Can dimpled fuselages reduce drag on aircraft?

[claytonh4]

First off, I'm not an engineer. I'm only in high school so this may sound like a stupid question to some. But, I watched this episode of mythbusters where they tested the gas mileage on a standard car, and one with golf ball-like dimples all over it, and there was a significant difference. Their explanation was that these dimples caused the airflow to create a smaller wake then on a smooth car. So I figured, why not apply this concept to an aircraft? After-all, winglets serve a similar purpose, right? So that the wing tip doesn't create so much turbulence? And I also heard that the paint on an aircraft can make a difference for weight reasons. I read something about a bare metal B747 vs a painted one being like a thousand pounds or so. And if something like paint can make a difference, I figure a dimpled fuselage would make a massive difference, especially with jet fuel prices. This isn't some crazy novel idea, I'm sure it's been considered before, but doesn't work for a reason. Could someone please tell me why that is? At super high speeds would it introduce a source of drag or something? Or would it interfere with lift?
Thanks!

 

[RandomGuy88]

No this would not work on an aircraft and it does not work on cars. The Mythbusters were wrong. They are entertaining but they are not really engineers or scientist.

Anyways to understand why this won't work you must know a little about a boundary layer and the difference between a laminar and turbulent boundary layer. So if you don't know those terms you should look into them. On a ball most of the drag is a result of the boundary layer separating from the surface and the resulting large wake and low pressure region behind the ball. On a golf ball, the dimples make the boundary layer transition turbulence which means it is more energetic. As a result the boundary layer separates further back on the surface and the wake is smaller. Smaller wake = smaller low pressure region and less drag.

On an aircraft most of the drag is from skin friction. The fluid drags on the body due to friction. This is different from the pressure drag on a ball resulting from the low pressure region in the wake. On the aircraft the wake is much thinner relative to the body than for the ball. Plus the boundary layer is probably turbulent already anyways. So putting dimples on the airplane will not have much of an effect on the wake and will not decrease the drag. If the dimples were on the wing they would make things worse actually.

 

[claytonh4]

No this would not work on an aircraft and it does not work on cars. The Mythbusters were wrong. They are entertaining but they are not really engineers or scientist.

Anyways to understand why this won't work you must know a little about a boundary layer and the difference between a laminar and turbulent boundary layer. So if you don't know those terms you should look into them. On a ball most of the drag is a result of the boundary layer separating from the surface and the resulting large wake and low pressure region behind the ball. On a golf ball, the dimples make the boundary layer transition turbulence which means it is more energetic. As a result the boundary layer separates further back on the surface and the wake is smaller. Smaller wake = smaller low pressure region and less drag.

On an aircraft most of the drag is from skin friction. The fluid drags on the body due to friction. This is different from the pressure drag on a ball resulting from the low pressure region in the wake. On the aircraft the wake is much thinner relative to the body than for the ball. Plus the boundary layer is probably turbulent already anyways. So putting dimples on the airplane will not have much of an effect on the wake and will not decrease the drag. If the dimples were on the wing they would make things worse actually.

Oh ok I didn't realize the difference in flow between those kinds of surfaces. Thanks for the response!

 

[xxChrisxx]

See: Bluff Body vs. Streamlined.

Golf ball dimples work on bluff bodies but not on streamlined (ie a plane and wing).

Dimples, or at least the principle of energizing the boundary layer certainly works on cars. On the trailing edge of some cars (certain saloon cars with a rear wing... usually of the Japanese variety) you'll see little shark fin type protrusions. This is designed to prevent flow separation, exactly what dimples do on a golf ball.

The reason you don't see dimpled cars is because:
a: They'd look crap and no one would buy one.
b: why stick dimples in an area that already had adequately energized flow that's attached to the body? All you would do in increase drag.You also occasionally see wings with vortex generators too, usually ones that operate at high angle of attack (near stalling). Also wing end plates are on aircraft to prevent the high pressure air spilling over the side to the low pressure region. Effectively increasing the efficiency of the wing. They also decrease wingtip vortices as you say.

 

[Aero51]

It would probably work at low speeds but as was mentioned earlier dimples are not effective a high Reynolds numbers. The idea is that these dimples force the flow to become turbulent (IE add rotation) so the air can remain attach to the body. For vehicles such as cars/planes, the regime where the flow is laminar (or smooth as in the case when your pour water into a glass slowly) is over a very small fraction of surface because the flow is already rotational. What will improve drag is:
1) Reducing skin friction - this may be possible with new nano materials someday.
2) Steamlining, as was mentioned earlier, involves curving the overall shape of the body such that their air does not separate from the surface. Note that in cars, sharp angles actually improve the aerodynamics.

 

[Enthalpy]

On a golf ball, the dimples make the boundary layer transition turbulence which means it is more energetic. As a result the boundary layer separates further back on the surface and the wake is smaller. Smaller wake = smaller low pressure region and less drag.

Nearly...

The dimples let the transition happen sooner, and this decreases drag. But i the first line, it makes drag more reproducible, which is even more important to golf players.

A large wake isn't necessarily bad. Its pressure can be lass bad than the one of a narrower wake, especially if the transition is clear.

On many planes, drag does result from pression rather than skin friction. That's why such planes (so many planes) have smooth forms behind their main section, instead of reducing their skin area as soon as possible.

