Winterthur, Switzerland
Winterthur, Switzerland

Time filter

Source Type

Schindler H.-J.,Materials Tec AG | Kalkhof D.,Swiss Federal Nuclear Safety Inspectorate ENSI
Journal of Testing and Evaluation | Year: 2015

An increased loading rate causes not only a shift of the ductile-to-brittle transition curve, but also a slight change of the shape of this curve. The latter tends to become steeper as the loading rate increases. This effect was observed even at loading rates that are considered to be quasi-static according to ASTM E1921 [ASTM E1921-13: Standard Test Method for Determination of Reference Temperature, T0, for Ferritic Steels in the Transition Range, Annual Book of ASTM Standards, ASTM International, West Conshohocken, PA, 2013. Actually, the coefficient 0.019 in the exponent of the MC turned out to be substantially higher at elevated loading rates. This means that a basic assumption of the evaluation procedure of ASTM E1921 is not met, which leads to an increased uncertainty of T0 or T0,x, respectively. This effect of is most pronounced in testing at elevated loading rates using the single-temperature option. An improved method to determine the reference temperature is proposed, where the exponent in the median transition curve in ASTM E1921 is considered to be an open parameter. The resulting reference temperatures, denoted as T100 and T100,x, respectively, are expected to be more accurate than standard T0, which is confirmed by comparison with experimental data. Furthermore, the validity requirements for T100 are less restrictive and more suitable than those in ASTM E1921, since they do not depend on the outcome of the tests. Based on the improved data, an improved estimation formula for the rate-induced shift of T0 is derived. Suggestions are made for improvement of ASTM E1921. Copyright © ASTM Int'l (all rights reserved).

Viehrig H.-W.,Helmholtz Center Dresden | Houska M.,Helmholtz Center Dresden | Kalkhof D.,Swiss Federal Nuclear Safety Inspectorate ENSI | Schindler H.-J.,Materials Tec AG
International Journal of Pressure Vessels and Piping | Year: 2015

The multi-layer beltline welding seam of the Biblis C reactor pressure vessel was characterized by hardness, tensile, ISO-V impact and fracture toughness testing. The reference temperature, T0, was determined according to the test standard ASTM E1921 at different thickness positions of the multi-layer welding seam. Additionally, the influence of the specimen orientation on the ISO-V ductile-to-brittle transition temperature and T0 was investigated. In contrast to the T-S orientation (crack extension through the thickness) the crack front of the T-L oriented specimens (crack extension in welding direction) penetrates several welding beads. By means of fractographic and metallographic analyses of the fractured surface of fracture mechanics SE(B) specimens was shown that the distribution of the crack initiation sites is not necessarily correlated to the structure of the different welding beads along the crack front. Furthermore, it was found that the scatter of the fracture toughness values at cleavage failure, KJc, determined with T-S specimens is significantly higher than in case of the T-L specimens. T0 values measured at different thickness locations of the multi-layer welding seam vary in a range of about 40 K.The evaluated T0 values are used to determine the reference temperature RTTo for indexing the lower bound curve KIc(T) according to the Regulatory Guide ENSI-B01 for the ageing surveillance of nuclear power plants in Switzerland. It could be shown that the KIc values converted from the KJc values are enveloped by the lower bound curves. © 2015 Elsevier Ltd.

Schindler H.-J.,Materials Tec AG | Kalkhof D.,Swiss Federal Nuclear Safety Inspectorate ENSI | Viehrig H.-W.,Helmholtz Center Dresden
Engineering Fracture Mechanics | Year: 2014

The fracture behaviour of a component or specimen that contains a sharp notch is governed essentially by the same theoretical relations known from cracks. The blunt notch root only causes an increase of the resistance against crack initiation, which depends on the fracture mechanism. In the present paper, the relation between fracture toughness and notch toughness is investigated by simple analytical models. The derived formulas were compared with experimental results obtained from fracture toughness tests on RPV-steel 24 NiCrMo 3-7 at various temperatures. 1T-CT- and 0.4T-SEB-specimens that contained a sharp notch with a root radius of 0.06mm introduced by spark erosion (EDM) instead of the standard fatigue crack were used. The predictions were found to agree well with the experimental data. The effect of the notch radius on fracture toughness is most pronounced in the brittle to ductile transition regime, where fracture toughness can be characterized by the master curve and the corresponding reference temperature T0 according to ASTM E1921. Accordingly, the effect of the notch radius can be quantified by a shift of T0. Since the shape of the transition curve depends on the notch radius, the procedure of ASTM E1921 to determine T0 is not applicable. An alternative is suggested. As limiting cases, ductile tearing and brittle fracture are also considered. © 2014 Elsevier Ltd.

