The influence of the heat input on the crack ratio (length of crack versus welded length) has been investigated using both the conventional and reverse Houldcroft type hot cracking tests for A5052 aluminum alloy sheet. In the conventional type test, where the temperature distribution around the molten pool is the main factor, the crack ratio decreased linearly when the heat input increased. In the reverse type test, where restraint is the main factor instead of the temperature distribution, the crack ratio increased when the heat input increased. This is the same tendency which is experienced actually in welding. Hence, the reverse type test is recommended as a better test. The study also showed that the crack would propagate more easily in the case of equiaxed dendrites than in the case of columnar crystals. The rotation angle at the tip of the crack was observed using.a video camera from the back side of the specimen. The rotation angle had a maximum value of about 0.08 rad near the start of the welding, and decreased gradually as the crack propagated. The crack stopped when the rotation angle reached from about 0.01 to 0.03 rad. It was made clear that the rotation angle greatly influences the propagation and arrest of the crack.
The nickel based self-fluxing alloys (S.F.A.) have an important role in the field of pressure, wear and corrosion resistant coatings. These alloys are mainly sprayed using conventional flame spraying process. An additional post heat treatment of these coatings is necessary to achieve desired properties by density the coating and improving the substrate bonding. This technique is expected to be one of the effective methods for the surface treatment. The slipping wear method was studied under atmospheric conditions for basic data, and the results were examined to see if the method was applicable to carbon steels. The material used in this study was mild steel (JIS type SS400) and bearing steel (JIS type SUJ2), and the material of the thermal spray was the self-fluxing alloy (S.F.A.). The size of ring type slipping wear specimen is 30 mm in outside diameter, 16 mm inside diameter and 8 mm in width. The constant of slipping facter is 9.09%. The contact load is 490 N and 980 N. This report shows the mechanical and metallurgical properties of the sprayed self-fluxing alloy coating after abrasion test changed the combination of upper and lower specimen used the powder flame spraying. The results are as follows : (1) In either pressure load, the wear resistance sprayed S.F.A. specimen is better than another specimen. (2) In the sprayed S.F.A. specimen, the abrasion loss of the lower specimen is bigger than the upper specimen. (3) The quenched SUJ 2 specimen is a little abrasion loss compared S.F.A. coated specimen. (4) The Vickers hardness of the substrate near the interface was changed from Hv 135 to Hv 250-500 after abrasion test. (5) The microstructure of the substrate near sprayed S.F.A. coated were squeezed after abrasion test.
Main causes that associate streaming in a weld pool are the electromagnetic, buoyancy, surface-tension and aerodynamic drag force. In this work, numerical calculations of heat and mass transfer in a stationary TIG arc weld pool were conducted considering the above four forces and their results were compared with actual penetration shapes of mild steel and steels containing active elements which were spot welded by a stationary TIG arc. The surface tension driven flow was the major factor to determine the penetration shape in a short arc process in the calculation. However, there is no experimental evidence that surface tension plays the dominant role in actual welding of various steels. The actual penetration shape is rather well fitted to the calculated shape without surface tension term. It has been thus concluded that the electromagnetic force is the major factor to determine the penetration shape and the aerodynamic drag force plays the secondary role in short arc welding. The results suggest that the value of surface tension measured in vacuum or inert atmosphere may not be the same with that of liquid exposed to plasma. While, in case of long arc welding, the aerodynamic drag force of arc plasma is the predominant force which leads to the "shallow center and deep peripheral penetration".
Molten pool phenomena of He-TIG arc welding were observed and electro-magnetic force near the anode boundary was considered. A new theory which combine the anode area, electro-maganetic force, surface tension and convection was developed. (1) On the bounadary of bright area, inner zone, was found floating slag patches, that means it is the colliding line of out- surface flows. (2) The boundary was found to be a specific line on which e-m force is maximum and its direction changes abruptly, under simple asuumtions and more elaborate one of Sozou and Pickering. (3) If the anode area coincides with inner area, the followings were considered. 1 Radial surface flow is not hindered by weak e-m force at the center. 2 Flow is suppressed near the anode boundary by the strong inward e-m force, and builds up high pressure. 3 Underward e-m force concentrates near the boundary of dual convection. 4 The liquid outside of anode is cold and is adsorbed, and desorption yields inward flow. 5 1-4 effects collaborate to drive both radial-inside and axial-outside flow. 6 R-A type binary convection and its penetration pattern was reasonably explained well through anode area. (4) Various factors influencing anode area were considered.
