As a study relative to the evaluation of corrosion characteristics in high temperature water of welds of ZrTN802D alloy (Zircaloy-2) tubes, continued investigations were made on the depth profiles of nitrogen (N) and oxygen (O) absorbed in HAZ by TIG welding in various atmospheres. A calculation was attempted to estimate N depth profiles in a portion heated to β phase temperature range in HAZ. Although the calculated depth profiles were somewhat different from those of SIMS analysis values at the portions nearer the surface, the calculated value was seen to decrease rapidly toward the value of N in base metal with increase of depth. It was also found out by additional chemical and SIMS analyses of N after surface etching of the welds that the depth of diffusion of N in HAZ was not deep under this welding condition. O was absorbed and diffused in the region heated within a phase range in HAZ, and it was also confirmed that the diffusion depth was shallower.
Large heat-input welding is increasingly employed because of its high productivity. For this welding, titanium and boron (Ti-B) bearing welding materials are suitable in acquiring high toughness at low temperatures. However, their properties at high temperatures have not been known. Preferable ductility at high temperatures is required when they are applied to the welding of fire-resistant steels. The present study examines the mechanical properties of Ti-B bearing welding materials for fire-resistant steels in large heat-input welding. Submerged arc welding with two electrodes in tandem was conducted with the heat-input of 13 kJ/mm. The mechanical and metallurgical investigation were v-notch Charpy tests, weld metal tensile tests, chemical analyses, and metallographic observation on ruptured surfaces. It was found that the addition of titanium and/or boron lowers the high temperature elongation. This reduction was caused by grain boundary fracture at the prior-austenite grain boundaries of the weld metal with titanium and boron. However, the high temperature properties were recovered in the weld metal with titanium as low as 30 ppm and nil boron, and this weld metal fractured in an intragranular mode with enough toughness.
The friction surfacing method entails the rotation of a filler rod under an axial load applied on a surfacing plate, which in turn gives rise to the generation of heat at the rotational plane and a softened rod metal is deposited as a surfaced metal on the plate. The surfaced metal of fixed thickness is attained during the heating hold process, then the plate moves laterally so that the surfaced metal is continuously deposited on the moving plate just like a bead of the arc welding. This method can be used for repairing local or surface damage of the work. After a detailed observation of the plastic flow at the friction surfacing parts using a tracer, this report revealed that metal at the central part of the filler rod formed the surfaced metal and one at the periphery did the flash. At that time, the central part of the rotational plane was travelling to the filler rod so that the surfaced metal was furishing. Then an author defined this central part of the rotational plane as the real rotational contact plane. The characterists of the real rotational contact plane was also revealed in subsequent experiments that the area of the real rotational contact plane decreased with increasing rotational speed and friction pressure. Calculations for the diameters of the real rotational contact plane were carried out from the mass of the surfaced metal and the length of the consumed rod.
In order to develop material for Divertor of Nucler Fusion Reactor, we investigated bonding method between C-C material (Carbon Fiber/Carbon composite) and three kinds of heat sink materials, the microstructure of the bonding layer, shearing strength, thermal shock resistance temperature and analysis of thermal stress. C-C material was manufactured with carbon fiber oriented in one direction. Thermal conductivity of C-C is 570 W/m·k along this orientation. For heat sink material, three kinds of material, Cu, Cr-Cu alloy and W-Cu alloy were used. Results are summarized below ; (1) As a brazing filler metal for bonding between C-C and copper alloy, ten kinds of brazing filler metals were investigated. As a result 2Ti-AgCu paste filler metal was selected. The brazing filler metal is excellent in wettability and as high shearing strength as a bonding layer. (2) Shearing strengths of the C-C/copper and copper alloy joints with 2Ti-AgCu paste brazing filler metal decrease in the order of C-C/Cu, C-C/Cr-Cu, C-C/W-Cu. Hardness of the bonding layer decrease in the order of C-C/W-Cu, C-C/Cr-Cu and C-C/Cu. Therefore, the shearing strength of the bonging layer increase with decreasing hardness. (3) By using a copperplate of 2-5 mm in thickness as a bonding layer between of C-C/Cr-Cu and C-C/W-Cu joints, shearing strength and thermal shock resistance temperature are improved to those of the C-C/Cu joint. Through thrmal stress analysis, the thermal stress relief of the copper plate was confirmed.
