The effects of material, welding procedure and weldment structure are used in the sense of joinability to analyse the complex problem of centreline solidification crack initiation during welding processing. The hot cracking test procedures with externally loaded specimen, based on the theories of Prochorov and Matsuda (MISO-technique, PVR-test and Varestraint-/Transvarestraint-test), are only partly true for the assessment of centreline solidification crack initiation during welding. The measurable condition and quantifying criteria of the tests vary from one to the next. They require a complicated calculation of the theoretical function “strain-rate”. A special measuring technique was developed, tried and tested to estimate the initiation of solidification crack in flat and large components. The combination of measurement and calculation was applied to put the reliability of new derived crack minimizing methods to the test with help of the Finite Element Analysis(FEA)simulation. It takes into account the speed of cross displacement during welding processing, effected by weld component dimension and weldment assembly. The critical speed of displacement during welding processing determines the crack criterion for that moment when the local initiation of a centreline solidification crack is prevented. That is, if no hot crack is diagnosed after the welding. This is the case when the speed of displacement, released by fusion in front of the weld pool and thermal expansion during the welding process, is in equilibrium with the speed of displacement within the brittleness temperature range during solidification behind the weld pool. The centreline solidification crack initiation is assessable by comparison of the critical speed of cross displacement in the welded material (determined by experiment) to the local speed of cross displacement in the weldment (calculation by FEA-simulation). An assessment diagram for centreline solidification crack initiation during one-side welding was arrived at the multitude of results within displacement measurement during welding processing in flat and large components. It was utilized for a predictive assessment of solidification crack initiation of weldment depending on component dimension and welding procedures.
Aluminum alloys that have high specific strengths have been widely used in constructing various space structures including the International Space Station (ISS) and launch vehicles. Considering space applications, a series of welding experiments were performed in vacuum on an aluminum alloy using the GHTA (Gas Hollow Tungsten Arc) welding processes under d. c., a. c., and d. c. - pulsed conditions. Results observed in the GHTA welding experiments performed in vacuum can be summarized as follows: (1) When GHTA welding was performed using a commercial a. c. welding machine, the arc discharge did not occur during the electrode positive polarity of alternating current. As a result, it was not possible to remove oxide films from the specimen surface resulting in the cleaning action. (2) During the d. c. - pulsed GHTA process, it was possible to mechanically shatter and remove the oxide films covering the molten pool by the impact of the arc force. (3) A study has been made to determine effects of pulse current during the d. c. pulse welding on characteristics of melting and bead appearances during the d. c. - pulsed GHTA welding. The above-mentioned results show that welding phenomena observed in the experiments performed under vacuum conditions were very different from those observed in the experiments performed in the air.
In this paper is described a 2D-numerical model of short-circuiting transfer process in GMAW. Time-dependent deformation and/or detachment of the molten drop at the tip of electrode wire were simulated with aids of the numerical solution technique based on finite-difference method, which was developed in order to investigate the effect of electromagnetic force and capillary force on the kinetics of metal transfer process. It is shown that the time required for breakup of liquid metal bridge, formed between the electrode and the weld pool, depends on the drop size before the contact and/or the short-circuiting current value.
The present paper describes the effects of inner diameter on friction torque during friction stage of friction welding with mild carbon steel pipes whose outer diameter is 16 mm. The experimental conditions included inner diameter (i.e., pipe thickness) and friction speed. A conventional friction welding machine was used to join all specimens. The following conclusions are given. (1) When various pipes with different inner diameters were welded under same friction welding conditions, the friction torque curves for each pipe is different in the first phase of the friction stage, particularly in the wear stage. That is, the initial torque and elapsed time for initial torque decreased with increasing inner pipe diameter. (2) The initial torque and elapsed time for initial torque decreased with increasing inner pipe diameter when the joints of solid bars to pipes were welded under the same friction welding conditions. (3) When pipes of the same inner diameter were welded: a) the friction torque in the first phase of the friction stage varied due to sparks flashing from the welded interface, and b) the initial torque and elapsed time for initial torque increased with decreasing friction speed. (4) The result in (3) indicates that the seizure temperature at low relative speeds at the welded interface was lower than that at high relative speed.