A novel laser beam shaping optics was developed, which can provide high shaping efficiency and high absorptivity to base metal. The optics consists of two cylindrical lenses, a pair of flat mirrors and a spherical lens. Shaping properties in terms of intensity distribution and shaping efficiency were analyzed on the basis of the geometrical and the physical optics where spatical intensity distribution and angular reflectivity of complex refractive index were taken into consideration. CO2 laser beam was shaped into a spot with uniform intensity distribution along y-axis with original intensity distribution along x-axis at variable aspect ratios, Dy/Dx, where Dx and Dy are spot size in the x-and y-axes, respectively. Shaping efficiency for different laser modes was determined as a function of angle φ between the polarization vector and the normal of the flat mirrors, and satisfactory agreement between experimental and theoretical values was obtained; negligible shaping loss was attained at φ=90 degrees in contrast to 40% of shaping loss at φ=0. Single CO2 laser beam was demonstrated to be split into 19 beam spots by utilizing the interference of the laser beam; the number of the spot can be easily changed by changing the ratio of length and distance of the flat mirrors.
This report discussed the new method to evaluate the brazed joint strength between cemented carbide alloy and sintered Cu-W joint by using the ultrasonic image. The brazing was performed in a vacum at 1213 K to 1333 K for 60 s to 1.2 ks to vary the joint strength. Although the joint strength made by filler metal containing Pd closed to that of the base metal, Cu-W, at 1213 K, it decresed at higher temperature heating and after prolonged holding time. The ultrasonic image of the brazed joint was measured through the cemented carbide alloy of 4 mm in thickness. In order to correlate a brightness of the ultrasonic image with the joint strength, the relative echo intensity, R, was defined by equation (1) R=1/16n nΣl=1 I/di/Iri (1) where Iri is reference data, Idi is sampling data and n is sampling number. R calculated from these data correlated to the shear strength. Good correlation with r=-0.86 was achieved between R and the joint strength. A microstructure in the joint cross section was investigated comparing with the ultrasonic image.
The relationships between the ferrite morphology and the fracture mode at low temperature have been studied about the weld metal of Fe-18Cr-8Ni type stainless steel used. Firstly, a new etching technique was used in order to observe the three dimensional ferrite morphology. The point of technique is that the austenite matrix was preferentially dissolved at certain potential by using a potentio-statt. The electrical potential was -0.15 volt vs HgCl electrode. Secondly, the technique was applied to fractographic observation of the fracture surface at low temperature. The results indicated that the cleavage fracture initiated at the continuous film like ferrite which caused the loss of fracture toughness. On the other hand, the dimple fracture occurred at the austenite matrix including the globular or discontinuously vermicular type ferrite.
The purpose of this study is to investigate the influence of extremely low levels of oxygen and nitrogen partial pressure, Po2 and PN2 in TIG welding atmospheres on the welding phenomenon and properties of welds of zirconium alloy tubing. In TIG welding of Zircaloy-2 tubing in welding atmospheres with various Po2 and PN2, the arc voltages were measured and the properties of welds (surface discoloration, oxygen and nitrogen contents) were examined. Although definite arc voltage change is not observed at welding in ≤ 4.1 Pa of Po2, a significant arc voltage drop with increase of PN2 is observed at welding in 0.4-16.9 Pa of PN2, and oxygen appears to inhibit this arc voltage drop. The surface of weld metal and heat affected zone (HAZ) in the atmosphere of 0.1 Pa of Po2 and 0.4 Pa of PN2 remains bright. The surface discoloration is observable slightly on weld metal and HAZ in the atmosphere of 1.1 Pa of Po2, and with increase of Po2 the initial straw color becomes darker until it gets partially blue. No surface discoloration is observable on weld metals and HAZ in the atmospheres of PN2 - 16.9 Pa with 0.1 Pa of Po2. The nitrogen content [N] in the weld metal increases linearly with increase of √PN2 and the increasing rate of [N] in inner part of weld metal is lower than that of [N] in outer part. The oxygen content [O] in the weld metal increases linearly with increase of √Po2 and shows same relations as [N], although the values of  in the weld metals fluctuate more than [N].
The depth profile of oxygen [O] and nitrogen [N] in TIG weld metals of Zircaloy-2 tubing were studied by two methods below. The calculative estimation from chemical analysis of [N] in weld metals before and after etching was applied to determine the [N] content in surface layers of the weld metals. And, the direct analysis by SIMS (Secondary Ion Mass Spectrometry) was applied also. Results obtained from the evaluation by two method avobe are summarized as follows. (1) The [N] content in a weld metal welded in an atmosphere with high level of PN2 is much higher at upper surface of it and decreases exponentially toward deep inside. (2) The [N] content at upper surface of weld metals decreases as the PN2 in the welding atmosphere reduces, and its depth profile changes until it becomes flat with weld metals welded in an atmosphere with 0.4 Pa of PN2. (3) The depth profile of oxygen [O] differs significantly from those of [N].
