Fig. 5 — Microstructure of HAZ of multipass butt-joint welding, hot pass, preliminary temperature 100°C (250×). equations of two- and three-dimensional heat-conducting paths are used to estimate the interrelation of modes of welding (heat input) and time of cooling (cooling rate) of welded connections. In particular, calculations of a threedimensional thermal field was applied to welding weld roots at low heat input. At a given mode of welding, there is no influence from the pipe wall thickness, d, and the equation reflects only the effect of heat input, E: (t8/5) = (0.67 – 5*10–4*To)*η*E* 1:(500 – To) – 1:(800 – To) *Κ3 (1) During welding of pipes using SAW with a large heat input, a two-dimensional thermal zone was considered. The equation demonstrates the influence of both pipe wall thickness and the level of heat input: (t8/5) = (0.043 – 4.3*10–5*To) *η2*E2/d2*1:(500 – To)2 – (1 : (800 – To)2*Κ2 (2) Table 3 shows the symbols and designations for Equations 1 and 2. The charts as presented in Fig. 9, which are based on corresponding calculations and experiments, Fig. 6 — Comparison of impact toughness vs. cooling rate dependences for testing the investigated steels at –30°C. allow estimations of the cooling rates from the peak temperature for every specific heat input. One of the corresponding charts for multipass welding will be presented later. During longitudinal welding with high heat input, the cooling rate of the HAZ is affected by the amount of heat input, a wall thickness, and a temperature prior to welding, meaning the temperature of the previous pass during two-pass SAW. The calculated cooling rate values, depending on the initial temperature of the weld during two-pass SAW are presented in Table 4 for pipes with wall thicknesses of 16.4 and 25.4 mm. During welding of the external joint, each thickness requires a specific optimal level of heat input, which ensures the necessary geometric parameters of the joints. Appropriate cooling rates of the external weld were defined both for the condition of full cooling of the internal joint (20°C), and for its incomplete cooling to 60° and 100°C. Phase transformations and microstructure. The study of phase transformations was performed using a fast operating, high-temperature dilatometer (DB-Chermet) capable of induction heating up to 1350°C at a heating rate from 10° to 300°C/s and cooling capacity with rates from 0.3° to 250°C/s. Microstuctures of dilatometric and weld-simulated samples were investigated using etching in 2% Nital and optical microscope Axiovert 40 MAT. Twelve to 15 samples were used to build each CCT diagram. The diagrams contain microhardness values against each cooling rate and corresponding product of phase transformation so those numbers can be used, in particular, for evaluation of hardness of martensite. Evaluation of resistance to brittle fracture. The investigated steel samples were subjected to induction heating in accordance with a specific thermal cycle of welding and subsequent cooling at a wide range of cooling rates. Specimens with simulated microstructure of the HAZ were machined to cut a sharp (Charpy) notch and subjected to impact testing in the temperature range 20° to –60°C. The usual determination of the temperature of ductile to brittle fracture transition, based on area fraction of shear fracture, is practically impossible on subsized samples due to the large plastic deformation of thin samples. Therefore, the estimations of resistance to brittle fracture were based on the following set of parameters, schematically shown in Fig. 2. • The “upper limit,” corresponding to the beginning (lowest temperature) of the “shelf toughness” (ShCVN) and signifying the beginning of the transition from ductile to brittle fracture (projection “T1” in Fig. 2). • The “average threshold” T50 ShCVN (here at ~110 J/cm2) corresponding to the decrease in impact toughness by 50% relative to the maximum (Shelf CVN) values, which corresponds to a mixed brittleductile fracture and, as shown by comparison with full-size samples, corresponds to 50–60% of the tear fracture pattern (projection “T2” in Fig. 2). JANUARY 2014, VOL. 93 26-s WELDING RESEARCH Table 3 — Symbols and Designations for Equations 1 and 2 Designation Units of Measure Parameter t8/5 seconds Time of cooling from 800° to 500°C η Dimensionless factor Parameter of the process effciency E J/sm Heat-input (E = U*I/V) U Volt Electric voltage of a welding arc I Amperage Electric current of a welding arc V sm/s Speed of welding To °C Temperature of preheating d sm Thickness of pipe wall
Welding Journal | January 2014
To see the actual publication please follow the link above