A general way of producing pulsed alternating current waveforms for metal inert gas welding is to reverse the polarity in part of the base current period from electrode positive (EP) to electrode negative (EN). In the case of reversing the polarity from EP to EN at the right beginning of a base current period, new cathode spots tend to be produced on the surface of elongated residual molten metal that remains on the tip of a welding wire right after the detachment of a droplet. A strong cathode jet breaks the elongated residual molten metal and the fragments deposit on both sides of weld bead as tiny spatters. Delaying the polarity reversing by a time of 1.5 ms at the beginning of a base current period can effectively reduce tiny spatters because in the delay time period the molten metal on the tip of a welding wire becomes steady with a hemispheric shape. At the end of EN current period the polarity is again switched from EN to EP and a new cathode spot is occasionally produced at a location away from the molten pool on the surface of base metal. Arc voltage becomes abnormally high so that the arc length control system extends automatically the base current period. If the cathode spot moves in a direction away from the molten pool arc interruption occurs in the extended base current period. Elimination of the abnormal arc voltage before it is used for arc length control can avoid the abnormal extension of base current period and probable arc interruption in the extended period as well. If the cathode spot moves more quickly in a direction away from the molten pool arc interruption occurs in the original base current period. In this case the rate of abnormal arc voltage increasing is very high. When the rate is higher than a preset value the present base current period is terminated immediately and arc current is switched from base current to pulse peak current. Consequently, the strong arc stiffness of pulse peak current brings the cathode spot back to the molten pool or its vicinities and arc interruption is avoided.
Uniform melting of fine wires in the bundle without cutting has been studied using YAG laser. Melting and cutting behavior for the twisted bundle in 3.5 mm pitch is observed, and the applied forces to the melting portion were discussed. Furthermore, laser absorption and heat transfer in the bundle are considered as compared with these in the single wire in the terms of melting length. The following results are obtained: (1)Process from melting to cutting of the bundle is almost the same as that of the single wire. After the molten pool reaches from the laser irradiation side to the opposite side, it extends to the longitudinal direction by heat conduction. Next, it is spherodized by surface tension. Then necking occurs by the growth of spherodizing, and it comes into cutting. (2)Surface tension is the most influenced factor of the force acting to the melting portion. Gravity appears to shift the spherodized molten metal downward, but does not cause cutting directly. Recoil force of evaporation and solid-liquid interfacial tension seem to be little for cutting. (3)Laser energy is easier to be absorbed in the bundle as compared with the single wire, but the absorbed heat is more difficult to transfer in the bundle than in the single wire. (4)In order to prevent the bundle from cutting, it must be straightened without elastic strain, and the pulse energy should be selected to optimize the melting length without necking.
Clad materials consisting of aluminum and titanium plates were produced using the vacuum roll bonding method. This method produces an excellent bond without the intermetallic phase formation observable with conventional optical microscope. Bond interfacial structures in the Al/Ti clad material were observed by TEM, and the bonding mechanism were discussed. Tensile fracture occurred in the aluminum base metal of the clad material on the tensile test and the maximum tensile strength was 95 MPa. In addition, no cracks near the bonded zone were observed in the tensile fractured specimen. Nano-crystals and ultra fine-crystals were observed within the bonded zone using TEM. Within the ultra fine-crystal zone, the compositions of Al and Ti were approximately 80 and 20 at%, respectively. Precipitates those identified as a super lattice structure of the L12 type and an intermetallic phase TiAl3 (D022 type), which are nano-meter size particles, were observed within the ultra fine-crystal zone having Al solid solution (FCC crystal) supersatulated with Ti. The bonding mechanism of aluminum to titanium on the vacuum roll bonding method can be considered that very small thin regions of aluminum surface to be bonded are melted by frictional heat due to the great relative sliding between aluminum and titanium plates during the roll bonding. After that the bonded zone is cooled rapidly due to the diffusion of the heat from the bonded zone toward each base metal and rolls, therefore, the nano-crystals and the ultra fine-crystals are formed in the bonded zone.
