Transient numerical simulations were conducted to investigate the influence of large amplitude and fast impact backpressure on a shock train. The fundamental problem consists of a shock train within a constant-area channel with a Ma=1.61 inflow and a pulse backpressure applied to the outlet. The pressure disturbance in the isolator has an intense forcing-response lag. From the moment of the backpressure peak appearance, it takes 36 times the backpressure duration for the pressure disturbance to reach the upstream end. It moves upstream with time in the form of a normal shock wave. As time progresses, the normal shock degenerates into a $\lambda $ shock and a compression wave behind due to the action of viscous dissipation in the boundary layer. Eventually, a multi-stage shock train is formed. The maximum backpropagation distance is a quadratic function of both the pulse backpressure peak and duration, and the relationship between these variables was determined by fitting. When the integral value of backpressure to time is fixed, reducing the backpressure peak while increasing the duration will reduce the backpressure pulsation at the isolator outlet, which will be more conducive to shortening the maximum backpropagation distance than reducing the duration and increasing the backpressure peak. The values of backpressure peak and duration are obtained from the detonation combustion case, which ensures the authenticity of backpressure characteristics. The relevant research conclusions can provide a reference for the design of the isolator of pulse detonation ramjet.

A new measurement technique to reconstruct the density field of the shock-wave/boundary-layer interaction (SWBLI) in a confined duct is proposed. With this technique, it is possible to quantitatively capture in detail the structures of the density field both in the regions of the shock-systems in the central core and boundary-layer flows near the duct wall concurrently. The novel feature of the proposed technique is to make use of the schlieren images with the rainbow filters of the vertical and horizontal cutoff settings and then to reconstruct the two-dimensional density field integrated over the line-of-sight direction using the corresponding filter calibration curves. The proposed technique is applied for the first time to a shock train in a constant-area straight duct under the upstream condition of the shock train: the freestream Mach number is 1.42, the incoming boundary layer thickness normalised by the duct half height is 0.175, and the corresponding unit Reynolds number $Re/m$ is $2.99 \times 10^7$ m-1. The calculated isopycnic field depicts the streamwise and transverse density variations inside the shock train, the mixing region after the shock train, and the boundary-layer of the interaction region. This technique is shown to be capable of identifying the locations of shocks in a shock train more precisely than a conventional approach measuring the static pressure distribution along the duct wall. In addition, various quantitative visual representations such as a shadowgraphy and a bright-field schlieren can be extracted from the density field acquired by the present approach, and the spatial evolution of the shape and strength of each shock constituting the shock train as well as the boundary layer flow properties can be quantitatively clarified.

The present study investigates the behaviour of the shock train in a typical Ramjet engine under the influence of shock and expansion waves at the entry of a low aspect ratio (1:0.75) rectangular duct/isolator at supersonic Mach number (M = 1.7). The start/unstart characteristics are investigated through steady/unsteady pressure measurements under different back and dynamic pressures while the shock train dynamics are captured through instantaneous Schlieren flow visualisation. Two parameters, namely pressure recovery and the pressure gradient, is derived to assess the duct/isolator performance. For a given back pressure, with maximum blockage (9% above nominal), the duct/isolator flow is established when the dynamic pressure is increased by 23.5%. The unsteady pressure measurements indicate different scales of eddies above 80 Hz (with and without flap deflection). Under the no flap deflection (no back pressure) condition, the maximum fluctuating pressure component is 0.01% and 0.1% of the stagnation pressure at X/L = 0.03 (close to the entry of the duct) and X/L = 0.53 (middle of the duct), respectively. Once the flap is deflected (δ = 8°), decay in eddies by one order is noticed. Further increase in back pressure (δ ≥ 11°) leads the flow to unstart where eddies are observed to be disappeared.