 

[Enthalpy]

Dimples on a golf ball serve to promote flow separation. But a well-designed plane fuselage is already better than a sphere improved by dimples.

Cars do promote flow separation at the rear. A first excellent reason is to avoid lift there at high speed, which would decrease the directivity of the rear wheels, making the car unstable. As second reason is to reduce noise, since the location of flow separation is then well defined and doesn't oscillate. The third, less obvious reason is that it reduces drag by providing a well-defined wake whose pressure isn't as negative.

Some cars separate the flow completely at the rear and keep the flow attached up to there using smooth forms; others separate the flow earlier and have a steeper rear.

As for energizing the layer: this works only if you create an exchange with the air outside the layer by provoking vortices with their axis parallel to the flow. Dimples don't do that. Small tilted plates near the leading edge of a wing do. Small protrusions before the leading edge do it only at high angle of attack, taking the best of both worlds.

Corrugated skins have long been studied because people imagined sharks and swordfish swim quickly thanks to their skin. It didn't work well and fish' explanation is nearly abandoned, but there is still research in this topic. An other direction is to have elastic skins (possibly piezoelectric with electric damping, or even active) that hinder the formation of turbulence. Seen no magic result up to now.

 

[boneh3ad]

On many planes, drag does result from pression rather than skin friction. That's why such planes (so many planes) have smooth forms behind their main section, instead of reducing their skin area as soon as possible.

This is wrong. On all planes, drag is a result of a combination of pressure drag and skin friction drag (and at increased Mach numbers, wave drag). Planes have smooth forms behind them to reduce pressure drag but this is completely independent of the attempts to lower skin friction. It is not an either/or thing.

On a typical commercial airliner such as a Boeing 737 or Airbus A320, skin friction still accounts for roughly 50% of the total drag on the aircraft¹. The rest is covered by things such as form drag, induced drag and wave drag. This is, in part, due to the fact that planes are streamlined and so the form drag is greatly reduced. Like I said, though, that is independent of efforts to reduce skin friction drag.

Dimples on a golf ball serve to promote flow separation.

Quite the opposite, as has already been discussed. Perhaps this is a typo?

But a well-designed plane fuselage is already better than a sphere improved by dimples.

It doesn't even need to be specially designed. The most common shape for a fuselage since the beginning of flight, the tapered tube, is just naturally a better shape for this than a sphere. You can improve it further, but it doesn't take a whole lot of special design considerations to get better than a sphere.

The third, less obvious reason is that it reduces drag by providing a well-defined wake whose pressure isn't as negative.

Assuming you are talking about differential pressure, I will let the negative pressure comment slide. However, this example does not reduce drag. In fact, it increases drag compared to letting the boundary layer separate later. The advantage, though, is maintaining a fixed separation point and therefore maintaining predictable controls, particularly in regards to accelerating and braking. If the separation point fluctuated, the car would be much more difficult to control. This was a problem in the Audi TT, which is why they had to add the rear spoiler.

As for energizing the layer: this works only if you create an exchange with the air outside the layer by provoking vortices with their axis parallel to the flow. Dimples don't do that. Small tilted plates near the leading edge of a wing do.

You can energize the boundary layer a number of ways, including streamwise vortices (which come with their own set of problems). You can also do it with dimples or bumps². Large dimples and bumps actually do create a hairpin vortex which can lead directly to boundary layer transition, feed energy into unstable modes that eventually transition or can do limited modification of the laminar boundary layer (which can also lead to transition). Small bumps or dimples can also help force unstable modes and lead to transition. Once the boundary layer transitions, the mixing that is the hallmark of a turbulent boundary layer will bring energy into the boundary layer from the free stream. This is essentially the mechanism of dimples on a golf ball. The promote boundary-layer transition and a turbulent boundary layer is more resistant to separation than a laminar boundary layer.

Corrugated skins have long been studied because people imagined sharks and swordfish swim quickly thanks to their skin. It didn't work well and fish' explanation is nearly abandoned, but there is still research in this topic.

Corrugated skins have some really interesting properties. They are very good at preventing the spread of turbulent spots once they form, so they can successfully help delay transition at least globally over the surface and reduce the drag increase that comes with a turbulent boundary layer. The trade-off is that they have a lot more surface area and therefore have a lot higher base level of viscous drag. Like you mentioned, I have yet to see anyone get a net gain out of this.

An other direction is to have elastic skins (possibly piezoelectric with electric damping, or even active) that hinder the formation of turbulence. Seen no magic result up to now.

This is a very bad idea. There has been a pretty large body of work on active cancellation of unstable waves in the boundary layer, particularly Tollmien-Schlichting waves, and it is really a fool's errand. It is a neat academic study and can be done, but it will almost certainly never be practical. It is much easier and more cost effective to design the surface to be subcritical to T-S waves through the use of natural laminar flow techniques. That just isn't as sexy.

References:
[1] Robert JP. 1992. Drag reduction, an industrial challenge, AGARD Report 786.
[2] Arcarlar MS, Smith CR. 1987. A study of hairpin vortices in a laminar boundary layer. Part 1. Hairpin vortices generated by a hemisphere protuberence. J. Fluid Mech.. 175:1-41.