Dynamics of Hydrofoils as Applied to Naval Propellers 



The experimental results obtained in the cavitation tunnel were compared 

 with theoretical results, using the superposition method referred to by the au- 

 thors as well as the rigorous method of conformal mapping. 



The experimental data showed indeed that for the NACA a = 1.0 mean line, 

 viscous flow effects resulted in an 25% shift in zero lift angle of attack, towards 

 smaller values. U satisfactory agreement between theoretical predictions of 

 cavitation-free angle of attack ranges and experimental data was to be obtained, 

 the viscous flow corrections had to be incorporated both in camber and angle of 

 attack. - ..y . . - 



At present, an experimental program has been started in the cavitation- 

 tunnel laboratory of Berlin Technical University where the cavitation-free angle 

 of attack ranges are to be assessed for oscillating foil sections. The latter con- 

 sist of NACA 16 thickness distributions cambered with the a = 0.8 mean line. In 

 total, 25 different foil sections will be tested in a two-dimensional test section, 

 over a wide range of cavitation numbers and reduced frequencies, covering 

 thickness -chord ratios 0.03 < t/c < 0.18 and design lift coefficients < Cli < 

 0.5. 



DISCUSSION 



C.-A. Johnsson 



Swedish State Shipbuilding Experimental Tank 



Goteborg, Sweden 



I should like to support Dr. Morgan when he claims that the lifting-surface 

 theory in its present stage is adequate for design purposes for conventional pro- 

 pellers in most cases. In his answer to Dr. Morgan, Dr. Castagneto refers to 

 some comparisons between calculations and experiments, reported recently by 

 me, where he has found, in some cases, very large differences between the cal- 

 culated number of revolutions and those obtained in open-water tests. Regard- 

 ing those two cases, where such discrepancies could be noticed, it can be clearly 

 seen in the report that they are characterized by very small values of the ad- 

 vance ratio /. Thus, the reason for the discrepancies is most likely that these 

 cases are outside the range within which the theory of moderately loaded propel- 

 lers can be applied. As this range is difficult to determine accurately enough, 

 low / values have to be considered with care, which, of course, is a limitation 

 of the applicability of the theory. Bearing this in mind, however, the lifting- 

 surface theory seems to be adequate for design purposes as long as the NACA 

 a = 0.8 mean line is used. Very good agreement between calculated and measured 

 number of revolutions can be expected without adding any empirical corrections, 

 provided the thickness effect is included. 



1043 



