Fabula 



to the additives were often 25%, nonnegligible spectral shifts just due to 

 Newtonian -fluid effects are to be expected. 



In order to test the similarity law, spectral measurements were made in 

 water at temperatures of 20.7°C and 11.7°C, giving a viscosity increase of 30%. 

 The results for (p(k)/(ev^)^^'* versus k/k^ are shown in the log-log plot of Fig. 

 18 for x/M - 10.7. The magnitude of the maximum separation between the two 

 curves is about 0.1 in terms of log [0(k)/(ev^)''''] , or in other words 1 db; 

 thus the expected similarity is found to within the typical measurement scatter. 

 Similar agreement was found for x/M =12.6 and 20.2. 



The magnitudes of the spectral shifts are not evident in Fig. 18 but are seen 

 in Fig. 19 where the spectral level ratio (cold water compared with warm water) 

 is shown for x/M = 10.7. Also included is the approximate prediction of Eq. (5) 

 (dashed curve), based on values of m determined graphically from Fig. 18, De- 

 spite the approximation made in deriving Eq. (5) it can be seen that Fig. 19 is 

 quite consistent with Fig. 18, so that the approximate relation can be used in 

 considering viscosity increase in the polymer solutions. 



Spectral Shifts in Polymer Solutions 



Some general characteristics of the spectral measurements in Polyox solu- 

 tions can be seen in the data already presented. For example in the P301-3 

 tests of Figs. 13 to 16 the distortion of the spectra due to signal raggedness de- 

 creased with solution age until finally any remaining signal raggedness effect 



10^ 



1 — I — I I I 



"i I 



-e-ioH 



10-2 



lO"' 



n 1 1 — I — 1 — I I r 



SOLID CURVE T= 20 7 °C 

 DASHED CURVE T= II 7 "C 



I 



_l I 1 I I I 1 I 



_I I I I I I I I 



Fig. 18 - Comparison of normalized spectra in water 

 for x/M = 10.7 for a 30% viscosity difference 



62 



