VIDEO: Mythbusters test golf ball-like dimpling effect on fuel economy (*Spoiler Alert!*)
Mythbusters test golf ball-like dimpling on MPG – Click above to watch video after the jump
The Mythbusters must be closet car fans, because the hour-long show on the Discovery Channel seems to be producing more and more experiments involving automobiles than ever before. Their latest again involves fuel efficiency, this time testing if a dirty car is more fuel efficient than a clean one because of the golf ball-like dimpling effect of the dirt. Turns out dirt doesn't make a difference, but Adam and Jamie went one step further to test if covering a car in actual golf ball-like dimples would improve its fuel efficiency. According to cable's most crack scientists, yes, it will.
The show's team completely covered a last-gen Ford Taurus with modelers clay and figured out that it would achieve about 26 mpg at a constant 65 mph. They then went about adding over 1,000 dimples to the car's exterior. To keep the experiment consistent, all 1,082 dimples removed from the clay exterior were put in a box and set in the back seat so that the car would weigh exactly the same as before dimpling. The theory is that, like a golf ball, the dimples would reduce the car's drag through the air, thus allowing it to travel the same distance at the same speed using less fuel. The result? Over 29 mpg.
Follow the jump to watch the whole episode for yourself, though if you're only interested in watching the dimpled car do it's thing, skip ahead to about 40 minutes in.
[Source: Megavideo]
Reader Comments (Page 1 of 2)
Jeremy 6:12PM (10/22/2009)
The Lexus LS430 (starting in like 2000 and then on all subsequent LS-series cars, as far as I know) employed a dimpled undertray that covers everything underneath the car. It GREATLY reduced drag, and helped it achieve one of the very lowest drag coefficients in the industry. Dimpling is a proven way to reduce aerodynamic drag. Unfortunately, it would be hard to make a car that was entirely dimpled look like anything anybody would want to drive haha
Dimpling undertrays? Sure.
Reply
DasBoese 9:27PM (10/22/2009)
Funny thing, that. Dimples actually increase drag by forcing the boundary layer to become turbulent. The net decrease in drag under some circumstances is due to the fact that turbulent flow "sticks" to surfaces much better and separates later.
I think it's highly unlikely that the dimples on those undertrays had anything to do with reducing drag, it's just not plausible. Same goes for the upper side of the car, from the perspective of the boundary layer, cars are full of "dimples" anyway: panel gaps, trim pieces etc. all serve the same purpose.
NmGfan 6:36PM (10/22/2009)
So those dimples on a Corbin Sparrow/Myers NmG might actually reduce drag a little? Pretty cool, thanks Mythbusters.
Reply
Sean 6:51PM (10/22/2009)
I would love for a major company to do an in depth study of this (such as size and placement of dimples). According to my limited understanding, the only dimples that actually help are the ones on areas that cause the air to flair out to make a wake wider than the vehicle itself.
http://wings.avkids.com/Book/Sports/Images/golf_01.gif
http://wings.avkids.com/Book/Sports/Images/golf_02.gif
@Jeremy: I'm not saying you're wrong about the dimpled under tray, I just don't understand it
Reply
Jeremy 8:08PM (10/22/2009)
Yeah I don't necessarily understand the exact physics of it either. My understanding is it's a laminar flow / surface tension (if that can be applied to air) sort of a thing and the dimples help the air flowing over the surface do so with less drag. A smooth, glossy surface may actually cause some 'sticking' of the air. I know this is the case with water, where a matte finish is slipperier from a gloss finish (used to wakeboard on a semi-pro level, and boards always had a gloss clear coat on them until companies figured out that a matte finish on the bottom made a noticeable difference in reducing drag. In fact, some wakeboards even had dimples on the bottom, although they claimed this was to "trap air" to make landings softer. Who knows).
Anyway, there ARE high pressure zones and low pressure zones (essentially pulling vacuum) underneath cars. Even cars with flat undertrays. I think dimpling gets rid of this and reduces drag because of it.
Truly, no idea exactly how though. It definitely works, however. Lexus absolutely did do this AND proved positive benefits.
My understanding is that it wouldn't work on an airplane because the velocities are too high. When I first heard about this on the Lexus, I thought "holy crap! they should dimple airplanes and reduce drag on those in a big way!!!" That would be awesome, of course, and totally fine to do since nobody cares if an airplane looks retarded. I'm still not going to drive a dimpled car for 3 mpg. But airlines could save $millions$. Anyway, sounds like the effect is limited to slower speeds.
DasBoese 9:13PM (10/22/2009)
No, you've got it backwards, the only dimples that help are those that -by making the boundary layer turbulent- cause it to stick to the surface longer.
"Flaring out" is undesirable because it essentially creates an area of low pressure at the back, sucking your shape back resulting in increased net drag. If the flow stays attached for longer, that negative pressure is smaller or can even become positive, i.e. "pushing" on your shape, thus reducing drag.
And yes, the effect does depend on speed -as well as geometry- by something called the Reynolds number which, among other things, can tell you wether you're in a laminar or turbulent flow regime.
GoodCheer 12:19PM (10/23/2009)
As dasboese points out, the Reynolds number is used to predict (or explain) the formation of turbulence in moving fluids (air is a compressible fluid for this discussion).
