Determination of local characteristics for the application of the Weakest‐Link Model

The Weakest‐Link model is based on defects that are statistically distributed within the material with local stress. The failure at least at one defect causes the failure of the total structure. Based on this model, the so‐called statistic size effect can be evaluated in the case of cyclic loading and in the case of static loading the failure behaviour of ferritic steel within the brittle fracture range is highlighted. The application of the Weakest‐Link model requires the allocation of the local characteristics ‐ surface and / or volume ‐ to the discrete points of the stress. By using the method “SPIEL” which is independent from the FE code used, the allocation of couples of values ‐surface and stress and / or volume and stress ‐ by a suitable choice of unit load cases is possible. In consequence of the method “SPIEL” particularities are to be taken into consideration. In the present paper these particularities will be described exemplarily for the FE programs ABAQUS¹ and ANSYS².     

Eine Methode zur Bewertung von Restspalten in Schweißnähten an T‐ und Kreuzstoßverbindungen

A method to assess root gaps in welded seams of T‐ and Cross joints. Design and welding codes require full penetration in case of complete butt splices. This often ends up to time‐ and cost‐intensive repair work for the executing steel work company. Quite recently, classification methods and along with these, a method of tolerating such partial penetrations (here: root gaps) to be considered right from the weld design or planning stage were actually missing, so that additional manufacturing work was preassigned from the outset. Within an AiF‐DASt‐research project [1] an assessment method has been developed basing on an accurate detection of root gaps by non‐destructive testing methods. That method guarantees the demanded structural reliability of a steel structure despite of leaving a weld that actually is to be rejected due to insufficient penetration according to the conventional codes.     

Sicherheit gegen Rissentstehung in statisch beanspruchten Stahlkonstruktionen

Die in den im Stahlbau zur Vermeidung von Sprödbrüchen neu eingeführten bruchmechanischen Regeln getroffene Voraussetzung, dass bereits ein Riss in der Konstruktion vorhanden ist, gibt für vorwiegend ruhend beanspruchte Konstruktionen die Veranlassung, ein Kriterium für die Rissenstehung zu suchen. Dieses ergibt sich als Kriterium für die erste Hauptdehnung, wobei deren kritischer Wert von Parametern abhängt, die den Bruch beeinflussen. Dazu werden bei Variation dieser Parameter Versuche und nichtlineare FE‐Berechnungen durchgeführt. Variiert werden: die Form und die Dicke der Versuchskörper, die Festigkeit und die Kerbschlagzähigkeit des Stahls sowie die Prüftemperatur. Es werden sowohl ungeschweißte wie auch geschweißte Proben untersucht. Daraus ergibt sich eine von der Mehrachsigkeit des Spannungszustandes, der Temperatur, der Bauteildicke sowie der Festigkeit und der Zähigkeit des Stahls abhängige Funktion für den Wert der ersten Hauptdehnung, bei der die Rissentstehung zu erwarten ist. Mit der Mehrachsigkeit werden auch die Schweißeigenspannungen erfasst. Safety against crack initiation in steel structures without fatigue loading. When fracture mechanics is used for selection of steels to avoid brittle fracture the indispensable crack assumption the question of crack initiation arises with structures without fatigue loading. The crack initiation is assumed to depend upon the maximum principal strain, where the critical value of the principal strain is a function of several determining factors. Experimental tests and accompanying nonlinear FE‐analyses are performed with unwelded and welded specimens with variation of: shape, thickness of specimens, strength and Charpy impact toughness of the steel and test temperature. The consideration of triaxiality of the stress state and maximum principal strain obtained from the nonlinear FE‐analysis for the load at crack initiation in the test together with the other test parameters leads to a criterion for crack initiation which gives the critical maximum principal strain as a function of triaxiality, temperature, strength, impact toughness and thickness of the material. This includes the effect of welding residual stresses by their effect on the triaxiality of the stress state.