[en] In creep tests within the temperature range 550-700deg C short regimes of elevated creep rate within the strain range 2% were observed at variants of the tempered martensitic chromium steel X 18 CrMoVNb 12 1 (DIN 1.4914). These results in undesirable time yield limit-curves which may even exhibit a retrograde shape. A correlation could be found between the creep anomaly and the ratio of the minor alloying elements Al and N. A recommendations is given for a necessary addition to the existing alloy specification. (orig.)

[en] The applicability of conventional strength hypotheses in the long-term creep range was examined by means of multiaxial loaded small and large-scale specimens of 1% CrMoNiV-steel at 550deg C. The experimentally determined deformation behaviour in the primary and secondary creep range was simulated by means of a finite element analysis. The creep deformation capacity is influenced by the degree of multiaxiality. The validity of a strength hypothesis depends on the type of load and on deformation resp. failure criteria. (orig.)

[en] The growth of short and long fatigue and creep-fatigue cracks was examined in various steels. The experiments were carried out to test if recent models for creep-fatigue crack growth can be applied to describe the influence of strain rate on the growth of large and small cracks. Damage parameters were derived from these crack-growth laws for the life prediction under complicated loading. These parameters were applied to predict the life of smooth specimens cycled at different temperatures and deformation rates. Compared to conventional life predictions better predictions were obtained presumed the growth mechanism of the short cracks did not change for the range for the range of temperature and strain rates in question. (orig.)

[en] Specimens transversely cut from weldments and tested up to failure in long-term creep tests were subjected to metallographic examinations. The weldments were mainly similar and dissimilar joints between CrMo- and CrMoV-steels. The investigations were focussed on the determination of fracture location and damage mechanisms. In the long-term range, where usually a reduction of strain to failure occurs, maximum creep damage as well as fracture location were observed in the fine-grained HAZ in most cases. At high stresses, a migration of the fracture location into other parts of the HAZ, or even into the base metal or weld metal can take place. Dependent on material, temperature and stress a reduction of the welds' creep strength in comparison to the base metal occurs. (orig.)

[en] The creep exposure of metallic parts usually occurs under time-variable loading conditions. For the prediction of creep behaviour under step loading a simple mechanical models is employed. The model is based on the time dependence of the material structure during creep. Therefore, the characteristic parameters of the model are considered to be functions of time and are correlated with the time dependence of the material structure parameters. For the definition of these functions a structure mechanical model has been employed. The mean internal material resistance has been considered as an appropriate structure parameter. The characteristic parameters of the model are defined in each step loading on the basis of the change of the internal stress derivative. For the verification of the model prediction the numerical results are compared with experimental results of the austenitic steel X8 CrNiMoNb 16 16 for the orthogonal cyclic stress function. (orig.)

[en] Due to thermal gradients and load changes steam-turbine casings suffer combined creep-fatigue loading. Creep and fatigue portions depend on service conditions. Special interest is focused on the behaviour of creep-fatigue cracks, originated from potential defects due to construction and manufacturing weldments, on residual lifetime assessment. The project deals with the crack-growth behaviour at 530deg C and r.t. of welded and unwelded heat-resistant cast steels: 1% CrMoV-cast (GS-17 CrMoV 5 11, two melts: low and high V-content) and 12% CrMoV-cast (G-X 22 CrMoV 12 1). The crack planes of CT-specimens investigated, taken from manual arc-welded plates, were situated in the heat-affected zone (HAZ). The fatigue, creep and combined creep-fatigue crack-growth behaviour was investigated by means of low-frequency (0.5 and 0.05 Hz) tests, holding-time (20 min) tests and constant load tests up to about 5 000 h. Metallographic investigations show that in the case of both 1% and 12% CrMoV-cast weldments the preferred crack path lies in the outest i.e., the high tempered and not α/γ transformed region of the HAZ. A strong increase on the crack-growth rate at 530deg C is exerted by dominant creep portions. In the case of the 1% CrMoV-cast this acceleration becomes important at stress intensity ranges higher than about 25 MPa √m with some higher sensitivity of the HAZ. For the 12% CrMoV-cast creep dominance begins at about 50 MPa √m. No differences for the HAZ and the base material were found for this material. (orig.)

[en] The influence of dual-phase structure (ferrite + martensite) on the impact toughness and rate of abrasive wear for as-cast and normalized steel (0.29 C, 1.15 Mn, 0.9 Si and 0.95% Cr) is investigated. The increased grain refinement and the well arranged lamellar packets of martensite and ferrite obtained during intercritical quenching (from 810deg C) of normalized and prequenched (from 880deg C) steel created a satisfactory combination between thoughness and abrasive wear resistance over those measured for intercritically quenched as-cast steel with or without prequenching. (orig.)

[en] For design and supervision of high temperature components for power plants the long-term creep-crack behaviour is important. Creep crack tests on different specimens are needed to obtain a description of the initiation and growth of creep cracks which is independent of geometry and load. Such tests were carried out in a wide loading range down to creep crack growth rates of 5.10^-5 mm/h and up to test durations of 14 000 h with specimens of different shape and size. As a result of these tests on steels of type 1%Cr-1%Mo-0.6%Ni-0.3%V, 12%Cr-1%Mo-0.3%V-0.22%C and 12%Cr-1%Mo-0.3%V-0.20%C, the 550deg C-creep crack growth rate is described in smaller scatter bands by the parameter C₂^* than by the parameters C₁^* or K_I. Only in the case of a 12%Cr-2%Ni-1%Mo-steel tested at 450deg C the parameter K_I gave a better fit. The creep-crack initiation can be estimated by means of a two-criteria diagram. For larger specimens, this diagram is too conservative and this is presumably also the case for larger components. (orig.)

[en] Creep-rupture specimen of similar welded joints of X6CrNi 18 11 (1.4948/AISI 304) and X6CrNiMo 17 13 (1.4919/AISI 316) show only low elongation after longtime testing. The reason for this loss of ductility was found by metallographic investigations. The weld metal of five joints had ferrite contents from 3 to 7.5%, due to a primary ferritic solidification. During creep testing in the temperature range from 500 to 800deg C carbide precipitation takes place at the austenite-ferrite grain boundaries, because delta ferrite is not in a state of equilibrium at these temperatures. After carbon has been used up, the remaining delta ferrite changes into sigma phase, if its alloying element content is high enough. In the upper temperature range, coagulation of sigma phase is dominating. At these large particles grain boundary migration is hindered. During the grain boundary sliding sigma phase particles break and initiate creep cracks. In fully austenintic weld metal sigma precipitation starts at austenite - austenite grain boundaries. During creep testing, void formation starts at sigma particles. Growing of voids leads to grain separations. (orig.)

[en] The various possibilities of localized corrosion attack which can occur at welds of unalloyed and low alloy steels in water are discussed. The discussion is based on the electrochemical and chemical principles of corrosion involving oxygen consumption and the formation of corrosion cells under these conditions. It is shown that the conceptions of the causes and mechanisms of such corrosion phenomena at welds pointed out in the literature are difficult to understand on the basis of corrosion involving oxygen consumption and must be incorrect. In contrast, in this paper reference is made to metallurgical factors that influence the corrosion at welds: chemcial composition, with the concentrations of both silicon and sulfur being particularly important, microstructure, and non-metallic inclusions of critical size and shape. To a large extent, the intensity of localized corrosion attack at welds is also determined by parameters of the aggressive environment, e.g., its neutral salt content and aeration. Considering these parameters in detail, one can easily understand the corrosion processes occurring and also establish suitable investigation procedures. It is also possible to develop effective protection measures and favourable welding techniques. (orig.)