Fig 3 - uploaded by Dénes Kollár
Content may be subject to copyright.
Deformation modes: a) transverse shrinkage, b) longitudinal shrinkage, c) rotational distortion, d) angular distortion, e) longitudinal distortion, f) buckling and g) twisting distortion.
Source publication
Local and highly nonuniform heating during welding and subsequent cooling cause complex thermal cycles inducing residual stresses and deformations in the weldment. There are numerous in-process control methods and post-weld corrective methods that can be used to mitigate welding-induced residual stresses and distortions. However, these can be expen...
Contexts in source publication
Context 1
... stresses are in equilibrium within a part without any external load, therefore, residual stresses are consequently mainly elastic and compressive in further regions. The fundamental deformation modes due to welding are sorted into seven classes (Figure 3) by their appearance [1, 2, 5]: a) transverse shrinkage perpendicular to the weld bead centre line, b) longitudinal shrinkage in the direction of the weld bead centre line, c) rotational distortion (planar angular distortion) due to thermal expansion, d) angular distortion close to the weld bead centre line caused by a nonuniform temperature distortion through the thickness, e) longitudinal distortion (bowing) in the normal direction of the plate, f) buckling distortion due to compressive thermal stresses in components, g) twisting distortion as a result of shear thermal stresses. In general, there are two design approaches for controlling structural and geometric imperfections. ...
Context 2
... temperature is not defined, the 2 nd weld pass is laid right after the 1 st weld pass with are erroneous as the elevated temperature is lower than liquidus temperature in the corresponding finite elements representing the weld bead (Figure 30). On the other hand, 100% ...
Context 3
... calibrated parameters for double-sided fillet welds with single weld passes are validated using an extended set of parameters in the case of the three types of T-joint; one from each configuration is presented here. Figure 35 shows the finite element models of the weldments contributing in the validation process. T-joints with double-bevel groove weld (JT1-1-03), with double-sided fillet welds using multiple passes (JT2-1-07) and with singlebevel groove weld (JT3-2-01) are studied. ...
Context 4
... consist of 26052, 23112 and 18632 finite elements, respectively. Welding sequences are also summed up in Figure 35 for the investigated cases. weld. ...
Context 5
... important to highlight that throat thickness varies for the three specimens. The top nodes, representing the maximum transverse deformations, are selected and the evolution of displacement are also given in Figure 38 showing the importance of different weld types. Initial temperature is 150 °C due to preheating. ...
Context 6
... time varies in a function of travel speed and number of weld passes for the validated cases. Transverse nodal displacement decreases after the 4 th weld pass for JT1-1-03 and it is alternating for JT2-1-07 because of the welding sequence shown in Figure 38. ...
Context 7
... aim of the experimental research program is to measure total strains during welding on nine corrugated web girder specimens having three different corrugation profiles. Several specimens with horizontal and vertical strain gauges are shown in Figure 43. Interpass temperature and lengths of cooling phases are determined according to the measured temperatures at the strain gauge positions during experiments. ...
Context 8
... notations in the figures of hexahedral solid elements are the followings: solid element type-keyopt(2)-number of divisions over the thickness (e.g., SOLID185-1-1). The 8-node SOLID185 low-order solid element type with full integration with B-bar method (Figure 93a) cannot be used for such analyses as it is overstiff to capture the realistic behaviour even with a small average element size and multiple divisions over the thickness. Neither reducing element size, nor increasing the number of divisions over the thickness from one to eight, i.e., mesh refinement (h-method), do not help convergence issues in this case. ...
Context 9
... reducing element size, nor increasing the number of divisions over the thickness from one to eight, i.e., mesh refinement (h-method), do not help convergence issues in this case. Similar phenomena can be observed by using uniform reduced integration with hourglass control (Figure 93b). ...
Citations
... The welding deposition speed was defined based on the finite element mesh size (5x5x5 mm in the denser zone) and in Table 5. The thermal transmission coefficient is obtained from the energetic parameters in the same table, and the emissivity coefficient of the material for heat dissipation was considered 0.9 [22][23][24]. ...
