Many papers have been reported about the tensile strength of steel brazed joints, but little about the tensile strength of copper brazed joints. We have found out the tensile strength of electric copper joints butt-brazed with brazing filler metals of Ag-P-Cu, Ag-Cu-Zn-Cd and Cu-Zn systems at the room and elevated temperatures. Then we have examined the relation between joint clearance and tensile strength of joints. We have also examined their fracture.
The conclusions are as follows :
1) As to the tensile strength at the room temperature, the joints brazed with Ag-P-Cu systems are of less strength than copper base metal. The joints brazed with Ag-Cu-Zn-Cd systems increase their strength with Ag-content in the brazing filler metals, and the joint brazed with BAg-la is the same with base metal in tensile strnegth. The tensile strength of the joints brazed with Cu-Zn systems is the lowest of all.
2) As to the tensile strength at elevated temperatures, the joints brazed with BCuP-2 and BCuP-5 are of less strength than base metal below 300°C, and are the same at 400°C. The joint brazed with BAg-la is as strong as base metal at any temperature.
3) The tensile strength of brazed joints is not related to joint clearance but to mechanical properties of brazing filler metals, i.e. when mechanical properties of brazing filler metals are the same with those of base metal, the joints with the brazing filler metals areas strong as base metal.
4) The joints whose tensile strength is less than that of base metal begin to fracture at the interface between base metal and brazing filler metal.

View full abstract

Weld cracking is influenced by both reaction stress during cooling and properties of weld metal and heat-affected zone. The problems of weld cracking, therefore, should be explored not only from metallurgical standpoint but also from standpoint of reaction stress. Researches on effect of reaction stresss during cooling or restraint on weld cracking are now undertaken by the authors, especially for first-pass-weld-cracking of restrained welds of some mild and high strength steels. At earlier stage of the research project, informations on what factors influence to reaction stress under restraint were wanted for reasonable research planning. The present paper was prepared as preliminary informations for the subsequent test planning.
In a restrainning system shown in Fig. 1 shrinkage force will be produced by hindered contraction or extension of first-pass weld metal during cooling. Analytical investigation was conducted on the extension of weld metal. The anaiysis is based on the assumptions that behavior of mother plate (B) and restraining members (C) is elastic during cooling, however, in the weld metal will be produced some plastic deformations under the stress-strain relation as shown in Fig. 3. The extension (λ_w) of weld metal at a given temperature below about 300°C is obtained by
p_wh_w/ph⋅λ_wⁿ+λ_w=S
where ph and p_wh_w are rigidity or force per unit weld length required for unit extension of mother plates and weld metal respectively, and S is free contraction between restrained length (l) of mother plate. Relations of λ_w versus rigidity ratio p_wh_w/ph are shown in Fig. 4 for a few values of S. Larger extension of weld metal will be produced under smaller rigidity ratio and larger free contraction.
Free contraction S will be influenced by heat input and overall cooling behavior of mother plate of length l. Under a given heat input the overall cooling rate or heat loss from the mother plate of length l will be increased as the decrease of heat capacity or length (l) and thickness (h) of mother plate.
It will be concluded, therefore, that under a given heat input reaction stress of weld metal at a given temperature is influenced by the following factors ;
1) restraint coefficient (p), which contains restrained length (l) and ratio of sectional area of mother plate to restraining members.
2) thickness (h) of the mother plate.
3) restrained length (l) of the mother plate.
4) rigidity of weld metal (p_wh_w).
Figs. 8 and 9 show some calculated examples of the effect of l and h on the extension of weld metal.

View full abstract

Control of weld cracking is one of the most important problems for welding engineers. Weld cracking is influenced by welding stress during cooling as well as properties of weld metal and heat affected zone. The problems of weld cracking, therefore, should be explored not only from metallurgical strandpoint but also from standpoint of welding stress. In this report research was focused on the influence of restraint on weld cracking.
A new test apparatus as shown in Figs. 3 and 4 was designed for evaluation of sensitivity for root cracking of steel welds. With this apparatus, "Rigid Restraint Weld Cracking test (RRC test)" as shown in Fig. 1 was developed, it is such that the length of a specimen is kept constant during cooling. The RRC test method is similar to the TRC test developed by Dr. Suzuki and his collaborators for the researches on root cracking of high strength steel welds. However the authors' test method differs from the TRC test as follows : In the TRC test a constant load was applied at a few minutes or immediately after welding because emphasis was laid on only delayed root cracking of high strength steel welds. The authors' test method is very significant, on the other hand, for the two reasons which (1) it gives gradual increase of reaction load by keeping the length of a specimen always constant during welding and cooling and also (2) it permits a wide degree for adjustability in the restraint intensity by changing the gauge length of restraining (l), (See Fig. 1). This may be much similar to the actual behavior of restrained welds and will be reasonable as the test method for weld cracking under restraint during cooling as well as after cooled to room temperature.
Experiments were carried out on the weldments of a mild steel and high strength steels of 60 and 80 kg/mm² tensile strength level.
The results obtained are as follows.
(1) In the RRC test, shinkage rate and magnitude of reaction stress were widely changed with changing the gauge length of restraining, and weld crack occurs in wide range of temperature. Reaction load increases gradully with cooling, and higher intensity of restraint usually favours crack initiation or decreases the time required for cracking.
(2) In mild steel weldments, weld crack occurs only during cooling and no crack initiation is observed after cooled to room temperature.
(3) In high strength steel weldments, on the other hand, crack initiation was observed not only during cooling but also after cooled to room temperature.
(4) This delayed cracking was observed even at lower reaction stress level. Minimum reaction stress level for the delayed cracking is 33 kg/mm² for HT-80 steel and 60 kg/mm² for HT-60 steel. The incubation period was measured 1-10 hours for HT-80 steel and 20 hours for HT-60 steel after welding.
(5) Critical gauge length (l_cr) of restraing, over which no crack initiation was observed, was 300-370 mm for mild steel, 450-500 mm for HT-60 steel and 550-600 mm for HT-80 steel.

View full abstract

In the welding or brazing, it is clear that the weldability of filler metal is one of the most important factors whine determine the bondability and workability. The wettability consists of the surface tension of molten filler metal or the interfacial tension between filler metal and flux, the interfacial tension between molten filler metal and base metal, and the fluidity of molten filler metal. In the case of chemical bondings, it is necessary to consider the chemical reaction between filler metal and base metal.
The authors have planned to study the above components of wettability separately. The present report is described the experimental apparatus constructed by the authors for the study of the wettability and the results of studies obtained for the surface tension of Pb-Sn alloy system.

View full abstract