The purpose of this study is to investigate the effect of Aluminum in Zn coating on electrode life of galvanized steel sheet. Three hot-dip galvanized and one electro-galvanized steel sheets were prepared for this study. Aluminum content in coatings was changed from 0.26 to 0.87 mass% among three hot-dip galvanized steel sheets. The approach used here included EDX, AES analysis of the coating layers, electrode life tests and EPMA analysis of electrodes after 900 welded. The obtained results are as follows: (1) Electrode lives of HDG materials were changed between less and more than 0.3-0.4 mass% Al content in Zn coating. Materials with low-Al coating content showed over three times longer electrode lives than materials with high-Al coating content. (2) Although thickness of Al oxide layers was in proportion to Al content in Zn coatings, the obvious correlation between electrode life and thickness of Al oxide layers was not observed. (3) In case of low-Al coating content, it was observed that Fe-Zn alloy grew from the steel-coating interface to Zn coating. It was considered that, in the initial stage of welding, the content of Fe in Zn coating increased immediately. (4) In case of high-Al coating content, Fe-Al Alloy was observed at coating-steel interface instead of Fe-Zn alloy. It was known that Fe-Al alloy suppresses Fe-Zn alloying reaction. Zn coatings was not alloyed with Fe on initial stage of welding. (5) From these results, it was concluded that Aluminum in coatings affected electrode life by changing the melting point of coating layers between electrode and steel. Melting point of low-Al content coating layer rose because of the diffusion of Fe into Zn coating. This phenomenon decreased electrode wearing and electrode life was long. In contrast, melting point of high-Al content coating layers remained low. This phenomenon caused electrode alloying easily and also increase electrode wearing. As the result, electrode life became shorter.
With the aid of ultrasonic vibration, brazing of SUS304 austenitic stainless steel using flux and silver-based filler metal was carried out, and the effect of the ultrasonic vibration on the tensile strength and the solidified structure of the joints was investigated. The following results were obtained in this study. The ultrasonic vibration applied during isothermal holding and during cooling from the brazing temperature of 700°C is significantly effective to increase the tensile strength of the. brazed joints, independent of the joint clearance and applying time of the ultrasonic vibration. Grain refinement of the solidified structure is achieved by the ultrasonic vibration, followed by increasing in the strength of the brazed joint with the joint clearance of 0.4 mm. The ultrasonic vibration accelerates the crystallization of Cu rich solid-solution and makes the Cu content decrease in the eutectic matrix of the brazed layer. As a result, the eutectic matrix is more easily strengthened by the tin which tends to form a solid solution with copper, followed by increasing in the strength of the brazed joint.
In the first step of the study, liquid surface around the ternary eutectic composition of Al-Ge-Cu system was surveyed by means of DTA to determine compositions of fillers having melting points lower than the solidus temperature of 2017 alloy. Auger electron spectroscopy (AES) analysis on 2017 alloy were also carried out to investigate the surface condition during the heating in brazing process. Based on the results of aforementioned investigations, Al-Ge-Cu-Mg and Al-Ge-Cu-Si-Mg fillers were developed and employed in brazing experiment. Brazing was carried out in a vacuum of less than 2×10-2 Pa at 773 K for 7.2 ks followed by diffusion treatment (solid solution heat-treatment of T4) at 773 K for 7.2 ks, 173 ks or 605 ks. Brazed specimens were then subjected to tensile test to evaluate the strength of the joints. 67% of joint efficiency was achieved in the specimen brazed with a Al-Ge-Cu-Si-Mg filler and diffusion treated at 773 K for 605 ks. Electron probe micro-analysis (EPMA) performed across the brazed joints revealed that longer diffusion treatment promotes homogenization in microstructure between the filler and the base metal, which homogenization will account for the higher strength in the joints with longer diffusion treatment.
