A close examination of the fractured surfaces of the specimens revealed a significant change in the orientation angles of the fibers protruding out in the horizontally and the vertically cast beams. Figure 3 depicts an average percentage value of the fibers visible from the fractured surface against their observed angle of orientation measured from the beam-axis.
In case of the horizontally cast beams (H), most of the fibers are observed to be aligned along the beam-axis with an orientation angle ranging from 0o to 30o. This percentage goes on reducing with the higher angles of the fiber orientation; only 5% of the fibers are found to have an orientation angle of 80o or higher. Whereas in case of the vertically cast beams (V), most of the fibers are found to be aligned at an orientation angle of 40o to 60o with a very small numbers orienting along the beam-axis. Only 13% of the fibers crossing the fractured beam surface were observed to be oriented along the beam-axis in the vertically cast beams against a value of 43% observed in case of the horizontally cast beams. This alteration in the fiber orientation caused by the mould position is expected to change the member post-peak strength parameters. The residual strength of SFRC depends only on the quantum of fibers that bridge the cracks being developed on the beam tension face. Their contribution towards the post-peak material response is expressed in the form of a parameter called as 'fiber-index' (β). This single parameter is reported to control the value of the crack-width that would develop on the tension face of the member and also, its residual tensile strength that it will exhibit in the post-peak range of load-displacement curves [Singh (2017)].
The residual tension/ flexural strength of SFRC specimen is a function of the three major constants, namely the constant γo that denotes the effect of fiber orientation with respect to the member-axis; the constant γd denoting their dispersion along the length and the width of flexural member caused during the concrete compaction and placement operations; and the length factor γl takes into account the non-linear distribution of the bond stress (τb) along the fiber length. The values of these three constants are reported as 0.8, 0.5 and 0.5, respectively, for the horizontally cast SFRC beams [Singh (2016 a)]. Accordingly, the final expression of the residual tensile strength (σth) of SFRC produced by using plain steel fibers in the pull-out failure mode is reproduced in Eq. 1. It must be multiplied by a suitable fiber-shape factor depending upon the fiber types, e.g. two for the hooked-end fibers and three for the wavy shaped fibers.
$$\:{\sigma\:}_{th}=2\:{\gamma\:}_{o\:}{\gamma\:}_{d}{\gamma\:}_{l}\left(0.75\sqrt{{f}_{c}}\:{V}_{f}\frac{l}{d}\right)=0.3\:{V}_{f}\frac{l}{d}\sqrt{{f}_{c}}$$
1
The value of the fiber orientation factor, γo in case of the vertically cast beams can be determined from the fiber distribution data shown in Fig. 3; it is obtained as 0.52. The values of other two factors (γd, and γl) will remain the same (= 0.5 each) as the fibers in the concrete will have equal probability to align themselves along the two dimensions of the cross-section during the pouring.
Equation 2 gives the modified value of the residual tensile strength (σtv) of SFRC for the vertically cast specimens by using the plain steel fibers. It should be multiplied by a suitable fiber-shape factor depending upon the fiber type used in the concrete production, i.e. two for hooked-end fibers and three for the wavy shaped fibers.
$$\:{\sigma\:}_{tv}=2\:{\gamma\:}_{o\:}{\gamma\:}_{d}{\gamma\:}_{l}\left(0.75\sqrt{{f}_{c}}\:{V}_{f}\frac{l}{d}\right)=0.2\:{V}_{f}\frac{l}{d}\sqrt{{f}_{c}}$$
2
Equations 1 and 2 show that the tensile strength exhibited by the vertically cast beams (V) reduces by a factor of 2/3 in comparison to the horizontally cast beams (H) having an identical size, fiber aspect ratio (l/d), fiber volume fraction (Vf) and concrete compression strength (fc). The change in the fiber orientation caused by the wall effect in the moulds seems to be a major influential factor that leads to the reduction in the strength exhibited by the SFRC specimens in the two casting positions. The change in the material residual tension strength will also lead to modification of the section flexural capacity of the members; it can be calculated from Eq. 3 [reproduced from Singh (2016a)] for the rectangular beams reinforced using steel fibers only.
$$\:\frac{{M}_{u}}{{f}_{ck}\text{B}{\text{D}}^{2}}=0.24{\left(\frac{{h}_{1}}{\text{D}}\right)}^{2}+0.5\beta\:{\left(\frac{{h}_{2}}{\text{D}}\right)}^{2}$$
3
In Eq. 3, the parameters h1/D and h2/D denotes the position of the section neutral-axis, which can be determined from the following expressions:
$$\:\frac{{h}_{1}}{\text{D}}=\left(\frac{2.38{\beta\:}}{1+2.38{\beta\:}}\right)\:\text{A}\text{n}\text{d}\:\frac{{h}_{2}}{\text{D}}=\left(1-\frac{{h}_{1}}{D}\right)$$
Fiber index (β) in above two expressions denotes the ratio of the material residual tensile strength (σt) and its compressive strength (fc) wherein; σt can be computed using equations 1 or 2 depending upon the mould position taken during the pouring.
The equations (1 to 3) show that the change in the fiber orientation caused by the mould position (H or V) changes the value of fiber-index (β), which directly influences the member flexural and shear capacity. The shear capacity of the beam section can be computed using the analytical model proposed by the author [Singh (2020)] to ensure that the designed beam specimens should not fail prematurely in shear. Table 3 illustrates the affect of the mould placement on different strength parameters of the test specimens, such as their flexural capacity (Mu), the value of the mid-span ultimate load (Puth) and safe shear strength calculated using the proposed equations 1 to 3.
Table 3
Strength parameters for two different casting orientations of SFRC beams (fiber aspect ratio = 80)
Orientation of mould while casting | Volume fraction, Vf | Fiber index, β | Mu (kNm) | Puth (kN) | Safe shear stress, MPa |
Horizontal (H) | 0.5 | 0.048 | 3.47 | 19.86 | 2.15 |
1 | 0.096 | 6.07 | 34.67 | 2.37 |
Vertical (V) | 0.5 | 0.032 | 2.43 | 13.90 | 2.09 |
1 | 0.064 | 4.42 | 25.26 | 2.23 |
Table 3 clearly established the changes brought in by different mould positions, for two values of the fiber dosage (i.e., 0.5% and 1%) used in the concrete specimens at the time of its production; the section flexural capacity reduces irrespective of the fiber dosage (Vf) used in the concrete; a similar trend is noticeable for their shear capacity though it is observed to be marginal. The flexural capacity and / or ultimate load of the beam sections reduce by ~ 30% when the position of the moulds during the casting process was changed from a horizontal to the vertical one. Interestingly, this change in the position of the moulds did not cause any significant change in the section shear capacity. The alignment and distribution of steel fibers along the span of the member and its depth are responsible for this reduction in the flexural capacity.
Figure 3 indicates the orientation of the fibers that become responsible for reduction in the flexural capacity as the fibers aligned along the beam-axis on its tensile face contribute maximum towards the flexural capacity; around 40% of all the fibers crossing the fractured surface are found to be aligned along the beam-axis, it reduces to 30% for an orientation angle of 20o and it further reduces to 20% for an angle of 40o for the beam specimens cast in the horizontal position; whereas, this percentage is only around 15% for the similarly cast beam specimens in the vertical moulds. On the other hand, the fibers aligned at angles of 40o to 50o contribute most towards the beam shear capacity and the average numbers of the fibers in this range are almost observed to be the same whether the beams were cast in the horizontally placed moulds or in the vertical moulds; thereby leading to an identical value of shear capacity being exhibited by them and as tabulated in Table 3.