PSI - Issue 42

F. Cesarano et al. / Procedia Structural Integrity 42 (2022) 1282–1290 F. Cesarano, M. Maurizi, C. Gao, F. Berto, F. Penta, C. Bertolin / Structural Integrity Procedia 00 (2019) 000 – 000

1288

7

3.2 Preliminary experimental results and their analysis through FEA In this phase, the two proposed experimental methods are analyzed in detail, obtaining support from numerical analysis to draw mathematical considerations on the phenomenon. Then, through testing combinations of time and temperature, an understanding of the most reliable method suited to study the SME in the light of programming parameters modification (e.g. geometric and printing parameters) is obtained. The first method of experimentation involves subjecting the specimen to thermal increase from room temperature to the target regime temperature (60°C, 70°C, 80°C), with a dwell time at the target temperature of several seconds (1s, 10s, 30s). Naturally, an increase in curvature can be seen as time and temperature increase. For a better understanding of the phenomenon, the actual values of curvatures (Table 3) and the plot in dependence on temperature and time are given below (Fig. 8a). Each experiment (curvature value) was repeated three times, and an arithmetic average of the results was calculated to obtain the values to be plotted.

Table 3: Curvatures - First experimental method with data as reported in Fig. 8a.

60°C 0.0306 0.0342 0.0410

70°C 0.0569 0.0622 0.0682

80°C 0.0961 0.0994 0.0999

Time/Temp

(t = 1s) (t = 10s) (t = 30s)

As can be seen from the graph (Fig. 8a), as the exposure time increases, the curvature concerning temperature converges towards a straight line; probably because the value of the curvature at 80° for 30 seconds appears to be a point of convergence since the two ends of the plate almost touch each other; this suggests that this first experimental method is not suitable for the analysis of the phenomenon for the purposes proposed in this work, since the exposure time for the entire thermal rise cycle is high (approx. 15 minutes are needed to reach 55° starting from the room temperature conditions) leading to high curvature values that can be confused with each other even in case of different times since they are in the range of seconds; this would limit the analysis of the problem. Therefore, a similar study using a second experimental method follows, with subsequent comparison of the two. The second method of experimentation involves subjecting the specimen to the target steady-state temperature (60°C, 70°C, 80°C), with a dwell time of several minutes (3m, 5m, 10m), without undergoing the complete thermal increase in the oven. Again, as in the previous case, an increase in curvature can be seen as time and temperature increase (Fig. 8b). The values of the curvatures obtained are given below (Table 4) (always considering the arithmetic mean of three values per experiment), using, as in the previous method, the geometric formula for calculating curvature (1).

Table 4: Curvatures - Second experimental method with data as reported in Fig. 8b

60°C 0.0029 0.0098 0.0166

70°C 0.0249 0.0311 0.0414

80°C 0.0471 0.0587 0.0813

Time/Temp

(t = 3m) (t = 5m) (t = 10m)

This second method shows an almost linear trend for the three-minute exposure time conditions (Fig. 8b). Therefore, given the short exposure time at a temperature above the glass transition temperature, it can be assumed that the deformation is almost entirely due to the release of residual stresses since the phenomenon of relaxation and creep does not generally have a linear trend. This consideration allows the phenomenon of residual stress release, and thus its deformation, to be considered linear with temperature. However, it should be emphasized that in this experimental approach, the residual stress process is not released entirely, but possibly only partially. Consequently, this method lends itself to future research, that is, studying the variation of deformation as a function of geometric parameters.