The internal microstructure, chemical composition, and hardness of M-A (Martensite-Austenite constituent) formed in weld CGHAZ (Coarse Grained Heat Affected Zone) of 780-980 MPa class HSLA steels have been investigated in order to reveal metallographic characteristics of the M-A constituent. The martensite was classified into lath and twin type. The massive M-A constituent had higher C contents, and included more twin type martensite and retained austenite than the elongated M-A. The cementite was found to be classified into a coarse rod type and a dendritic or fine needle type. The coarse rod type is considered to be precipitated directly from the austenite. The dendritic or fine needle type is probably precipitated by the self tempering of martensite. The C content of the M-A constituent was increased with Δt8/5 (cooling time from 1073 to 773 K) and reached 1.3-2.2% for Δt8/5=100-1000s. The hardness of the M-A constituent was increased with C content. The hardness of the massive M-A constituent (Hv-950) was generally higher than that of the elongated M-A constituent (Hv-700). The hardness and C content of the M-A constituent can be reduced significantly by a post weld heat treatment at temperatures from 623 to 773 K.
In order to elucidate the characteristics of residual stress and plastic strain produced by the dissimilar friction welding operation, detailed FEM modelling was carried out. The obtained main results are as follows : (1) During numerical modelling of residual stress and plastic strain formation in dissimilar joints, the distribution of δz (axial component) in the radial direction produced by uniform cooling from 800°C, in material adjacent to the bondline in both substrates, provides an effective means for evaluating the accuracy of the computational procedure in dissimilar materials welds. (2) Heat transfer into a narrow zone, z=±20 mm on either side of the bondline promotes residual stress formation. The peak temperature occurs in the AISI304L substrate. (3) Adjacent to the bondline, σr (radial component) is tensile in the higher thermal expansion substrate (AISI304L), and is compressive in the substrate (titanium) with the lowest thermal expansion value. σz is tensile in the central region of the component (where the stiffness is large) and is compressive at the periphery of the component. (4) The σz distribution in the axial direction, in material far from the bondline, indicates that σz is compressive in the central region of the component and is tensile at the component periphery. Specially, σz is tensile in titanium substrate, except in regions extremely close to the bondline at the periphery of the component. (5) Considerable plastic strain is produced perpendicular to the bondline and in the radial direction. Limited plastic strain occurs in the circumferential direction. (6) Severe plastic strain is produced in titanium substrate close to the bondline. This region in titanium close to the bondline has a hardness 1.5 times higher than in the base metal.
The information of the creep behavior of the thick welded joint is very important to secure the safety of the elevated temperature vessels like the nuclear reactors. The creep behavior is very complex, thence it is difficult to practice the experiment and the theoretical analysis. A simple accurate model for theoretical analysis was developed in this study. The simple model is constructed of several one-dimensional finite elements which can analyze three-dimensional creep behavior. The model is easy to treat, and needs only a little labor and computation time to predict the creep curve and local strain of the thick welded joint. It was verified by comparing them with experimental results that the model can accurately simulate the creep behavior of the thick welded joint. By using this model, the various important creep behaviors of the thick welded joint will be cleared to improve the creep characteristics of the joint.
This paper treats with the estimation on the flexural rigidity of adhesive-aided joints in steel (e.g. adhesive joint, weldbonded joint, etc.) based on experiments using three-point bending tests and calculations treating the adhesive joint as composite beams. As main conclusions for the adhesive joint, it is proved that in the case that thicknesses of both steel sheets of the joint are the same, the flexural rigidity increases with the increase of the adhesive part thickness, hence lightening weight of about 30 percent is possible under the condition that the flexural rigidity of the adhesive joint is the same as one of the spot-welded joint, moreover the width of the adhesive part has not so much effect on its flexural rigidity, and calculation results agree very well with experimental ones ; in the case that thicknesses of both steel sheets of the joint are different, the increase of the flexural rigidity using the adhesive joint is in general lower compared with the case that thicknesses of both steel sheets are the same. As main conclusions for the weldbonded joint, it becomes clear that its flexural rigidity is governed by the average thickness of the adhesive part, and decreases with the increase of numbs of spot-weld because the average thickness decreases.
