Effect of welding variables and solidification substructure on weld metal porosity

Porosity is defined as cavity-type discontinuities formed by gas entrapment during solidification. Causes of porosity in fusion welds are the dissolved gases in weld metal and welding process variables that control the solidification rate. To study the mechanisms of porosity formation in weld metal, single-pass gas tungsten-arc weld metal was produced using the bead-on-plate technique on three nickel-copper alloys (80 wt pct Ni-20 wt pct Cu, 65 wt pct Ni-35 wt pct Cu, 35 wt pct Ni-65 wt pct Cu). Four different welding speeds were used under various amounts of nitrogen content in argon-shielding atmosphere. A qualitative model was proposed to characterize the effect of welding variables and solidification substructure on bulk and interdendritic porosity formation. Increasing amounts of nitrogen gas (from 0.2 pct to 6.0 pct in volume) introduced in argon-shielding atmosphere increased the amount of porosity in weld metal. The amount of bulk and total porosity increased as the solubility of nitrogen in the weld metal alloy decreased. The solidification rate of the weld pool is the most important factor controlling the mechanism of porosity formation. The observed amount of bulk pores in this study increased with the increase of welding speed; that is, if the time is insufficient for dissolved and evolved gases to escape during solidification, porosity will result. However, a decrease in the amount of interdendritic pores was observed with increasing welding speed in the 80Ni-20Cu and 35Ni-65Cu alloys. This decrease can be related to the effect of solidification rate on the balance between the disjoining pressure, resistance of the liquid film to be disrupted, repulsion of the bubble from the solidification front, and the hydrodynamic force resisting the movement of the bubble. This balance determines the ability of the cellular solidification front to “equilibrium” capture the pores. Furthermore, the observed decrease of interdendritic porosity with increasing welding speed (80Ni-20Cu and 35Ni-65Cu alloys) can also be related to the time for nucleation and growth of pores in the molten weld metal and their entrapment in the interdendritic channels of a dendritic solidification front. This phenomenon is considered a “nonequilibrium capture” of pores. On the other hand, the 65Ni-35Cu alloy that exhibited a structural transition in solidification substructure with the variation of welding speed showed a slight increase in the amount of interdendritic pores. This increase was correlated to the change of pore-capture mechanism from an equilibrium to a nonequilibrium mode as the solidification substructure changed from cellular to cellular dendritic. To substantiate that the controlling mechanism of interdendritic porosity formation is the nonequilibrium capture, a good correlation between the measured mean pore radius and the interdendritic arm spacing was found.     

Effect of hydrogen,restraint and welding conditions on weld metal cold cracking of HSLA steels in the IRC test

The effect of heat input and hydrogen partial pressure of the TIG process without filler wire and of restraining conditions in the IRC test on hydrogen induced cracking have been investigated for weld metals of StE 380, 15 NiCrMo 10 6 (HY 80) and 12 Ni 19. The weld metals are fused from the roots of the plate materials exclusively. The continuously measured combined nominal stresses transverse to the welds are strongly dependent on transformation stress release prior to cold cracking of the respective weld metals. Higher restraints are increasing nominal stresses. Critical heat inputs for prevention of hydrogen weld metal cracking are increased with restraint, hydrogen potentials and carbon contents of the weld metal, 12 Ni 19 revealing no fracture.     

Effect of submerged arc welding parameters on microstructure of SA516 steel weld metal

Abstract

The influence of submerged arc welding (SAW) process parameters on the microstructure of SA516 grade 70 steel weld metal (WM) was investigated in the present work. Steel plates of 17 mm thickness were submerged arc welded using welding currents of 700–850 A and welding speeds of 5·3–15·3 mm s^?1. The morphologies and volume fractions of the various ferrites in the WM were studied using optical microscopy and the morphologies and chemical compositions of the WM inclusions were examined using scanning electron microscopy and energy dispersive X-ray spectrometry. The results showed that the WM grain structure coarsened but the grain width of prior austenite grains decreased with increasing heat input. Also, the proportion of acicular ferrite (AF) in the WM increased initially, while the volume fractions of grain boundary ferrite and Widmanstätten ferrite decreased with increasing welding current. However, with further increasing the welding current above 800 A, less AF was produced. The weld nugget area decreased with increasing welding speed at all currents, but did not affect the amount of AF produced.

