Fully Cavitating Propeller for a Hydrofoil Ship 



the correction to the camber only. This reasoning is endorsed by the work of 

 Oba in Ref. 12, where he has shown that in the case of a two-dimensional cas- 

 cade of fully cavitating foils, the minimum incidence required to ensure clear- 

 ance between the back cavity and the face of the section is less than that re- 

 quired in the case of the isolated hydrofoil. Tulin's description of the advantage 

 to be gained by using high camber in fully cavitating propellers, given in Ref. 6, 

 reinforces this conclusion. 



A further factor that was instrumental in producing large angles of attack 

 in these circular-arc face screws was the manner in which allowance was made 

 for the structural thickness of the sections themselves. This was done by speci- 

 fying a control thickness at 20 percent of the chord from the leading edge of each 

 section, and then installing the initial incidence (and camber), before applying 

 the aforementioned corrections, so that the back cavity would clear this point. 

 This procedure led to relatively large values of incidence being installed. The 

 cavity height calculations used in this approach were made using Wu's theory, 

 and are given in Ref. 13. • i 



Up to the time of the W257 experiments, the extreme leading edges of the 

 fully cavitating screws designed at NPL were made relatively sharp, the struc- 

 tural thickness being controlled at the 20 percent chord position as previously 

 described. Because this region of these propellers is so critical from the 

 strength and vibration standpoints, it was decided to conduct a few simple exper- 

 iments to determine the effect of increasing the extreme leading edge thickness. 

 The screw used for these experiments was T95, as described in Ref. 8. The 

 leading edges were initially sharp and were thickened for these experiments by 

 adding soft solder to the backs of the blades in the leading edge vicinity. The 

 leading edge roundings had a diameter of one -half percent of the local chords 

 and the solder was then faired in to zero thickness at a postion 20 percent of the 

 chord length from the leading edges. The modified screw was then rerun and 

 the results compared with those obtained earlier. This comparison is shown in 

 Figs. 14, 15, and 16, where it is seen that in the normal operating range the ef- 

 fect of the thickened leading edges is mainly manifested as a reduction in thrust. 

 The effect on efficiency is small— the efficiency with the thickened leading edges 

 being about 97-1/2 percent of that with sharp edges. Clearly, the small loss in 

 efficiency and thrust arising from this modification is worth incurring for the 

 added strength that will result. 



A further experiment was conducted on the modified T95 screw with the ob- 

 ject of gauging the heights of the cavities above the backs of the blades. For 

 this purpose three pins, each of different length, were attached to the blades, 

 one to each blade at 70 percent radius and 20 percent of the chord length from 

 the leading edge. The arrangement is similar to the method used by Johnson in 

 Ref. 11 on a three-dimensional hydrofoil in a tank. The relative positions of the 

 back cavities on the propeller were then observed under stroboscopic lighting, 

 leading to the results given in Table 3. 



A reasonable J value for the high-speed operating condition of this screw 

 would be in the range 0.9 to just over 1.0, i.e., at cj = 0.3, J = 1.0, then t? = 

 0.59 and k^ - 0.09. In this case the height of the cavity above the blade surface 

 is about one and one -half percent of the chord and, therefore, the thickness of 

 the blade at this position could be increased by this amount without adversely 



979 