Schindler H.-J.,Materials Tec AG | Kalkhof D.,Swiss Federal Nuclear Safety Inspectorate
13th International Conference on Fracture 2013, ICF 2013 | Year: 2013

The inherent scatter of fracture toughness of ferritic steels in the brittle-to-ductile transition regime require statistical methods to be applied for testing and evaluation. However, for engineering purposes lower bounds of KIc such as the ASME-reference curve are often preferred since they allow deterministic worst-case predictions to be made. So the question is how to derive lower bounds of a quantity that is governed by weakest-link-statistics. Actually, neither the MC-approach nor the empirical ASME-reference curve deliver well-founded lower bounds for components of relatively small thicknesses. A theoretical model is suggested to fill this gap. The key element of the approach is the hypothesis that the weakest-link-effect is saturated at a certain thickness. The corresponding upper limit of size-dependence turned out to be close to the minimum thickness required for plane-strain conditions at the crack-front. The derived mathematical relations enables KIc to be calculated from KJc as measured on a smaller specimen. In reverse, from a lower-bound KIc as provided by the ASME-code a thickness-dependent lower bound of KJc can be obtained. The proposed model is shown to yield predictions that are consistent with experimental data as well as with the ASME-lower bound. Copyright © (2013) by International Conference on Fracture.

Leinenbach C.,Empa - Swiss Federal Laboratories for Materials Science and Technology | Schindler H.-J.,Materials Tec AG | Baser T.A.,Empa - Swiss Federal Laboratories for Materials Science and Technology | Ruttimann N.,ETH Zurich | Wegener K.,ETH Zurich
Engineering Failure Analysis | Year: 2010

Brazed components are widely applied in industry and are often subjected to complex loading conditions. Even if such components often contain brazing defects, no failure assessment procedures for brazed joints are reported in the literature. In this work, the deformation and the quasistatic behaviour of brazed joints of the martensitic stainless steel X3CrNiMo13-4 were investigated. This includes the determination of stress-strain-curves as well as of the fracture toughness. In addition, the mechanical behaviour of components such as specimens with T-joint geometry under tensile loading were characterized. In order to consider the effect of brazing defects on the structural integrity, typical defects with different sizes and geometries were introduced in the brazing zone. The experimentally determined material parameters were used for additional numerical deformation analyses by means of finite elements (FE). Both the experimental and the numerical results were in good agreement with predictions according to the R6 method and provided a basis for the engineering defect assessment of brazed components based on failure assessment diagrams. © 2009 Elsevier Ltd. All rights reserved.

Leinenbach C.,Empa - Swiss Federal Laboratories for Materials Science and Technology | Koster M.,Empa - Swiss Federal Laboratories for Materials Science and Technology | Schindler H.-J.,Materials Tec AG
Journal of Materials Engineering and Performance | Year: 2012

This work aims at the development of lifetime estimation procedures for defect-free and defect-containing brazed joints. Preliminary investigations were performed to measure the influence of specimen geometry on the joint strength. To estimate the influence of defects on the fatigue lifetime, defect-free specimens were compared with specimens containing defects. The experiments show that defect-containing specimens provide considerably lower joint strengths than defect-free specimens. The decrease of the fatigue strength with increasing defect size can be shown, but the direct comparison of different defects is not possible with S-N-curves. Based on the experimental results and on theoretical investigations, a procedure was developed to estimate the lifetime of defect-free and defect-containing brazed joints based on the stress intensity caused by a defect © ASM International.

