Weissinger and Maass 



-2 



-0.5 



\ 



Fig. 23 - (See (157)) Angle of 

 attack contribution to net shroud 

 pressure coefficient (n = 3, fi, = 

 0.956, d/Rp = 0.213, Xp/c = -0.219, 

 X = 0.5, J = 0.344) 



Of course, there remain problems, even within the framework of the linear 

 theory. For example, the problems connected with the higher harmonics of the 

 finite-bladed propeller might be investigated more thoroughly. It might be de- 

 sirable to drop the assumption of slenderness for the propeller blades and to 

 investigate the flow in the neighbourhood of the propeller (supplemented, per- 

 haps, by guide vanes) by a lifting-surface theory. Also problems of interfer- 

 ence or of free surface might be sought. An important problem, though tran- 

 scending the realm of nonviscous theory, is a practical determination of the 

 propeller tip clearance for use in the above theory. 



For the ducted propeller at angle of attack, the rough theory of Sec. 6 should 

 be refined, at least for checking this comparatively simple theory. 



Finally, two problems will be mentioned which are beyond the scope of this 

 review. The first one arises from the fact that a ducted propeller does not usu- 

 ally operate in a uniform stream. As a matter of fact, the NSRDC computer 

 program includes the possibility of a radial dependence of the inflow. One might 

 also include the higher harmonics induced by a ship as analysed in (4). Isay (40) 

 developed a theory for ducted propellers in a wake based on two-dimensional 

 theory with three-dimensional corrections. But, by the methods discussed in 

 this review, the problem had to be solved by the methods of potential theory, 

 though the flow in a wake usually is not a potential flow. This leads to the prob- 

 lem of nonviscous nonpotential flow (Euler equations), which has been treated 

 only rarely in comparison with the extensive literature concerned with solutions 



1246 



