This study presents new measuring apparatus which can be applied to in-process measurement of plume reaction force in laser materials processing. The principle of measuring apparatus is based on the load-cell device of parallel beams structure type which has a high stability and high detection sensitivity. In this paper, an effectiveness of new measuring apparatus, the relationship between plume reaction force and irradiation conditions of pulsed Nd-glass laser and target materials were investigated. As the results, it was found that the measuring apparatus developed by us appears to be useful for the analysis of plume reaction force. The plume reaction force increased with increasing output power, decreased with increasing pulse duration. Furthermore the force depended on the value of focal length, focal condition and the kinds of target material.
The force curves were measured in the ultra high vacuum (UHV) chamber of an Auger electron spectroscopy (AES) which is generally used for surface analysis. Contacts were made between a gold sphere of about 200μm radius and a silicon wafer. The specimens were sputtered by argon ion to clean their surfaces before the contact. No adsobates were observed in Auger spectra obtained from the sputtered surfaces. Contact theory for adhesion between an elastic sphere and a rigid plane is described. This theory is based on Johnson, Kendall and Roberts (JKR) theory and Takahashi theory. The total energy in this system is a sum of the elastic energies in the sphere and the stiffness for the measurement system, the change of the surface energy within the contact region and surface-surface interaction outside of the contact region. The force curves are determined by the stiffness and the characteristic distance parameter of Lennard-Jones potential. Experimental force curves were compared with the theoretical curves. Experimental force curves were influenced by vibration of experimental apparatus. It was suggested that surface-surface interaction was negligible under the order of radius of a gold sphere used to measure. Surface-surface interaction does not become negligible as radius of sphere becomes small. The developed measurement system reveabled informations of interatomic interaction and surface roughness.
The characteristics of the GHTA (Gas Hollow Tungsten Arc) welding of aluminum alloy in a vacuum are shown in the present paper. The results are summarized as follows, (1) There is a shifting phenomenon of arc discharge in which transient arc discharge shifts via the transitional arc discharge to the stationary arc discharge under GHTA welding in a vacuum, and the characteristics of current and voltage of these discharges are shown. (2) The period of transient arc discharge could be shortened by increasing the flow rate of argon, using the 2%La2O3-W electrode and/or by decreasing the initial current. (3) The surface of electrode and collet body is eroded by the vaporization of the material during the transient arc discharge. The amount of vaporized material increases linearly with increase of the period of transient arc discharge. (4) Taking account of the repair of the space station and space structures which may be made of an aluminum alloy with a thickness about 3-4 mm, the arc current at 50-70 A may be enough to weld aluminum alloy of plate thickness 3-4 mm by the GHTA method in a vacuum. (5) The characteristics of hardness distribution and strength of joints of aluminum alloy by GHTA welding in a vacuum are clarified.
An experimental study was performed examining the sources of temperature variability of free burning arcs by some spectroscopic measurements. In this paper, quantitative spectroscopic methods, shown as follows, were applied for temperature determination of pure argon or 90% argon -10% hydrogen gas tungsten arc. (1) Fowler-Milne method, (2) Method of relative intensity ratio of ArII to ArI line, (3) Method of relative intensity ratio of ArII to ArII line, (4) Method of relative intensity ratio of ArI to ArI line, and (5) Method of relative intensity ratio of Hβ to Hβ line. In method (3)(4) or (5), it was shown that a slight difference of data in measured spectra intensities or tansition probabilities caused remarked one of temperature. Temperature distribution by method (1) or (2), howevewr, was not remarkably affected by those differences. In pure argon arcs, it was concluded that temperature distributions measured by all methods (1), (2), (3) and (4) approximately coincided with each other. This conclusion led that the great variability of temperature data in argon arcs published previously was not necessarily valid. It was found in 90%Ar-10%H2 arcs that the temperature distributions by methods (2) and (3) were in good agreement, whereas those distributions were different from both distributions measured by ArI intensities and by HI intensities in methods (1), (4) and (5).
In the present work, a three-dimensional numerical model for circumferential GTA welding of pipes is developed. It consists of a heat conduction model and a molten pool balance model. In the heat conduction model, the transient temperature distribution and the molten pool size have been numerically analyzed by using a finite difference method and enthalpy method in order to take the influence of latent heat into account. In the molten pool model, using the molten pool size derived from the heat conduction model, the transient theoretical configuration of molten pool is calculated, considering of the balance of gravity and surface tension. The model is classified into the following two models. One is the rotational pipe model and another is the fixed pipe model. The influence of material properties and pipe dimensions on welding process has been discussed on the basis of numerical analysis. As a result, it is made clear that the molten pool in fixed pipe welding is more sensitive to a burn-through than the pool in rotaional pipe welding.
