Mythbusters vs. Common Sense part 2371213

October 25, 2009 5:10 PM

Why aren't cars dimpled like golf balls if it increases fuel efficiency?

In the most recent episode of Mythbusters, in testing the "a dirty car gets better mileage" myth, they seem to have proven that a car which has been dimpled like a golf ball gets better mileage than a car of the same weight and general shape without dimples. I'm aware of why other aerodynamic surfaces like airplane wings aren't dimpled, but with this empirical evidence of the benefits of dimpling (and the results of other similar experiments which have surely been done at some point), why aren't the bodies of cars dimpled? Is there a scientific explanation for the lack of dimpling on modern cars or one rooted in economics (time and manpower required) or aesthetics (ugly car)?

posted by tehloki to technology (28 comments total)

People don't buy cars based on scientific fact.
posted by cellphone at 5:13 PM on October 25, 2009 [6 favorites]

other similar experiments which have surely been done at some point

What makes you so sure it has been done before?

Calling asavage....
posted by jozxyqk at 5:16 PM on October 25, 2009

Someone else answered this recently in a very thorough way. I can't remember all the details, but it basically boils down to the old equation How Much Gain vs How Much Effort.

Since the dimpling really only manifests a benefit under pretty specific circumstances (highway speeds, no other cars around messing up the airflow, etc.), and the pain in the ass factor of dimpling your car (manufacturing issues, bodywork & maintenance issues, etc.) is quite high, I think it's pretty obvious that dimpling doesn't make sense for the large majority of car owners.

However, Messrs. A. & J. have quite probably lit quite a few large bonfires under some NASCAR asses, I'm guessing.
posted by Aquaman at 5:24 PM on October 25, 2009

OK, here's the link to the Reddit poster who was specific in the details. Sorry about the Reddit link. Please don't kill me.
posted by Aquaman at 5:27 PM on October 25, 2009

I'm not sure this is an issue for Formula 1 cars, as they need to maintain a ludicrous amount of downforce to stay on the road. I'm pretty sure whichever scientists that are responsible for the body design of ultra-high-performance cars know what they're doing; I'm more concerned with the impact of this principle on consumer automobiles and fuel efficiency.
posted by tehloki at 5:28 PM on October 25, 2009

It would make it damned near impossible to do body repair on the car without outright replacing the entire damaged body section.

It would be a lot more difficult to manufacture. Producing the sheet metal would only be mildly more difficult, but painting it would be very tough. Paint would accumulate in the dimples in irregular ways and it would look crappy.

And, quite frankly, it would look really stupid and few would want to buy such a car.
posted by Chocolate Pickle at 5:29 PM on October 25, 2009

The guy in the reddit link uses his expert knowledge to explain why planes aren't dimpled, but he's just making educated guesses when it comes to the first three reasons regarding why cars aren't. I'm still not convinced.
posted by tehloki at 5:32 PM on October 25, 2009

Dimpling leads to boundary turbulence, which reduces wake drag by delaying flow separation from the body, which in the case of small objects like a golf ball more than makes up for the increased skin friction drag from the rough surface. But for large objects like a car the airflow is going to be turbulent anyway, so there would still be losses but no benefits.

That reddit comment was okay, but there was a much better link in the thread:

Golf Ball Dimples and Drag
posted by roystgnr at 5:44 PM on October 25, 2009 [2 favorites]

Oh, almost forgot IAABSME, but IANYBSME. Always wanted to say that.
posted by roystgnr at 5:45 PM on October 25, 2009 [1 favorite]

FYI Zipp manufactures dimpled bicycle wheels.
posted by spikeleemajortomdickandharryconnickjrmints at 5:54 PM on October 25, 2009

roystgnr, to me it looks like you didn't click the links in my more inside, from your reposting one link in your comment and making statements in it which are disproven in the other link.
posted by tehloki at 5:56 PM on October 25, 2009

Believe me, if Mercedes-Benz or the boys in Maranello thought cars with dimpled sheet metal (or carbon fibre) would sell, everyone would be making them and driving them by now.
posted by Zambrano at 6:16 PM on October 25, 2009

I'm pretty sure whichever scientists that are responsible for the body design of ultra-high-performance cars know what they're doing; I'm more concerned with the impact of this principle on consumer automobiles and fuel efficiency.

