 is smaller than unity including the case where n1/Nf1 is small. This may be due to the work softening under the first stress level.
    (4) The results of the pre-strained specimens agreed with the previous results showing that n1/Nf1+n2/Nf2>1 when σ12 and n1/Nf1+n2/Nf2<1 when σ12. Therefore, the validity of the Miner's rule depends upon the existence of work hardened layer.
    Download PDF (2710K)
  • Hironobu NISITANI, Satoru ARAKI, Tooru TAKASE
    1984 Volume 33 Issue 369 Pages 673-677
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    The previous fatigue tests on annealed and electropolished 0.23%C steel specimens under two stress levels have shown that Miner's rule holds good for both crack initiation and crack propagation processes.
    In this study, rotating bending fatigue tests were carried out on age-hardened aluminum alloy specimens under constant loading or under two-step loading. The behavior of fatigue cracks was observed successively by the replication method and the differenc between an age-hardened aluminum alloy and a carbon steel was examined.
    The main results obtained are as follows:
    (1) The crack propagation curves show that Miner's rule does not hold when analyzed based on the total fatigue life. Moreover, these curves explain the phenomenon that the value of the cumulative cycle ratio, n1/Nf1+n2/Nf2, is larger than unity when σ12, but becomes smaller than unity whenσ12.
    (2) Miner's rule holds good for the crack propagation process where the crack propagetes from 0.5mm to final fracture.
    (3) Miner's rule does not hold for the initiation process of small cracks (e.g. length of 0.1mm to 0.5mm). This is different from the case of a carbon steel. This may be explained based on the characteristics of fatigue behavior of an age-hardened aluminum alloy.
    Download PDF (1290K)
  • Kazunari FUJIYAMA, Kazushige KIMURA, Masamitsu MURAMATSU, Hiroshige IT ...
    1984 Volume 33 Issue 369 Pages 678-684
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    Low cycle fatigue (LCF) tests were carried out at 839K on Cr-Mo-V forgings and castings taken from various power plant components after long-term service.
    Δεe-Nc and Δεp-Nc properties of the used and artificially deteriorated materials were compared with the virgin material.
    The effect of deterioration on mechanical properties and its correlation with LCF properties were examine.
    The results obtained are summarized as follows.
    (1) Cr-Mo-V forging and casting became soft and/or brittle through long-term heat. Material deterioration (softening and embrittlement) changed LCF properties of these steels as follows.
    i) Δεe-Nc relationship shifted downward by softening.
    ii) Δεp-Nc relationship shifted downward when embrittlement became dominant.
    (2) Vickers hardness had good correlation with Δεe-Nc coefficient C1, as long as hardness took the value from 200 to 260, while Δεe-Nc expenent α1, Δεp-Nc coefficient C2 and exponent α2 were insensitive to the change of hardness.
    (3) True fracture strain had no clear correlation with C1, α1, C2 and α2.
    (4) FATT (50% fracture appearence transition temperature) had good correlation with C2 of casting. Embrittled casting had relatively high FATT and low C2. De-embrittled casting decreased FATT remarkably and increased C2 to some extent. By fractographic observation the former showed intergranular cracking, and the latter showed ductile striation pattern. The values of α2, C1 and α1were insensitive to the change of FATT.
    As the effect of deterioration on LCF properties is complicated, further investigation is necessary.
    Download PDF (1999K)
  • Saburo USAMI, Yoshio FUKUDA, Shigeru SHIDA
    1984 Volume 33 Issue 369 Pages 685-691
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    Strain controlled time-dependent fatigue tests were carried out on plain specimens of 304 stainless steel at 550°C in air and the initiation and propagation behavior of main cracks were observed. The strain wave shapes were widely varied, including one day hold at tensile peak strain.
