Abstract
Shock vector control (SVC) is a technique to control airplanes and spaceplanes by means of fluid
injection into the nozzle flow. SVC is part of fluidic thrust vectoring, a technique whereby a
control force can be generated by means of fluid injection in the engine exhaust.
For this thesis the application of shock vector control to a converging-diverging nozzle and a
hypersonic vehicle was investigated. The experiments with a converging-diverging nozzle with
an injector took place in an experimental facility which houses extruded two-dimensional nozzle
profiles of 8 mm thick. The profiles are sealed by two windows on either side of the nozzle for
optical access. The nozzle is a linear converging-diverging nozzle with a sonic injector placed at
71% length of the diverging section. On this geometry two-dimensional steady state simulations
have been performed, as well as time accurate simulations and a large eddy simulation. It is
shown that the injected mass creates a blockage for the incoming flow generating a shock wave.
This blockage causes the core flow to separate before the injector, meaning that there is a high
pressure region before the injector creating the sideforce. It was demonstrated numerically and
experimentally that the sideforce, relative to the axial force, has a linear relationship with the
injected mass until the shock wave hits the opposite nozzle wall causing a shockwave boundary
layer interaction that is detrimental to the net sideforce.
It is shown that this type of nozzle flow can be modelled with analytical relations based on
blast wave theory. The established analytical methods were calibrated based on an orifice injector
and therefore the calibration constant needed to be changed from 1.2 to 2.36 because of the full
width injector. In an orifice injector the cross flow can go around the injected fluid, while in a
two-dimensional flow case all the nozzle flow is being turned by the injected fluid, and therefore
the calibration constant needs to be increased.
It was demonstrated that for the small scale experiment the force establishment time is 0.2
msec, which is faster than most solenoid valves available on the market. Analytically it is shown
that the establishment time will increase with the size of the nozzle, but it will remain fast enough
to be used as a control system for a full scale rocket or scramjet.
The experimental supersonic geometry was modelled with Large eddy simulation (LES) and
compared to Reynolds averaged Navier-Stokes (RANS) simulations. It was shown that the flow
shows three-dimensional effects at the point where the sidewall meets the injector, but is more
iii
two-dimensional towards the middle of the nozzle.
To test the hypersonic implementation of SVC a scramjet of similar shape to the X-43 was
designed with a single slot injector close to the trailing edge of the nozzle. The pressure in the
hypersonic environment is orders of magnitude lower than in the supersonic experiment and therefore
a large expansion of the Mach disk at the injection point was seen. An increase of sideforce
and moment is present with an increase in injection massflow for the experiments and simulations
up to the point where the size of the separation zone does not anymore match the size of the Mach
disk at high injection pressures. For the current experimental setup it is therefore shown that efficient
control can be established with an injection pressure lower than 39 times the static flow
pressure.