DYNAMIC MATERIAL PROPERTIES OF THE HEAT-AFFECTED ZONE (HAZ) IN RESISTANCE SPOT WELDING

Transcription

1 International Journal of Modern Physics B Vol. 22, Nos. 31 & 32 (28) World Scientific Publishing Company DYNAMIC MATERIAL PROPERTIES OF THE HEAT-AFFECTED ZONE (HAZ) IN RESISTANCE SPOT WELDING JI-WOONG HA 1, JUNG-HAN SONG 2, HOON HUH 3 School of Mechanical, Aerospace & System Engineering, KAIST, 335, Gwahangno, Deadoek Science Town, Daejeon, KOREA, hajiwoong@kaist.ac.kr, jhsong_me@kaist.ac.kr,hhuh@kaist.ac.kr JI-HO LIM 4,SUNG-HO PARK 5 Automotive Steel Research Center, POSCO, 699, Gumho-dong, Gwangyang-si, Jeonnam, KOREA, jiholim@posco.com, sunghopark@posco.com Received 15 June 28 Revised 23 June 28 This paper is concerned with a methodology to identify the dynamic material properties of the heataffected zone (HAZ) near the base metal in a resistance spot weld process at various strain rates. In order to obtain the dynamic material properties of the HAZ in the spot-welded steel sheet, specimens are prepared to have similar material properties, hardness and microstructure to the actual HAZ. Such thermally simulated specimens are fabricated with the material thermal cycle simulator (MTCS) and compared with the real one for the hardness and microstructure. Dynamic tensile tests are then conducted with a high speed material testing machine. Stress strain curves of the thermally simulated HAZ are obtained at various strain rates ranged from.1/sec to 1/sec. Obtained material properties are applied to the finite element analysis of the spot-welded tensile-shear specimen in order to verify validity of the proposed testing methodology and obtained results. Analysis results demonstrate that the material properties obtained are appropriate for the FE analysis of spot-welded specimens. Keywords: Resistance spot weld; material thermal cycle simulator (MTCS); thermally simulated HAZ; dynamic material properties. 1. Introduction The electric resistance spot welding process is an indispensable assembling process of steel auto-panels in the automotive industries. As a modern auto-body contains several thousands of spot welds 1, the strength of spot welds becomes extremely important in the crashworthiness assessment of auto-body members. During the spot welding process, heat generated by electrical resistance induces inhomogeneous microstructures around Corresponding Author. 58

2 Hardness [Hv] Dynamic Material Properties of the Heat Affected Zone (HAZ) in a Resistance Spot Weld 581 HAZ2 HAZ Fusion zone HAZ Base metal HAZ Fusion zone HAZ 5 5 (c) (d) Fig. 1. Hardness distribution of the spotwelded SPRC34R. Fig. 2. Optical micrograph for cross-section of the spot-welded SPRC34R: overview; base metal; (c) HAZ1; (d) HAZ2. the welded nugget. These inhomogeneous properties should be considered in the FE analysis to characterize the actual mechanical behavior of a spot-welded specimen. In case of automotive simulation, material properties of inhomogeneous regions can be applied to a simplified FE modeling of the welded part described as a beam element instead of a rigid link to restrict two sheets. However, in most numerical simulations, inhomogeneous material properties are not considered due to the lack of both an effective testing methodology and available data for describing the material properties of the HAZ. Although Zuniga and Sheppard 2 proposed the relationship to approximate the yield and ultimate stress of HAZ as a function of a hardness value, few studies reported regarding appropriate testing methodology to acquire accurate stress strain curves of the HAZ. Moreover, there have been little studies about the dynamic material properties of the HAZ, which are important to evaluate the dynamic failure load of a spot weld. In this paper, testing methodology is newly proposed to identify the dynamic material properties of HAZ near the base metal in a resistance spot weld. At first, the thermally simulated HAZ specimen which has similar hardness and microstructure as the actual HAZ is fabricated using the material thermal cycle simulator (MTCS). In order to determine the thermal cycle of MTCS, the peak temperature was changed from 1 C to 13 C keeping other conditions. Dynamic tensile tests are then conducted with high speed material testing machine at various crosshead speed. Stress strain curves of the thermally simulated HAZ are obtained at the strain rates ranged from.1/sec to 1/sec. Finally, the material properties obtained are applied to the finite element analysis of the spot-welded tensile-shear specimen in order to evaluate the validity of the proposed testing methodology and obtained results.