The rail is a typical steel material of the high-carbon steel and the improvement of weldability is an important technological problem. In narrow gap gas shielded arc welding of rail, effects of process parameter, carbon content in weld metal and post heat treatment were examined by observation of defect and microstructure, hardness, tensile tests of welded joints. By adjusting temperature of the water-cooled copper block surrounding the rail welds, it is possible to control heating to the base metal and thus to improve the weld quality. With post heat treatment and suitable welding condition, properties of welds are improved. Control of carbon content in weld metal is considered to be a factor of hot crack prevention. The MnS type inclusion exerts the influence on the hot cracking in the heat-affected zone and it is possible to prevent hot cracking by applying the suitable process parameters.
New laser carburizing quench process using hydrocarbon gas as the assist gas was proposed to strengthen ultra low carbon steel sheet. In this paper, the effect of CH4 concentration and various laser radiation parameters on hardness and area of fusion zone was investigated. Tensile strength of laser treated specimen were measured. The results of the present work were summarized as follows, (1) By using CH4, fusion zone with vickers hardness of more than 400 was obtained. (2) Hardness and area of fusion zone were well controlled by CH4 concentration, laser power, focal position, beam scanning rate and so on. (3) Tensile strength of laser treated specimen was about 20% higher than that of base metal. (4) CO2 laser carburizing apparatus consisting of the nozzle head and the exhaust of CH4 was newly developed, which enabled to strengthen ultra low carbon steel without breaking focal lens and the combustion of CH4.
A model (algorithm) for the interfacial contacting process in the solid state diffusion bonding has been developed to establish the support system for prediction of the solid state diffusion bonding. By taking into account the change in the void spacing during bonding, the model was updated from the existing model which Takahashi et al. had proposed. The method (overlap process) for estimating the void spacing on the bond-interface is firstly proposed. Four fundamental bonding mechanisms are introduced in order to update the bonding model. The calculated results are compared with experimental ones and they agree well. The present model predicts the latter stage of bonding (the percentage bonded area greater than 50%) which largely depends on the void spacing, although the time required to complete the full contact has not been exactly estimated.
The algorithm for predicting the intimate contact process in the solid state diffusion bonding has been developed. The numerical model proposed in our previous study (Part 1) was updated to predict the void shrinkage process in the final stage of bonding. The influence of dispersion of void spacing on the contacting process was examined based on the calculated results. The plural measurements of surface profiles was performed to estimate the variation of the bonding process. The distribution of void spacing estimated by the overlap method was also examined. It was useful to take into account the largest void spacing and the largest void volume for the prediction of the final stage. The algorithm was verified experimentally. It was indicated that the algorithm predicts the time taken to attain the full contact.
An algorithm for estimating the bonding process variation is developed to establish the selection system for the optimal solid state bonding conditions. The bonding process variation (dispersion) due to surface roughness is discussed by applying the statistical analysis. The distributions of void spacing and void volume are estimated by the overlap method which was proposed in our previous paper (part 1). The standard deviations are calculated from the distributions. The numerical calculations are performed by taking into account the standard deviations. Also, the hazard rate is newly introduced to comprehend the bonding dispersion. Experimental bonding tests suggest that the algorithm developed in the present study can predict the bonding process variation.
An in-process monitoring system was developed to monitor CO2 laser welding quality with two photo sensors set at different aiming angels to detect long-term power stability as well as underfill and pitting defects, and has been used in production line for tailored blank welding of automotive parts since 1993. Stability of the laser power can be monitored by long-term recording of the average intensity of the plasma emission. The underfill can be recognized by decrease in the light emission detected by the low angle sensor, since the hot core of the plasma plume is hidden by the surrounding material. Small pits can be detected by sharp decrease in the light intensity recorded by the high angle sensor.
A low energy γ-ray source, Ytterbium-169 (Yb-169), has been produced in the Japan Atomic Energy Research Institute to apply to the industrial fields in the near future. In this paper, the results of experiments on the characteristics of the source for industrial radiography and the image quality of a radiograph including calculation were described. The used material is the stainless steel, SUS 304, up to 12 mm in thickness. The number of the perceptible wires of the JIS type image quality indicator using Yb-169 is smaller by 4-5 for the specimen of 1 mm in thickness, 2-3 for 3 mm and 1-2 for 5-12 mm than that perceived using Iridium-192 (Ir-192).
