This article seeks for an effect of glass fiber reinforcement in the construction of glass fiber reinforcement material. Furthermore, ultrasonic welding of F.R.P.E by batch method has been experimented, and behavior of weld under load is sought for. When the fibers are dispersed uniformly in matrix and basic unidirectional composite material is subjected to tension in the direction of fibers, law of mixture holds true on the assumption that fibers and matrix do not slip each other but receive the same strain. Same stress is transmitted to matrix as transmitted to fiber. But, since the fiber has extremely high modulus compared with matrix, elongation of matrix is much greater than that of fiber under the same tensile force. In consequence, difference in elongation of fiber and of matrix and shearing strain are considered to be accumulated and become greater and greater toward the ends of fibers. This condition is ascertained in the form of stress distribution by means of photo-elastic experiment based upon photo-elastic effect of matrix. Furthermore, the relationship among fiber length, clearance between fibers, lap length of fiber, and reinforcement effect of glass fiber is sought for by mechaincal test, as well as in correlation with photo-Blast:c experiment. Behavior of weld of F.R.P.E is examined on the other hand by mechaincal test and photo-elastic experiment. Tension test was carried out with universal testing machine with electronic tube, Autograph IM-100, and tenisle stress-strain and change in shape until failure have been observed. On the other hand, Autograph IS-2000 was combined with photo-elastic experiment device for photo-elastic experiment, with which interference fringe of double-refracted light under different loads has been photographed and dynamic distribution of internal strain has been obtaiaed. Material used for the experiment was polyethylene of low density and E-glass formed into F.R.P.E by hot press. Welding was carried out by ultrasonic welding method using Sonopet 1000 B. Relationship of L and Ts can be given in straight line, and so is relationship of Cc and Ts so long as Gc is less, than 30%. Relationship of C and Ts is on the other hand something like what is expressed by a curve of second degree, and C is desired to be as small as possible. No fiber reinforcement effect is expected unless Land Care with ineffective range. Failure in the course of tension occurs first at the tip of fiber and propagates to the side of tip. Tranasmission of stress by the effect of fiber reinforcement can be observed in photo-elastic phenomenon. In order to judge whether the welding is good or not, ratio of weld thickness serves as an effectual criterion. Welding joint efficiency of F.R.P.E was approximately 100%.
Solute distribution and solidification structure in weld metals were investigated. The welding was performed by following two methods. The first method was that bead-on-plate welding was performed on specimens after some pieces of Al-4.40%Mg alloy were arc-spot-welded on 99.999% Al sheets at intervals of about ten millimeters. The othermethod was that butt-welding was performed on 99.999% Al and Al-Mg alloys. The distribution of magnesium in the weld metals was measured by X-ray diffraction method and with electron probe microanalyser. In the weld metal which had been welded by the first method the amount of magnesium increased abruptly near the fusion boundary and nearly the same amount of magnesium existed in the region where a value of parameter y was larger than approximately thirty per cents. The values y which equal zero and one hunderd per cents represent the fusion boundary and the center of the weld metal, respectively. Although only planar growth was observed in the weld metal of 99.999% Al which had been bead-on-plate welded, subgrains were developed in the weld metal which had been welded by the first method. When the butt welding was performed, the amount of magnesium increased abruptly near the fusion boundary of 99.999% Al side and after that increased gradually. Besides, it increased abruptly also near the fusion boundary of Al-Mg alloy side and reached the amount in the base metal of Al-Mg alloy. It would be concluded from the above phenomenon that the mixing of the solute occurred considerably near the center of the weld metal but was inadequate near the fusion boundaries. When the welding is performed at a low speed the elliptical-shaped puddle is developed and the shape of the puddle changes to tear-drop shape when the welding speed becomes higher. However, the distribution of magnesium in the weld metal welded at a high speed was almost the same as the one welded at a low speed. By performing the butt welding, magnesium in Al-Mg alloy transfered to 99.999% Al. Consequently the competitive growth occurred also near the fusion boundary of 99.999%Al side and many stray crystals were observed, although neither the competitive growth nor stray crystals were observed in the weld metal when only 99.999% Al was bead-on-plate welded.
