tion amplitude had a more significant effect on decreasing the bulk hardness of deposited Al3003-H18 foil, whereas increasing normal force may increase hardness in top layers, as seen in sample TB-10-38-8000. Furthermore, the average bulk hardness values of the foil after heat treatment at 343ºC for 2 h were almost the same for all samples. This implied that the difference in bulk foil hardness in different layers of VHP-UAM samples in as-processed conditions was neutralized by this annealing condition. The bulk hardness contours of Al3003-H18 foils in the VHP-UAM samples fabricated from the SL7200 machine are illustrated as hardness maps in Fig. 4. Each hardness map has a line dividing the upper layers (above the 50th layer), in which the weld speed was higher (42.7 mm/s) as compared to the lower 50 layers, where the weld speed was equal to the former value (35.6 mm/s) in the TB machine. It was found that at the lower vibration amplitude (28 μm) in sample SL-66-28-5340, the hardness was more uniform (or less change in color contour) from top to bottom than at the higher vibration amplitude (34 μm) in samples SL-80-34- 4000 and SL-80-34-5340, which yielded a large drop in hardness in the bottom layers of the foils. This result is similar to Fig. 3, where a large increase in vibration amplitude eventually affected the bulk foil hardness, especially in bottom layers. Along the horizontal position of each layer, the bulk hardness slightly deviated from left to right. It was speculated that this horizontal variation in hardness is likely due to the different localized normal force values applied on the left and right sides of the sample. From Fig. 4A and B, the hardness was lower in the bottom left corner than the bottom right corner, meaning different levels of plastic deformation occur in these regions resulting in some different bulk hardness values. Figure 5 shows the trend of average bulk hardness from the bottom to top of VHP-UAM samples made from the SL7200 machine. Although the average hardness values may fluctuate up and down from one layer to the next, it can be observed that the trends of increasing hardness from bottom layers to top layers of VHP-UAM builds exist, especially in samples processed at a higher 34-μm vibration amplitude. It was speculated that the 28-μm vibration amplitude was not large enough to have an accumulative effect on lowering the bulk hardness of the underneath foils and almost all hardness drop took place in a single pass. The accumulative effect of thermomechanical loading conditions during ultrasonic additive manufacturing has already been studied in previous works (Refs. 19, 20), which report that this mechanism was related to relative shear displacement of bonded and matrix WELDING RESEARCH regions well below the current layer that was being welded. In contrast, the 34-μm vibration amplitude from the SL7200 machine was high enough to cause an additional input of energy or power into the bottom layers where the accumulative effect occurred. It was also worth noting that although the process parameters of samples TB-10-28-5340 and SL-66-28- 5340 are similar, their overall bulk hardness values were different. However, this experiment did not provide concrete evidence of why the SL7200 machine caused a larger drop in hardness as compared to the TB machine. In order to assess the effect of ultrasonic power on the change in the bulk hardness of VHP-UAM samples, Fig. 6 demonstrates the measured electrical power drawn from the TB machine during VHP-UAM of the three samples. It is known that during UAM and VHP-UAM, the measured electrical power drawn from the UAM and VHP-UAM machine increased with the higher levels of normal force and vibration amplitude (Ref. 20). It was seen that the amount of electrical power drawn was greatly affected by increasing the vibration amplitude but was less affected by increasing the normal force. When the vibration amplitude of 28 μm was used in sample TB- 10-28-5340, the average power was 550 W as compared to as high as 1189 W in sample TB-10-38-4000 and 1234 W in sample TB-10-38-8000 when vibration amplitudes were set at 38 μm. It was also noticed that in all three samples, the power seemed to reach the maximum around 2.5 s. This was possibly related to the increase in rigidity or stiffness during vibrations in VHP-UAM where the sonotrode had a solid grip on the foil and the base plate, and larger power were necessary to maintain the same amount of vibration amplitudes. Since the levels of power vs. time oscillated without a unique trend in each weld pass, the average power used to bond each layer is plotted in Fig. 7 against the layer number from bottom to top of VHP-UAM samples for better comparison. It can be seen that the actual average power level used gradually decreased with increasing build height or in the higher layers bonded, i.e., it took less power to produce the same vibration amplitude JUNE 2016 / WELDING JOURNAL 189-s Fig. 7 — Average ultrasonic power used to weld each layer during VHPUAM of Al3003H18. Fig. 8 — Correlation between average hardness and average ultrasonic power used to weld Al3003H18 tape at a constant weld speed of 35.6 mm/s using the TestBed and SonicLayer7200 machines.
Welding Journal | June 2016
To see the actual publication please follow the link above