Pulsed MIG welding phenomena are studied fundamentally in order to make clear the effects of welding current, supplied from the transistor power source. Correlation between current waveform and metal droplet transfer is investigated with the optical metal droplet detector, which is consisted of He-Ne laser, optical lens and filters, and photo-transistor. Results obtained are summarized as follows; (1) In pulsed MIG welding, it is found out that there exists critical frequency fcr. At the frequency of fcr, metal droplet transfers synchronously with respect to each pulse current. (2) In the frequencies lower than fcr, if pulse peak current is set enough higher than the level for spray transfer, metal transfer mode becomes spray transfer mode in the duration of peak current. (3) In the frequencies higher than fcr, if the average current level is lower than critical current, metal transfer is carried out in irregular drop transfer mode. (4) Critical pulse frequency fcr, sustaining one pulse-one droplet transfer, varies with respect to the pulsive peak current level and the size of wire used. (5) In the case that average current is lower than critical current, metal transfer mode and bead formation mechanism vary in accordance with pulse frequency. (6) In extremely low pulse frequency range, pulsed MIG welding is effective to improve the shape of welds in a repeating sequence of melt and solidification of pool. (7) Even if the average current is kept constant, weld bead penetration becomes deeper as the level of peak current becomes higher.
Type 405 stainless clad steel plate has been widely used as a corrosion resistant material, especially as a stress corrosion cracking resistant material. However, occurrence of transverse cracking due to hydrogen in the first layer welds of clad side and apperance of carburized layer in the heat affected zone of first layer welds of clad side are reported in the welds of this steel. Investigation has been made to clarify the effect of chemical composition of base metal and welding condition on these problems mentioned above. Results show that the largest effect wasobtained with the reduction of C content in base metal on the improvement of these problems. Therefore, the application of newly developed low-C, Al-B treated low alloy steel plates for base, metal of Type 405 stainless clad steel plates is expected to be available.
In our previous report IIW IX-1232-82, a new and more generalized formula PHA for cold cracking parameter has been introduced, considering local accumulation of hydrogen, so that the conditions leading to various types of hydrogen-induced cracking in welded high strength steels can be analyzed and predicted using the data obtained from the standard JIS-y (Tekken type) cracking test. In this report, the formula for the PHA has been made more accurate with additional JIS-y test data. The idea of equivalent intensity of restraint RFy, which directly concerns the magnitude of local stress concentration near the root of a one-pass weld in a butt-joint, has proven to be very useful in explaining quantitatively the effects of various groove shapes and eccentric weld locations on the values of critical cooling time for the prevention of root cracking. Moreover, experimental findings on heel cracking (HAZ root cracking in a short fillet weld in a T-joint) of HT50 steels are satisfactorily explained by PHA analysis.
Ti treatment has been developed to apply to heavy plate and line pipe steel. Ti treatment disperses fine TiN particles in steel, and improves welded H.A.Z. toughness of the steel. Recently, by the advance in steel making technology, it becomes easy to make low Nitrogen content steel. Accordingly, we investigated the effect of Ti treatment on low Nitrogen content steel. The results were as follows; (1) On synthetic H.A.Z. toughness test, the toughness was improved remarkably through the finegrained H.A.Z. structure by the effect of Ti treatment. (2) Consequently, on low N and low Ti content steel, the synthetic H.A.Z. toughness was inferior, because the amount of dispersed TiN was few and there was few effect to make H.A.Z. structure fine-grained. (3) On synthetic H.A.Z. toughness test, the optimum range is about 0.015-0.020%Ti and 60-80 ppm N content range. (4) On actual multi-pass welding, multi-thremal cycle itself has the effect of making H.A.Z. structure fine-grained, therefore, the fine-grained structure effect of Ti treatment didn't appear clearly. Low N and low Ti content steel had good H.A.Z. toughness, too.
The Application of supplementary heat to a weld joint befor or after welding have been studied using a mild steel, medium carbon steels, 50 to 100 kg/mm2 high tensile steels, low alloy steels, martensitic stainless steels and cast steels. The results indicate that (1) When preheating is applied for preventing cold cracking in heat-affected zone, a new carbon equivalent, i.e. Pre. Ceq.=C+0.06Mn+0.03Ni+0.06Cr+0.10Mo, is useful in predicting the cracking susceptibility of a wide variety of steels above-mentioned, (2) the critical preheat and interpass temperatures are reduced by using postheating, (3) these temperatures must be held till postheating at the temperatures above the critical point (T2c) at which hydrogen-induced cracking can not occur, (4) the T2c depends on the chemistry of steel, and the T2c values may be given by T2c(°C)=455 Post Ceq. -110, Post Ceq.=C-0.08Si+0.20Mn+0.03Ni+0.05Cr+0.12Mo+0.84V, and (5) it is desirable for avoiding hot cracking that the preheat and interpass temperatures are lower than Ms point.
