Fabula 



Perhaps, therefore, small-eddy suppression is essentially irrelevant to 

 turbulent drag reduction but is due rather to Dr. Fabula's "raggedness," which 

 might be absent in very weak concentrations. 



DISCUSSION 



Richard L. Hummel 



University of Toronto 



Toronto, Canada 



This work was well conceived and carefully conducted, and the "signal rag- 

 gedness" which seems to have frustrated the original aims of determing a fre- 

 quency distribution curve for comparison with that predicted theoretically by 

 certain models may suggest important aspects in the behaviour of these solutions 

 which had not been sufficiently expected. Ideally, possibly one -half of our ex- 

 perimental work should come out in the form we are prepared for, or possibly 

 expect, so that it may be tied up in a neat package for the scientific community, 

 and the remainder, which at first does not seem as satisfactory, may eventually 

 lead us to new experiments and insights. 



The author's suggestion that clusters of macromolecules may exist and con- 

 tribute substantially to the raggedness is certainly justified from the high at- 

 traction of water molecules for each other relative to their interaction with 

 ether linkages as would be present in the Polyox WSR-301. Thus the water 

 molecules would squeeze the polymer molecules together into clusters. It is 

 well known that much smaller molecules, such as soaps and detergents will 

 cluster under similar conditions into relatively large micelles. The ratio of 

 polar bonds to hydrocarbon segment is high and, therefore, more favourable for 

 these macromolecules than for most soaps and detergents, and the larger mole- 

 cule considerably reduces the opposing entropy effect of molecular association. 

 Thus, larger molecules are statistically more likely to be associated with each 

 other. 



Whether or not raggedness is occasioned by the sensor meeting sufficially 

 large agglomerates or not, another effect should also be considered. We have 

 conducted tests with a fluidized bed where a Newtonian liquid flows vertically 

 upward, suspending to some degree of porosity hard spheres of uniform diam- 

 eter, and we have optically chosen planes of these spheres for study. =■= Far from 

 obtaining the random movement postulated for spheres not in contact, we have 

 found that to a large extent substantial numbers of these spheres move in con- 

 sort. For example, these spheres tend to move upward in planes and downward 



*E,J. Ryan, J.W. Smith, and R.L. Hummel, "Dynamically Preferred Particle 

 Orientations in Liquid Fluidized Beds," submitted to International Symposium 

 on Fluidization, Eindhoven, 1967. 



72 