This paper deals with the resistance spot weldability of an aluminum clad steel sheet to a steel or an aluminum sheet, and the dissimilar metal joining of a steel to an aluminum sheet with an aluminum clad sheet as an intermediate layer. The clad sheet was produced by hot rolling of steel and aluminum sheets with a direct resistance heating process. The mechanical properties of the clad sheet were between those of the steel and the aluminum sheet, and the peel strength was satisfactorily high. Materials used in spot welding were a 0.8 mm-thick EDDQ steel sheet, a 1.0 mm-thick aluminum-magnesium alloy sheet and the 0.8 mm-thick clad sheet which has a thickness ratio of aluminum/steel 1.1. The spot welding of the steel sheet to the steel side of the clad sheet and the aluminum sheet to the aluminum side of the clad sheet were successful. The spot welding of the steel sheet to the aluminum sheet by using the clad sheet as intermediate layer was also successful. However, thin intermetallic compound layer was formed at the steel/aluminum interface of the clad sheet due to the spot welding thermal cycle. The suitable welding current ranged between the values suitable for steel to steel and for aluminum to aluminum. The tensile shear strength and the U-tension strength of steel-clad-aluminum joints were higher than those of steel-aluminum direct joints and close to those of aluminum-aluminum joints. These results from mechanical tests suggest that the existence of intermetallic compound layers at steel/aluminum interfaces of the clad sheets were not detrimental to the strength of the steel-clad-aluminum joints. The fracture mode of these joints varied with the welding current.
In off-line teaching of a robot, motion data is generated on model space in a computer. The data is then to be converted to the coordinate system in real working space. This paper presents a simple stereo method for this coordinate transformation. The stereo method runs as follows. Each stereo image is divided into small rectangle sub-areas for fast feature extraction. In each area, a local direction histogram (LDH) is calculated from intensity gradient. By the shape of LDH, sub-areas which include vertices of the object are selected as feature areas. Then 3-D location of vertices are calculated by stereo-corresponding of these feature areas. The transformation matrix between model space and real working space is obtained by selecting corresponding vertices as reference points from stereo data and from the object model. The method was implemented on a personal computer. A result of simulation experiment of reference points teaching is shown.
An optimum structure of the substrates system for power semiconductor has been investigated based on the themal elastic-plastic finite element method and a performance of substrates system has been verified experimentally. The thermal stress in ceramics and the thermal conductivity of the substrates are selected as the design objects. The design acceptable region of the substrates is determined by a dielectric break down voltage, a current capacity, a thermal conductivity and a deformation. In this region, the optimization of the substrates with molybdenum-copper layered conductors has been inventigated taking into account the residual stress at bonding process. An example of the optimum structure of the substrates is presented. The themal conductivity and the reliability under temperature cycle have been evaluated by experimentally. It is confirmed that the degradation of the themal conductivity according to increase of bonding interface is negligible. The reliability of the substrates with the layered conductor has more than ten times compared with the conventional substrates. It is also shown that the optimum structure estimated by the computer simulation agree well with the structure obtained by the experimental results.
A good quality of weld is obtained by controlling the weld pool width and the cooling time. The cooling time t815 (from 800 [°C] to 500 [°C]) is related to the hardness of welding metal. The welding metal is influenced by rapid heating and rapid cooling. Therefore, various kinds of crystalline structures are formed by cooling time. It is important how to select cooling time for gaining good crystalline structure and the proper hardness after the welding. The cooling time is influenced by heat input (arc voltage U, welding current I, welding speed S, heat input efficiency η). The numerical simulations in the TIG welding of pipe are performed by using the finite difference method on the basis of heat conduction theory. The weld pool width and the cooling time have been calculated for various kinds of welding current and welding speed. The above data are used as the teaching data of neural network model. A weld pool model is constructed by neural network. The welding speed and the welding current are determined to obtain the desired pool width and cooling time by using the neural network model.