The Reynolds number is
[fluid density] * [speed] * [length] / [viscosity]
Air density: ~1.2 kg/m^3
Highway speed: ~ 30 m/s
Car length: ~ 1 m (use different numbers if you like)
Air visocisty: ~ 1.8 E-5 kg/(m*s)
So 1m back on the hood, the Reynolds number is already 2E6. Turbulent transition is usually about an order of magnitude less than that... so everything back from the hood ornament will be turbulent (at highway speed) anyway.
stevezilla 8:18PM (10/22/2009)
Dimples don't decrease drag, as I always understood; they create high pressure below the ball which pushes it upward. This is similar to a pitcher in baseball throwing a curve ball.
Reply
DasBoese 8:57PM (10/22/2009)
No, that's total bullshit.
What Dimples do is force the airflow over a surface to become turbulent. Normally you'd say that's not desirable, you want laminar flow to have as little drag as possible, however at high speeds that laminar flow detaches early and creates a zone of negative pressure aka suction at the back of the ball, in addition to vortices that make the flight path unstable.
Now, a turbulent flow on the other hand does have more drag, but this also allows it to stay attached to the surface much better so that it separates farther back, leading to a zone of positive pressure that "pushes" the ball forward. The net result is a greatly reduced overall drag.
Note that this penomenon is more pronounced at small scales. For something as large and covered with flow obstacles (panel gaps, trim etc.) as a car, the boundary layer turns tubulent quickly anyway so covering your car (or parts of it, say, your mirrors) in dimples is rather pointless.
Gordio 9:49PM (10/22/2009)
steve zilla, you're thinking of bernolli's principal, where an airplane wing gets lift b/c the speed of the wind is faster above the wing than below the wing.
http://en.wikipedia.org/wiki/Bernoulli%27s_principle
it's not the same as dimples.
Gary Reysa 11:06PM (10/22/2009)
The mileage they got for the dirty car was about 24 mpg -- this increased to 26.4 mpg when they cleaned the car. I was amazed that it made that much difference.
When they added the dimples, which added something like 800 lbs to the cars weight, they gas mileage went up to 29 mpg -- I was amazed again.
Hard to imagine that something this simple could make such a big difference, and that it has not been used. I'd be happy to drive a car with dimples for a 10% gain in fuel economy.
As has been said, dimples transition the boundary layer to turbulent, and this delays separation on the back of the car, reducing pressure drag. On a golf ball, you have to dimple the whole ball to get this effect, but it seems like on a car you could dimple the area just ahead of the point where separation would start without the dimples, and continue them only to the point where separation occurs even with a turbulent boundary layer -- this would probably not be a whole lot of area. On airplanes, small vortex generator fins are often used for the same purpose -- the vortex brings high velocity air down into the boundary layer and delays separation. These are very simple little vanes that stick up from the surface about an inch -- seems like they might work as well.
Gary
Reply
Peter Rockwell 12:56PM (10/23/2009)
Like the little vanes at the rear edge of the roof of the last-gen Evo?
DasBoese 9:57PM (10/22/2009)
Gordio:
That's a common misconception that I fear will never die out. The lift generated by an airplane wing has nothing to do with flow speeds or Bernoulli's principle. Yes, I know that's how it's explained even in otherwise good science textbooks, but it's wrong.
Reality is much simpler: Airfoils generate lift by deflecting the airflow downwards ;)
Reply
Chris M 10:34PM (10/22/2009)
Yes, but the wings use the Bernoulli effect to deflect the airflow above the wing downward, so both Bernoullian and Newtonian principals apply!
A aerofoil that only deflected the air below the wing wouldn't generate as much lift and would have more drag.
DasBoese 11:35PM (10/22/2009)
Yeah, I was a bit careless with the wording. Of course Bernoulli's principle does apply and there are differences in velocity, just wanted to point out that that's not what makes lift.
A (subsonic) airfoil that only deflects air on the lower surface doesn't generate any lift at all, that's what happens when a wing stalls ;)
Nick Martin 12:50PM (10/23/2009)
Hmm. What kind of Aerospace Engineering degree do you have? I have a real one, and I say that Bernoulli's principle has everything to do with generating lift at cruise, which is the majority of the flight. During take off and landing, a major portion of the lift is generaterd by deflecting air; however, this comes at the expense of much increased drag and noise.
Treehugger 12:46AM (10/23/2009)
Well that's an outstanding results indeed, and make us wonder what the car makers are doing... as said above you don't need to dimble the whole car but just the rear part where the air start to detach frpm the car. The effect on the look is quite good, I mean it gives the car a special look, the main problem then is how do you clean the car with a surface like that ?? But that's probably one of the best Mythbuster ever, I am pretty impressed.
Reply
Scatter 9:00AM (10/23/2009)
My understanding of the dimples is that they actually increase drag slightly but they're there to increase lift. So this was a bit of a non-starter i reckon .Qutie cool though.
Reply
Unknown 1:01PM (10/23/2009)
They don't know about rolling resistance apparently.
Reply
Matt Peckham 3:09PM (10/23/2009)
Just think how much fun it would be to wax, let alone wash, what with hundreds of little bowls to hold in the water and sediment.... and how the water would collect after a rainstorm, and the hummingbird nests, and the neighbor's kid filling all of those holes with sand....
But at least you can play mancala anywhere you want!
Reply