PSI - Issue 37

Liting Shi et al. / Procedia Structural Integrity 37 (2022) 351–358 L Shi et al/ Structural Integrity Procedia 00 (2022) 000 – 000

353

3

(2020b). After welding, all RSW specimens were baked in an oven at 175 °C for 35 min to simulate ELPO bake operations that all vehicles undergo during production to cure adhesives and sealers.

Table 1 Welding schedules of AA5754-AA5754 and AA5754-HSLA RSWs as used on a medium frequency direct current (MFDC) welder

Preheat

Weld time (ms)

Hold time (ms)

RMS weld current (kA)

Stack-ups

Materials

Polarity

Weld force lb(N) 700 (3115)

AA5754 to itself

1.1mm AA5754

Al positive

10ms at 16%VS

103

250

22.7

Positive polarity Al steel Negative polarity Al steel

1.1mm AA5754 to 2.0mm HSLA 1.1mm AA5754 to 2.0mm HSLA

Al positive

1250(5562)

40ms at 7.5kA, 10ms cool

1955

250

11.2

Al negative

1250(5562)

40ms at 7.5kA, 10ms

1995

250

12.0

2.2. Weld characterization

Macrostructures of the RSWs were obtained using optical microscopy of the standard metallographic specimens as described by Shi, et al. (2020b). Weld features including nugget diameter, intermetallic layer thickness, and the notch root angle were measured and reported. 2.3. Tensile and fatigue testing Tensile tests of tensile shear and coach peel specimens were conducted on an MTS Landmark hydraulic test frame at a test speed of 3 mm/min. Seven (7) tensile tests were conducted for each stack-up and specimen configuration. The average maximum tensile loads of the seven tests were calculated and then used as reference load levels for fatigue testing. Load controlled fatigue testing was performed at a load ratio, R of 0.1. The fatigue life was defined as the point at which the material completely separated or when the load dropped to zero. After mechanical testing, the fracture specimens were collected for fracture mode observation and the same metallographic procedure was applied as for the untested specimens for notch root angle measurements. 3. Finite element analysis Finite element analysis (FEA) was performed using ABAQUS 2020 Implicit Analysis. Three dimensions (3D) brick element C3D8R with reduced integration points was used in all the models. Enhanced hour-glass control was used to mitigate hour-glass in stress-concentrated areas. The measured material properties and dimensions of weld nugget and the heat affect zone (HAZ) together with that of the base material were input into the finite element models. Symmetric and asymmetric notch root angles were created in the FEA of tensile shear specimens to highlight their effect on local stress and strain states at the notch tip. In the models with symmetric notch root angles ranging from 10  to 36  , the notch root angle was assumed to be uniform around the weld nugget, similar to Shi et al. (2020b). Tensile shear FE models with asymmetric notch root angles are shown in Fig. 1. In Fig. 1a, the notch root angle is smaller (e.g. 10  ) in the loading direction (i.e. left-hand side) or leading edge and larger (e.g. 36  ) in the opposite direction (i.e. right-hand side) or trailing edge. In contrast, the notch root angle in Fig. (b) is larger (e.g. 36  ) in the leading edge and smaller (e.g. 10  ) in the trailing edge.