Film structure is widely utilized in various products from electronic parts to daily necessaries. Evaluation of adhesion of thin film is therefore an important issue. In this paper, the applicability of scratch test to the evaluation of adhesion of thin film is discussed. Al and Cu were vacuum evaporated on glass substrates. Those specimens were subjected to scratch test in which a diamond stylus scratches film on the substrate and the resultant scratch patterns were observed by SEM and optical microscope. Scratch patterns were classified into four categories ; 1) uniformly thinned track, 2) uniformly film detached track, 3) irregularly film detached track and 4) intermediate shape of those three. Most of the scratch patterns of Al film were classified into category-1. On the other hand, those of Cu film were varying from category-1 to category-4. Category-2 corresponds the clear-track which was defined by Benjamin et. al., and shear strength of film adhesion can be calculated correctly in this case. It should be noted that category-1 and category-2 cannot be distinguished each other by the optical inspection alone, it requires careful inspection including composite analysis by SEM.
When hot pressing is used to fabricate the fiber reinforced materials (FRM), hot pressing process parameters should be selected so as to satisfy the following two necessary conditions, i.e., (1) the deformation and the adhesion of the matrix foil must be completed, (2) the thickness of the reaction layer at the matrix/fiber interface must be smaller than the critical value which is derived from the Griffith's theory. In order to predict the deformation of matrix foils, a model was proposed, where plastic and creep deformation were taken into account. The thickness of the reaction layer was predicted on the basis of the diffusion controlled mechanism. These calculations provide a consistent and unified interpretation of experimental data for Ti-matrix FRM. An algorithm of the determination of hot pressing process parameters was proposed. According to the algorithm, Ti3Al matrix FRM was formed to demonstrate the applicability of the algorithm. The tensile strength of the FRM was found to be about 1200 MPa at the temperature range from 300K to 1000K.
For investigating plasma ion nitriding process, we used two kinds of metals as an objective material ; titanium and an austenitic stainless steel (SUS304). We observed the D.C. glow discharge morphology and also investigated the nitriding situation of metal surface with X-ray diffraction analyses. We measured Vickers hardness of nitrided surfaces. Titanium was able to be nitrided if we only elevated the temperature of specimen and wherever we placed it (as cathode or anode) or exposed it into positive column of the glow plasma. On the other hand, SUS304 was nitrided only when we placed it as a cathode. It is necessary for SUS304 to remove oxide film from the surface. It is concluded that role of plasma in nitriding process is to facilitate removal of the oxide film by means of sputtering.
In order to make a further understanding of plasma ion nitriding process, we observed the relationship between the cathode fall voltage of D.C. glow plasma and the nitriding situation of metal surface. We used a stainless steel (SUS304) as a specimen, because it had the strong oxide film on the surface. If we nitrided SUS304 which was placed as a cathode then we could obtain the nitride layer without oxide on the surface of SUS304 under the voltage condition of 600-800 volts, which we selected as an optimum voltage range for nitriding. This voltage range corresponded to the conditions for absolutely sputtering the oxide surface film on metal and prevent oxygen to diffuse into bulk.
A tungsten/copper graded composite is newly developed as a new material against high heat flux, for example, plasma, ion beam, electron beam and so on. A sintering and infiltration technique is proposed for fabricating process of the tungsten/copper graded composite. The technique is very useful for combining two materials with much difference in melting point as a combination of the tungsten and the copper. In this paper, each process condition for preparating the tungsten/copper graded composite is established. It is shown that a sintered tungsten with graded pores is fabricated by sintering stacked tungsten powders from small size to large size, and that a capsule-free HIP (hot isostatic pressing) treatment of the sintered tungsten with graded pores is very useful to reduce closed pores, and that a molten copper is well infiltrated into open pores of the sintered tungsten with graded pores by HIP treatment.
Laser alloying process has been investigated to make a thick hardened alloyed layer on the surface of commercial aluminum alloy (A5052) plate using the powder mixtures of Al with 20, 50, 70 and 100 mass% Fe. The measured hardeness and wear resistance of alloyed layer have been discussed in relation to the microstructure of alloyed layer. The optimum traveling speed to obtain a smooth surface decreases with the increase in Fe content in the powder mixture. There were four types of the microstructural morphologies in alloyed layer depending on Fe content in the powder mixture and traveling speed, namely cellular dendrite, needle-like FeAl3, fine needle-like FeAl3 and lump-like Fe2Al5. The hardness of alloyed layer with needle-like FeAl3, fine needle-like FeAl3 and lump-like Fe2Al5 were HV100-200, HV300-500 and HV700-900, respectively, though cracking occurred in the case of lump-like Fe2Al5 structure. The wear resistance of alloyed layer increased with the increase in hardness. The most desirable structure of alloyed layer for wear resistance was that with fine needle-like FeAl3 having 29-34 mass% Fe.