PSI - Issue 28

Ezio Cadoni et al. / Procedia Structural Integrity 28 (2020) 933–942 Author name / Structural Integrity Procedia 00 (2020) 000–000

940

8

Table 3. Dynamic results of UHPFRCs and UHPC (matrix) in direct-shear test. Material Loading-rate Shear strength

Fracture time

Max shear-slip

L

˙ γ

( s − 1 )

( GPa / s )

( MPa )

( µ s )

( mm )

( mm )

UHPC 0%

164 (32) 128 (36) 175 (26) 216 (30) 154 (29)

15 . 9 (3 . 1) 20 . 1 (3 . 7) 22 . 9 (3 . 7) 23 . 3 (5 . 6) 25 . 6 (3 . 5)

222 (36) 249 (48) 199 151 (14) 274 (20) (3)

0 . 27

10 . 61 (2 . 53)

194 (90) 141 (35) 139 (24) 297

(0 . 12)

UHPFRC 1%

0 . 24

9 . 19

(0 . 09)

(2)

UHPFRC 2%

0 . 22

8 . 17

(0 . 11)

(0 . 24)

UHPFRC 3%

0 . 17

5 . 44

(0 . 03)

(1 . 47)

(121)

UHPFRC 4%

0 . 30

8 . 07

134 (40)

(0 . 16)

(0 . 38)

Fig. 8. Dynamic shear strength in function of the fibre reinforcement.

The influence of the fibre percentage can be highlighted observing Fig. 8 where the comparison of the average shear strength between the di ff erent UHPFRCs are made. It is possible to observe as the shear strength increases with increasing of the percentage of fibres. In comparison to the matrix shear strength, increments of 26%, 44%, 46% and 61% in shear strength were obtained for 1%, 2%, 3% and 4% UHPFRCs, respectively. Fig. 9 shows the comparison between representative dynamic shear versus shear-slip curves in function of the percentage of fibre reinforcement obtained by using equations (1) and (2). It can be observed how the post-peak shear strength is hardly governed by the percentage of fibre in the UHPFRC. As well in direct tensile test the post-peak residual strength depend from the distribution and orientation of the fibres [Cadoni et al. (2009); Cadoni and Forni (2016); Cadoni et al. (2019); Caverzan et al. (2012, 2013); Luccioni et al. (2018)].

5. Concluding remarks

The dynamic response of UHPFRCs in direct-shear tests was investigated at high strain rates by using a modified Hopkinson bar device. Four di ff erent fibre reinforcement (1%, 2%, 3% and 4%) were tested at the same high strain-rate level. Prior to the experimental tests, a deep numerical investigation was performed in order to find the most appro priate specimen shape. Three di ff erent shear sample configurations were numerically studied. The numerical analyses highlighted that only one shape configuration can be an optimal solution for the dynamic tests (Fig. 2). Moreover, on the selected axial-symmetric shape configuration, a second numerical study was performed to understand the better notch position and depth (Fig. 3). It was numerically demonstrated that in the final sample geometry the main and