A class of self-adjusting injectors for spraying liquid oxidizers and/or fuels into combustion chambers has been proposed. The proposed injectors were originally intended for use in rocket-engine combustion chambers, but could also be used to improve control over flows of liquid propellants in other combustion chambers.
In some applications, there is a need to vary the combustion power. Usually, this requirement is satisfied by throttling either or both propellants injected into a combustion chamber. Most prior injectors work well over small throttling ranges; attempts to use them over wider throttling ranges can result in undesired injection pressure drops that can compromise combustion performance, including giving rise to combustion instability. Some prior injectors have fixed inlet areas. A fixed inlet area is optimum at one flow rate only; if the flow rate is changed, then the pressure drop also changes to a value that, typically, is not optimum. Some other prior injectors have been equipped with externally actuated mechanisms to vary their inlet areas, but, in each case, either the mechanism has been excessively complex or else combustion performance has still been compromised.
The basic idea of the proposed injectors is to use simple mechanisms, inside the injectors themselves, to adjust inlet areas so as to keep injection pressure drops at or near optimum values throughout wide throttling ranges. These mechanisms would be actuated by the very pressure drops that they are intended to regulate.
A typical injector according to the proposal (see figure) would include an injection tube containing multiple inlet slots cut tangentially to the inner surface of the tube. The tangential orientation of these slots would channel the fluid entering the tube through them into a swirling motion. (The tangential orientation of the slots is not an essential element of the proposal; it is mentioned here because swirling injectors are common.) The dimensions and arrangement of these slots would be chosen, in conjunction with those of other components, to optimize performance.
The upper end of the injection tube would be covered with a cap that would contain a number of outer slots equal to the number of injection slots. The cap would translate axially (up and down in the figure). Two retaining pins (of which one can be seen in the figure) in holes in the cap would protrude into grooves in the injection tube to prevent the cap from coming off the injection tube while allowing the cap to slide freely only within limits. The pins would also keep the outer slots in the cap aligned with the injection slots in the tube.
A coil spring would lie in an annular recess in the injection tube and would be compressed between the bottom of the recess and an inner flange at the bottom of the cap. Small tangential holes could be included, in addition to the outer slots, to allow initial flow at the lowest power level. The cap and the exposed portion of the injector tube would protrude into a manifold containing the fluid to be injected, and outer flanges on the cap would contribute drag between the cap and the fluid to damp any oscillatory motion of the cap, thereby helping to suppress instability. Labyrinth-type seal grooves would prevent gross leakage, so that most of the flow must enter the injection tube either through the tangential slots or the small tangential holes.
In operation, the pressure drop between the manifold and the inside of the cap (which pressure drop would be part of the total pressure drop from the manifold to the combustion chamber) would create a force that would push the cap downward against the spring. This downward motion would cause the outer slots in the cap to partially expose the tangential slots in the injection tube, thereby limiting the pressure drop by increasing the cross-sectional area for flow into the tube. The number and dimensions of the tangential slots would be chosen in conjunction with the stiffness and preload of the spring to obtain the optimum pressure drop as a function of the rate of flow (and, hence, as a function of the combustion power level).
This work was done by Huu Trinh and William Myers of Marshall Space Flight Center.