A nonlinear model has been developed to describe the relationships between the output (welding current and voltage) and the input (torch height) of through-the-arc sensor (simply called arc sensor) in DC MIG/MAG welding in open arc mode, and it is expressed by following five equations: Ut=uao+(Ra+Re)I+(Eai+aiI)(Lt-Le) LwdI/dt+(Ks+Rc)I+Ut=Us dLe/dt=Vf-AI/(1-BJe) Je=∫t t-ΔtI2(τ)dτ Re=∫Le o r(Jz)dZ All values of the parameters in the model were identified in the case of welding with both mild and stainless steel wires (1.2 mm diameter) and the shielding gas of 100%Ar. The simulated results based on the model have shown quite good agreements with the experimental. On ther other hand, a linearized model of the arc sensor has also been drived from the nonlinear model under the hypothesis of any variable only showing a smaller variation around a given operating point, and it's form is shown as follows: dUt/dt-Kw1dI/dt-Kw2dLe/dt=Kw3dLT/dt Lwd2I/dt2+Kw4dI/dt+Kw2dLE/dt=-Kw3dI/dt d2Le/dt2+Ke1dLe/dt=-Ke2dI/dt Of course, some other features, for instance, the dynamic characteristics of electrode wire melting in GMA welding have also been made clear by the present work.
In order to improve the sensitivity and reliability of the arc sensor, it is necessary to know how its characteristics change with the welding circuit conditions and the welding process variables and so on, because the sensor simply uses the electric signals — the weling current and/or the welding voltage as it's outputs. The present work has theoretically analyzed these relationships based on the nonlinear and the linearized models of the arc sensor. The analyses were made in time-domain and in frequency-domain. The results have revealed that the welding loop inductance Lw shows a little effects on the characteristics of the arc sensor if the variation frequency (f) of the torch height is lower than about 6 Hz, but its effect becomes greater in higher frequency range and increases with increasing f. The equivalent output resistance Ks (the slope of U-I (voltage-current) characteristic) of welding power supply exhibits greater effect on the characteristics of the arc sensor in full frequency range. The change in compositions of electrode wire or in welding conditions also effects the characteristics of the arc sensor but these effects are small in extent compared with Lw and Ks. Finally, the real-time behaviour of the arc sensor during the welding for a V-type joint groove was simulated, and the results were discussed by combination of the frequency-characteristics of the arc sensor and real welding conditions. Based on these results, it may be a better selection to use the welding current rather than the voltage as a detected signal in lower frequency range, but it does not necessarily remain good way in high frequency range for current wave-form generally exihibits greater asymetry with respect to the center of groove and current response also normolly declines with increasing f for a conventional welding system.
The microstructure and Charpy impact energy of simulated weld HAZs of low alloy steel SQV-2A have been investigated with particular reference to the effect of the formation of M-A constituent and its morphology. Weld HAZ thermal cycles having cooling times Δt8/5, from 6 to 1000 s, which reflect various welding heat inputs, were simulated with Gleeble 1500 thermal/mechanical simulator. The formation of M-A constituent and its morphology were examined with TEM and SEM. For this, steel, the M-A constituent formed at cooling times Δt8/5 more than 20s. With increasing cooling time, the M-A constituent formed not only near the grain boundary but also in bulk region, and changed from elongated shape to massive shape. Both enlongated and massive M-A constituents seem to impair the HAZ toughness of SQV-2A steel. The M-A constituent acts as the initiation site of microcracking to induce the toughness loss, but the toughness may be affected by not only M-A constituent but also the matrix, because the matrix itself seems to become brittle with increasing cooling time.