The crack initiating temperature and time of rehaet cracking were investigated on three types of commercial heat resistnig steels. Two modified implant test machines were employed, one of which was constant strain type (stress relaxation type) and the other was constant load type. The crack initiating curves for those steels were drawn on the stress-temperature (time) diagrams. The results were summerized as follows. (1) The crack initiating temperatures obtained by the constant strain test were 550, 510 and 540°C for 1/2Mo, 1Cr-1/2Mo and 2 1/4Cr-1Mo steels, respectively, when the stress was high enough (σPW>30kg/mm2). The crack initiating temperature elevated with the decrease of the stress. (2) The crack initiating temperatures obtained by constant load test agreed with those by constant strain test, only when the stress was high. The crack initiating curves of the former situated generally in higher stress-and longer duration side than that of the latter. This difference will come from the difference of the stress during the preparation stage prior to crack initiation. (3) The crack initiating temperature was not so much varied by the variation of Cr and Mo contents of a very wide range.
Improvement of toughness at low temperature in Cr-Mo weld metal is one of the most important problems. Welding materials which could satisfied rather high toughness after step cooling had been developed. In order to obtain much improved toughness, further fundamental studies on the submerged arc weld metal in 2 1/4Cr-1Mo steel have been carried out. Effect of chemical compositions on weld metal toughness after stress relief treatment and step cooling were investigated. As the result, it could be possible to reduce temper embrittlement and to obtain high toughness at low temperature by means of lowering Si, P and oxygen content as well as controlling C content in weld metal. These optimum chemical composition coud be obtained by welding materials which are combination on solid wire and highly basic bonded flux.
The transformation behavior of nitrogen bearing 25Cr-13Ni-1Mo and 21Cr-11Ni-2.5Mo steel weld metals containing 1.5% to 5% delta-ferrite has been studied after heat-treatment in the temperature range 700°C to 1000°C. The extent of the delta-ferrite transformation was estimated magnetically using a ferrite scope. The microstructural changes occurring in the delta-ferrite were examined using a transmission electron microscope and a scanning electron microscope, and it was found that the delta-ferrite transformed to sigma-phase, chi-phase and austenite during aging in the temperature range 700°C to 1000°C. The rate of transformation was significantly high at temperatures 850°C to 900°C, and was, however, rather low at lower or higher temperatures. The transformation behavior of delta-ferrite in the temperature range examined could be described by the Johnson-Mehl's equation y=l-exp (-ktn). Based on these results, a time-temperature parameter has been developed for representing by means of a single variable, the combined effects of time and temperature on the transformation of delta-ferrite during aging. Available. data are analyzed, which result in indicating that it is possible to predict the transformation behavior of delta-ferrite in the weld metal after heat-treatment or long time service at elevated temperatures using the newly developed time-temperature parameter.
Hydrogen-induced disbonding of austenitic stainless steel cladding was investigated. The used test specimen was made from the heavy section 2 1/4Cr-1Mo steel plate weld-clad using type 309 strip with 75 mm width and type 347 strip with 150 mm width. And it was heated at 455°C in autoclave containing high pressure H2 for hydrogen charging. The results indidate that (1) the delayed cracking occurs when the specimen is rapidly cooled from hydrogen charging temperature, but it cannot occur above 100°C, (2) this type of cracking takes place in grain boundaries of cladding adjacent to the fusion line, (3) the susceptibility to cracking decreases by application of post-heating or by reducing the rate of cooling after hydrogen charging, (4) numerical analysis applied to hydrogen distribution may explain the behaviour of cracking.
Information about welding transient and residual stresses produced in welded joints is indispensable to investigate weld cracking and the safety of welded structures. Such information can be obtained either from a theoretical analysis or an experiment. Nevertheless, for estimation of welding residual stresses in a very thick plate produced by multipass welding, the analysis takes a long computation time and the experiment requires much labor. The aim of this research is to contrive an analysis method in order to accurately estimate such welding residual stresses in much shorter computation time and with far less labor than ever. It is the theory of thermal elastic-plastic analysis based on the finite element method that is always adopted as the analytical theory. Applying this theory to the analysis of transient and residual stresses due to multipass welding, the analysis method (specimen, mesh division, analytical procedure, etc.) is simplified. This analysis method excluding the analytiacl theory is hereinafter called the analytical model. Applying the same theory of thermal elastic-plastic analysis to this analytical model which is as accurate as possible, a theoretical analysis is performed. From the result of this analysis, the production mechanism and the distribution of the above mentioned residual stresses are investigated in detail. Based on these results, the analytical model is simplified to accurately estimate (1) the distributions of transient and residual stresses in the whole cross section and (2) the distribution of residual stresses only near the finishing bead (in this vicinity, maximum tensile residual stresses are generally produced and such weld cracks as longitudinal ones and transverse onse may occur). It is shown that highly accurate solutions can be obtained by the proposed simplified analytical models in a very short computation time. It is also shown that the experiment can be similarly simplified, which is very effective especially for residual stresses near the finishing bead.
