Search

The independent analysis of flow measurements is frequently hampered by incomplete characterisation of the working fluid. This problem is particularly acute in situations which require working fluids with identical properties, such as the development of scaling laws for the turbulent flow of drag-reducing liquids. In this paper, we demonstrate that the viscometric viscosity, loss and storage moduli for two of the most common polymers used for flow experiments, carboxymethylcellulose (CMC) and xanthan gum (XG), are practically insensitive to the chemistry of the tap water used as a solvent, to the method of mixing, and to the biocide added. However, the properties of CMC from two different manufacturers were found to be significantly different, whereas there was no difference between XG solutions prepared from different batches from the same manufacturer. Our conclusion is that for a given concentration in water, the properties of certain non-Newtonian liquids, such as CMC and XG, are essentially fixed and reproducible. Although the situation is less than ideal, comparisons of fluid-flow data from entirely independent laboratories can thus be made even in the absence of direct rheological measurements.

The paper concerns a computational and experimental study of fully developed laminar flow of a Newtonian liquid through an eccentric annulus with combined bulk axial flow and inner cylinder rotation. The results are reported for calculations of the flowfield, wall shear stress distribution and friction factor for a range of values of eccentricity e, radius ratio κ and Taylor number Ta. For fully developed flow the radial/tangential motion is decoupled from the axial component of velocity. However, the axial component of velocity is directly affected by the radial/tangential velocity field and rotation of the inner cylinder is found to have a strong influence on the axial velocity distribution, ultimately leading to two maxima in the case of a highly eccentred inner cylinder at high rotation speeds, a feature not reported hitherto. This influence of rotation on the axial velocity is mirrored in the behaviour of the shear stresses on the inner and outer cylinder walls and hence on the friction factor. An unexpected result is that (at fixed Reynolds number) as the Taylor number is increased the friction factor for high values of e(>0.9) increases rather than decreases.

Measurements of the radial distributions of the axial and tangential components of velocity and rms velocity fluctuations are presented together with friction factor versus Reynolds number data for two liquids, one Newtonian, the other a shear-thinning polymer, in laminar, transitional, and turbulent flow in an annular geometry with a rotating centerbody of radius ratio 0.506. In all flow regimes, the friction factor is increased by centerbody rotation. However, the influence is slight and most apparent for laminar flow of the Newtonian fluid. Laser Doppier anenometry (LDA) measurements of the tangential velocity reveal three distinct regions across the radial gap with a central region of almost constant angular momentum, which diminishes in magnitude as the Reynolds number increases. Axial velocity measurements show only slight deviations from what would be expected for the case without centerbody rotation. In turbulent flow, the axial velocity fluctuations decrease progressively with increasing Reynolds number for all fluids. For the polymeric liquid, the tangential velocity fluctuations are somewhat suppressed, especially at high Reynolds numbers where the influence of centerbody rotation is reduced. Over a limited range of (low) Reynolds numbers and rotation speeds, the generation and advection of Taylor vortices produces complex flow patterns. Limited measurements are reported for the vortex advection speed.