The behavior of the superficial oxide film during the diffusion bonding of Al-Mg-Si series 6063 alloy and its influence on the bond strength have been investigated mainly by TEM observations in order to explain the effect of alloying elements on the bondability of aluminum alloys. When faying surfaces were finished by electropolishing, crystalline oxide particles less than a few 10-1μm in size were found to be dispersed on the joint interface of 6063 alloy, whereas rather continuous film of amorphous oxide remained at the joint interface of commercially pure aluminum. When faying surfaces were finished by wire brushing, an interfacial zone of a few μm thickness involving a number of fine oxide particles were found in the joint of 6063 alloy. A similar interfacial zone was also observed in the joint of pure aluminum, but its thickness and density of oxide particles were much greater than those in the joint of 6063 alloy. The crystalline oxide particle at the joint interface of 6063 alloy was identified as Al2MgO4 by SAD patterns. The formation of Al2MgO4 can be thermodynamically explained as a result of the following reaction between the superficial oxide film (Al2O3) and Mg : Mg+4/3 Al2O3→ Al2MgO4+2/3 Al. The tensile strength of the 6063 alloy joint was much higher than that of the pure aluminum, suggesting that the above difference in the behavior of the oxide film at the joint interface improved the bondability of 6063 alloy.
In order to decrease the completion time during TLP-bonding, an iron-based filler metal, (IM-7) comprising MA956 base metal containing 7 mass%Si and 1 mass%B was developed to join MA956 base material. TLP-bonded joints free of microvoids and bondine intermetallic phases were obtained using a bonding temperature of 1563 K, a holding time of 2.16 ks and an applied pressure of 7.0 MPa. The bondline region in TLP-bonded MA956 base metal had a bamboo-like microstructure. In order to make clear the formation mechanism of this microstructure, the growth mechanism of solid phase during isothermal solidification was evaluated. During tensile testing at 923 K, the joints TLP-bonded at 1563 K-for 2.16 ks fractured in the base metal zone and consequently, the mechanical properties of the joint region and the base metal were similar. The creep rupture properties of the joint regions were close to the base metal properties in the transverse direction.
Various study have been performed in order to relax residual stress and enhance bonding strength of bonded dissimilar materials. It is well-known that improvement of the interface shape is one of the most effective factors for achieving high bonding strength. To estimate the effect of interface shape on bonding strength, it is necessary to consider both theoretical thermal stress analysis and actual bonding strength of bonded dissimilar materials. In this paper, the correlation between the order of stress singularity (λ power) and experimental bonding strength was investigated under a appropriate bonding temperature condition. The investigation was carried out with a cylindrical TiB2 -Ni joint by varying the interface edge angle. Actual bonding strength was evaluated by the tensile test, and the λ power was calculated by substituting secant stiffness moduli into Bogy's equation for each interface angle condition From these results, the relationship between λ power and actual bonding strength was clarified, and the most desirable interface shape for achieving high strength joint was discussed. It was clarified that residual stress could be reduced and actual bonding strength increased by employing a desirable interface shape. The interface edge angle was related to the appropriate value of A and the highest bonding strength.
To establish a simple design method for a desirable interface shape based on theoretical analysis, the relation between the order of stress singularity (λ power) and actual bonding strength is evaluated with various combinations of bonded dissimilar materials. With the analysis method of secant stiffness modulus, the calculation model at the different bonding temperatures can be replaced by applying calculation model of secant stiffness modulus in varied combinations. Thus, in this paper the correlation between λ power and experimental bonding strength was investigated under several bonding temperature conditions. The investigation was carried out with a cylindrical TiB2-Ni joint by varying the interface edge angle. Actual bonding strength was evaluated from the tensile test, and λ power was calculated by substituting secant stiffness moduli into Bogy's equation for each interface angle condition. From these results, the dependency of material combinations on the relationship between λ power and actual bonding strength was clarified. Desirable interfaces with arbitrary material combinations were also discussed for achieving high strength joint. It was clarified that residual stress could be reduced and actual bonding strength increased by employing a desirable interface shape. The most desirable interface edge angle was shown for several combinations of bonded dissimilar materials.