Au cours de ce travail, on a examiné l’influence des paramètres du procédé de soudage à l’arc sous flux en poudre (SAW) sur la microstructure du métal soudé (WM) de l’acier SA516, Gr. 70. On a soudé à l’arc sous flux en poudre des tôles d’acier de 17 mm d’épaisseur en utilisant des courants de soudage de 700 à 850 A et des vitesses de soudage de 5·3 à 15·3 mm s^?1. On a étudié la morphologie ainsi que la fraction volumique des diverses ferrites du WM en utilisant la microscopie optique et l’on a examiné la morphologie ainsi que la composition chimique des inclusions du WM en utilisant la microscopie électronique à balayage (SEM) et la spectroscopie à dispersion d’énergie (EDS). Les résultats ont montré que la structure de grain du WM grossissait mais que la largeur de grain des grains antérieurs d’austénite diminuait avec une augmentation de l’apport de chaleur. Également, la proportion de ferrite aciculaire (AF) dans le WM augmentait initialement, alors que les fractions volumiques de ferrite des joints de grain (GBF) et de ferrite de Widmanstatten (WF) diminuaient avec l’augmentation du courant de soudage. Cependant, avec une augmentation supplémentaire du courant de soudage au-dessus de 800 A, moins de ferrite aciculaire était produite. La région du noyau de soudure diminuait avec une augmentation de la vitesse de soudage à toutes les valeurs de courant, mais la quantité d’AF produite n’était pas affectée.     

Control of δ‐γ Transformation during Solidification of Stainless Steel Slabs in the Mould

In order to obtain satisfactory workability properties required for defect free slab and strip production, the parameters of the casting process, e.g. cooling rate at the initial solidification for the alloy in question must, on the one hand, be carefully adjusted. On the other hand, controlling the characteristics of the solidification structure by chemical composition then takes on particular significance. The main aims of the work were to find out the influence of the phase transformation δ‐γ during the initial solidification of different variants of AISI 304 slabs cast in the industrial process on the ferrite distribution on slab surface, and how this relationship could favour the improvements of the surface slab quality. This report contains the joint contributions of the collaborative ECSC project among Centro Sviluppo Materiali (CSM), Krupp Thyssen Nirosta (KTN) and the Department of Ferrous Metallurgy of Rheinisch Westfälische Technische Hochschule Aachen (RWTH).     

The isothermal δ-γ transformation of ferritic-austenitic iron-chromium-nickel alloys

Quenching from ferritizing temperatures of 1350 and 1250 °C the isothermal δ-γ transformation of two iron-chromium-nickel alloys with 26.0% Cr, 6.9% Ni resp. 30.0% Cr, 7.05% Ni at temperatures between 700 and 1200 °C is studied. Three different transformation stages are revealed indicating grain boundary diffusion controlled growth in the first and last stage while volume diffusion seems to act at least partially in the second stage. Isothermal TTT diagrams are established and activation energies calculated.     

Effects of surface active elements on weld pool fluid flow and weld penetration in gas metal arc welding

This article presents a mathematical model simulating the effects of surface tension (Maragoni effect) on weld pool fluid flow and weld penetration in spot gas metal arc welding (GMAW). Filler droplets driven by gravity, electromagnetic force, and plasma arc drag force, carrying mass, thermal energy, and momentum, periodically impinge onto the weld pool. Complicated fluid flow in the weld pool is influenced by the droplet impinging momentum, electromagnetic force, and natural convection due to temperature and concentration gradients, and by surface tension, which is a function of both temperature and concentration of a surface active element (sulfur in the present study). Although the droplet impinging momentum creates a complex fluid flow near the weld pool surface, the momentum is damped out by an “up-and-down” fluid motion. A numerical study has shown that, depending upon the droplet’s sulfur content, which is different from that in the base metal, an inward or outward surface flow of the weld pool may be created, leading to deep or shallow weld penetration. In other words, it is primarily the Marangoni effect that contributes to weld penetration in spot GMAW.     

Fließspannung im Bereich der (γ → α)-Umwandlung von kohlenstoffarmen Stählen

Auswertung von Fließkurven aus den Warmflachstauchversuchen und quantitativer Gefügeanalyse. Anomalie auf der Fließspannung-Temperatur-Kurve im (γ → α)-Bereich bei Umformung während der anisothermischen Umwandlung. Bilanz der Einflußfaktoren.     

The continuous δ-γ transformation during cooling of ferritic-austenitic iron-chromium-nickel alloys

Following previously published isothermal TTT diagrams, continuous TTT diagrams for the δ-γ transformation of a series of different iron-chromium-nickel-alloys containing 23.0–35.0 Cr and 6.45–15.6 Ni are presented. The microstructures and transformation behaviors indicate the same nucleation and growth process as in the isothermal reactions. The critical upper cooling rate for complete undercooling of the δ-γ transformation decreases with increasing chromium- and decreasing nickel contents.     

