City Research Online

Forcing of the primary and secondary crossflow instability

van Bokhorst, E. (2018). Forcing of the primary and secondary crossflow instability. (Unpublished Doctoral thesis, City, University of London)

Text - Accepted Version
Download (96MB) | Preview


With fossil fuels diminishing and fuel prices rising the development of more efficient aeroplanes is needed from an economical and environmental point of view. One possibility to increase the aerodynamic efficiency is to reduce the skin friction. This reduction can be achieved by having large areas of laminar flow. To be able to design laminar flow wings a thorough understanding of the transition from laminar to turbulent flow on a swept wing is needed.

The current experimental investigation focuses on the influence of environmental disturbances on the development of the crossflow instability. In order to study this influence, in the low turbulence wind tunnel at City University of London, an experimental model had to be designed. It was chosen to have a 45 degrees swept flat plate with a displacement body placed above it which created a favourable pressure gradient on the plate. With the sweep angle and favourable pressure gradient the conditions to have a transition process dominated by the crossflow instability were met. To characterize the flow detailed single hot-wire scans were carried out with a custom made data acquisition system in LabVIEW.

Prior to the experiments on the crossflow instability, measurements to characterize the freestream environment and to obtain the pressure distribution on the flat plate were carried out. The turbulence intensity was found to be around 0.02% for freestream velocities from 10 to 18m/s. The pressure distribution was obtained with two methods, with static pressure ports embedded in the plate and with a custom made pressure belt which spanwise location could easily be varied. Both methods gave comparable results and showed that a moderate favourable pressure gradient was created on the plate compared to other studies. The pressure belt results and panel code results were compared to a three-dimensional RANS computation of the set-up. A similar pressure gradient was found for all three pressure distributions. The magnitude of the pressure coefficient obtained from the panel code was slightly higher due to the inviscid character of this code.

The main experiments focused on two stages of the transition process, the development of the primary and secondary crossflow instability. The goal of the primary crossflow instability experiments was to investigate the influence of the wavelength content of different roughness distributions while previous studies focused more on the roughness height. A cylindrical and pyramidal roughness elements were studied for this purpose. The roughness elements were spaced at a spanwise distance of λ, which was the wavelength of the most amplified stationary crossflow wave following linear stability analysis. From the calculation of the Fourier coefficients it followed that the pyramidal roughness distribution had stronger forcing at λ and weaker forcing at λ/n compared to the cylindrical roughness distribution.

The experiments on the secondary instability first investigated the characteristics of this instability during different stages of its development. While in previous studies the characteristics in the early growth stage were studied here the characteristics of the secondary instability during breakdown are reported. Next, the experiments focused on excitation of the secondary instability by wall-forcing and freestream forcing. The secondary instability was forced from the wall with a small speaker. The results with different amplitudes of the excitation signal showed that the secondary instability is receptive for wall-forcing. The secondary instability was forced from the freestream through a small pipe which was connected to a speaker placed in the displacement body. The first experiments showed that the freestream forcing did not interact with the secondary instability in the boundary layer. Analogously to the travelling primary instability a thin roughness strip was placed on the swept flat plate at the location where the secondary instability started to grow. With the roughness in place an interaction between the freestream disturbances and secondary instability did occur and the phase structure of the secondary instability could be obtained.

Publication Type: Thesis (Doctoral)
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
T Technology > TL Motor vehicles. Aeronautics. Astronautics
Departments: Doctoral Theses
Doctoral Theses > School of Mathematics, Computer Science and Engineering
School of Mathematics, Computer Science & Engineering > Engineering > Mechanical Engineering & Aeronautics

Actions (login required)

View Item View Item


Downloads per month over past year

View more statistics