I'm pretty sure it'd matter a lot in NASCAR, and they don't do it. High speed ovals put a LOT of emphasis on going very fast with as little drag as possible, and the fully enclosed body, as opposed to open wheeled racers, gives them a lot of opportunities for smoothing airflow along the sides of the car. At 200+ mph, you're burning a crap load of fuel, and anything you can do to reduce consumption will save you time in the pits.

Maybe they haven't thought of it, or maybe it's against some obscure rule, but I'd think they'd be the first to try it. On your car, it might cost $1000 to get .1 mpg this way. That's ludicrous for you or I to spend, but is nothing to people who stand to make millions on a tenth of a second faster lap times.
posted by paanta at 6:17 PM on October 25, 2009

Also, at the risk of a backlash from the asavage-loving fraternity here, and without putting too fine a point on it:

Mythbusters ain't science. Nowhere near it.

"and the results of other similar experiments which have surely been done at some point"

It's so obvious that you might think so - but then, nobody had undertaken a study on the causes of belly-button lint until Dr. Kruszelnicki.
posted by Pinback at 6:22 PM on October 25, 2009 [1 favorite]

I always thought it worked because the golf ball is spinning.....not something you want your car doing:-)
posted by patnok at 6:23 PM on October 25, 2009

The Corbin Sparrow had dimples.
posted by gyusan at 6:43 PM on October 25, 2009

The reason I can come up with is that all those dimples would increase the weight of the car, canceling any gas mileage gain.

The surface area of a sphere is given by 4*pi*r^2. Using 1 cm as r, dividing by 2 (since it's actually the surface of a hemisphere we are concerned with), gives approximately 6 as the SA. You also have to add in the extra surface area between the dimples (4 cm^2 - 3 cm^2 = 1 cm^2) for a total of 7. So for every 4 cm ^2 in surface area, you need 7 cm ^2. To make the numbers easier to work with, lets just say that its approximately double the surface area. This essentially doubles the body weight of the panels. Body panels on a car weigh approximately 200 kg, so about 400 lbs. So you're adding 400 lbs to the car. According to http://fueleconomy.gov/feg/driveHabits.shtml, every 100 lbs added decreases fuel efficiency by 2%. So while you might gain 11% over a car with the same weight, you're really only gaining 3% compared to a car without dimples. The cost of all that extra metal would also have to be factored into the car price as well. Also, all those dimples would really decrease the rigidity of the body panels, so you'd have to increase the thickness (and therefore weight) even more.
posted by wayofthedodo at 7:30 PM on October 25, 2009

all those dimples would really decrease the rigidity of the body panels

The opposite is true, if done properly. Impressions in a flat sheet of steel make it stiffer, like corrugations in cardboard.
posted by yesster at 8:01 PM on October 25, 2009

Car design is not driven by fuel efficiency, it's driven by aesthetics and manufacturing requirements.
posted by Dasein at 8:18 PM on October 25, 2009

The cost of all that extra metal would also have to be factored into the car price as well. Also, all those dimples would really decrease the rigidity of the body panels, so you'd have to increase the thickness (and therefore weight) even more.

If only someone would make a car with body panels made from plastic...

Oh, right. Way to go, GM.
posted by bhayes82 at 9:03 PM on October 25, 2009

Honestly, there are a lot of much easier improvements that the manufacturers don't do. The first generation Honda Insight is a great example: it has rear wheel skirts and a very small grill opening. Very minor, cheap changes that give 5% or better MPG. But people hate the look of wheel skirts, so the 2nd generation Insight dropped them. Dimples all over the car? Nobody would buy it.
posted by smackfu at 10:06 PM on October 25, 2009

Roystgnr has it right. Dimpling reduces wake drag, but increases skin friction drag. This results in a net reduction of drag in certain cases, but an increase in drag in other cases.

like Roystgnr, IANYBSME, but soon.
posted by JauntyFedora at 12:41 AM on October 26, 2009

Mythbusters ain't science. Nowhere near it.

