Controlled accelerations for Rayleigh-Taylor instability

TL;DR

This work addresses the dependency of Rayleigh–Taylor instability on acceleration history and initial conditions by introducing CAMPI, a programmable, low Atwood number RT facility with fully miscible fluids and high-resolution in-situ diagnostics. CAMPI's moving-frame diagnostics, closed-loop winch acceleration, and torque-feedforward control enable precise replication of complex acceleration histories, including Accel–Decel–Accel sequences. The study demonstrates expected RT behaviour: single-mode growth during acceleration, mixing-region shrinking during deceleration, and re-energised multi-scale growth upon re-acceleration, providing ground-truth data across regimes for validation of numerical models. The apparatus and methods offer a versatile resource for investigating variable-acceleration RT, with potential impact on understanding mixing in industrial and astrophysical contexts.

Abstract

The dynamics of turbulent mixing induced by Rayleigh-Taylor instability are heavily dependent on the acceleration experienced by the fluids and the frequency content of the initial interface between them. Both are readily controllable in numerical simulations, but in experimental studies are difficult to influence and adequately diagnose. In this paper we present the CAMPI apparatus, an experimental facility for study of low Atwood number Rayleigh-Taylor instability with highly controllable, complex acceleration histories. The apparatus provides unique and novel capability for the experimental study of variable acceleration Rayleigh-Taylor instability with fully miscible fluids and at a scale suitable for high resolution optical diagnostics. We present experimental results of initially single mode instability evolution through two stepwise acceleration reversals, a case termed Accel-Decel-Accel, demonstrating the ability of the apparatus to accurately generate a prescribed acceleration history. We observe the behaviour predicted by previous numerical studies, with instability growth reaching a terminal velocity in the first episode of acceleration, followed by a shrinking and homogenisation of the mixing region throughout deceleration, and unstable growth from a multi-frequency initial condition during the second acceleration. We present the CAMPI apparatus to the field as a much needed. resource of ground truth data on the behaviour of Rayleigh-Taylor instability across a broad range of regimes.

Figures (10)

• Figure 1: The variation of Rayleigh--Taylor growth with apparent gravity during a maximal simple acceleration profile spanning the full track length.
• Figure 2: The CAMPI variable acceleration apparatus. The propulsion is provided by a closed loop winch system, with drum, motor, and brake located on a single shaft at the base of the apparatus, and return pulley located at the top. The apparatus is mounted on a pair of universal columns that span the full height of the Hele-Shaw Laboratory. The experimental setup is contained within a shuttle assembly that runs along a track spanning the full height of the apparatus; $12.6 \mathrm{\; m}$ between the two rotational axes.
• Figure 3: The moving reference frame onboard diagnostic systems.
• Figure 4: The closed loop control block diagram showing the improved system (black), the original system (black + red) and the implementation of torque feed forward (blue).
• Figure 5: The acceleration time-history as experienced by the fluid before (red) and after (black) the addition of torque feed back and gain scheduling control methods.