I’m a fan of Mythbusters. There’s nothing better (or worse) than a Mythbusters Marathon on a Sunday afternoon. In a recent episode they did an experiment that involved dimpling car like a golf ball to see if it got better gas mileage (Dirty Car vs. Clean Car). In an extreme case they discovered the theory was plausible…and that got me thinking, would dimples make a Hot Wheels car go faster?
The Mythbusters Golf Ball car
The theory
The whole concept behind the dimples was creating less drag on the car, making it more efficient. And in theory, the less drag a car has the faster it can go. It’s one thing when applied to a car with an engine, it’s another thing applied to a gravity car like a Hot Wheels car. Not to mention, you have the scale to worry about. Nonetheless, it sounded like a fun little experiment.
However, when I say “experiment” I need to disclose that this was quite possibly the most un-scientific experiment ever. I don’t have the tools or time to do a Mythbusters-quality job. But it was fun all the same and the results honestly surprised me.
The control
The Nissan GT-R seemed like a good fit
To start, I picked out two of the exact same car, the Nissan GT-R, and labeled them A and B. The Nissan looked like an average car that had a lot of metal surface to dimple. To get a control (of sorts), I ran both cars unmodified down the track 10 times, alternating lanes each time to help eliminate any effects the track the may have on the car. The control races found that both cars were pretty well matched, as one would expect. However, Car B proved to lose two more races which meant it was the slower car and it would thus get the dimpling in an attempt to make it faster than the undimpled Car A.
I also weighed the cars before the races and dimpling, each car weighing 32 grams. For every race I used the official Redline Derby race track which is 18-feet long with a 4-foot drop.
Both cars were (pretty much) equal
Results of the control races: Car A, 6 wins; Car B, 4 wins
- Race 1, Car A
- Race 2, Car A
- Race 3, Car B
- Race 4, Car A
- Race 5, Car B
- Race 6, Car B
- Race 7, Car B
- Race 8, Car A
- Race 9, Car A
- Race 10, Car A
Lets get dimpling!
All dimpled
Once the control was set, I went to work on dimpling Car B. I had good intentions to keep the dimples evenly spaced and as close to golf ball pattern as possible, I even dotted the car with a Sharpie to help guide my dimples. But as I started the dimpling process I quickly realized that with the tools I had – a Dremel and drill – that type of control was not going to be possible.
I used a 1/8″ bit with a Dremel to make the dimples. I did my best to maintain a golf ball pattern and overall, I think it turned out pretty well. After 20 minutes of drilling, Car B had gained 126 dimples. Again, I must disclaim that this was all “by eye” drilling, so the dimples are not all the same size, depth, or distance from each other…but they’re close (this isn’t hard science, folks). I’m sure the dimples are also not the right scale compared for the car, they should be a lot smaller.
PHOTOS: Check out more photos of the Hot Wheels golf ball car
I weighed Car B after all the drilling was done to find I had shaved off one gram of weight, the car now being 31 grams. Like the Mythbusters, to keep things as close to fair as possible, I took some clay and added back the one gram of weight lost to the bottom of the car so once again both cars were 32 grams. Then it was time to race.
Final race, dimpled vs. smooth
I ran 10 more races like I did before, alternating lanes between runs. Considering all the sloppy science going on here, I was honestly surprised by the results.
Results of dimple races: Car A, 2 wins; Car B, 8 wins
- Race 1, Car B
- Race 2, Car B
- Race 3, Car B
- Race 4, Car B
- Race 5, Car B
- Race 6, Car B
- Race 7, Car B
- Race 8, Car A
- Race 9, Car A
- Race 10, Car B
The results surprised me
I expected the results to be similar to the control races, both cars still being pretty much equal. I figured aerodynamics and drag at such a small scale wouldn’t make much impact. I don’t think I proved that it does, but I’m still surprised that the dimpled car won so many races. It was far more lopsided than I expected. It seemed the dimpled car was faster.
Do dimples make a difference?
Lots of room for error
Do dimples make a Hot Wheels car go faster? Sure, why not? Based on this simple experiment, it looks like a dimpled car performs better than one that is not dimpled. Is it because of aerodynamics and drag? I guess so…that’s the theory, anyway. However, this little project was way too casual to say anything for sure. Since we didn’t use the exact same car for the pre-dimpling and post-dimpling, and the dimpling process wasn’t as accurate as it should have been…so there is lots of room for error. But it was fun to try all the same.
