Investigation of methods of delaying or controlling ventilation on surface piercing struts

The problem of undesired ventilation of partially submerged foils and appendages on
hydrofoils is traced from the earliest hydrofoils which appeared toward the end of the last
century. Ventilation is assessed in the context of other problems in the development of
hydrofoils, and is found to be still a serious problem. The elements of the mechanism of
ventilation inception are identified as a region of low energy and low pressure flow, a surface
seal, and the existence of a means of rupturing the surface seal. Two modes of rupture are
identified, corresponding to nose and tail ventilation. The mechanism is used to explain the
use of taper, dihedral, camber, and fences to prevent ventilation. The extension of these
techniques is discussed, as well as the possible application of some novel techniques and some
techniques presently in use in the aerodynamic field. These include separation control by
boundary layer suction and blowing, air bleed, solid and liquid fences, flaps, and controlled
ventilated cavities. Observed differences between model and prototype behaviour is the basis
of a discussion of modelling ventilation phenomena. Most existing data for similar geometries
can be reduced to functions of the dimensionless cavitation and Froude numbers. An
empirical correlation between surface drawdown and a dimensionless velocity is shown to have
wide applicability independently of geometry. Two experiments were performed to gauge the
effect of parameters suspected to influence model tests, primarily roughness, wettability and
speed. It was found that the effects of roughness and size and speed may be analogous in
certain respects. High speed cine photography revealed a mixed subsurface mode of ventilation
not previously observed. It also enabled a description of the interaction of separation,
cavitation and ventilation. The present methods for prediction of separation are presented and
used as the basis for semi-empirical predictions of the amount of blowing and suction required
to eliminate separation from struts of finite length in the presence of the free water surface.
Suction and blowing in the amounts predicted were used to retard the inception of ventilation.
The greatest success was achieved with a roughened model, which it is believed simulated full
scale conditions more accurately. The results of the experiments with separation nearly completely eliminated call into question the hypothesis that separation is a necessary precursor
of ventilation. The results of an experiment with a liquid fence barrier were mixed, but were also more successful with a roughened model.