Der Einfluß der Keimbildung auf die irreversiblen Dimensionsänderungen von Reinsteisen bei zyklischen α/γ/α-Umwandlungen

Versuchsanordnung mit symmetrischem Temperaturprofil. Messung der relativen differentiellen Längenänderung nach α/γ-, γ/α- und α/γ/α-Umwandlungen. Einfluβ der Aufheiz- und Abkühlungsgeschwindigkeit und der Vorverformung der Probe auf die irreversible Längenänderung. Beobachtung der Struktur durch thermische Ätzung. Untersuchung der Substruktur durch chemische Ätzung, Laue-Aufnahmen und elekironenoptische Aufnahmen. Zusammenhang zwischen irreversibler Längenänderung und Keimbildung im flachen Teil des Temperaturprofils.     

Verformung von Manganstahl mit 16% Mn während der γ/ε-Umwandlung

An Stahl mit 0,09% C, 1,0% Si und 16% Mn Untersuchung des Einflusses von Zugspannungen unterhalb der Fließgrenze auf die Verformung und bleibende Dehnung beim Durchlaufen der γ/ε-Umwandlung. Erörterung der Ergebnisse.     

Simulation of the γ–α‐transformation using the phase‐field method

Phase transformation is a powerful tool to change the properties of steels. Of the known transformations especially the γ–α‐transformation is utilised. It occurs in a temperature range relevant for heat treating and hot deformation processes. In this paper an approach is presented in which the γ–α‐transformation is simulated with Micress. This software applies the multicomponent multiphase‐field model, which is based on the reduction of total free enthalpy. Two different steels have been selected for the simulations, an ULC and an IF steel. Dilatometric tests serve as a basis for the simulations. These tests have shown that the transformation behaviours of the two steel grades are governed by two different kinetics. The transformation kinetics of the IF grade is influenced by the microalloying concept applied, resulting in a very slow start of the transformation. This has also been incorporated in the simulations by choosing two different grain boundary mobilities, one main parameter of the simulation. The simulation results of the ULC grade show the huge influence of nucleation undercooling as another one of the main parameters. Both simulation results are satisfying. They show that the phase‐field method offers a strong simulation tool in the area of phase transformation.     

Untersuchungen zum Einfluß des Kohlenstoffs und des Stickstoffs auf die δ-γ-Umwandlung ferritisch-austenitischer Chrom-Nickel-Stähle

Einflußsteigender Kohlenstoff- und Stickstoffgehalte (bis O, 13 % C bzw. 0,27 % N) ferritisch-austenitischer FeCrNi-Legierungen mit rd. 24,5% Cr und 7% Ni auf die Phasengrenzen und Konodenrichtung bei 1200°C. Glühtemperaturabhängigkeit des δ-Ferrits und M₂₃C₆-Carbids. Zeitlicher Verlauf der Ferritisierung und Kornvergröberung. Darstellung der δ-γ-Umwandlung in kontinuierlichen ZTU-Diagrammen.     

Untersuchungen zur Ermittlung des δ-Ferritgehaltes in austenitischen Stählen

Thermomagnetische Analyse und ihre Verwendbarkeit als Meßverfahren zur Ermittlung des δ-Ferritanteils in austenitischen Stählen. Die auf die Masseneinheit bezogene Sättigungsmagnetisierung des δ-Ferrits in Abhängigkeit vom Chrom- und Nickelgehalt. Ermittlung des δ-Ferritanteils aus der spezifischen Sättigungsmagnetisierung bei Raumtemperatur und dem Gehalt an Chrom und Nickel im Stahl. Genauigkeit des Verfahrens.     

Einfluß der δ-γ-Umwandlung des Eisens auf den Wärmeübergang zwischen Block und wassergekühlter Kupferkokille

Gießversuche mit Reineisen sowie mit ferritischen Fe-V- und Fe-Si-Legierungen im Vakuuminduktionsofen. Thermische Messungen des Erstarrungsablaufes in einer wassergekühlten Kupferkokille. Messung der Wärmeflußdichte zwischen Block und Kokille bei Gießen und Erstarren unter Vakuum ohne Gaswärmeleitung im Spalt. Ermittlung der Wärmeübergangskoeffizienten. Aussagen zur Spaltbildung mit einem Strahlungsmodell und anhand der dilatometrisch bestimmten Längenänderung bei der Abkühlung. Einfluß der δ-γ-Umwandlung des Eisens auf die Spaltbildung in unterschiedlich zusammengesetzten Stählen und mögliche Folgerungen für die Vergießbarkeit in Stranggießkokillen.     

