13th International Conference on Fracture June 16–21, 2013, Beijing, China -1- Improved Exploitation of High-strength Steels in Pressure Vessel Design by Simulations Based on Damage Mechanics Victoria Brinnel1,*, Christian Schruff1, Sebastian Münstermann1 1 Department of Ferrous Metallurgy, RWTH Aachen University, Aachen D-52072, Germany * Corresponding author: Victoria.Brinnel@iehk.rwth-aachen.de Abstract Pressure vessels are subjected to regulation by European standards, which define design procedures and nominal material properties that are guaranteed by the steel producers. Over-conservative safety factors for materials with high yield-to-tensile ratios hinder the application of modern high-strength steels despite their excellent toughness and possible economic and ecologic benefits. Damage mechanics enable a more adequate failure description for these steels since they can consider ductile crack initiation as limit state. Such concepts should refer to nominal material characteristics to enhance their applicability in design procedures. A method is presented to correlate the nominal Charpy impact toughness to simulations with the Gurson-Tvergaard-Needleman-model (GTN). The resulting GTN parameters are used in cell model simulations to derive strain-based empirical failure criteria as nominal ductility measure for simulations on component level. A sensitivity study on the GTN parameter choice shows that the initial pore volume is the most relevant parameter, but other non-unique parameters also influence the failure prediction. The impact of the individual parameters is discussed. By the definition of a reliable calibration scheme for the GTN parameters this approach enables an identification of adequate lower bound failure criteria for high-strength steels in pressure vessels as a preliminary pressure vessel simulation demonstrates. Keywords Strain-based Design, Pressure vessel, Damage mechanics, Damage curve 1. Introduction In Europe, pressure vessels are subjected to regulation by the European Standard EN 13445, which contains the corresponding design rules. Additionally, the material requirements for pressure vessel steels are part of EN 10028. Therein, nominal material properties are defined, which specify minimum values for the mechanical properties. These have to be guaranteed by the steel producers and are often exceeded, especially by high quality steels. The nominal characteristics form the basis of the design process for pressure vessels. The required wall thickness is determined in dependence of the materials’ nominal yield or tensile strength. The relevant design stress is specified in EN 13445 as the minimum of the yield strength divided by a safety factor of 1.5 or the tensile strength divided by 2.4. Therefore, the yield-to-tensile (Y/T) ratio determines the magnitude of the applied safety factor. From a Y/T-ratio of 0.625 on, the tensile strength with an increased safety factor of 2.4 is relevant for the design. Modern high strength steels are often characterised by such an increased Y/T-ratio, e.g. steel grade P500Q with Y/T = 0.85 and P690Q with Y/T = 0.9. By the application of the increased safety factor, only a poor exploitation of their load bearing capacity can be achieved. In this safety factor domain, the design stress is only 41% of the yield strength. Consequently, these rather expensive steel grades are rarely applied in pressure vessel design. However, their application would foster substantial economic and ecological benefits by e.g. thinner walls, which result in lower welding, energy and transportation cost. The current safety factors are mainly based on experience and do not consider the actual failure behaviour [1]. Since EN 13445 requires brittle fracture to be excluded by the toughness requirements, an improved prediction of ductile crack initiation is needed. Hence, to improve the exploitation of high strength steels in pressure vessel design it is necessary to include a description of their ductility. The methods of damage mechanics are able to provide such predictions [2]. Although research on this field has been conducted for several decades, the methods of damage mechanics are not used regularly in industrial applications, let alone considered standard design procedures. One reason is that the application of damage mechanics models requires considerable
RkJQdWJsaXNoZXIy MjM0NDE=