440 



FIGURE 13. Photograph showing transition (T) from laminar to turbulent boundary layer on blunt nose 

 (s^/D = 1.68). The flow is from left to right, V = 8 m/s . 



from 0.333 to 1.0. The calculations showed that 

 none of the blunt noses were subjected to laminar 

 separation. The present observations are in 

 agreement with these theoretical predictions. 



Non-Newtonian Flow 



The influence of polymer additives on the boundary 

 layer flow about the models was investigated by 

 injecting a 500 ppm (parts per million by weight) 

 Polyox WSR-301 solution from the nose of the models. 

 To visualize the flow, the injection fluid contained 

 2 percent sodium chloride. For the SST hemispher- 

 ical nose , the holograms showed that laminar flow 

 separation was no longer present. An example is 

 given by the photograph presented in Figure 15. At 

 or shortly downstream from the location where 

 Newtonian flow separation occurred, transition from 

 laminar to turbulent boundary layer flow is observed. 

 From the holograms made in the velocity range 4 to 

 20 m/s, it could be derived that transition to 

 turbulence occurred close to the location of 

 Newtonian flow separation. It was difficult, however, 

 to indicate the precise location of transition. 



Another important observation was that the sodium 

 chloride was not completely mixed in the turbulent 

 region, but was still able to show the existence of 

 waves and streaks further downstream, till the end 

 of the hologram. An example of this phenomenon 

 has been given by Van der Meulen (1976b) . For the 

 Teflon hemispherical nose it was found that the 

 influence of polymer additives on laminar flow 

 separation was the same as for the SST hemispher- 

 ical nose. Although the observations made with the 

 bltmt nose were somewhat obscured by the irregular 

 outflow from the nose of the model, the main con- 

 clusion to be derived from the holograms is that 

 the polymer causes early transition to turbulence. 

 The approximate locations of transition are plotted 

 in Figure 14. 



The polymer concentration used during the above 

 observations is rather high when compared to the 

 most effective concentration for turbulent- flow 

 friction reduction. From Figure 16, where the 

 friction factor, f, for flow through a circular 

 tube is given as a function of the Reynolds number, 

 it can be derived that a Polyox WSR-301 concentration 

 of about 20 ppm gives a maximum friction reduction. 

 Additional holograms for the SST hemispherical nose 



FIGURE 14. Strearawise distance to boundary 

 layer transition over diameter, s„/D, as a 

 function of Reynolds number for blunt nose. 



ni E 



0.5 0.6 0.7 0.8 10 



Reynolds Number x 10"^ 