Effect of local heat treatment on residual stresses of 13 Cr cast steel (SCS5) is investigated by an experimental analysis of the transient thermal stresses of weld. The transient thermal stresses are obtained by using both-ends-fixed bar subjected to double thermal cycles simulating welding and local heat treatment. The experimental results are compared with the longitudinal residual stress distributions of butt welded joints which are subjected to local heat treatment. The residual stresses at welds after local heat treatment of which temperature ranges from 500°C to Ac1 transformation point become tensile stresses not less than 400 MPa. Vickers hardness decreases with increase in holding time and rise in holding temperature. It is considered that the variation of residual stresses are mainly attributed to relaxation of thermal stress and reduction in yield stress. From these results one dimensional analysis method is proposed to evaluate residual stresses under local heat treatment. This analysis shows good agreement with the experimental results.
In this paper, in order to develop the model of mathematical treatment of metal during transformation, a series of experiment and theoretical analyses were conducted. Thermal elastic-plastic analysis was performed using the variously idealized mechanical properties in the region of phase transformation. The validity of the idealization was examined by comparing the residual stress produced in RRC test and slit weld specimen and clarify the production mechanism of restraint stresses. The main results are as follows. 1) When restraint stress before transformation acts on occurring or developing transformation, the transformation super plasticity phenomenon appears. Thus the formerly produced restraint stresses are relaxed for a time. At the same time, compressive stresses are produced by transformation expansion. Restraint stresses stagnate until welding heat transfers out of the gage length. Afterwards, restraint stress rapidly increase to be residual stress at a room temperature. Therefore, if the detailed information on the production process of restraint stress is necessary, the analysis needs to take into account the above-mentioned phenomenon in addition to transformation expansion. 2) In consideration of the above-mentioned phenomenon, the most accurate model (model M2) assumes that the material which is cooled to the starting temperature of phase transformation does not possess rigidity for a time while rigidity gradually recovers and transformation expansion occurs in the region of phase transformation. The simplified model (model M3) which assumes that the material does not possess rigidity until it is cooled to the starting temperature of phase transformation is also accurate. If only welding residual stress is necessary, the analysis on the assumption that the weld metal and HAZ do not possess rigidity until the material is cooled to the terminating temperature of phase transformation is adequate enough that its result roughly coincides with the experimental one (model M4). 3) Based on the above outcomes, the already developed analytical calculation theory for estimation of restraint stress-strain perpendicular to the weld line of the weld metal due to slit weld is extended so as to apply to the material (HT-80) in which the effect of phase transformation is remarkable.
TRC and RRC tests are basic testing methods for studying cold cracking sensitivity of materials. The restraint condition of RRC test is clearly defined but that of TRC test is not. In contrast with this, the testing method of TRC test is much easier than that of RRC test, and an arbitrary loading diagram after weld can be applied in the test. Therefore, it is very important and useful to clarify the dynamical characteristics of both tests and their corelation. For this purpose, a series of thermal elastic-plastic analysis by the finite element method was carried out. The results are summerized as follows. (1) There is no essential difference between dynamic al characteristics of TRC test and of RRG test if the loading process in TRC test is chosen as similar as the process of development of restraint stress in RRC test. (2) In TRC test, stress distribution in the throat section may be determined only by the final restraint load, but plastic strain distribution is affected very much by the loading process. (3) In order to get stress and plastic strain distribution in the throat section, which are very similar to those of RRC test, it is necessary to impose in TRC test a loading process as indicated below, (a) the load may be applied when heat begins to be transmitted out of the restraint length. (b) the load may reach at the final value at a temperature which the corresponding RRC test ends. (c) the final load may be equal to that of RRC test. (4) The corelation between two tests is obtained as follows, (a) the free contraction S is measured. (b) a tensile test is conducted for a weld joint of 2l length. (c) the elongation of the specimen should be the same as S.
The fracture toughness of the welded joint measured by three point bending COD test usually results in unignorable variation. Especially low critical COD value is often obtained when the fatigue crack tip of COD test specimen is located at the coarse grain region i.e. partially embrittled region. In this paper, the reason to obtain such low critical COD value and fracture behavior of the COD test specimen are investigated in view of welded joints of low temperature service steels. To make examination into this subject, welded joints furnished with embrittled regions artificially are prepared. Welded joints with partially embrittled regions are reproduced in the following two types, which are welded joints contained regions of hot straining embrittlement and mixed with low toughness material. As a result, it was evidenced that critical COD value obtained by COD test specimen depended on the partially embrittled region, and that a crack initiation in any embrittled region had potential to lead to a fracture, but there was a tendency of fracture behavior showing not a one-step fracture but a multi-step fracture.