Macrostructures which are usually developed near the center of weld metal of aluminum alloys are columnar crystals and equiaxed dendrites. The development of the macrostructures depends on the welding conditions. A feathery crystal is sometimes developed when welded by GTA in a low welding speed. In this study, we investigate characteristics of the feathery crystal using both a scanning acoustic microscope (SAM) and a transmission electron microscope (TEM). It is possible to observe structures without etching by SAM. By observing normal and parallel section to the welding direction, we made clear the growth feature of the feathery crystal. When we measure the velocity of elastic surface wave of the feathery crystal using SAM with an anisotropic acoustic lens, the velocity is about 100 m/s faster in parallel to the welding direction than in normal direction. That is, the feathery crystal is acoustically anisotropic. We could observe the feathery crystal even in electron beam welds using SAM. It is not generally easy to observe macrostructure of electron beam welds, because the bead width is narrow. Moreover, we confirm that the feathery crystal is a twin by analyzing electron diffraction patterns obtained using TEM.
The sensitization behavior in the weld metals of Type 308 stainless steel was theoretically studied in order to determine the heat-treating condition to simulate low temperature sensitization (LTS). The theoretical parameter ΔS was introduced to evaluate the amount of chromium depleted zone adjacent to the grain boundaries. Calculation of ΔS has shown that the value of ΔS increased at first and then decreased from a certain time, which was corresponding to the deterioration of intergranular corrosion resistance and its healing in the weld metal of stainless steels during sensitizing heat-treatment. Moreover, the value of ΔS was found to agree fairly well with the degree of sensitization measured by EPR test. On the basis of the results of calculating ΔS, the sensitizing condition to simulate the low temperature sensitization was considered. The typical sensitizing condition was determined 773K × 30 ks, which is corresponding to the condition of 573K×1.26×109s.
The temper embrittlement arising in the heat affected zone (HAZ) of 1 1/4Cr-1/2Mo steel was studied in the tempering temperature-time range of 775 to 900 K and 5 to 10000 hours. The temper embrittlement was detected as the rise of energy transition temperature. Three types of embrittlement were recognized; they are the second, third and fourth types of embrittlement which have been observed in 2 1/4Cr-lMo steel. The third type of embrittlement is same as the reversible temper embrittlement (RTE), but other two types are new ones observed by the authors. They are characterized by their own shapes of transition curves and appearances of the fracture surfaces. The third and fourth types in this steel appear later than in the case in 2 1/4Cr-1Mo steel. The third type in this steel continues to appear at a higher temperature than in the case in 21/4Cr-1Mo steel. The influence of stress in the tempering process was examined. It increases the magnitude of the third type of embrittlement, but does not affect the second type. Examinations were also made on the SR-treated specimen. SR treatment turns the HAZ specimen into the de-embrittled state, but reheating it at 825K induces two different types of embrittlement. One of them is the third type of embrittlement (RTE) which exhibits the intergranular fracture. The other type exhibits the cleavage fracture; it is not yet known whether it belongs to the third type or not.
The behavior of elements composing ruby and brazing filler metals at the brazing interface was examined by EPMA, and the construction of reaction layer formed at the brazing interface was examined by X-ray diffraction method, in order to investigate the corrosion phenomina of ruby by active metal brazing filler (Ag-Cu-Ti). The corrosion depth of ruby brazed under various conditions were evaluated by measuring surface roughness, in order to examine the relation between the corrosion depth of ruby and brazing condition. The examination of brazing interface by EPMA showed that the reaction layer was constituted by two distinct layers with different titanium content. According to the X-ray diffraction method, the reaction layer formed adjacent to ruby was mainly constituted with TiO. Aluminum made from corroded ruby diffused outside of TiO-layer and formed the secondary reaction layer constituted by Ti3AI, Cu2Ti4O, AICu and CuxTiy. These experimental results showed that ruby was reduced by titanium contained brazing filler metal. According to the corrosion test using Ag-Cu-4 mass%Ti, the isothermal corrosion depth of ruby increased proportional to square root of brazing time in the temperature range of 1123 K to 1223 K. The temperature dependence of the rate constants fits an Arrhenius equation yielding activation energies of 224.2 kJ/mol.