The microstructure and grain growth behavior were investigated in the hyper interfacial bonded joint of ultra fine-grained steels. The influence of M-A constituents on hardening and/or HAZ softening in hyper-interfacial bonded joint was also examined. Microstructure of the hyper interfacial bonded joint of ultra fine-grained steels was changed melted zone, coarse-grained zone, mixed-grained zone to base metal region in order from bond interface. Martensite and bainite were formed in coarse-grained zone, and ferrite existed at bond interface. M-A constituents and fine carbides were formed in the lower temperature region of HAZ. Grain growth in joint adjacent to bond interface was restrained below 12.0μm. The maximum hardness in the hyper interfacial joint was about Hv380, and HAZ softening was suppressed by M-A constituents formed in AC1-AC3 region. Hyper interfacial bonding process was superior to fusion welding processes such as TIG, LBW and EBW from the viewpoint of the restraint of grain coarsening.
Recently, the discussion on prolonging life of steel bridges such as life cycle cost and long life bridge is widely done, and main factors of the prolonging life are countermeasures against fatigue and corrosion. As a countermeasure against corrosion, galvanizing is one of the effective methods, and many galvanizing steel bridges have been constructed until now. It is also considered that residual stress of welded portion is reduced by the annealing effect due to immersing welded member into hot dip galvanizing tank of about 450°C and the stress concentration at the weld toe as an origin of fatigue crack is reduced if a part of stress is transmitted in the galvanizing layer. These are correspondent to the two methods for increasing the fatigue strength. However, the effect of the galvanizing on the fatigue strength of welded joints has not been made clear sufficiently at the present. This study aims at investigating fatigue strength of welded joint with galvanizing. For this purpose, fatigue tests have been carried out on out-of-plane gusset welded joints which are made of 3 kinds of steel and are galvanized in several conditions. Residual stress measurements and three-dimensional finite element stress analyses for them are also carried out.
For two series of API 5L X65 linepipes (Pipes A and B), the critical condition for ductile cracking of the linepipe steel and the applicability of the critical condition to an axially notched linepipe were investigated. Static 3-point bending tests for Charpy V-notch specimens were conducted in order to evaluate the critical condition of ductile cracking from the notch tip by using FE- analyses. At the position of ductile cracking from the notch tip for the Charpy type specimens, the stress triaxiality was approx. 0.6 for both linepipe steels, however the equivalent plastic strain (εp) was different on each linepipe; the εp for the ductile cracking was approx. 0.65 for Pipe A and approx. 1.47 for Pipe B. Hydraulic burst tests were then conducted for internally patched linepipes with an axial through-wall (TW) notch. The results of the FE-analyses for the hydraulic burst tests indicated the following: 1) the position of the ductile cracking at the TW notch tip was not the center of the wall-thickness (WT), but a slightly shifted position to the inner surface from the center of WT, 2) the equivalent plastic strain at the position where a ductile crack was initiated for the TW notched linepipe was almost the same as that obtained from the 3-point bending test result for the Charpy V-notch specimen. The present study revealed that the critical strain for ductile cracking from a notch tip for a Charpy type specimen was in good agreement with that for an axially notched linepipe. It was therefore clarified that the critical condition for ductile cracking for linepipes with an actual flaw could be predicted from the results of a small-scale test and FE-analysis to evaluate the relationship between the stress triaxiality and the equivalent plastic strain at the position of the ductile cracking.
The relationship between bonding conditions and bond strength was investigated for A3003/SUS304L clad material produced using vacuum roll bonding, with observation and analysis of the bond interface conducted by means of a transmission electron microscope, and consideration of the bonding mechanism. An amorphous phase of 2–5 nm thick was observed on the A3003 side of the bond interface, as well as patch structures. These structures are thought to have been formed due to the fusion and rapid freezing of the bond interface during the roll bonding process. The direct bonding mechanism of A3003 and SUS304L in vacuum roll bonding is considered to be as follows. Due to friction heat generated during the roll bonding process, an ultra-thin layer (2–5 nm) on the A3003 bond surface fuses with high concentration regions of Cu and Si in the A3003 material. At this time, structural constituents of SUS304L diffuse into the molten layer, thus forming the amorphous layer and patch structures as a result of rapid freezing. This leads to strong bonding between A3003 and SUS304L.
