溶接割れ試験体の拘束係数と拘束力について

抄録

It is well known that reaction stresses of weldments produced during cooling under external restraint will be one of the serious factors affecting on weld cracking particularly on cold cracking. That is the reason why various types of restrained specimen have been used for weld cracking test. Reaction stresses in these specimens will be changed by the two factors ; the one is the relative rigidity of weld metal to restrained specimen and another is free contraction due to welding. Weld cracking test results may be influenced by the rigidity of specimen used even if the same conditions of materials and heat input are used. Therefore, it will be important for the evaluation of weld cracking test results to obtain informations on the rigidity of cracking test specimens.
In order to consider the rigidity of restrained specimens, restraint coefficient was used, the definition of which is such that the restraint coefficient is equal to average value of transverse stress along weld line necessary to produce average elastic change of root gap of unit magnitude along the weld line. The restraint coefficient thus defined will be calculated with elastic theory for restrained specimens of simple shape, however, the calculation will be difficult for more complicated specimens and practical welded joints. For the measurements of restraint coefficient a new apparatus called "Restraint Meter" was designed by the authors. Principle of the restraint meter is that separating force is applied to both beveled edge planes of a weld groove before welding by the action of a wedge pushed into the groove and then the change of root gap is measured with displacement gauges. The separating force per unit change of root gap will be increased with the increase of rigidity. The restraint coefficient can be obtained from the separating force.
Measurement of restraint coefficient by the restraint meter was performed for various cracking test specimens such as Lehigh type, N.R.L. type and Tekken type, which is widely used in Japan. Restraint coefficient of Lehigh type specimens ranges between 33 and 75 Kg/mm². mm for 75 mm weld length and 11 and 45 Kg/mm². mm for 125 mm weld length. Tekken weld cracking specimen shows 59 Kg/mm². mm restraint coefficient. Reftraint coefficient of N.R.L. type specimen is much smaller than the above two because a end of weld groove is opened. (See Column (4) in Table 1.)
The restraint coefficient obtained has close relationship to reaction force due to welding by the equation
R=phlΔ_l
where R: reaction force due to welding
p: restraint coefficient
h: thickness of plate
l: weld length
Δ_l: average elastic dislocation along weld line
The relationship was confirmed for Lehigh cracking test specimens as shown in Table 2. (See Columns (5) and (8) in Table 2.)