The oxidation and recrystallization behaviors of Ni-base single crystal superalloy, CMSX-2 were investigated to determine the condition of the preparation for TLP-bonding operations. The Paying surfaces of CMSX-2 were worked by the shot peening, fine cutting and mechanical polishing treatments and the degree of working of treated surfaces was evaluated by the hardness test and X-ray diffraction method. CMSX-2 was heat-treated at 1173-1589 K for 3.6 ks in vacuum of 4 mPa. The mechanically polished surface was slightly oxidized after heat treatment even in the vacuum atmosphere of 4 mPa. The thickness of an oxide film increased with increasing the heating temperature and the surface roughness of the faying surface. Recrystallization occurred at the surface after heat treatment at above 1423 K when the hardness was increased more than Hv600 by the shot peening treatment, while the mechanically polished or fine cut surfaces didn't recrystallized. Based on these results, it was clearfied that the mechanically polishing with fine abrasive grit could be used for the preparation of faying surface of CMSX-2 before bonding operation.
The dissolution of base metal and isothermal solidification behaviors of Ni-base single crystal superalloy, CMSX-2 during transient liquid phase (TLP) bonding was investigated employing MBF-80 and F-24 insert metals. TLP-bonding of CMSX -2 was carried out at 1373-1548 K for 0-19.6 ks in vacuum. The (001) orientation of each test specimen was always aligned perpendicular to the bonding interface. The dissolution width of base metal at the bonding temperature increased when the bonding temperature and holding time were increased. The dissolution phenomena of base metal into melted insert metal could be expressed by Nernst-Brunner's theory. The eutectic width diminished linearly with the square root of holding time during isothermal solidification process. Borides were formed in the bonded layer during TLP-bonding operation. The amount of microconstituents in the bonded layer after isothermal solidification decreased with bonding temperature, and the microconstituents in the bonded layer disapperaed at the bonding condition of 1523 K × 1.8 ks when MBF-80 insert metal was used.
The authors have applied the laser speckle method to the in-situ dynamic strain measurements during welding. The principal objective of this report is to detect the strain behavior during phase-transformation in welds by using the method. Specimens were thin plates with 4 mm thickness of 9%Ni steel and SUS304 stainless steel, the former of which undergoes martensitic phase-transformation at about T=325°C in the cooling period whereas the latter does not. A specimen was heated by a moving GTA along its center line. A laser beam illuminates a spot just below the heating line on the bottom surface of the specimen and the speckle patterns formed on a pair of linear image sensors were continually recorded. Several kinds of specimen in different shapes were adopted to compare the effects of the restraint on the strain during phase-transformation. The laser speckle method could detect the strain behavior during phase-transformation successfully. Differences in the strain curves due to the direction, i.e., along or perpendicular to the heating line, were clearly detected. The amount of expansion during transformation differed considerably according to the heat input, measuring direction or specimen size. In some cases where expansion in one direction is strongly restrained, expansion in the other direction exceeded greatly the value in the case of a stress-free sample. Such anisotropic strains due to phase-transformation were qualitatively in agreement with the predictions by an elasto-plastic finite element analysis.
The numerical results of stable crack growth simulation (3D FE analysis) using equivalent plastic strain showed good agreement with the experimental results irrespective of the heterogeneous specimen presented in the previous papers. From the results, it can be seen that the equivalent plastic strain is a characteristic value that is independent of the existence of strength mis-match in ductile fracture process. However, in order to obtain the equivalent plastic strain, it is needed to perform 3D FE calculation. On the other hand, the local displacement obtained from the experimental measurement is beneficial regardless of numerical simulation. The local displacement is influenced by the measurement distance from the initial crack tip. The value of local displacement measured at the shifted location of crack tip does not change with an increase in stable crack growth. A similar relation between equivalent plastic strain and stable crack growth was also observed in the previous publication of authors. The local displacement measured at the mid-thickness is independent of the existence of strength mis-match. While, in practice, the local displacement is measured on the surface, discussion is about the effect of the strength mis-match on the ratio of local displacement at the surface to its value at the mid-thickness. The strain estimated by the local displacement can be used as a substitute for the equivalent plastic strain. Moreover, it is expected to express the stable crack growth resistance of mis-matched joint in terms of the local displacement measured at the surface of specimen.
