99 



the values of the wall temperatures at S/D = 0.4, 

 0.6, 0.8, 1.0 were of the most interest. The total 

 heat flux was set at 250 watts (about 3W/cm^) at 

 which the schlieren effect was observable. The 

 wall temperatures were then measured at increasing 

 values of velocity. It was found that the maximum 

 wall temperature between S/D =0.4 and 1.0 varied 

 from 3°C to 5°C above the ambient temperature. 

 However, it must be emphasized that these are very- 

 conservative values since the thermocouples are 

 actually somewhat below the surface in a region of 

 a high temperature gradient. When this gradient 

 is accounted for our estimate of the surface excess 

 temperature is from 1-3°C, a smaller but not neg- 

 ligible amount. van der Meulen avoided the tem- 

 perature effect by injecting a two percent salt 

 solution. On the whole this method and the present 

 one agree quite favorably (Figure 22). There is, 

 however, the possibility of instability via a de- 



OD 



3 



g 

 I 



> 04 



< 



IxlO' 2KI0' 3x10' 4x10' 



BODY REYNOLDS NUMBER - UD/I/ 



FIGURE 30. Cavitation inception on the NSRDC body. 



FIGURE 29. Schlieren photographs of the Schiebe body 

 with and without salt water injection. The top photo- 

 graph of each group is without injection; the bottom 

 photograph shows the injection of MgSOij solution having 

 a specific gravity of 1.02. The Reynolds number is 

 1.67 X 2 ' 

 in (c) . 



stabilizing density gradient. This point was ad- 

 dressed experimentally and in Figure 29 matched 

 pairs of schlieren photographs , without and with 

 salt injection are presented. It was found that 

 although the appearance of the transition changed 

 markedly the location of transition did not change 

 significantly. 



6. PRESENTATION OF CAVITATION INCEPTION RESULTS 



Freestream Turbulence Level 



The data on the influence of freestream turbulence 

 level upon cavitation inception is limited because 

 of the low maximum water speed in the LTWT of about 

 8m/s but more importantly because the turbulence 

 generating grids located at the entrance to the 

 test section cavitated themselves before the test 

 models did. Consequently, only the 0.05 and 0.65 

 percent turbulence level configurations could be 

 used. The NSRDC body was the only one to be so in- 

 vestigated. Some of these inception data are sum- 

 marized in Figure 30 where they are compared with 

 Brockett's (1972) data. Inception on the NSIRDC 

 body was always of the band type which occurred 

 suddenly without any precursor bubble type cavita- 

 tion. As can be seen in Figure 30, inception oc- 

 curred at the same value of the inception index for 

 both turbulence levels, but as illustrated in Fig- 

 ure 31 the subsequent developed cavitation was much 

 less steady at the higher turbulence intensity. 



The Effects of Polymer Solutions 



Hemisphere Nose Body 



The type of cavitation and the value of the incep- 

 tion index were found to be strongly dependent on 

 the amount of polymer present in the boundary layer. 

 For a fixed polymer solution concentration and free- 

 stream velocity the following changes in inception 

 were observed to take place: at zero injection 

 rate, incipient band type cavitation as illustrated 

 in Figure 32(a) always occurred. At injection rates 

 less than the critical value (the injection rate 

 at which the separation would disappear) , band type 

 inception still occurred but as can be seen in Fig- 

 ure 32(b) the surface of the developed cavitation 



