Quote:
Originally Posted by Vman455
None of the textbooks on vehicle aerodynamics I have claims this. Quite the opposite, in fact; most of the flow around a car is turbulent (aside from a thin laminar layer next to the body surface), and transitions to a turbulent layer based on Reynolds number, which is proportional to the distance from the leading edge of the body. But this is a good thing, since a turbulent boundary layer will follow curves that a laminar one won't!
(Hummel, Dietrich, "Some Fundamentals of Fluid Mechanics," in Aerodynamics of Road Vehicles, 4th ed., ed. Hucho [Warrendale: SAE International, 1998], 66-68).
(Barnard, R.H., Road Vehicle Aerodynamic Design, 3rd ed., [St. Albans: MechAero, 2009], 9-10).
Further, researchers recognized as far back as the 1970s that there is an additional problem of cars operating in a turbulent atmosphere, i.e. the free stream flow, rather than being laminar as in a wind tunnel, is in fact turbulent. P.W. Bearman outlined the problem and attempted to begin to address it in his paper "Some Effects of Free-Stream Turbulence and the Presence of the Ground on the Flow Around Bluff Bodies" at the 1976 GM conference:
(Bearman, P.W. "Some Effects of Free-Stream Turbulence and the Presence of the Ground on the Flow Around Bluff Bodies," in Aerodynamic Drag Mechanisms of Bluff Bodies and Road Vehicles, ed. Sovran et al, [New York: Plenum Press, 1978], 112).
But it is a problem that still vexes engineers. How do we model dynamic atmospheric turbulence--which by its nature is random--in a systematic manner? Various turbulence models are used in CFD analysis, but these are approximations. I don't know of any wind tunnels that have some device for creating free stream turbulence, but perhaps there are some.
Maintaining laminar flow as far back from the front of the car as possible by smoothly rounding the front end is one strategy to achieve lower drag, where laminar flow can be encouraged (given acceptable environmental conditions) before the inevitable transition to turbulence:
(Barnard, R.H., Road Vehicle Aerodynamic Design, 3rd ed., [St. Albans: MechAero, 2009], 80).
Lastly, it is quite easy to observe on the road whether the flow at any point on a car is laminar or turbulent: simply tape wool tufts onto the body surface and then observe their behavior. If a tuft points perfectly in one direction with no movement, the flow is laminar; in laminar flow, pathline = streamline = streakline, and there is no mixing. If it moves, vibrates or fluctuates at all, the flow is turbulent. In fact, now that I think about it, it may be possible to observe the point of transition at the front of the car on a calm day--but you would need a road completely free of other traffic and a day with absolutely no wind.
|
* any mention of the boundary layer taken within the context of laminar flow, would have to do with going out of bounds on what the turbulent boundary could tolerate as to pressure rise in the flow direction of the aft-body.
* Due to the 'size' of an automobile, once to 20-mph, it's going to be fully immersed in a turbulent boundary layer. And it's Cd constant up to 250-mph.
* it's technically impossible to have a laminar boundary layer for more than an inch or so on a production passenger car, even on a perfectly calm day.
* At the time of Hucho's 2nd-Edition publication in 1987, forebody separation was basically, already a thing of the past. Minimum edge radii were already sufficient to provide attached laminar flow up to the A-pillars and windshield
header.
* mirror turbulence has always been an issue. That's why the DOT has been petitioned to allow carmakers to go to camera systems.
* recessed window glass triggers separation
* wheel openings
* pseudo-Jaray 'fastbacks'
* you'll see that the 2011 Audi A7 Sportback had nearly full rear separation over the car.
* notchbacks
* It was from the A-pillars aft that's the issue, however plenty of cars maintain attached, laminar flow all the way to the rear separation edges.
* Tufts can reveal some aspects of attached flow, however cannot differentiate vortex flow and downwash flow. Typically, it takes smoke flow visualization or tuft screens behind the vehicle to discern that.
--------------------------------------------------------------------------------------
Automotive aerodynamics has to do with reducing or eliminating separation. Separation is caused by the body profile ( hence 'profile drag').
And by definition, if the body profile is not 'streamlined' it will create an unfavorable pressure gradient in the direction of flow which will cause the TBL to detach under the conditions illustrated in boundary layer theory.
Once separated, all kinetic energy that might have been harvested with a 'streamlined' body will be lost to the atmospheric heating of turbulence.
--------------------------------------------------------------------------------------
Any vehicle with a drag coefficient above Cd 0.09 is demonstrating drag due to separation, linked to the turbulence it creates, which cannot recover kinetic energy which IS recovered with a 'streamlined' body.
Energy lost to the turbulent boundary layer is a non-negotiable consequence of air viscosity.
-------------------------------------------------------------------------------------
We have no control over atmospheric turbulence, crosswinds, or gust. The natural 7-mph crosswind yaw effects, statistically experienced by all vehicles, is measured in CFD or wind tunnel for EPA certification purposes.
Turbulence 'screens' can be introduced in a wind tunnel. CFD uses an accepted Kappa- Epsilon (K-E) turbulence model as part of its makeup last I looked.