Correlation between Heat-Affected Zone (HAZ) microstructures and the mechanical properties during Post-Weld Heat Treatment (PWHT) on two types of Thermo-Mechanical Control Process steels (TMCP steels), especially the relation between the stress relaxation behavior and high temperature deformability was investigated. Simulated-weld heat treatment was carried out using a weld-heat cycle simulator with a maximum temperature of 1623 K. PWHT was done at a heating rate of 55.6 K/ks. The mechanical properties during the PWHT process were evaluated by means of the constant-strain rate test on heating and the constant-load test on heating. The influence of PWHT on the reducing tendency of ductility was discussed from the viewpoint of precipitation behavior of inter-or transgranular carbides and the deformability of the matrix on each HAZ microstructure. The results are summarized as follows : (1) The stress relaxation behavior of the specimens of bainitic structure on both AcC type (Controlled rolled+Accelerated Cooled) and Non AcC type (only Controlled Rolled) TMCP steels were delayed clearly above 600 K. The stress relaxation behavior of the Non AcC type TMCP steel was further delayed to the higher temperature range during PWHT than that of the AcC type TMCP steel. (2) High temperature deformability of the specimens of bainitic structure on both TMCP steels had a remarkable reducing tendency of ductility in the range of 850-900 K. The Non AcC type TMCP steel had a remarkable reducing tendency of ductility than that of the AcC type TMCP steel. (3) The reducing tendency of ductility was caused by the difference in precipitation behavior of inter-or transgranular carbides in each HAZ microstructure during PWHT process. The tendency of the Non AcC type TMCP steel which contained more alloying elements occurred more strongly because of precipitation of alloying carbides. (4) When the high temperature ductility or embrittlement of HAZ microstructure of the TMCP steels was evaluated, it was very important to measure the deformability of the material during the PWHT process. An evaluating method of ductility of TMCP steels was proposed using the relation between deformability of material and the displacement of each microstructure during PWHT process.
The laser speckle strain measurement is a non-contacting and highly accurate method. It has been developed mainly for the measurement of static strains. In the previous report, the authors proved the applicability of the laser speckle method to the dynamic strain measurement in welding. However, we indicated in the report that there was a range of distance between the heat source and the measuring point in which the strain could not be measured as the heat source approached the measuring point and passed through the point. It was caused by a rapid change of the speckle patterns and the slow responsibility of the measuring instruments. In this report, the authors improved the sampling rate of the measuring system up to 8 ms/frame and could follow the change of the speckle patterns through almost all the period during and after welding. The obtained strain curves generally agreed not only with qualitative prediction but also with quantitative testing results obtained by sequential photographs. Only the strain perpendicular to the weld line deviated from them when the heat source was in the vicinity of the measuring point. High refreshing rate of the reference frame in the strain calculation and the local deformation of the sample surface were considered as possible causes for the discrepancy, whereas the oxidation of the surface seemed have little effects.
Two joining methods were investigated to develop the low resistivity joint of silver and Y-Ba-Cu-O superconducting oxide (YBCO) with the columnar crystal structure of 123 phase (YBa2Cu3O7-x), which was obtained by unidirectional solidification from liquid+211 (Y2BaCuO5) phases. One method was a solid state processing by the use of hot-press, in which no detectable interaction layer was formed. The interface resistivity (ρc) at 77K measured by the four-prove method decreased with the increase in joining temperature (T) or joining pressure (P), and the minimum value : ρc=2×10-12Ω·m2 was obtained when the joining conditions were P=32MPa, T=873K, t=3600s. The other method was a liquid state processing by using molten silver, in which remarkable interaction layer was formed. Since the reaction was caused by the decomposition of 123 phase to liquid+ 211, the interaction layer was recovered to 123 phase by slow cooling and solidification. This recovering treatment remarkably decreased pp, and the minimum pc value was as low as about 10-12Ω·m2 at 77K, which satisfied the stability criteria for composite conductors.
