Experimental steels similar in composition to structural grades were prepared from weld metal deposits to study the formation of acicular ferrite under conditions experienced in the heat affected zone for a range of welding processes. The formation of acicular ferrite under these conditions is found to be dependent on the presence of a suitable distribution of oxide inclusions > 0·4 μm in size. The characteristics and proportion of acicular ferrite in the microstructure also depend on the prior austenite grain size and cooling rate. The relationship between these factors is presented in a simplified quantitative model, which is supported by data from limited welding trials. Metallographic observations suggest that acicular ferrite forms in two stages. The first involves the formation of relatively large primary acicular ferrite plates by multiple nucleation at intragranular inclusion sites, and the second involves the formation of many smaller acicular ferrite grains that grow sympathetically from the primary plates.

A theoretical model is developed for determining the optimum notch position in an implant test used for predicting the susceptibility to hydrogen cracking during welding. Using a microcomputer for processing the equations for weld-bead geometry and heat flow during welding, a microstructural cross-section, with the notch positioned at the center of the grain-growth zone, and an implant testing diagram, showing the notch position and microstructure as a function of welding parameters, can be generated. A single bead-on-plate weld is used to determine the unknown kinetic and geometrical constants in the equations. It is shown that notch position is very sensitive to the type of welding process employed and that implant diagrams thus can be used to position the notch with greater reliability and hence reduce the scatter in the fracture loads measured in this test.

The paper improves and extends kinetic models to include: precipitate coarsening; the use of semi-empirical equations including carbon-equivalence to predict microstructure and hardness; a comparison between the theory and data obtained from different types of real welds; and an alternative, more easily used diagram.

Data for carbide dissolution and grain growth in the heat affected zone of a weld can be assembled into a diagram showing the extent of each, for different weld cycles, and at different points in the zone. The diagrams are based on elementary kinetic models for grain growth and carbide dissolution, integrated over the weld cycle. The sets of kinetic constants which appear in such a treatment are determined by fitting the equations to data from real or simulated welds, at certain fixed points. Diagrams are presented for six steels. As well as summarizing much data they allow the effect of change in weld procedure, or of preheat, to be predicted.

Summary form only given. The authors present an interim report on the effect of various alloying additions in the weld deposit and base material on the precipitate dispersion and stability, the microstructure and toughness of microalloyed constructional steels, in both the as-welded as well as welded and stress-relieved conditions. Materials investigated include welded AlTi, AlTiV, AlTiNb and AlV steels with the use of electrodes either with or without Mo additions

Criteria based on minimum energy and maximum stress considerations have been developed for the first stages of growth of martensite in steel. It is shown that the critical size of nucleus at which dislocation-assisted growth can occur is about 25 nm diameter. The mode of growth governing low carbon lath martensite forming on {111} austenite habit planes is found to be fundamentally different from that of plate martensite which forms on irrational habit planes. A mechanism for lath thickening and growth is advanced in which dislocations nucleate at the thickest section of the lath and then follow after the coherent leading edge of the lath to generate ledges. Plate growth on the other hand can occur by the formation of twins which nucleate and thicken laterally behind the coherent leading edge of the plate.

Nous présentons des critères basés sur des considérations d'énergie minimale et de contrainte maximale, pour les premiers stades de la croissance de la martensite dans l'acier. Nous montrons que la taille critique des germes pour la croissance par un mécanisme de dislocations est d'environ 25 nm de diamètre. Le mode de croissance de la martensite en latte à bas carbone qui se forme sur le plan d'accolement {111} de l'austénite est fondamentalement différent de celui de la martensite en plaquette qui se forme sur des plans d'accolement irrationnels. Nous proposons un mécanisme d'èpaississement et de croissance des lattes, selon lequel des dislocations germent sur la plus grande section de la latte, puis suivent la marche cohérente de tête pour former d'autres marches. La croissance des plaquettes peut par contre se produire par la germination et l'épaississement latéral de macles derrière la marche de tête de la plaquette.

Für die Beschreibung der ersten Stadien des Martensitwachstums in Stahl wurden Kriterien entwickelt, die auf Betrachtungen der Minimalenergie und der gröβten Spannungen beruhen. Es wird gezeigt, daβ die kritische Gröβe eines Keimes, an dem versetzungs-unterstütztes Wachstum auftreten kann, ungefähr 25 nm beträgt Es ergibt sich, daβ die Bildung niedrig gekohlten lattenförmigen Martensits auf {111}-Austenit-Habitebenen von einer Wachstumsmode bestimmt wird, die sich grundlegend von derjenigen plattenförmigen Martensits, der sich auf irrationalen Habitebenen ausbildet, unterscheidet. Es wird ein Mechanismus für das Wachstum des lattenförmigen Martensits in Dicke und Länge vorgeschlagen. Bei diesem entstehen Versetzungen an der dicksten Stelle der ‘Latte’ und folgen dann nach der kohärenten Führungskante der Latte und bilden Vorsprünge. Plattenwachstum andererseits kann durch Bildung von Zwillingen erfolgen, die seitlich hinter der kohärenten Führungskante der Platte entstehen und dicker werden.