This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
Air streaming over a wing, oil being pushed through a long pipeline, water flowing over a mega-tanker hull and winds blowing over the earth’s surface: In each case, highly turbulent flow is created next to a surface by the fluid rubbing against it. Wall turbulence greatly increases frictional drag on surfaces, and the fuel spent to overcome it costs commercial aviation, America’s one-quarter million-mile long system of pipelines and maritime shipping and transportation hundreds of billions of dollars, year after year. Flow science and engineering have long sought techniques to reduce turbulent drag on walls, either to save some fraction of this huge expenditure or to increase the speed of transport. For example, heating and pumping oil through the Trans-Alaska pipeline initially consumed about as much fuel as the oil being transported. By adding minute concentrations of long-chained polymers to the oil, a highly successful drag reduction technique discovered accidentally by Toms, the rate of transport was increased approximately 50%.
Several other methods of drag reduction (DR) have been developed since Toms’ discovery - passive patterns on the surface, active motions of the surface, air-bubble injection, super-hydrophobic surfaces and others. Because wall-turbulence is complex, multi-scale, and not well-understood, and because DR can be achieved by many seemingly disparate techniques, there are few established principles of drag reduction. Consequently, new techniques are motivated by rudimentary principles of wall turbulence followed by cut-and-try exploration and optimization. Advancing our understanding of how wall-turbulence is created, sustained and reduced is necessary to establish new DR concepts that can motivate and guide discovery, development and optimization of more effective drag reduction techniques.
The principle outcomes of this project are:
It was found that the origin of non-drag-reduced turbulent flow in a pipe is the triggering of individual hairpin vortices near the wall by inlet disturbances. A single hairpin develops into a packet of hairpins that move together, and the packet creates multiple new packets ahead of it that multiply and grow into a pattern known as a turbulent spot. This finding suggests that the creation of turbulence depends on near-wall disturbances strong enough to form hairpin packets.
By comparing two very different types of DR flow, it has been shown that both reduce drag by inhibiting the formation of hairpin vortex packets. Inhibition is caused by weakening the vortices close to the wall relative to the mean stream. This cuts off the process that auto-generates new hairpins in a packet, and the number of hairpins is significantly diminished. With fewer hairpins, turbulence throughout the flow is weakened, and the drag is reduced. Thus, a common mechanism between these two flows is the inhibition of auto-generation. If further studies of other DR flows show it to be a common mechanism in their operation, as well, the inhibition of packets could become a new principle for optimization of known DR surfaces and invention of new DR surfaces.
The most interesting outcome came from a study of turbulent thermal convection in a horizontal fluid layer lying between a warm lower-plate and a cool upper-plate. In this flow, turbulence is created and maintained by warm fluid rising and cool fluid falling, a situation very different from wall turbulence. It was found that, despite their differences, the key structures occurring in each flow and the sequence of events responsible for their formation were surprisingly analogous. Hairpins from the wall correspond to mushroom-shaped thermals rising(falling) from the warm(cool) plates; packets of hairpins correspond to packets of thermals that are swept along the plates by horizontal flows; Sheets and plumes of warm(cool) fluid consisting of aggregated packets of thermals correspond to sheets of aggregated hairpin packets moving towards and away from the wall. The impact of a wall-ward-moving sheet on the wall causes it to turn and flow horizontally along the wall, until it collides with another horizontal flow from a different sheet. The impact from the collision causes the flows to turn downwards in a sheet until it impacts the lower wall and turns to the horizontal again, thereby creating the wall flow that swept the thermal packets towards the rising sheet at the beginning of this cycle. The result is a closed cell that is driven by the sheets which are, in turn, formed by the aggregation of small wall structures caused by the horizontal wall flows of the cells.
Thus, in both wall turbulence and turbulent free convection, there is a direct coupling between the small-scales near the wall and the large-scale circulations, each sustaining the other in a tightly coupled manner. The concept of energy flowing from small to large-scales appears to contradict the famous eddy cascade model of Kolmogorov, but that model pertains only to hom*ogeneous turbulence, whereas turbulence with boundaries is necessarily inhom*ogeneous.
Last Modified: 09/18/2017
Modified by: RonaldAdrian
