Jorma Salonen

Accuracy requirements for life assessment

Abstract Several methods can be regarded as routine tools for creep life assessment of power plant components. The selection and extent of applying a technique, as well as interpretation of the results, varies, but the main challenges in reliable predictions remain quite similar. In general, apart from small tubes inside the boiler, the final failure is rarely expected in components and locations where destructive sampling would be easy. Representative sampling, e.g. for creep testing of welded joints of heavy components, is particularly problematic, although miniature specimens can be helpful. However, the methods based on creep testing appear to have potential for long-term life prediction. Nondestructive metallographic inspection with replicas combined with other NDT is suitable for welds, but the present judgement on safe life is limited to short- or medium-term predictions. Improved monitoring of the service history in terms of mechanical, thermal and environmental loadings would deserve more attention. To fully profit from this requires more sophisticated instrumentation than the present average, and a systematic approach in dealing with the accumulating service records. This is likely to be justified in terms of the life cycle costs of most old plants.

Creep Damage, Ductility and Expected Life for Materials With Defects

Defects can pre-exist and grow by creep in structures subjected to loading at high temperatures. As structural integrity is not necessarily conveniently predicted and managed by applying design and life assessment techniques intended for nominally defect-free material, it is important that methods are available for quantified and safe assessment of defects. In addition to the assessment methods, also materials behaviour will affect the likely outcome. In particular, ductility of the materials is important, and unfortunately ductility tends to decrease when shifting from short-term testing to long term creep conditions. In this paper, two examples are shown of materials with such ductility effects when combined with defects. The first example involves 316H stainless steel subjected to creep loading with an extensive crack-like defect, resulting in a transformation from microscopically ductile to brittle intergranular cracking within a relatively modest time span. The second example will demonstrate a corresponding shift in OFP copper that shows a radical ductility and life reduction in creep when including so small weld defects that they would be undetectable in conventional NDT.Copyright