No abstract

Two solutions for the detailed mechanics of tube extrusion are presented. One is based on the theoretical velocity field, and the other on the flow field observed experimentally. The theoretical solution makes use of admissible velocity fields containing no velocity discontinuities. Experimental flow patterns are obtained for commercially pure lead and a superplastic alloy of the eutectic of lead and tin. The two solutions are compared in terms of velocity components, grid distortions, and strain and stress distributions, and very good agreement between the two solutions is revealed.

The weld overlay process has been developed and applied to repair of nuclear reactor pipe girth welds for many years in BWR plants. The objectives of such repairs were to induce compressive axial residual stresses on the pipe inside surface, as well as increase the pipe thickness with a weld material that is not susceptible to stress-corrosion cracking. Hence, understanding the residual stress distribution is important to evaluate the reliability of pipe joints with weld overlay repairs. In this paper, a six-inch diameter Schedule 120 stainless steel pipe with an overlay thickness of 7.87 mm (0.31 inch) was picked as a validation case. Weld sequencing effects were thoroughly studied. The residual stresses were calculated by using thermal elasto-plastic finite-element analysis (FEA). After comparing results using different weld sequences, it was found that the calculated weld residual stresses on ID surface were very sensitive to weld sequencing in FE analyses as well as internal cooling rate. The influence of the weld sequencing was relatively secondary to the pipe distortion. An optimum (producing compressive residual stress on the ID surface) weld sequencing was obtained and applied to a 711.2 mm (28-inch) diameter pipe-to-elbow girth weld with an overlay thickness of 24.9 mm (0.98 inch) and a pipe thickness of 29.5 mm (1.16 inch).

The finite element method was used to obtain the stress and strain distributions in sheet metal on nonaxisymmetric flat punch heads in stretching. Displacement boundary conditions were assumed. The experimental investigation was carried out by stretching securely clamped grid blanks with suitable punch heads; boundary displacements and strains were obtained from the observed deformation pattern on punch heads. A comparison with an experiment in terms of effective strain and thickness strain distributions and strains along the longitudinal axis reveals excellent agreement. The finite element method was also applied to bore expanding in a circular plate. Strain distributions were obtained for two types of materials, one isotropic and the other material having thickness anisotropy. The results showed excellent agreement with exact solution and experimental results.

When the ASME Code fatigue curves (S-N curves) are used in the assessment of high frequency cyclic stresses (such as those produced by flow-induced vibrations), the question arises as to the need for an E correction (i.e., multiplying the calculated cyclic stress by the ratio of the E value at the room temperature and the E value at the temperature used in the stress analysis). This question becomes significant for materials such as stainless steels when the two sets of S-N curves up to the 2007 Edition of the Code [1] are specified: i) the first curve covered the cyclic range of 10 to 106 cycles and specified an E value. This curve covered mostly the strain controlled fatigue data for which the correction for E is required. ii) The second curve covered the cyclic range of 106 to 1011 cycles and didn’t provide a specific E value. This curve covered mostly the load controlled fatigue test data for which the correction for E is not required since the stress was independent of E (stress was either P/A for axial loading or Mc/I for bending). However the 2010 and subsequently the 2013 Editions of the Code [2] combined the two curves into a single curve with a cycle range of 10 to 1011 cycles with E value specified at room temperature. This means that the E correction applies across the board for the entire cyclic range of 10 to 1011 cycles including the high cycle end where the test results are independent of E. The inclusion of the E correction for the high cycle fatigue range presents a problem for the evaluation of components with vibratory loading. The present paper describes the results of a thorough review of the past technical basis papers for the ASME Code S-N curves and examines the necessity for E correction at the high cycle end of the Code S-N curves for stainless steels.