    The slow strain rate and long hold time in tensile strain enhanced both crack initiation and propagation, while those in compressive strain did not affect them, because the time-dependent fatigue in the test condition was dominated not by creep damage but by environmental oxidation. A good correlation existed between the period of tensile straining in one cycle τt and the fatigue-oxidation interaction characteristics under various strain wave shapes. The number of cycles to initiate 0.05mm long crack Nc0.05, the number of cycles to failure Nf and the fatigue crack growth rate were proportional to the power of τt when τt was longer than about 100sec.
    The value of Nc0.05 was frequently less than 10% of Nf and the crack propagation behavior governed the failure cycles. Although the fatigue crack growth rate da/dN was large when the crack was smaller than a grain diameter, 0.05mm, it was proportional to the effective crack length ae for large crack sizes at each test condition. The normalized fatigue crack growth rate k[=(da/dN)/ae] can be obtained from Nf of a plain specimen through the relation expressed as k=10/Nf.
    Download PDF (1592K)
  • Kenzou MIURA, Makoto KUMADA, Toshitugu OHI
    1984 Volume 33 Issue 369 Pages 692-696
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    The effect of frequency on the fatigue strength and patterns of fracture surfaces of Cr-Mo cast steel were investigated in fresh water and also in water Containing sodium nitrite inhibitor at 88°C.
    The fatigue strength in fresh water at 88°C had no fatigue limit and became markedly low at low frequency, 170cpm, in the long term tests. The fatigue strength in fresh water containing sodium nitrite inhibitor had no effect on frequency and had the same level of fatigue limit as that in air. In the long term tests, it was found that sodium nitrite inhibitor was stable and was effective for the inhibition of corrosion fatigue failure at any frequency. Fracture surfaces in fresh water were of an intergranular pattern in the crack initiation area, and of a plain and brittle pattern in the crack propagation area. However, the effect of frequency on patterns of fracture surfaces was not clear. Intergranular facets were partially observed in the crack initiation area, and ductile striations were observed clearly in the crack propagation area on fracture surfaces in fresh water containing inhibitor, which were of the same patterns as those in air. From the observation of fracture surfaces, it was found that sodium nitrite inhibitor was very effective for the inhibition of corrosion fatigue failure.
    Download PDF (2051K)
  • Hitoshi UCHIDA, Keiji KOTERAZAWA, Iwao YAMADA, Hideaki KAWABE
    1984 Volume 33 Issue 369 Pages 697-703
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    The stress corrosion cracking (SCC) of SUS 310S stainless steel having coarse grain structure in boiling MgCl2 solutions was studied fractographically by a scanning electron microscope. The crystallographic orientation of transgranular fracture mode in SCC was determined with the aid of etch-pitting technique.
    The mode of crack propagation during SCC-tests depended on the orientation of coarse grain, and the crack growth showed a delay of several ten minutes at grain boundaries, twin boundaries or crack branching points. On the other hand, the mode of transgranular fracture was characterized by a fan-shaped pattern, which was formed mostly by the crack growth along <110> direction on {100} plane. On this fracture surface, a striation-like pattern was found locally as the result of slip lines formed perpendicular to the direction of crack growth.
    The above results imply that the transgranular fracture in SCC occurs crystallographically by the interaction of sessile dislocation with corrosive environment.
    Download PDF (4052K)
  • Yasuyuki SEGUCHI, Taro OCHIAI
    1984 Volume 33 Issue 369 Pages 704-710
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    In this paper, the following problems are discussed in order to establish a design procedure based on the reliability when the crack growth yields fatigue failure; how to construct the life time distribution models and the related equations which give design values and what characteristics the models and equations have against parameter variation.
    Download PDF (812K)
  • Kiyotake MORIMOTO, Toshio SUZUKI, Ryutoku YOSOMIYA
    1984 Volume 33 Issue 369 Pages 711-717
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    Thermal and photo-degradation of glass fiber reinforced rigid polyurethane foams (FRU) with different glass fiber lengths and different expansion ratios of matrix (PUF) have been studied.
