PSI - Issue 49

Teiba Ahmed et al. / Procedia Structural Integrity 49 (2023) 37–42 Teiba Ahmed et al. / Structural Integrity Procedia 00 (2023) 000–000

40

4

at which load was applied and weighed (Fig. 3. (b)). All experimental data was statistically analysed using three replicates for each specimen and reported as mean ± standard deviation (SD).

Fig. 3. (a) Loading of sample in cantilever jig; (b) cutting and weighing of samples.

3. Results and Discussion The averaged stiffness per gram at 5 mm displacement can be shown in Fig 4. The force at 5 mm displacement value was chosen as this was in the near-linear section of the force-displacement curve for all specimens. As expected, thicker samples were always stiffer than thinner samples. For all thickness groups, the straight specimens were the stiffest and increasing the frequency of the sine waves (reducing the wavelength) reduced stiffness. The thick straight specimen exhibited a specific stiffness of 17.97 ± 0.11 N.g -1 whilst the thin straight specimen exhibited 11.45 ± 0.03 N.g -1 . At a wavelength of 5 mm, the thick specimen exhibited 5.15 ± 0.32 N.g -1 and the thin specimen 1.63 ± 0.04 N.g -1 . The thick specimen had approximately 1.5x the stiffness of the thin specimen for the straight specimens, but almost 3.5x greater stiffness for the specimens with the lowest wavelength. This difference in the relative changes should be considered when designing the toolpath for a specific application, based on the range of stiffnesses required. It may be the case that differences in the relative effects of wavelength for specimens of different thicknesses are linked to the ratio of amplitude:thickness, which could be investigated in future work studying multiple amplitudes. Each thickness group was normalised by the control specimen (straight) to find the normalised specific stiffness (Fig 5.). It was found that the straight specimens were stiffest for all thickness and increasing the frequency of sine waves (reducing wavelength) reduced stiffness. This was a clear and consistent trend for all three thicknesses. However, the earlier finding in relation to Fig 4., regarding different relative effects of wavelength for different thicknesses, are also apparent in this figure since the thick specimens have greater relative stiffness towards the left hand side of the figure. For design applications, the desired stiffnesses can be used to define the thickness and the range of wavelengths using the normalised plots. The sinusoidal wave design clearly had a greater influence than thickness. Even though it would be possible to user a greater range of thicknesses, it is not expected that they could offer the potential to modify stiffness to the same degree as adding sinusoidal fluctuation. However, sine waves increase physical size, whereas variation of thickness does not. For applications where space is available, sine waves may be the best approach. Alternatively, for applications where space is critical, the potential to reduce thickness gradually within a single wall print could be considered (Moetazedian et al., 2021). Exploration of double-wall designs, varying amplitude of the sinusoidal wave, increased thicknesses, and varying extrusion width within a single printed line could be considered for future work.