Creep properties of thin wall tubes with a circumferential joint subjected to high temperature internal pressure have been discussed experimentally and analytically. From the results, a method of evaluating the creep rupture property of the tube from the creep data of weld metal and base metal has been proposed. And, the creep rupture properties of Hastelloy X thin wall tubes with TIG welded joint and EB welded joint by internal pressure have been evaluated by the above method.
In the conventional spot welding process, the heat generation in welds makes use of resistance heating of workpieces and the workpieses are jointed by nugget formed at the interface of them. However, in this process, the indentation of workpiese increases with the size of nugget. The problem described above is improved by means of using the pulsed high current density with very short duration, because the contact resistance at the interface is used efficiently to form nugget and the thermal loss is obstructed. In this welding process, tensile shear strength and peel strength of welding joint are sufficiently large, and fusion layer is very thinner than one of the conventional spot welding, and its indentation is very small. It seems that the joining of workpieses is occured mainly due to Joule's heating at the interface by contact resistance.
Effect of faying surface treatment on the tensile shear strength (TSS) of joints has been studied in the weld of aluminum alloy (0.8t 5052) to mild steel (0.8t SPCC) using uni-directional weld current from SPCC to 5052. Welded joints had 380 kg/spot TSS under the condition of wire-brushing on both SPCC's and 5052's faying surfaces. Slightly reduced TSS value was obtained under the condition of wire-brushing on SPCC's faying surface alone. But on the contrary, wire-brushing on 5052's faying surface alone, gave TSS value merely about 100 kg/spot. Therefore, to give wire-brush or not on SPCC's faying surface before the weld affects strongly on TSS of joints. Further, peeled fracture surface of the joint was observed by scanning electron microscope and it was found that joints under SPCC wire-brushing showed ductile deep dimple region while the other presented relatively flat bisttle face. The above mentioned results indicate that the surface treatment of metals with higher melting point is important in the welding of dissimilar metals.
This investigation is carried out to clarify the factors influencing the weldability of beryllium-copper alloy, using the 25-alloy and 10-alloy. 25-alloy is copper-base alloy containing 1.80-2.00% Be, more than 0.20% (Co+Ni) and less than 0.6% (Co+Ni+Fe). And 10-alloy is also copper-base alloy containing 0.4-0.7%Be and 2.35-2.70% (Co+Ni). From the Auger analysis of the fracture surface of weld, it is supposed that the poor weldability of this alloy is due to formation of beryllium oxide on the weld interface. The results obtained as follows. (1) Difference of weldability between several heat treated alloys is not noticeable. (2) Weldability of 10-alloy is superior to that of 25-alloy. This is supporsed to be due to the difference of beryllium quantity segregating on the weld interface. (3) Weldability in high vacuum is superior to that in air. (4) Copper and nickel plating are expected to be very effective to improve the weldability of berylliumcopper alloy, because of establishment of intimate contact of weld interface, reduction of beryllium quantity on the weld interface, and its rpotection action from oxidation of that.
Tensile properties of electron-beam-welded pure molybdenum, MO-0.4%Nb, Mo-0.05%Zr and Mo-5%Re alloys were measured at -70°C and 20°C. Furthermore effects of a small amount of alloying elements on the ductility of weldments were discussed from the structural points of view (grain size, precipitate size and distribution). The results were summarized as follows. (1) The strengths of EB-welded Mo-Nb and Mo-Zr alloys were enhanced compared with those for EB-welded pure molybdenum. However the ductilities of these alloys were only slightly improved. The strengths of EB-welded Mo-Re alloy were reduced, in contrast. However its ductility was relatively improved. (2) Postweld recrystallization induced marked reduction of strengths of the weldments, except for the fracture strength of EB-welded Mo-Re alloy. The ductility of the weldments were appreciably improved by the postweld recrystallization. Above results were reasonably interpreted from the differences in grain size, precipitate size and distribution, and from the softening effect by Re addition.
This paper is concerned with the effect of the reinforcement on the fatigue strength of butt welded joints. Since the fatigue strength of butt welded joint is strongly influenced by the geometrical conditions of reinforcement, it has been pointed out that the flank angle and the root radius of weld toe are responsible for the fatigue strength. However, the fatigue limit of butt welded joint of SM50 steel plates is nearly independent of the root radius of weld toe. This result is explained from the characteristics of fatigue behavior of notched specimens made of the same materials.