    The results obtained are summarized as follows:
    (1) The reduction in flexural strength and flexural modulus by heat treatment was more remarkable for the samples containing shorter length glass-fiber.
    (2) No significant difference was observed in activation energy of thermal degradation between FRU and PUF: about 12-21kcal/mole for flexural modulus, and about 12-17kcal/mole for flexural strength. With a few exceptions, the samples of higher apparent density tended to show slightly higher activation energy.
    (3) The similar tendency was observed also for the decline in interlaminar shear strength by heat treatment. Probably the degradation of this type arises from both the reduction in shear modulus of matrix and that in bonding strength between fiber and matrix.
    (4) The reduction in flexural properties by WOM exposure was nearly the same for FRU and PUF, but was slightly larger for the samples of higher apparent density.
    (5) IR analysis proved a decrease of urethane linkage after WOM exposure.
    Download PDF (873K)
  • Kiyotake MORIMOTO, Toshio SUZUKI, Ryutoku YOSOMIYA
    1984 Volume 33 Issue 369 Pages 718-723
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    The thermal conductivity and thermal expansion curves of glass fiber reinforced rigid polyurethane foam (PRU) were determined experimentally, and the interaction between matrix and fiber was examined to analyze the thermal expansion behaviour.
    (1) The thermal conductivity of molded FRU in the thickness direction was independent of the glass fiber content (8.5 to 9.5 volume%), and the value was same as that of matrix PUF itself.
    (2) The thermal expansion behaviours of FRU and PUF were essentially the same, but their behaviours were peculiar because the matrix had a foam structure.
    (3) The thermal expansion coefficient increased with increasing apparent density of the matrix for FRU having the same volume content of glass fiber, and decreased with increasing volume content of glass fiber or with increasing fiber length for the matrix having the same density.
    (4) Young's modulus was calculated by using the efficiency factor according to the theoretical equation of Shapery. The results agreed well with the experimentally measured tensile moduli. From this fact, it becomes apparent that the thermal expansion behaviour can be analyzed with the mechanism analogous to that of the mechanical tensile behaviour.
    (5) The critical fiber length of glass fiber was calculated theoretically from the thermal expansion coefficient according to the method of Marom. The values obtained agreed approximately with the values calculated by using the interlaminar shear strength of FRU.
    Download PDF (876K)
  • Nozomu TAKEUCHI, Katsuyuki INABE, Shoichi NAKAMURA, Hidehito NANTO
    1984 Volume 33 Issue 369 Pages 724-728
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    The effects of Cr3+ and Fe3+ impurity concentrations on the kinetics order of the first glow peak of thermoluminescence in MgO single crystals have been examined. It was found that the concentration of Cr3+ ions affected the kinetics order of the red component of the luminescence and that of the Fe3+ ions the blue one. When these impurity concentrations are relatively low, the second order kinetics is expected, while with increasing concentration the kinetics order should decrease and become the first order. Several such examples have been observed for both red and blue emissions. This may support the consideration that the Cr3+ and Fe3+ ions work as the luminescent centers for red and blue emissions, respectively. The annealing temperature before irradiation is considered to control the valence conversion of the impurities.
    Download PDF (761K)
  • Ryuichi OHTANI, Jun-ya NISHIMOTO, Ryuichi HIRATSUKA
    1984 Volume 33 Issue 369 Pages 729-735
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    A number of techniques have been used in recent years for the measurement of crack length in notched laboratory specimens during fatigue or fracture at room temperature. However, at elevated temperatures, useful and simple means of monitoring crack growth are limited. The optical measurement is the most widely used technique, and it has been used by the authors for studying creep and monitoring high-temperature fatigue cracks in ductile materials and is still being used together with a TV monitoring system. However, this method is not always advantageous. In this study, the applicability of the d. c. electrical potential method was examined to the crack initiation and propagation tests at elevated temperatures. The kinds of tests and the results of the measurement are as follows.
