Seyer and Metzner 



spread appears continuous owing to the many small-scale high-frequency oscil- 

 lations which do not appear to be present in the viscoelastic systems. In agree- 

 ment with the above are some hot-wire data reported by Merrill (36), although, 

 as will be shown in the next section, it is not clear that these measurements are 

 correct quantitatively. 



EXPERIMENTAL TECHNIQUES 



This section considers briefly the major tools or methods available for 

 studying the flow fields of interest and in so doing points to some of the signifi- 

 cant results, limitations of techniques, and areas in need of further work. 



Heated Probes 



Limitations, owing to viscoelastic properties, on the use of hot probes have 

 been recently discussed (38) and will not be considered in detail. Of interest is 

 the fact that for many cases of practical interest the flow near the probe tip is 

 characterized by large values of the Deborah number (Eq. (5)), which is pre- 

 dicted to result in "thick" boundary layers or even stagnant regions of fluid 

 near the leading edge of the probe. Practically, these results suggest that the 

 frequency response of the probe will be significantly reduced and that the probe 

 may become insensitive to velocity at sufficiently high values of the Deborah 

 number. These predictions find some support in the experimental results of 

 Acosta and James (1) and of Leathrum (33); as the cutoff velocities are fre- 

 quently of the order of 1 fps or even less, the restriction is a very serious one. 



It is not clear at this point how the above problems can be alleviated, as 

 reductions in velocity or increases in probe size in order to circumvent the 

 Deborah number limitation increase the uncertainty of the measurements due to 

 natural convection effects, and take the measurements out of the ranges of pri- 

 mary interest. 



Impact Probes 



In the case of impact probes several effects arise which may, in principle, 

 serve to alter or limit the utility of the probe. First, the boundary -layer con- 

 siderations noted in the previous section suggest that here too the response of 

 the probe may be significantly reduced, and more importantly Metzner and 

 Astarita (38) have shown that a boundary layer of the same order of thickness 

 as the probe size may exist at the leading edge of the probe. This problem is 

 particularly acute in the case of small probes and imposes a severe restriction 

 on use of the probes for measurements near the wall. 



Secondly, in laminar flows the impact reading is directly dependent on the 

 magnitude of the deviatoric normal stress terms, as well as on the fluid mo- 

 mentum (4,49). Consequently, unless the magnitudes of the deviatoric stresses 

 are negligible or known beforehand one cannot obtain velocity information. 

 Similar problems arise under turbulent conditions with an additional contribution 



30 



