Behaviour of a Ground Effect Machine 



unstability would require the execution of forced roll and heave experiments 

 over a range of frequencies and using various base pressures. Such an investi- 

 gation was not included in the present research. 



When fitted with flexible trunks the model in the over water hovering condi- 

 tion only suffered from self-induced roll oscillations, with amplitudes smaller 

 than when the jet exits were rigid. The heave and pitch damping seem to have 

 increased also, in spite of the enlargement of the air cushion by means of flexi- 

 ble trunks. 



Comparison of the extinction curves over land and over water learns that 

 the heave and pitch damping is larger over land than over water. The natural 

 periods of these motions were smaller over land. 



FLYING BEHAVIOUR OVER SMOOTH WATER 

 AND OVER LAND 



The behaviour of the model with both rigid jet exits and flexible trunks was 

 quite satisfactory over land. It was dynamically stable in roll, pitch and heave. 

 It skimmed smoothly over the ground with no appreciable change of trim at 

 speeds up to 20 knots. 



Flying over smooth deep water, rolling decayed with increasing speed. The 

 motion returned above the hump speed and it decayed again with further increase 

 of the forward speed. In shallow water the picture was the same. This behav- 

 iour is shown in the Figs. 6 through 9. 



The resistance curves had their highest hump at speeds between 10 and 12 

 knots, corresponding with Froude numbers between 0.40 and 0.45. These are 

 speeds for which also the highest specific wave resistances of ship hulls are 

 found. Apparently the water depth did not largely affect the speed where the re- 

 sistance showed the highest hump. It affected primarily only the height of the 

 hump. 



BEHAVIOUR OF THE MODEL PROCEEDING 

 OVER REGULAR WAVES 



The natural periods of the pitch, heave and roll motions at zero speed lie 

 between 1.8 and 2.5 seconds. It is reasonable to assume that these quantities do 

 not change much with increasing speed. So the speed range and simulated wave- 

 lengths assure that in many cases the period of encounter was equal to the natu- 

 ral period. 



Figure 10 shows only slight humps in the curves of the pitch and heave am- 

 plitudes. This indicates that these motions were well damped. The curves of 

 the roll amplitude have a hump at the speed for which the period of encounter is 

 expected to be about equal to the natural period. This picture of the dynamic 

 properties is in accordance with that obtained from the motion extinction curves 

 of the model hovering over smooth water. In these conditions the motions are 



701 



