PSI - Issue 62

Giulia Rossini et al. / Procedia Structural Integrity 62 (2024) 347–354

350

4

Giulia Rossini et al. / Structural Integrity Procedia 00 (2019) 000 – 000

Fig. 1. (left) Details of Saw-Cut method -Web, dimensions in cm. (right) View in the horizontal plane of the 4 geometric configurations: SCw: GC_[distance]_[depth]. The dimensions of the 30 mm strain gauges are indicated in magenta. For all tests, two strain gauges (SGs) were simultaneously applied at the same point along the direction of prestressing to ensure result repeatability. The values obtained from strain gauge measurements were subsequently processed to derive using Hooke’s law ( = ⋅ ). In each test, this result was then compared with both a numerical and an analytical model, considering prestressing, applied vertical load, and position on the beam as variables. To assess the efficacy of the release methods, the ratio between the mean experimental stress (derived from the pair of strain gauges) [ , ] and the stress predicted by the FEM program VecTor2 (Wong et al., 2013) [ , ] was calculated. A ratio approaching 1 indicates that the release test provides a value consistent with the estimation from finite element analysis. Considering the prestress from analytical calculation [ , ], note that the ratio , , ⁄ would be nearly close to 1. 3. Results, comparison and discussion Table 1 presents a comparison of all the methods applied in this experimental program, summarizing the statistical values of the ratio , , ⁄ . For tests conducted using the Core Trepanning method and the Saw-Cut-Intrados method, the results obtained are considered unreliable and therefore the statistical parameters are not reported. The same consideration applies to tests conducted with the geometric configuration of Saw-Cut - Web method on a limited number of specimens. To illustrate the comparison between experimental and numerical values, Fig. 2 - left shows the results of only the blunt pyramidal specimens and the Saw-Cut Web methods. This figure shows the agreement between the experimental and numerical values for these specific test methods.

Table 1

Results of different stress release methods. Core Trepanning Saw-Cut Intrados Blunt Pyramidal specimen

Saw-Cut Web

SCw GC_10_2

SCw GC_10_3

SCw GC_6_2

SCw GC_6_3

C

SCi

R

n° Specimen

11

9

16

2

4

1

8

, ,

Mean

- - -

- - -

1.40 0.58 42%

0.34

0.62

0.63

0.91 0.19 21%

St. Dev.

- -

- -

- -

CoV