160 



has been made and the resulting curve is seen in 

 Figure 2 to be in better agreement with the measure- 

 ments. 



Theoretical results obtained with the simple 

 representation of a propeller included in the 

 method indicate that the propeller can produce 

 large local changes in the boundary layer flow. An 

 example for which measurements are also available 

 is shown in Figure 3 where results are presented 

 for various stations along a body of revolution 

 having a fine tail and contra-rotating propellers. 

 The measurements are of total velocity and were 

 made with rakes of probes fixed to the body, the 

 rake at the forward propeller plane being removed 

 when the propeller was fitted. The velocity pro- 

 files on the unpowered body are well predicted 

 except close to the tail where boundary layer 

 separation appears to be present: it is noted 

 earlier that the calculation procedure will not 

 predict separation. The changes produced by the 

 propellers are in surprisingly good agreement with 

 predicted changes considering that an actuator 

 disc representation of the propellers has been 

 adopted. The velocities in a region close to the 

 hull are under-predicted at the two rearmost 

 stations and at the station very close to the tail 

 the velocities near to the edge of the boundary 

 layer are also underpredicted. Apart from these 

 discrepancies the effect of the propeller is well 

 represented . 



Velocity profiles close to the propeller plane 

 with and without propeller operating have been 

 reported by Huang (1976). A typical example is 

 shown in Figure 4 where it can be seen that the 

 Myring theoretical prediction for the unpowered 

 body gives velocities which tend to be too low. 

 Nevertheless the discrepancy is less than 4% of the 

 measured values. 



Reliability of Harmonic Analyses of Measured Flow 

 Fields 



The comparisons between theoretical prediction and 

 measurement indicate that the calculation method 

 gives a good approximation to velocity profiles 

 measured on powered and unpowered bodies of 

 revolution. The method is not expected to pre- 

 dict velocities to better than 4% in absolute 

 terms but this is satisfactory for the purpose of 

 deriving a simple means for modifying model measure- 

 ments to represent full-scale values. It is re- 

 quired to obtain a representative flow field over 

 the propeller disc area and an essential starting 

 point is to have reliable model data not only in 

 the sense that velocities can be measured accurately 

 at a given point, but also that, if a Fourier anal- 

 ysis is made of the velocities measured during one 

 revolution at a given radius , then a good approxi- 

 mation to the magnitudes of wake harmonics is 



+ 0+ o 



0-2 



^'l= 0-898 

 PLANE OF FORWARD PROPELLER 



I I I I 



3C/ L = 0-873 



_| I 1_ 



o^**^o^ 



•^'l = 0-936 



I 



(»/.) 



BODY LENGTH 



FIGURE 3. Velocity predictions and measurements on a torpedo-like body. 