It has been emphasised by the writers that a low density grain boundary containing micro cracks or impurities maybe important in regard to the mechanism of mottling. The mottled appearance is frequently observed on radiographs of austenitic stainless steel weldments. Concerning the idea, however, there is a question how the diffraction of anx-ray beam can happen when an X-ray beam passes such a low density grain boundary. A specimen that contains weld cracks was chosen. The specimen was subjected to a radiographic examination and a mottling that originated in a weld crack was confirmed. The result obtained seems to suggest the soundness of putting emphasis on the importance of a low density grain boundary in the mechanism of the mottled appearance.
Effect of the welding conditions, such as flashing time, flashing speed, upset distance, upset time and shielding atmospheres on the formation of decarburized zone in 0.5 C steel flash welds were discussed. It was interesting to note the results of experiment that the width of decarburized zone tended to increase in upset time, whereas the change in other conditions did not affect it. Specimens that had narrow and wide decarburized zones made with flash welding were tested on fatigue strength. The main results could be summarized as folllows. (1) Both widthes of decarburized zone had no effect on the static tensile strength. (2) The width of decarburized zone could not be made to zero by welding conditions (3) The width of decarburized zone were 1.6—1.8 mm in wide zone and 0.2—0.4 mm in narrow zone. (4) The increased width of decarburized zone decreased the fatigue strength.
Temperature curve of any point on thin mother plate due to linear moving point heat is compared with that due to the heat delivered along a straight line instantenuously at t=0. Fig. 10 shows the former for points near arc starting point inform of n-X scale which are defined in equ. (1), (2) and (30). Fig. 2 shows shows the latter in form of m-t/τ scale defined in equ. (9) and (14). The curves in Figs. 2 and Fig. 10 are expressed using a common scale in Fig. 12, the relation of conversion of scales is given in equ. (32). Fig. 13 shows those of points along bead line. From Fig..12 and 13 we can clearly see the difference of temperature curves between moving point heat source and instantenuously delivered heat. It is well known experimentally that the maximum temperature of point near bead end is higher for arc stopping side compared with arc starting side. The graphical explanation is given using Fig. 14 and 15.
As the weld length is generally short in the welding of steel structures of building, welding is completed in many cases in continuous multi-pass welding sequence. Consequently, it is expected that the cold weld cracking can be prevented by depositing the succeeding weld passes within the incubation time for crack initiation. Confirming that the incubation time for cracking is affected by the welding positions, weld heat input, kind of electrode and intensity of restraint, the new method of preventing weld cracking has been established by welding the succeeding passes within the incubation time for crack initiation without employing preheating.
It was shown in the previous report that the toughness of as-solidified steel was significantly influenced by the sulfur-segregation through the Fe-C peritection reaction. For this reason, 0.1% of carbon content is a critical value for weld metal toughness. In this report, the influence of the peritectic reaction of the Fe-Ni-C ternary system on the toughness and sulfur-microsegregation of as-solidified steel were investigated. It was concluded that this ternary peritectic reaction had also the critical significance for the toughness of as-solidified steel and the nickel and carbon content of steel weld metal shoule be controlled under this ternary peritectic reaction, i.e. so as to solidify into only δ phase to get higher toughness.
Copper alloy clad steels have been produced by the hot rolling process with insert metal to prevent the interfaces to be bonded from the oxidation. Nickel is most popular for the insert of copper alloyc lad steel. But in the case of aluminum bronze clad steel, an intermetallic compound forms necessarily by the interdiffusion between alumium bronze and nickel when nickel insert is used. This intermetallic compound sometimes gives rise to the separation of clad steel during rolling, or the reduction of bond strength even if the separation does not occur. Therefore, it is necessary to avoid the formation of intermetallic compound in order to produce the clad steel. A counteract for this compound is to insrt a copper layer between the two metals. Investigation on the diffusion phenomena at the bonded region of clad steel, especially, on the formation and growth process of intermetallic compound was carried out using aluminum bronze nickel and aluminum bronze/copper/nickel diffusion couples. The following are result of this investigation. 1) The intermetallic compounds formed by interdiffusion between alumnum bronze and nickel are β' phase and α' phase which are respectively NiAl and Ni3Al type compounds both containing Cu and Fe. The β' phase is very hard and brittle. 2) Incubation time of the formation of β' ophase becomes linearly longer with a square root of the thickness of copper layer. 3) A banded α phase formed by diffusion becomes thicker with a thickness of copper just as much as the thickness of copper layer before diffusion. 4) Thickness of the α band at the incubation time of β' phase is independent of diffusion temperature but depends on the thickness of copper. Linear relationship is existing between the thicknesses of α band and copper layer. Above result will be available to the industrial production of aluminum bronze clad steel.