I’m gonna call it: plausible
I had two cars that were the same model out of the box, and after racing they were about even. After one car was dimpled, it won more races than the one that wasn’t dimpled. That’s all I know.
And a special thanks to the Mythbusters for giving me the inspiration for this fun little project, and for providing us all with scientific entertainment every week.
I’m also surprised that dimpling would have much of an effect at the sizes we’re talking about. The aerodynamic resistance (drag) is a function of the surface area, but velocity is the only variable that is squared:
F(d) = -1/2pv^2AC(d)u
F(d) is the force of drag,
p is the density of the fluid
v is the velocity of the object relative to the fluid (air, in this case)
A is the reference area,
C(d) is the drag coefficient (a dimensionless parameter, e.g. 0.25 to 0.45 for a car)
u is the unit vector indicating the direction of the velocity
Also, I would think that dimpling would lower the mass of the dimpled car, and that mass is one of the primary determinants of the speed of a gravity-powered vehicle like a Hot Wheels car. If two cars are identical except for mass, the one with greater mass has more potential energy, if both cars are at the same starting height.
But your results speak for themselves; the dimpled car out-performed the un-dimpled one, in spite of its lower mass.
If aerodynamics makes a big difference in a Hot Wheels’ performance, I’m wondering if I should be looking for exceptionally aerodynamic models when I purchase them in stores, if I’m after performance?
And yes, if you haven’t guessed, I’m a Rocket Scientist. Or at least, that’s what I tell the neighborhood kids that play with my Hot Wheels. It’s almost true, since I work at NASA.
— Charlie
Wow, a real rocket scientist…welcome! Glad to have your input, regardless how much over my head everything you said is! I’m not a scientist but I’m glad to have one checking this stuff out, I think testing effects on Hot Wheels can be very interesting and challenging.
To be fair, I didn’t do a whole lot of testing/racing with the dimpled car. I should do a ton more races versus the non-dimpled car to see if results are consistent. Like I said in the article, there is a lot of room for error the way I did things.
I explored weight distribution too a few weeks ago (check out the article under the Racing section) but haven’t really been able to do any testing with it. Again, it’s one of those cases where I would have figured the scale would negate efforts but maybe the dimple experiment suggests otherwise.
However, in your thoughts about looking for aerodynamic cars…I wouldn’t change my habits based on this experiment. Hot Wheels are so variable when it comes to performance, even from the exact same model of car. A car that looks (and could be) aerodynamic may suck in a race because it’s too light, the axles are weak, the wheels wobble, chassis is narrow.
Albeit, the slick/aero cars *look* faster, I’ve found that they rarely are. But finding a good racer as a base, then applying science (like the dimples) might just be the combination to get the fastest car. Hmmmm…
If I had the tools and equipment, I should do the dimple test with a timer. Get time to finish without dimples, then with dimples on the same car just like the Mythbusters did. That would eliminate variables between car modelings…but hey, it’s Hot Wheels…it’s fun!
Brian,
I’m no rocket scientist, but I am a nuclear physicist and a professor at Ohio State University. I’m teaching an undergrad course this term, and will direct them to your page. Very nice work!
Everything Charlie says about drag is correct. The thing of course is that while that formula captures *most* of the velocity dependence of the force, the drag coefficient itself depends on velocity, for nontrivial reasons. (I know, you guys both know this. I just have to say *something*.)
Very cool.
Rocket scientists AND nuclear physicists…geezoo. I think more brain power has commented on this article than all others combined…ever! haha..
Thanks for the insight Mike. And hey, OH-IO! I’m an OSU grad myself.
… though I should say that I, too, am somewhat surprised. For a 4′ (1.2 meter) drop, the cars would be moving at most ~4.8 m/s ~10.5 mph. Naturally, track friction and air drag reduce this. According to drag coefficients for golf balls that I’ve seen, the reduction in drag due to dimples doesn’t really kick in until Reynolds number ~4*10^(-4), or about 16 m/s ~ 35 mph for a sphere in air, unless I’ve screwed up somewhere.
But:
(1) experiment trumps expectation
(2) drag coefficients are VERY sensitive to details of the object.
so… I still think your thing here is cool.