Bit of a derail, but this is not the case. Mythbusters is mostly entertainment of course, and the dudes that make it have never denied that. However; it remains the case that the basic story for each myth goes "hypothesise", "design experiment", "run experiment". That is science.
It is not well-conducted science because:
1) They experiments aren't usually that well designed (as experiments, as entertainment the design is often superb)
2) There isn't as much use of prior work, mathematical modeling, etc. to design the experiments as there is in real science.
3) Real experiments are often refined and rebuilt several times

Fine, they don't do those things because they have a tight budget/schedule and because it wouldn't make as compelling viewing as the explosions.
But all the same, they're following an empirical method of hypothesis testing.

Personally I don't much care for it, but Mythbusters is the most scientific show on the Discovery Channel. I don't think there's a single other regular show on Discovery that actually shows the method of science in operation. The fact that the experiments aren't perfect is an absurd nitpick and to say that it is "Nowhere near it" is mind boggling.
posted by atrazine at 3:06 AM on October 26, 2009

Can you imagine caring for a car that's dimpled? A car that gets noticeably heavier every time it rains or snow, one where you have to get the ice out every time it freezes.
posted by jedrek at 3:32 AM on October 26, 2009 [1 favorite]

Guess what! Some cars are dimpled like golf balls. However, that aerodynamic modification takes place on the various plastic panels on the underside of the car. The belly of a car is the source of lots of aero drag and manufacturers have been paying increasing attention to this by attempting to smooth it out with plastic panels or covers. The ones I've seen have a very shallow golf ball texture so presumably they are finding it makes at least some difference.
posted by Jon-o at 3:42 AM on October 26, 2009

One obvious reason, assuming the premise is even correct, is a dimpled car would be essentially impossible to paint conventionally. Which would mean huge expenditures developing new paint finishes. Something that even now isn't 100% though much better than 20 years ago.

Though it would be cool (in an oh the horror kind of way) to see cars roll off the factory line with intentional orange peel finishes.

Aquaman writes "However, Messrs. A. & J. have quite probably lit quite a few large bonfires under some NASCAR asses, I'm guessing."

NASCAR works to a formula to contain costs and safe speeds. Super speed way cars are all ready as fast as they are going to get, spending money dimpling your car,if it isn't banned already, is going to result in possibly slightly faster times for one race immediately followed by, if it doesn't work: banning and if it does work: banning or some other change in the formula to slow your car down.

atrazine writes "Personally I don't much care for it, but Mythbusters is the most scientific show on the Discovery Channel. "

That’s like saying the Corleones are the friendliest Mafia family. And I'm of the opinion that the sloppy at best science that Mythbusters does is actually harmful. How many times have you seen "Mythbusters proved yaddy yaddy yaddy" pulled out on discussion groups as if it actually proves anything. One particularly egregious example is when they were trying to light a rich fuel air mixture and when it wouldn't go boom they poured more fuel in. Or the surfing on the explosive wave episode where they completely ignored the shape of the liquid's container.
posted by Mitheral at 6:43 AM on October 26, 2009

Nthing that Mythbusters isn't science. I wasn't able to watch the video, but from reading forum posts about it, they ran the car only over a one mile stretch, possibly only once per test (though someone said 5x per test), and got 26mpg for no clay, 26mpg for clay and 29 for dimpled.

Over such a small duration and small number of tests, any possible number of variations could have crept into the test; heat (hotter engine, hotter tires), section of the road driven over, slight differences in speed, wind. Hell, they put the clay from the dimples into the back seat to keep the weight the same, but perhaps that changes the weight distribution.

In addition, I'm not hearing anyone throw out any numbers regarding the standard deviation of the tests. What if they ran it 5x in each configuration and got 26 +/- 10 mpg for clay and 29 +/- 10 mpg for dimpled. That wouldn't be a significant result at all. And if they did run it only once in each configuration, then forget about it, it can't tell us anything for sure.

I'm not saying that the effect isn't impossible, but I would use Mythbuster's result as a justification for a more rigorous study to be performed, not as a proof in and of itself.
posted by cali59 at 6:04 AM on October 27, 2009

They did it five times.
posted by smackfu at 7:19 AM on October 27, 2009