Most cars are not panels of sheet metal on a frame. For the most part, the sheet metal **is **the frame in today’s unibody or monocoque build techniques.
Well, one difference between the Mythbuster’s car and those tow is that they used much bigger dimples. I’m not sure of the exact size but they looked to be 3-4 inches wide and maybe 1/2 inch deep.
The purpose of dimpling would be to trip the boundary layer, delaying its separation from the vehicle’s skin and allowing it to follow further around the backside curvature before peeling away. This decreases the vacuum (or increases the pressure, depending on how you care to express the phenomenon) on the rearward-facing surfaces of the object.
This is critically important for something like a sphere, where (if smooth) the boundary layer would separate shortly after passing the fattest part of the sphere; tripping that boundary layer allows it to stick to the surface farther past that point.
Note that the dimpling increases skin drag (due to localized turbulence, but decreases form drag (due to increased backside pressure) by a greater amount.
Maybe now we can see why this won’t help cars or airplanes:
-
cars, for the most part, have a pretty flat backside with a relatively sharp transition between “side” and “rear.” Dimpling won’t help the airflow get very far around that corner at all, so it’s just not worth it. Never mind that the slipstream of a car isn’t very clean at all to begin with: things get pretty messy due to exposed wheels/fenders, window trim, windshield wipers, panel seams, and so on. It doesn’t matter if you trip the boundary layer into turbulence with dimples because the whole slipstream is already ridiculously turbulent.
-
Airplanes already have an extremely aerodynamic shape. Most of them taper to a point at the back end, and so there’s little opportunity for the slipstream to separate and form a recirculation zone behind the aircraft. To add dimples to the skin of an aircraft would be to address a problem that does not exist.
But is it still correct to say the panels are providing strength? Seems a technicality that they just happen to have combined the two.
If I took a metal cutter and cut out all the metal panels on a car (leaving the frame intact) I think the car would hold together just fine and not collapse from the loss of that metal.
Cars are sold in a certain way that appeals to the emotional side of the buyer. There are many brands that specialize in specific demographics. Most likely a lot of tricks like this are dismissed early on as unsellable because they are unstylish or clash with the brand philosophy.
Why not go whole hog? How about go-kart-like wheels for less spinning mass? Or a 3-seater? Or an engine that can barely do 55 mph?
This is a marketing problem not an engineering one.
They did their test with a Ford Taurus that already looked like a jellybean. That may have something to do with the results they observed, and it would not have done as much for your average SUV or other more boxy vehicle.
The tests in the Popular Mechanics article varied speeds and switched drivers over a long run to be more “real world” rather than purposely testing carefully measured amounts of fuel at a constant speed like Mythbusters did. That could make a huge difference.
That had occurred to me as well.
During the show, we were discussing whether they should use the giant dimples (IIRC, they made them proportionally the same size relative to the overall size as they are on the golfball) or use actual golf-ball-dimple size, or maybe combine both.
I wondered about this. In the scale-model wind tunnel test they did, it looked to me like the dimpled car fared worse than the original. They said it was slightly better, hence the need to upscale, but I thought they were just looking for an excuse to go fullsize.
Can’t put easily legible ads on a dimpled surface.
Another point, I haven’t seen this mentioned.
Golf balls are dimpled in different patterns for different purposes. I don’t know that they have the engineering to the point where they can predict what pattern will produce what purpose or that they put a bunch of different patterns on balls and identified their in flight behavior patterns.
But the point is, maybe the mythbusters hit on a good pattern and Pop. Mech didn’t.
Another point:
Since when is Mythbusters considered science rather than “fun science-like entertainment”?
You mean like the B-2 and F22 at the National Air Force Museum. 
I could see dimpling used to change the lifting characteristics of a wing but I would be surprised if it did anything but add drag to the body of an aircraft. An 11% change in mileage is too much of an out-of-the-ballpark number to ignore.
I think you’re right about the airplanes, but based on the Mythbuster’s experiment I can’t agree about dimpling not helping on cars.
With cars (I’m guessing) it’s a fashion issue more than anything, although it would likely make buffing and waxing more difficult as well. If people actually cared more about mileage than design, all cars would be as low-drag as the Prius.
Another dimple relevant area is artillery shells. It’s not a field I know anything about, and I’m curious what kind of pros and cons dimpling offers there.
The reason race cars at the high end (LMP, Indy, F1) don’t have dimples is because dimples aren’t needed to keep the airflow attached to the skin of the car. The cars are almost a flat plane, so you don’t have any places where the airflow separates from the body of the car. Except, of course, at the back end of the car, where the wings make a tremendous change to the airflow (you can see it in F1). Since the flow over the skin is already laminar(ish), dimpling would only increase skin drag with no resulting benefit.
It might help to dimple the bottom of the car to increase suction if there lifting qualities to using it on a wing.
Body in white manufacturing expert here. Unibody cars are not just sheet metal on a frame. There really isn’t such a thing as a frame. The body is the structure. Given that, most of the the outer skin is completely non-structural, so I would allow a layman to say that it’s just sheet metal on a frame (even though it’s vastly more complex than that).
Not having seen the episode, I have no idea on the dimensions of the dimples or where they were on the vehicle, but I’d say that it would be no problem to dimple the sheet metal on the vast majority of the surface area of the vehicle. Things like fenders, decklids, hoods, doors, and roofs are no-brainers. It gets considerably more difficult at the perimeters of these panels, of course. The pillars would be tricky, too.
Here’s a picture:
Not dimples, but many aircraft have vortex generators to delay flow separation.
Where inside the car?
If they put it where it puts additional weight onto the driving wheels of the car, that may have increased the traction and thus the mileage. Here in Minnesota, during the winter, many people added weight, like sandbags, in the trunk of their rear-wheel-drive cars. This decreases spinning tires and can get you better mileage, despite having extra weight in the vehicle.
They put it in the back seat. And it only looked to be about 10 lbs of clay. Adam didn’t put too much effort into picking up the box.
I’m not sure the box they showed him hefting was the only one they put on the back seat - that didn’t look like 1082 divots. At any rate, yeah the redistribution of weight could have been a factor, but they did not stick it over the drive wheels, and I’m not inclined to think it would explain the results. Note that the majority of late model cars are front wheel drive, and you can’t just dump a cement block in the trunk anymore.
I grew up in Chicago, and we put the bags of sand or an extra body in the trunk to cut down on spinning the tires and skidding on ice. Nothing to do with mileage - just putting some weight on the tires because some cars were just too light in the read end.
An unrelated aside: Those maniacs were about half a mile from my house doing their tests.