PSI - Issue 28

Shayan Eslami et al. / Procedia Structural Integrity 28 (2020) 659–666 Shayan Eslami/ Structural Integrity Procedia 00 (2019) 000–000

662

4

Process optimization Generally, in the FSW technique, tool design and its variables have the main influence on the weld quality and strength. In order to optimize the welding process, instead of testing the entire welding parameters range, a Design of Experiment (DOE) method was selected and implemented to reduce the number of experimental runs during the study. The Taguchi method is one of the most popular methods for quality control that has been widely used for process optimization, to keep the number of experiments under control, avoiding the evaluation of all the parameters combinations. After selecting the right tool and choosing the welding parameters range, the welding parameters were optimized using Taguchi method, and the optimized parameters were used to fabricate several specimens to evaluate the weld strength. The Taguchi DOE using an L4 orthogonal array was used as presented in the Table 2. The rest of the parameters kept constant for all the test combinations.

Table 2. The chosen L 4 orthogonal array for three parameters each in two levels.

Experiment Rotation Speed (rpm) Welding Speed (mm/min)

Axial Force (N)

S1 S2 S3 S4

2000 2000 2800 2800

20 40 20 40

800

1000 1000

800

3. Results and discussions The aim of this work is to develop a technological solution for FSW of composites. The tool design proved to have a significant importance in the weld quality. As previously claimed for welding polymers (Eslami et al., 2015), it is very hard to produce quality welds using conventional welding tools. At the beginning, a conventional welding tool was tested, which is a combination of a rotating probe attached to a larger rotary shoulder, as shown in Figure 1a. Similar to polymeric materials (Eslami et al., 2016), flash defects were the main reason for poor quality welds, as presented in Figure 1c. For low content glass fiber thermoplastic composites, it was concluded that the stationary shoulder is essential to forge the plasticized material inside the weld seam. However, with absence of a rotating shoulder, the welds suffer from insufficient heat generation, and further investigations regarding the tool design and process optimization is required. The second tool concept was the same tool design that was developed for welding polymeric materials. This patented stationary shoulder welding tool (Eslami et al., 2017) is made of PEEK material, equipped with a copper sleeve around the rotating probe to compensate lack of the frictional heat due to absence of a rotary shoulder, Figure 1b. Low thermal conductivity of the PEEK preserves the generated heat inside the copper sleeve, heating up the base material in advance. The rotating probe stirs the plasticized material under an axial force, while the PEEK shoulder creates a smooth surface by forging the material inside the stir zone, Figure 1d. The welds produced by this tool were much smoother that the conventional tool, with no harsh transition or flash between the weld and the base material. It is also possible to see the sleeve effect on the surface of the welds, which refers to the forging marks that hot sleeve creates on top of the base materials. This tool design is much more suitable for welding thermoplastic composites. A statistical approach was used to optimize the welding parameters, as presented in Table 2. Each welded plate was cut to prepare 5 specimens for each welding condition. The welding temperature recorded for all the welding combinations using a thermocouple shaft locating inside the copper sleeve, as close as possible to the rotating probe. Also, the active forces during welding were monitored to analyze the effect of each welding parameter on the generated forces using the sensitized clamping device.