3 582 J. W. Ha et al. (c) Fig. 3. Fabrication of the thermal simulated HAZ specimen: material thermal cycle simulator (MTCS); dimension of a specimen and a guide bar; (c) experimental set up for thermal simulation using MTCS. 2. Thermal Simulation Tests for the Simulated HAZ 2.1. Classification of the spot-welded region The HAZ of a spot weld was investigated with a high strength steel sheet of SPRC34R whose thickness was 1.2 mm. Spot welding of steel sheets was performed using a static spot/projection welding machine. The welding current of 7.6 ka was imposed during the welding time of 12 cycles at 6 Hz with the holding force of 4. kn. Hardness distribution and optical micrographs for produced spot-welded regions are shown in Fig. 1 and Fig. 2, respectively. From the hardness profile and micro-structures shown in the figures, spot-welded region can be classified into three heterogeneous metallurgical zones as follows 3 : the base metal (BM) where no metallurgical transformation occurs; the heataffected zone (HAZ) where regeneration and transformation zone conveying thermal and structural gradients between the melted zone and the base metal; the fusion zone or nugget (FZ) where melting and solidification occurs. The range of hardness values at BM, HAZ and FZ are 165~17 Hv, 17~4 Hv and 4~41 Hv, respectively. As the hardness profile explains, the HAZ is the transition zone between the FZ and the BM with steep gradients in the hardness. The varying mechanical properties of the HAZ are resulted from non-uniform thermal histories during the spot weld. The HAZ also can be divided into several subzones which have distinct microstructures and mechanical properties. In this paper, the HAZ is divided into two subzones of the HAZ1 which is close to the BM and the HAZ2 which is close to the FZ. The material properties of the HAZ1 are examined in this paper since failure of the spot weld occurs at the interface between the HAZ and the BM when a large load is applied to the spot-welded components. The average grain size in the HAZ1 is 9.22 µm and the range of the hardness value is 17~19 Hv. Compared with the grain size of µm in the BM, grains are refined due to the thermal histories of melting and fast solidification during the welding process. The grain size mentioned above was measured by Planimetric method 4 explained in ASTM E112 standard.

4 Tmeperature [ Dynamic Material Properties of the Heat Affected Zone (HAZ) in a Resistance Spot Weld 583 o C] Input thermal cycle Real thermal cycle Time [sec] Fig. 4. Input thermal cycle for thermal simulation using MTCS. 5 Fig. 5. Optical micrograph of the specimen after thermal simulation Fabrication of the thermal simulated HAZ The goal of the simulated HAZ development effort is to generate tensile specimens of simulated HAZ materials that have similar hardness values and microstructures to those of the HAZ1 in actual spot weld and uniform microstructure throughout the tensile specimen. This preparation was carried out with the assumption that material properties would be the same as that of the HAZ1 when the simulated HAZ specimen has similar hardness values and grain sizes that match those of the HAZ1. MTCS shown in Fig. 3 is used to fabricate the thermally simulated HAZ specimen. Although the MTCS has been widely adopted to fabricate the simulated HAZ in arc welding of thick plates, sufficient eddy currents to elevate the temperature up to melting points cannot be fully induced in the steel sheets. As a remedy to induce sufficient eddy current, two guide bars are attached onto the sheet specimen as shown in Fig. 3. The specimen with guide bars is mounted on MTCS to fabricate the thermally simulated HAZ specimen as shown in Fig. 3(c). Fig. 4 represents the thermal cycles imposed on the specimen to fabricate simulated HAZ specimen. The peak temperature is assigned as 12 C with the holding time of 5 sec considering the thermal histories of HAZ1 in actual spot welding process. Cooling rate from 8 C to 5 C is 35 C/sec. During the operation of the MTCS, a function generator transmits the thermal cycles into the feedback controller in the MTCS to control the temperature by comparing the measured temperature with the thermal cycles inputted. Microstructure and hardness distribution of the simulated HAZ specimen are shown in Fig.5 and Fig. 6, respectively. Average grain size obtained from the thermal simulation is 9.62 µm with the hardness of 186 Hv. Compared with the grain size and hardness value in HAZ1 of the actual spot weld, the simulated HAZ specimen can be fabricated with the thermal cycles shown in Fig Dynamic Material Properties of the Simulated HAZ 3.1 Dynamic tensile tests of the simulated HAZ specimen Using the simulated HAZ specimen, dynamic tensile tests were conducted at the strain rate ranged from.1/sec to 1/sec. A high speed material testing machine 5 is utilized in the dynamic tensile tests. The simulated HAZ specimen is mounted onto the high