Kohärente Ausscheidung in Substitutionsmischkristallen des α-Eisens

Bildungsbedingungen, Merkmale und Formen kohärenter Ausscheidungen in Nichteisen- und Eisenmetallegierungen. Beispiele für die kohärente Ausscheidung vor allem in den Systemen des α-Eisens mit Kupfer, Gold, Aluminium, ferner mit Beryllium, Chrom, Molybdän und Wolfram. Änderung von mechanischen und physikalischen Eigenschaften (vor allem Härte, Kennwerte im Zugversuch; elektrischer Widerstand, Koerzitivfeidstärke) in Abhängigkeit vom Ausscheidungszustand.     

Numerical Modeling of δ‐Ferrite Dissolution Kinetics in Austenitic Stainless Steels

In the present paper a numerical model that describes the diffusion processes of Ni and Cr between ferrite and austenite in Fe‐Cr‐Ni ternary systems has been developed by using the finite difference method (FDM). The model employed the tie‐line relationships given by Kajihara and Kikuchi. The moving ferrite/austenite interface was determined by solving the mass balance equations using the Newton‐Raphson method. The model has been applied to predicting the dissolution kinetics of ferrite in austenitic stainless steels, which were processed by welding, continuous casting and strip casting, respectively. The accuracy of the model has been confirmed to be quite satisfactory as compared with measured values, with advantages over another numerical model developed by Vitek et al. This model was reproduced in the present work.     

Effect of oxygen on hydrogen cracking in high-strength weld metal

The effect of oxygen content on the susceptibility of high-strength weld metal to hydrogen cracking is examined. Increasing oxygen content had a detrimental effect on the cracking susceptibility of weld metal containing a dψusible hydrogen content of 4.7 ppm. In weld metal containing a much lower dψusible hydrogen content (0.87 ppm), increasing weld metal oxygen content had no detrimental effect on hydrogen cracking susceptibility. These results are explained by a model which proposes that hydrogen cracking occurs when a critical oxide inclusion density promotes intergranular fracture at prior austenite grain boundaries and when a critical level of hydrogen is present in the weld metal. For the same level of hydrogen (moisture) contamination, high-strength weld metals containing oxygen contents greater than 200 ppm will be much more susceptible to hydrogen cracking than deposits made using inert gas-shielded or vacuum-operated welding processes. Formerly Visiting Scientist, Department of Metallurgy and Materials Science, University of Toronto     

δ-γ phase equilibria in iron-rich iron-chromium-nickel alloys at temperatures between 1200 and 1350 °C

The δ-γ phase equilibria of iron-rich iron-chromium-nickel alloys with 23 to 38% Cr, 7 to 22% Ni are established at temperatures between 1200 and 1350 °C. The effect of chemical compositions and annealing temperatures on equilibrium δ ferrite fraction is discussed.     

Effect of oxygen on hydrogen cracking in high-strength weld metal

The effect of oxygen content on the susceptibility of high-strength weld metal to hydrogen cracking is examined. Increasing oxygen content had a detrimental effect on the cracking susceptibility of weld metal containing a dψusible hydrogen content of 4.7 ppm. In weld metal containing a much lower dψusible hydrogen content (0.87 ppm), increasing weld metal oxygen content had no detrimental effect on hydrogen cracking susceptibility. These results are explained by a model which proposes that hydrogen cracking occurs when a critical oxide inclusion density promotes intergranular fracture at prior austenite grain boundaries and when a critical level of hydrogen is present in the weld metal. For the same level of hydrogen (moisture) contamination, high-strength weld metals containing oxygen contents greater than 200 ppm will be much more susceptible to hydrogen cracking than deposits made using inert gas-shielded or vacuum-operated welding processes. Formerly Visiting Scientist, Department of Metallurgy and Materials Science, University of Toronto     

Cold rolling texture development of α/γ duplex stainless steels

The cold rolling texture development of two α/γ duplex stainless steels (DSS) with similar volume fractions of both phases but with totally different microstructures were investigated. Due to the limited number of available pole figures using X-rays, for the calculation of the ODFs both a direct method and a recent iterative series expansion method were used. The results were checked by neutron diffraction measurements. The austenitic phases of both DSS behave similarly to single phase materials with a low stacking fault energy which develop a brass-type rolling texture. In contrast, the texture development of the ferritic phases strongly differs from those of single phase ferrites. Instead of a fibre type texture the α-phase in both DSS exhibits a peak dominated texture regardless of whether it is the matrix phase or not. These differences, as well as the sharpness of both phases, are explained by the presence of the second phase.