Revisiting the reattachment regime: a closer look at tandem cylinder flow at

Abstract

Tandem cylinder flow comprises several different flow regimes. Within the reattachment regime, the development of the gap shear layers is of utmost importance to the flow, but has received little attention so far. Through direct numerical simulations at $Re = 10^{4}$ , for a gap ratio of 3.0, we have discovered that the shear layers are significantly altered with respect to a single cylinder. These differences include early onset of separation, crossflow stabilising, delayed transition to turbulence and little meandering of the transition region. Vortex pairing in the gap shear layers is reported for the first time. The interaction between the recirculating gap flow and the shear layers was investigated. Asymmetrical, large-scale gap vortices influence the position of transition to turbulence through direct contact and through secondary flows. The occurrence of transition in the gap shear layers has consequences for both the reattachment mechanism and the development of the downstream cylinder wake. The reattachment points are unsteady with large amplitude fluctuations on a fine time scale. Reattachment is seen to be a combination of impingement and modification of the upstream shear layers, which causes a double shear layer in the downstream cylinder near-wake. Buffeting by and interaction with the gap shear layers likely cause transition to turbulence in the downstream cylinder boundary layer. This leads to significant changes in the wake topology, compared with a single-cylinder wake.

Asymmetric cellular bi-stability in the gap between tandem cylinders

“…We have discovered that bi-stability may indeed be cellular, as suggested by Aasland et al. (2022). Local manifestations of the mode switch emerge clearly from the velocity time traces in figure 8.…”

“…This idea was put forth by Aasland et al. (2022), who suggested that bi-stability may occur in spanwise cells, based on the observation of spanwise inhomogeneity at and . In the present study, we have investigated this hypothesis.…”

“…However, in our previous study with a Reynolds number of 10 000, a grid cell size of gave good results (Aasland et al. 2022). Therefore, we are confident that a similar cell size will be sufficient also at the present Reynolds number, which is nearly two orders of magnitude lower.…”

“…The existence of spanwise phase differences implies the possibility of variation between different tandem flow regimes along the cylinder span, near the drag-inversion spacing. This idea was put forth by Aasland et al (2022), who suggested that bi-stability may occur in spanwise cells, based on the observation of spanwise inhomogeneity at L/D = 3 and Re = 10 000. In the present study, we have investigated this hypothesis.…”

Flow topology in the gap and wake of convex curved tandem cylinders

“…A similar result was recently reported by Wang et al (2021), for channel-confined tandem cylinders. The results of Aasland et al (2022b) indicated the possibility of spanwise localised bi-stability near the critical spacing, where shedding in the gap occurs in spanwise cells. This was confirmed by Aasland et al (2023), who found that bi-stability manifests not only spanwise localised, but also asymmetrically in the gap; shedding may occur repeatedly from just one gap shear layer.…”