This paper deals with a basic reseach on the characteristics of welding distortion and residual stress of HT80 high strength steel. Single-Vee groove welds have been made in rectangular plates of the steel by covered arc welding, and a series of measurements have been made of the thermal cycles and mechanical strains at some points of the base metal during the welding and cooling as well as the various types of distortions and the macro- and micro-residual stresses in the welded plates. Some points of interest revealed are as follows: 1) The transient mechanical strains in both longitudinal and transverse directions, which are correlative to the welding stresses, can be explained well by the temperature distribution in the plate at every moment during the welding and cooling. 2) A part of fan-like elastic deformation of the base metal plate on each side of the weld line that occurs due to the thermal stress during the welding remains after the cooling in addition to the conventional contraction occurring in the weld zone, whence the transverse shrinkage distributes archwise along the weld line and the longitudinal shrinkage is larger in the outside of plate than in the weld zone. 3) The micro-initial stress of base metal distributes irregularly, and so do the macro- and micro-residual stresses in the welded plate except in and near the weld zone, though the longitudinal macro- and microresidual stresses caused by the welding process alone have the normal pattern such as tension in the weld zone and compression in the other. 4) The longitudinal plastic contraction occurs in the weld zone of about 40 mm width, where the temperature rises above about 300°C in the welding, and its comparatively small magnitude results in much smaller tensile residual stress therein than the yield point of base metal.
A pulsating tension fatigue test has been made of 80 kg/mm2 high strength steel double-strapped side fillet welded joints (weld length 1=20-80 mm, fillet weld design ratio r=Aw/Ap=0.22-0.8l) with the following results. (1) Fatigue failure characteristic of side fillet welded joint is widely different from its static failure characteristic ; the side fillet weld or the plate fails, depending on the number of cycles (N) which in turn depends on r or 1 and cyclic stress (fillet weld : τ ; plate : σ). Putting the critical fillet weld throat stress as τc, the critical plate stress as σc, and the critical number of cycles as Nc, when τ≥τc the fillet weld failure can occur (fillet weld throat stress being relatively high) and N≤Nc; when σ≤σc, the plate failure does (plate stress being relatively low) and N≥Nc; and both the fillet weld failure and the plate failure are likely to happen when τc≥(τ, σ)≥σc. Meanwhile, Nc, is the larger, the smaller r or 1 (Table 4); and using r and Nc, it is possible to express the fatigue characteristic and the static characteristic denfinitely (Fig. 8). (2) From the S-N diagrams (Fig. 5, Fig. 6), the fatigue strength of welded joint does not seem to improve very much even if the value of r exceeds 0.81. Thus, 1.0≥r≥0.8 would be desirable for side fillet welded joints in the high cycle range. From the standpoint of static performance, r≥ 1.4-1.7 is found desirable. After all, allowance for statics would assure sufficient fatigue performance. (3) For the side fillet weld, the fatigue strength at 2.106 cycles would be about 15 kg/mm2; and for the side fillet welded joint with r=0.4-0.8, the fatigue strength at 2.106 cycles (plate failure, plate stress) would be 5.5-9 kg/mm2. (4) The first fatigue crack in a side fillet welded joint is likely to initiate at the root of the side fillet weld end whee the maximum shear stress is concentrated. For this reason τc, has to be increased to improve the fatigue strength of the side fillet welded joint and the effect will be the greater, the larger the value of r. (5) The fatigue failures of side fillet welded joints have been broadly classified into four modes, which can be regionally indicated on the S-N diagram (Fig. 9).