Sevik 



periodic effect. However it is possible that there is a relation between the two 

 that shows up in the present results, in particular which is concerned with the 

 discrepancy in the figures exemplified by the hump in Fig. 12. Simple calcula- 

 tions indicate that the propeller rotational speed in that case was 19 rps, and the 

 blade rate associated with this case for a ten-bladed propeller would then cor- 

 respond to 190 Hz. The appearance of the hump in the spectral density presented 

 in Fig. 12 could possibly be due to the existence of a periodic blade-rate signal, 

 which would be obtained via a wave analyzer with the bandwidth characteristics of 

 the equipment used in this study. It is thus possible that the blade-rate signal 

 occurs, since the fluctuations in the propeller thrust force should be due to ve- 

 locity fluctuations about a steady value, corresponding to the inflow speed, which 

 should be constant circumferentially at any radius. Even though the grid is as- 

 sumed to be symmetric, and also the tunnel as well, it would be best to survey 

 the flow field that is entering the propeller disk to be certain that no harmonics 

 of blade rate are present in the oncoming flow. A detailed consideration of this 

 point is essential if any random process is analyzed, since the analysis must al- 

 ways refer to a base reference, and the characteristics of that reference should 

 be known and used in analyzing the characteristics of any random response. It 

 is suggested that careful consideration of these points may eliminate spurious 

 results in the future. 



REPLY TO DISCUSSION 



Maurice Sevik 



The hump in Fig. 12 was a matter of concern, and the reason for its pres- 

 ence was investigated after completion of the experiments. It was felt that vi- 

 brations of the propeller blades might be the cause. The first two natural fre- 

 quencies of the propeller blades were established by calculation and verified 

 experimentally. The fundamental frequency occurred at 127 Hz; the associated 

 mode shape is shown in Fig. Dl. The second mode occurred at 231 cps and its 

 modal shape was essentially torsional. The hump in Fig. 12 is located at about 

 220 Hz, and it appears that it is due to the second mode of vibration of the indi- 

 vidual iDlades, whose resonances are spread in the vicinity of this frequency. 



However, I agree that Dr. Kaplan's explanation is also a reasonable one. 

 It is unfortunate that the blade- rate frequency and the natural frequency men- 

 tioned above fall so close together that a separation of the two effects is not 

 possible. 



312 