5 584 J. W. Ha et al. 22 Longitudinal direction 22 Lateral direction 2 2 Hardness [Hv] Mean Hardness : Hv Hardness [Hv] Mean Hardness : 186.Hv Distance from the center of a specimen [mm] Distance from the center of a specimen [mm] Fig. 6. Hardness distribution of the simulated HAZ specimen: longitudinal direction; lateral direction. (c) Fig. 7. Dynamic tensile tests of the simulated HAZ specimen: high speed material test machine; dimension of the specimen; (c) fixture set for mounting the specimen onto high speed material testing machine. True stress [MPa] Simulated HAZ 3 Strain rate (/sec).1 2 Experimental Results.1 Fitted Curves True strain True stress [MPa] SImulated HAZ 3 ε=yield ε=.5 2 ε= Strain rate [/sec] Fig. 8. Experimental results of the simulated HAZ specimen: true stress-strain curves at various strain rates; strain rate sensitivity. speed material testing machine using the fixture set as shown in Fig. 7. The stress strain curves of the simulated HAZ are obtained from the tests at various strain rates as shown in Fig. 8. Compared with the yield stress of MPa for the BM at quasi-static states, the simulated HAZ shows higher strength due to the grain refinement caused by the thermal histories. The results also indicate that the flow stress gradually increases as the strain rate increases. The strain rate sensitivity of the flow stress is also plotted in Fig. 8 with different plastic strains. The figure represents that the strain rate sensitivity of the flow stress decreases with increasing plastic strain.

6 Dynamic Material Properties of the Heat Affected Zone (HAZ) in a Resistance Spot Weld 585 Fig. 9. FE modeling of the lap-shear specimen: overview; detailed FE model near the spot weld. 12 Loading speed: 5x1-5 m/sec 12 Loading speed:.5 m/sec 1 1 Load [kn] Experiment 2 w/ considering HAZ w/o considering HAZ Displacement [mm] Load [kn] Experiment 2 w/ considering HAZ w/o considering HAZ Displacement [mm] Fig. 1. Comparison of the FE analysis results to the experiment: quasi-static loading; dynamic loading. 3.2 Application to FE analysis of the spot-welded specimen In order to verify the testing methodology and experimental results using the simulated HAZ, material properties obtained from the simulated HAZ are applied to the FE analyses of the spot-welded lap-shear specimen. A finite element model for lap-shear specimen is shown in Fig. 9. The specimen is modeled with three-dimensional solid elements. The HAZ is divided into five element groups. The HAZ close to BM is named as HAZ-A and that close to FZ is named as HAZ-E as shown in Fig. 9. The stress strain curves shown in Fig. 8 are given to HAZ-A and the properties from HAZ-B to nugget are estimated with the assumption that stress in plastic deformation region is proportional to hardness profile. Fig. 1 compares load displacement curves obtained from the analyses to those from the experiments. The comparison represents that the results obtained from the analyses are close in coincidence with those obtained from the experiments. The discrepancies between analyses and experiments are less than 3% when the material properties of HAZ are considered in the FE analysis. Analysis results indicate that material properties obtained from the simulated HAZ specimens are valid for the FE analysis of the spotwelded specimen. 4. Conclusion This paper proposes a methodology to identify the dynamic material properties of the HAZ in a spot weld. Using the MTCS, the simulated HAZ specimen is fabricated as the representative of the HAZ1 in actual spot weld. Stress strain curves of the thermally

7 586 J. W. Ha et al. simulated HAZ are obtained at various strain rates ranged from.1/sec to 1/sec. Compared with the BM, the simulated HAZ shows higher strength due to the grain refinement caused by the thermal histories. The material properties obtained from the experiment are applied to the FE analyses of the spot-welded lap-shear specimen. The analyses results fully demonstrate that testing method proposed are valid to evaluate the material properties of the HAZ. References 1. Y. J. Chao, J. Eng. Mater.- T. ASME. 125 (23) S. M. Zuniga and S. D. Sheppard, Modeling Simul. Mater. Sci. Eng. 3 (1995) S.-H. Lin, J. Pan, S.-R. Wu, T. Tyan and P. Wung, Int. J. Solids Struct. 39 (22) ASTM E112, Standard Test Methods for Determining Average Grain Size, (26). 5. H. Huh, S. B. Kim, J. H. Song and J. H. Lim, Int. J. Mech, Sci. 5 (28) 918.