    (1) Creep crack propagation tests were conducted using CT specimens of the brittle aluminum alloy (2014-T6) at 150°C. Although the detection of crack tip was difficult by the optical method due to very small COD, the d. c. potential drop method was successfully used to monitor not only the crack length but also the crack initiation time.
    (2) In ductile steels, the effect of creep deformation on the change in potential drop was quite large. The increase in d. c. potential with strain was found in the creep tests of the smooth specimens of 304 stainless steel at 650°C.
    (3) For the same 304 stainless steel, crack propagation tests of low cycle fatigue were carried out at 650°C using CCP specimens with large amplitudes of push-pull load conditions. It was found that the contribution of cyclic deformation to the change in potential in each cycle was significant, and that the calibration curve must in general be determined for each test condition. On the other hand, the crack initiation time could be determined reliably by positioning the potential leads close to the notch root.
    Download PDF (1068K)
  • Yosohiro SUGIE, Chiaki SAKAMOTO, Satoshi FUJII, Masatoshi NIIHAMA
    1984 Volume 33 Issue 369 Pages 736-741
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    Factors involved in the accelerated crevice corrosion test of stainless steel in 3% NaCl solution containing activated carbon have been examined by using a separate anode-cathode system. An anode having a crevice and a cathods were coupled through a zero resistance ammeter and the short circuit current was measured.
    The agreement of the weight loss of the anode specimen with the calculated value from the current demostrated that the separate electrode system was useful. The kind of activated carbon affected the corrosion rate, and X7100 gave the maximum value among various kinds. The crevice width was also an important factor and the maximum corrosion rate was observed at the width of 7μm. The location of initial attack and its way of propagation were clarified. The crevice corrosion rate was controlled by the cathodic reaction taking place on the surface of activated carbon. By using the optimized test conditions, stainless steels of different grades were tested, giving a reasonable order of resistance. The effects of the important factors, i. e., the anode/cathode area ratio and the torque with which the crevice was tightened, obtained in the accelerated test were in accordance with those observed in 2 years tests conducted without activated carbon.
    Download PDF (1689K)
  • Katsushi SARUKI, Takashi ASANO, Shigetsune AOYAMA
    1984 Volume 33 Issue 369 Pages 742-748
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    Instead of the conventional Charpy test using a 10mm square bar specimen with U-notch of 1mm radius and 2mm depth[(a)-specimen], the Izod test using a round bar specimen has been sometimes used by the present authors, for convenience of machining and heat treatment of its specimen. The standard specimen being used by them is a round bar of 14mm diameter with a circumferential U-notch of 1mm radius and 2mm depth[(b)-specimen]. But it is frequently required to use slenderer specimens. In the present experiment, two kinds of selenderer round bar specimens of heat treated 0.55% carbon steel were investigated for their impact properties by the instrumented impact test. One was a round bar of 11.5mm diameter with a geometrically similar notch to the standard specimen[(c)-specimen]. The other was a round bar of 11.5mm diameter with a different shape notch, but the minimum diameter and the stress concentration factor of the notch were equal to those of the standard[(d)-specimen].
    The results obtained are as follows:
    (1) The individual impact values (Total, crack initiation and propagation) Ir1 of (c)-specimen were approximately related to those Ic of (a)-specimen by the equation Ir1=0.8Ic. This relationship has been well established between (b)-specimen and (a)-specimen.
    (2) In the case of impact value Ir2 of (d)-specimen, the crack initiation impact value was relatively large. The relation, Ir2=2Ic, was found to exist between the crack initiation impact values of (d)-specimen and (a)-specimen. However, the crack propagation impact value was related to that of (a)-specimen by the equation, Ir2=0.8Ic, as similar to the case of other round bar specimens.
    Download PDF (910K)
  • R. Koterazawa
    1984 Volume 33 Issue 369 Pages 749-755
    Published: June 15, 1984
    Released: June 03, 2009
    JOURNALS FREE ACCESS
    Download PDF (2666K)
feedback
Top