Schindler H.-J.,Materials Tec AG | Leinenbach C.,Empa - Swiss Federal Laboratories for Materials Science and Technology
Engineering Fracture Mechanics | Year: 2012

Experimental investigations have shown that cracks in a bonding interface such as adhesives or brazings under cyclic loading behave fundamentally different from cracks in homogeneous metals. In particular the slope of da/. dN-vs.-Δ. K-curves is much steeper and the threshold value higher than in homogeneous metals. In the present paper, the general case of a fatigue crack located in a thin layer of elastic-plastic material embedded in an elastic surrounding is considered analytically. Since the layer is thin compared to the size of the plastic zone, linear-elastic fracture mechanics is not applicable. Therefore Paris' law is formulated in terms of CTOD. It was found that a non-linear shielding mechanism reduces CTOD of a crack in the layer compared to a crack in bulk material under the same stress intensity. The shielding decreases with increasing stress intensity factor. This effect leads to the steep da/. dN curves. Based on the hypothesis of a linear damage accumulation in the active plastic zone, an analytical formula is derived to estimate the fatigue crack growth rate as well as the threshold values of cracks in thin interlayers. The predictions agree well with the experimental data. © 2012 Elsevier Ltd.

Schindler H.-J.,Materials Tec AG | Leinenbach C.,Empa - Swiss Federal Laboratories for Materials Science and Technology
Journal of ASTM International | Year: 2010

The fatigue crack growth behavior of a brazed joint is characterized by a Paris-exponent that is much higher than the one of bulk metals, which means that brazed components have only a short residual fatigue life after the initiation of a fatigue crack. Therefore, the threshold of fatigue crack growth of brazed joints is of particular importance in the fatigue analysis of brazed components and needs to be understood well. Residual stresses have to be considered in crack growth analysis in the threshold regime. The corresponding effects are explored experimentally and theoretically. The stress intensity factors due to the residual stresses in a brazed T-joint of compressor impellers were measured by using the cut compliance method. It was found that residual stresses are present but relatively small. They do not affect the endurance stress significantly. A concept for an endurance analysis of a brazed T-joint that includes imperfections and residual stresses is suggested. Copyright © 2010 by ASTM International.

Schindler H.J.,Materials Tec AG
Proceedings of the Biennial International Pipeline Conference, IPC | Year: 2010

Under special circumstances a fast propagating crack can be triggered in pipelines by just a small local damage. In order to assess the safety with regard to this type of catastrophic failure, an analytical model was established to describe the process of large ductile tearing mathematically. Using a simplifying kinematical model and the hypothesis that the tearing fracture process is governed by a constant CTOA, the total energy dissipation rate can be quantified and compared to the available fracture energy. Since the Charpy fracture energy KV is often the only available toughness-related material parameter of existing pipelines, the toughness parameters used in the model, particularly CTOA, has to be determined from KV. This was achieved by using the same 2-parameter model of ductile tearing to analyse the fracture process in bending. A closed form relationship was obtained between the minimum pressure required for fast ductile crack propagation and the system and material parameters. It could be shown that rapid ductile tearing requires the hoop stress to exceed a certain limit, which depends on the geometrical parameters of the pipeline and KV of its material. The analysis was verified by comparison of the results with the experimental data of full- scale burst tests available in the literature. Unlike empirical correlations, the derived analytical formula seems to be universally valid, regardless of pipe dimensions or steel grades. Copyright © 2010 by ASME.

Schindler H.-J.,Materials Tec AG
Procedia Engineering | Year: 2014

For physical reasons fracture toughness KJc of ferritic steels in the brittle-to-ductile transition regime is affected by a pronounced scatter, which requires statistical methods to be applied for evaluation of test results as well as for application in safety analysis of structures. For this purposes the probabilistic Master-Curve-Approach according to ASTM E1921 is often used. However, for engineering purposes like a screening safety analysis of a defect-containing component it is usually preferable to use a deterministic lower bound. However, within the framework of ASTM-standards there is no possibility to determine KIc experimentally. So KIc needs to be determined indirectly from the reference temperature T0. In the present paper it is shown how lower bounds of fracture toughness - either plane strain KIc for large components or thickness-dependent KIc for smaller ones - can be derived from T0, and how the latter can be determine with sufficient accuracy from one or a few KIc values. © 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license.

Loading Materials Tec AG collaborators
Loading Materials Tec AG collaborators