A3003/SUS304L two-layer clad materials were produced using the vacuum roll bonding, and subjected to heat treatment at a holding temperature of 500°C with various holding times. Bond strength and the bond interface structure of the clad material were investigated, and were compared with those as produced. In the range of heat treatment conditions with up to 60 min, the bond strength of the heat-treated material was found to exceed that as produced. Heat treatment at 500°C induced the mutual diffusion of the structural components of A3003 and SUS304L through an amorphous layer at the clad material bond interface, thus forming a new structure. That is, the patch structure formed at the bond interface of the as produced was changed into a feather structure due to 500°C×10 min heat treatment, formed through the erosion of A3003 crystals. In the case of 500°C×30 min heat treatment, a fine columnar structure was precipitated at the interface between the amorphous layer and the SUS304L. Accordingly, the bond strength is considered to have increased by the heat treatment at 500°C.
The possibility of a welding between ADC12 and SS400 by means of friction stirring was fundamentally clarified. The primary process parameters, such as a rotating speed, rotating direction of rod and offset of pin periphery from materials interface were optimized, respectively. The welding was performed only in the clockwise rotating direction while that was not in the counter clockwise. Too large figures of the rotating speed failed in the welding due to the broken of rod itself. Optimum condition in the offset of pin periphery from the interface was given as it is on the interface or inside of the hard material. Consequently, the welded joints showed higher tensile strength than that of the base aluminum material when the welding was pursued under the moderate conditions. Intermetallic compound phase was not detected obviously at the interface region of the welded joint. Welding mechanism between aluminum alloy and steel by means of friction stirring with clockwise rotating direction was proposed.
Dissimilar diode-laser brazing of A5052/IF steel and A5052/SUS304 has been conducted using BA4047 filler metal with Nocolock flux (KAlF4+K3AlF6). The processing parameters such as laser power, wire feeding speed and travel speed were varied while defocusing distance and shielding gas (Ar) flow rate were constant. Strength of lap joints of A5052 on steels was evaluated by tensile shear test. Fracture strength of A5052/steels joints was increased with increasing laser power and reached the maximum strength more than about 80% of the A5052 base metal strength at laser power of 1300 W. Fracture position shifted from at the A5052/braze layer interface to at the steels/braze layer interface with increasing laser power. The voids and incomplete penetration of filler metal were observed at the A5052/braze layer interface when the laser power was below 1100 W. The Fe-Al intermetallic compounds were formed at the steels/braze layer interface and grew drastically when the laser power exceeded 1300 W. The superior brazability of A5052/steels was indicated at the brazing conditions corresponded to the temperature of melted droplet of BA4047 filler metal being 1050-1250 K during laser brazing process.
Feature extraction of Pb free solder fillet profiles for QFP (Quad Flat Package) lead connection is discussed for the purpose of making it easy to recognize the front fillet profiles. Kinds of Pb free solder used in the present study were Sn-3.5Ag-0.7Cu, Sn 3.5Ag-3.0Bi-1.0Cu and Sn-3.5Ag-2.5Bi-2.5In systems (contents in mass %). Au/Pd/Ni plated leads connected with Pb free solder were mainly investigated and compared with Sn-10Pb solder plated leads. Wetting angles of the top fillets in the vicinity of lead edges were measured by changing the camera angle. The wetting angle and the scattering were different between kinds of solder. The highlight area of front fillet images was, therefore, influenced by the scattering of wetting angles. The luminance distribution affected by the surface roughness of solder fillets was also measured and the scattering of luminance (highlight) was investigated to reduce errors for the front fillet pattern recognition. It was suggested that two factors; the proper camera angle to obtain a highlight area at the top fillet zone and the luminance scattering (standard deviations) were very important for developing the image processing of Pb free solder fillet. A simple image processing method (for averaging the luminance distribution and decreasing the highlight scattering)proposed in the present study was useful for obtaining the stable characteristics extraction of Pb free solder fillet profiles.