Wire bonding is used as a method of the electrical connections between an electrode terminal on a chip of a semiconductor and an outer lead terminal. Cu wire has a good corrosion resistance and high electrical and thermal conductivity. Furthermore, Cu wire can be bonded directly to Cu alloy lead frame as a 2nd stitch bonding. For these reasons, a practical application of the Cu wire bonding process is anticipated. But, for application of Cu wire stitch bonding onto Cu alloy lead frame, there are a lot of difficult problems to be solved. In this report, the influence of the bonding conditions and surface states on the bondability of Cu wire stitch bonding is investigated. It is pointed out that the control of the wire deformation behavior is necessary for obtaining the good bondability in thermosonic Cu wire bonding. Furthermore, it is clarified that the surface state on Cu alloy lead affects on the bondability. If the maximum surface roughness on Cu alloy lead is more than 0.4 μm, or if the thickness of the oxide film on Cu alloy lead is more than 10 nm, the bondability of Cu wire stitch bonding is inferior.
For the direct bonding between Cu wire and Cu alloy lead frame, the wire deformation behavior and the temperature rise at the interface between a wire and a lead terminal are very significant. In this report, the operation of the load on the bondability and the correlation between the wire deformation behavior and bondability are clarified. The wire deformation is controlled by load control, and the plastic flow and the temperature rise at the interface are depended by the equilibrium of the load and the ultrasonic vibration. By setting the higher value of the initial load, the working load can be settled the smaller value, and so the wire deformation becomes smaller and bondability is superior than that in bonding by applying the constant load.
TAB (Tape Automated Bonding) technology is the main method in IC interconnection technology, because it can be applied to fine pitch, under 100 μm pitch. In the conventional TAB technology which is the gang bonding technology, bonding under the uniform loading and heating are very difficult. So, single point TAB technology which is to bond gold bumps one at a time on the IC pad with TAB lead is proposed. For the thermosonic bonding used in the single point TAB technology, the deformation behavior of the bonding material and the working of the ultrasonic vibration are very significant. An applied load that is one of the bonding parameters in the single point TAB process has influence on the deformation behavior and the working of the ultrasonic vibration. So, in this report, we designed and manufactured by way of trial the single point TAB bonder with the control mechanism of the loading pattern in order to elucidate the single point TAB process. The loading force is controlled by the moving of the tool tip. Then, the control ability is dominated by the spring constant of the loading system, moving velosity and moving resolution of the tool tip. A spring constant of the loading system is very important factor for the design of a loadig system, because it has influence on the response or the stability of applied load for the change of the bump height or the position of a tool tip. The developped bonder had the following ability; 1) a spring constant of the loading system is 0.00175 N/μm, and so the loading fluctuation for the change of the bumpheight or the position of a tool tip is under 0.01 N. 2) the loading force is controlled in resolution of 0.007 N and monitored by ICP (Integrated Circuit Piezoelectric) in process.
Bonding under the conditions of low temperature and low loading is required so that an IC chip won't be damaged while being bonded for an inner lead bonding between an electrode terminal on an IC chip and an outer lead terminal. For the thermosonic bonding used in the single point TAB technology, the deformation behavior of the bonding material and the working of the ultrasonic vibration are very significant. An applied load that is one of the bonding parameters in the single point TAB process had influence on the deformation behavior and the working of the ultrasonic vibration. In this report, a correlation between the bonding parameters and the bondability is discussed, and it is pointed out that the control of the loading profile is important for attaining the low temperature and small deformation bonding. The bonding process in the thermosonic bonding is divided to the initial contact process dominated by the material plastic deformation and the adhesion process dominated by the plastic flow at the interface between a gold bump and TAB lead. In the initial contact process, the minimum loadingg force exists to obtain the contact between bonded materials, and the other hands, in the latter adhesion process, the contribution of the ultrasonic energy into the interface becomes larger under the lower loading force. Furthermore, it is clarified that the bonding part with a smaller deformation and a higher strength is obtained by setting the higher value of the initial load and the lower value of the working load in a single point TAB process.