Experimental Investigation of Clear-Water local Scour at pile groups

         Experimental setup       

The experiments were carried out in a 4 m long, 0.41 m wide, and 0.25 m deep flume. The flume had a working section in the form of a 0.08 m depth recess below its bed that was filled with movable sands. The properties of sediments used for the experiments are presented in table 1S. In this table, d₁₆ = particle size for which 16% are finer, d₅₀ = median particle size of sediment bed, d₈₄ = particle size for which 84% are finer. details of the flume and experimental setup are shown in Fig. 1S. The water depth in the flume is adjusted by the position of the sluice gate at the downstream end of the flume. An electrical flow meter was used to measure flow discharge. As shown in Fig. 2S, eight types of pile group arrangements were tested, including the side–by–side, tandem, and 2×2, 2×3, 2×4, and 3×2.

Table S1. Properties of sediments used in experiments.

FIG. S1: Experimental set-up.

FIG. S2: Pile Groups Used in Experiments. 

The test conditions together with the experimental results are listed in Table 2S. In this table, D is pile diameter, h is upstream water depth, d₅₀ is the mean particle diameter, U is mean approach flow velocity, Uc is velocity at the threshold condition for sediment movement, G is pile spacing, and S is maximum scour depth.

               Table 2S: Summary of test conditions and experimental results.

For all of the experiments, the test duration was kept more than 7 h. for some of experiments, this time lasted up to 15-18 h depend on the flow conditions and bed materials. The experiments were stopped at a stage when the rate of scour did decelerate to very small values, or there was no scour at all. Some of previous experimental works are in the range of this duration. Vittal et. al (1994) adopted a duration of 6 h for their experiments on pier group. Salim and Jones (1998) kept the experiment duration at 4 h for most of the tests conducted on exposed pile foundations. Kumar et. al (1999) carried out their experiments for a duration that the scour did not change by more than 1 mm over a period of 3 h. Mia and Nago (2003) stopped their experiments at a stage when there was less than 1 mm scour by 1 h or no scour at all (2.33 hr <t_e <5hr).

Fig. 3S shows the time development of the maximum scour depth, S, in the front of piles in each row for the case of 2×3 pile group with G/D = 2, and U = 0.18 m/s. As seen from these figure, the scour develops toward the equilibrium stage through a transitional period. A calibration run carried out for more than 45 hours for 2×4 pile group with G/D = 2 and U = 0.21 m/s to investigate the duration needed to achieve equilibrium scour.  As shown at Fig. 4S, 86% of the equilibrium scour depth is attained at 5.5 hours and 92% of the equilibrium scour depth is developed in a time 8 h. The maximum scour depth does not change after 29 h of test duration.

Fig. 3S. Development of scour depth with time: U = 0.18 m/s, G/D = 2 (Exp. No. 62).    

Fig. 4S. Variation of scour depth with time for 2×4 pile group: U = 0.21 m/s, G/D = 2. 

Fig. 5S compare the scour depth obtained for different pile groups for U = 0.246 m/s. Maximum scour depth obtained for 3×2 pile group. Maximum scour depth obtained for 3×2 pile group. This may be due to the increased effect of compressed horseshoe vortices. Minimum scour depth obtained for tow–pile, tandem arrangement (1×2 pile group). This indicates that the effect of compressed horseshoe vortices is larger than reinforcement.

Fig. 5S. Effect of pile group arrangement on scour depth.

Before and after test photographs for 2×3 pile group with G/D = 2, is shown in Fig. 6S. When a group of piles is exposed to flow action, tow kinds of scour patterns emerge: (1) the local scour around the individual piles, and (2) the global scour in the form of a saucer-shaped depression (the general lowering of the bed level over the entire area of the pile group) (Fig. 7S). The local scour is caused by the horseshoe vortex, the vortex shedding, and the contraction of streamlines, associated with the individual pile. (Note that the resulting local scour characteristics may not be the same as those in the case of a single pile due to the possible influence of the global scour). The global scour, on the other hand, is caused by the following tow effects: (1) the change in the flow velocity in the gap between the piles, and (2) the turbulence generated by the individual piles (Sumer and Fredsøe 2002).

FIG. 6S: Photograph of the bed profile around piles before and after scour test. D = 2.2 cm, d₅₀ = 0.98 mm, U = 0.26 m/s. 

Fig. 7S: Scour hole contour and surface plots for 2×4 pile group. D = 2.2 cm, h = 3.5 cm (Exp. No. 44 and 36).

A brief Review of Research about Local Scour at bridge piers and references 

| SUT |C.E | Home |Links |

B. Ataie-Ashtiani July 2005