In car body, many types of high strength steel sheets are used. Because the strength of many body parts are mainly dependent on the strength of spot welded joints, estimation of joint strength becomes important. JIS Z 3136 prescribes tensile shear test method as the representative test for spot welded joint, and tensile shear strength (TSS) has been regarded the most important property for thin sheet structure. Up to this time, many experimental formula are known to estimate TSS from factors of tensile strength of base metal, sheet thickness and nugget diameter. However, in those estimation the applicable range was limited, and there was no way to break the limit. The reason was seemed that there was a few consideration to roles or meanings of above three factors on TSS. Then, in the present study the new estimation are tried to lead by a simple fracture model which has practically satisfactory precision to low carbon steel sheets used for car body, and which has a appreciate easy form related above factors. The estimation formula is as the following. TSS=2.05t·TS(1+0.0059EL) (ND+2.09). TSS is in N, t is thickness in mm, TS and EL are strength in N/mm2 and elongation in % of base material respectively and ND is nugget diameter in mm. The value EL was the factor concerned with the width of stress distribution. Fracture was occurred outside the hardened region, then TSS was not in proportion to ND simply.
The recently proposed method based on the inherent strain theory has proved to have better accuracy than those traditional methods proposed by Stablein, Daridenkov et al. Some layer removal methods tend to produce additional strain or stress in specimen which leads to estimated residual stress error. In the first part of this work, the inherent strains near the machined surface of a mild steel specimen, produced respectively by fraise milling, shaping, electric spark machining and emery paper wet-polishing, were examined. In the second part of this work, electropolishing, wet-polishing and fraise milling were employed to measure the residual stress distribution in a clad plate respectively. The following conclusions are obtained: (1) Electric spark machining can not be used as a layer removal method. The heat produced in the machining not only introduces additional inherent strain in the near-surface, but also tends to change the original inherent strain distribution through a large thickness into the specimen. (2) Fraise milling and shaping produce large additional inherent strains which lead to large estimation error. These methods can be used only when the mechanically strained layer is furtherly removed by electro-polishing or wet-polishing. (3) Wet-polishing by emery paper produces only a negligible inherent strain and therefore is a reliable method for layerremoval like electro-polishing.
In order to investigate the corrosion fatigue characteristics of the coated structure materials, the rotating bending fatigue tests were conducted for carbon steel specimen with Zn, Al and paint coated layer partially under accelerated corrosive environment cycled of salt water spraying, drying with warm air and holding at room temperature. This report shows the mechanical and metallurgical properties of the coated specimens with Zn, Al and paint after corrosion fatigue tests. The results are as follows: (1) In the thermal sprayed and painted specimens, the fatigue limit under corrosive environment is the same as that of the blasted specimen tested in air. (2) The specimens partially sprayed with axial and circular direction had the almost same fatigue limit as the sprayed specimens under corrosive environment. (3) In the macrostructure of the fatigue fracture surface, the crack propagation have been observed on the specimen non-sprayed circular direction. Furthermore, the micrography of the fatigue fracture surface shows mainly dimple and striation-like pattern.
A smoothing of fillet toe results in an improvement of its fatigue strength, but at the same time it seems to develop its liability to root failure occurrence from root defect. The purpose of this study is to investigate the relation between root failure and stress concentration and to find the control condition of root failure on the fatigue of a toe-smoothed non-load-carrying type cruciform fillet welded joint. In fatigue tests, size of fillet (s) are 5-23 mm. After welding, toe are TIG-dressed and circular holes are made at the root of fillet as defect by electrical discharge machining. Diameter of circular holes (φd) are 0.3 and 0.6 mm. The maximum stress of fatigue cycle was set at yield stress of the material (S=σmax-σmin, σmax=σY). As the results of finite element analyses and fatigue tests, it was found that root failure could be prevented in the ratio of about s/d≥53-33 from a point of view of stress concentration factor (in practice: s/d≥36-13) in d=0.3-0.6mm, but fatigue strength decreased on excess size of fillet on the contrary.
In this paper, the J-integral for an interfacial crack by thermal stress that developed during the cooling of the dissimilar material joint is studied. The thermoelastic analysis of J-integral for the bonded joint with an insert layer is performed using the finite element method. The effect of insert layer on the J-integral under the thermoelastic condition is discussed. It is found that the J-integral have three types of variation with an increase of insert layer thickness. The variations of J-integral with an increase of insert layer thickness are dominated by the nondimensional parameter (αC-αB)/(αA-αB) that expressed the relation in magnitude of coefficients of linear expansion and the ratio of Young's modulus (EA/EC). As a result, it is possible to predict the variation of J-integral with an increase of insert layer thickness by using these dominant parameters.