Weld induced residual stresses largely contribute to premature failures in welded structures, especially in the presence of weld defects. PWHT is used to reduce the risk for such failures by its effects on residual stress and microstructure.
This study concerns the effectiveness of PWHT in reducing welding stresses to acceptable levels. The project involves extensive experimental and analytical investigations.
Experimentally, stress relief of three large multi-pass welded pipe-on-plate specimens have been fabricated and examined. Residual stresses are determined in the parent materials, 12-50 mm thick, at weld toe and weld root. Experimental centre-hole drilling and block sectioning techniques are used to measure the surface and through thickness residual strains.
Theoretically, an axisymmetric representation of the pipe-on-a-plate specimens in a commercial available elastic-plastic-creep finite element package provides good agreement with the experimental results.
The study shows that PWHT does not totally relieve residual stresses at weld-ments. It suggests that PWHT should at least meet the minimum basic standards of temperature and time recommended in BS 5500 in order to reduce residual stresses of yield magnitude to below 20% yield stress. However, the code concession (BS 5500) should be applied with care as less stress relief will be achieved. The study finds the unacceptable use of joint or material thickness as the determining factor as to whether a joint needs PWHT.
The analytical study highlights creep as the main mechanism by which stresses relax during PWHT. The final state of stress of a heat treated joint is not very much affected by the heating rates of the joint is held at temperature for an hour or more. This has led to a simplified creep-relaxation model which broadly supports the BS 5500 concession formula for PWHT.