The manufacturing process of steel plate girders generates residual stresses and geometrical imperfections in the flange and web plates that constitute them. These imperfections must respect the manufacturing tolerances that are presented in the regulation. In the present work several stages of the manufacturing process are studied, being these the thermal cutting of the plates, the assembling of the beam and the flange‐web welding process. This work is also accompanied by several results in terms of residual stresses and geometric imperfections of an experimental study and is made the confrontation of those results with the theoretical information available on the prEN 1993‐1‐14 and with the results obtained by numerical models that simulate the thermal processes of the plates cutting and welding.
... Large deflection effects are included in the mechanical analysis. Background and additional details of the developed manufacturing framework can be found in Ref. [26]. ...
... To accurately predict the distribution of the weld temperature field, García et al. [10] obtained a design of relevant parameters affecting weld temperature measurements through statistical analysis and experimental design. Kollár et al. [11][12][13] analysed the welding residual stresses of straight web and corrugated web Ibeams, and displayed and evaluated the distribution and types of residual stresses in corrugated web beams. It was concluded that the "accordion effect" of corrugated steel webs was the main reason why the residual stress was significantly different from that of straight webs. ...
This paper proposes a new approach for obtaining the welding temperature field and stress distribution in a corrugated steel web girder. This approach was validated by comparing the control temperatures (obtained from finite element modelling of CO 2 gas shielded welding) of a corrugated steel web girder with experimental findings. Moreover, the effects of different welding speeds on welding residual stress were analysed. The control point temperatures obtained using modelling corresponded closely with those of the observed data. The welding residual stress was primarily affected by the longitudinal residual tensile stress and increased as the distance to the weld decreased. The welding speed had an insignificant effect on the residual stress distribution but a significant effect on the residual stress peak. The findings of this study can serve as a reference for the analysis of the welding temperature and stress fields of corrugated steel web girders.
... Welding heat sources induce heat generation which is defined as element body force load during the three-dimensional transient thermal analysis. The double ellipsoidal heat source model introduced by Goldak et al. [37] is implemented in the numerical framework, while an automated Frenet-Serret frame is used for modelling the movement of heat sources [38]. Eqs. 1 and 2 determine the power density distribution in the front and rear quadrants for a double ellipsoidal heat source model, respectively. ...
... Large deflection effects are included in the mechanical analysis. Background and additional details of the developed manufacturing framework can be found in [38]. ...
Deformations of bottom flanges in the vicinity of bridge bearings, i.e., in bearing areas, due to manufacturing and loading can result in serious problems in service life and damages in bridge superstructures and structural bearings coming from nonuniform stress distribution. The paper focuses on the out-of-flatness measurement of bearing areas using Coordinate Measuring Machine (CMM) in typical steel and composite bridges having box and open cross-sections. Advanced manufacturing simulations are also carried out in addition to site measurements to analyse imperfections due to welding during assembly and strengthening of a typical bridge superstructure. A qualitative comparison is made with measurement results showing that the magnitudes of simulated and measured distortions are in good agreement. Results are evaluated in accordance with the permitted total deformation limit in EN 1337-2 recommended for spherical structural bearings. The current study uses a novel approach in the field of bridge engineering; the analogy of Abbott-Firestone curve is applied for describing the deformed shape and evaluating the extent of defects in the contact area since the magnitude of out-of-flatness imperfection is not inevitably sufficient for classifying the surfaces. Based on the obtained results it is demonstrated that using additional transverse bearing stiffeners for strengthening the superstructure can result in large distortions. Even the magnitude of out-of-flatness due to welding of transverse bearing stiffeners can exceed the permitted limit resulting in nonuniform stress distribution in the sliding elements affecting wear resistance and service life of spherical structural bearings.
O processo de fabrico de vigas I de aço de secção soldada introduz tensões residuais e imperfeições geométricas nas almas e nos banzos das vigas, que devem cumprir as tolerâncias de fabrico. No presente artigo é estudado o fabrico de vigas de aço de secção soldada que en-volve o corte das chapas, a montagem da viga e o processo de soldadura banzo-alma, a partir dos registos experimentais relativos às imperfeições geométricas e às tensões residuais obtidos nestas fases do fabrico das vigas. Estes resultados são comparados com os padrões de tensões residuais propostos na nova prEN 1993-1-14 e com os obtidos por modelos numéricos que pro-curam simular os processos de fabrico de corte e soldadura.
