Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Car Aerodynamics: Experimental trials not agreeing with lift theory, help

  1. Mar 21, 2007 #1
    Hi, I'm currently doing a research study on the effect of the presence of a rear car spoiler on the lift of a vehicle at low speeds

    As far as my understanding, with an increase in air velocity, there should be an increase in lift on the car. My trials however show that with an increase in air speed, the greater the down force, even with the control which has no rear spoiler. (I measured this using a wind tunnel and my model car above an electronic balance)

    I am currently unclear on the theory: should a car (specifically the Honda Integra), with no spoiler, see an increase in lift as speed increases, or vice versa?

    Hope you can help, thanks!
  2. jcsd
  3. Mar 21, 2007 #2
    I am by far not qualified to really answer this but just from thinking about it. Wouldnt the air deformed by the front of the car move downward to fill the area of lower pressure between the roof and trunk on the rear of the car. Thus creating a downward force on the car which I assume would be greater than any lift created by a spoiler. But Im just speculating.
  4. Mar 21, 2007 #3


    User Avatar

    Staff: Mentor

    Does the car have a ground effects package....or just a rear that is higher than the front? That'll create downforce.
  5. Mar 21, 2007 #4
    Considering the car as a wing, the air passing below the bottom of the car will not change direction. On the other hand, the air flowing over the top of the car will have to curve around the body of the car. So it would appear to act exactly like an airfoil. Faster flow on top, and slower flow on bottom. So that would result in increased lift on the car as you go faster.

    I dont understand what you mean by this. A wing creates downforce, not lift.

    I thought of that as well. The air impacting the front of the car will change its momentum, thus driving the front of the car down, creating a downforce.

    I suppose the question is which force wins, the momentum change or the 'wing effect'. If either of those are occuring.
  6. Mar 22, 2007 #5
    Nevermind, carl your right I retract my statement I was thinking the oppossite. But then why would the tests show differently.
  7. Mar 22, 2007 #6
    Sorry, I might not have been very clear in my post

    Basically I'm wondering: does a typical passenger car, with no spoilers/wings, travelling at low speeds (say 20-30 kmh) experience lift or downforce?

    Because my trials show that the car experiences downforce, even without a spoiler. I understand that a spoiler is meant to create downforce and hence it is reasonable if the car still experiences down force if it has a spoiler attached. However, without a spoiler, what is the expected outcome?

    Thank you
  8. Mar 22, 2007 #7
    Your trials seem to indicate that the purpose of a spoiler is to *increase* that downforce.
  9. Mar 22, 2007 #8
    I'll put my two cents worth in..

    Re: "Considering the car as a wing, the air passing below the bottom of the car will not change direction. On the other hand, the air flowing over the top of the car will have to curve around the body of the car. So it would appear to act exactly like an airfoil. Faster flow on top, and slower flow on bottom. So that would result in increased lift on the car as you go faster."

    This makes sense at first, but please consider: Why are you considering the car as a wing? A car is not an airfoil, it's a bluff body. Also, an airfoil is not intended to generate lift by its shape i.e. "faster flow on top, slower flow on bottom". If that was true, how do aircraft with symmetrical airfoils fly, and how can an aircraft fly upside down? Even a flat rectangular board with no camber will produce lift at angle of attack thanks to newtons 3rd law. The airfoil shape is designed to keep the airflow hugging the wing thanks to the Coanda effect, which is how an airfoil with a hump at the top can generate lift at zero angle of attack - it keeps forcing air downwards. The pressure differential created by the acceleration of air over the top of the wing does help push the wing up, but it is not the complete picture of reality. An airfoil generates positive lift by deflecting the air flow downwards, and newtons 3rd law does the rest. .

    On the topic of spoilers for saplingg: Have you researched the type of spoiler you are using? There are different types of spoilers. Some are for low speeds and others for high speeds. Have you considered the type of spoiler in your study? Generally, spoilers are most effective at fast speeds where they can improve handling of the vehicle and improve fuel efficiency (by reducing drag). I suggest that at low speeds you won't see much effect of lift at all on the vehicle caused by the spoiler, so perhaps you should observe just the spoiler by itself to see whether it generates any lift and how much.

    The primary purpose of spoilers is to "spoil" the turbulent air at the back of the car, to reduce drag of the vehicle, resulting in improved fuel efficiency and improved handling qualities. This is the main difference between spoilers and automotive wings (used on race cars) which are primarily designed to keep the racecar hugging the ground. Most spoilers will help keep the back of a car down by generating downward force, but this is not their primary intended function. This negative lift effect due to the spoiler is something I suggest won't be observed unless at high speeds.

    What about a car with no spoiler? I suspect that it is as others have stated: the air hitting the car at the front forces the car into the road. Also consider, the turbulent air flowing over the roof of the car and any spilling down onto the rear of the car, which helps force the back down. Are you able to make observation of the flow in a wind tunnel by using coloured smoke or something like that?

    So to answer this question "Basically I'm wondering: does a typical passenger car, with no spoilers/wings, travelling at low speeds (say 20-30 kmh) experience lift or downforce?"
    I am thinking that there is minimal lift generated if any at all due to the spoiler, but mostly downforce is generated due to drag.
    All the best with your research.
  10. Mar 22, 2007 #9

    First I'd like to say thanks for your response: it shed some light on my research, but I have a few questions I'd like to ask..

    1. You suggest that cars, and airfoils too, experience lift/down force due to conservation of momentum, but it is well understood that Bernoulli's effect contributes a large amount to the lift and downforce of a car or an airfoil (as Carl wrote). I am no expert on aerodynamics, but during my research I have found that Bernoulli's effect plays a large role in LIFTING the car off the ground.

    Are you able to disprove this? Because while my results agree with you that the higher the speed, the greater the force that conservation of momentum pushes the car into the ground with, (downforce due to the shape of the car e.g. the angled windshield) my research has in fact stated the opposite, that at high speeds, Bernoulli's effect outweighs any conservation of momentum and actually lifts the car off the ground.

    --- see "The Issac Newton school of driving: physics and your car" B. Parker.

    I quote from this source,

    Lift is normally of little importance in passenger cars as their speed is usually too low to produce much lift. It was noticed early on that something strange happened at high speeds: the car seemed to be lifting off the ground........

    Lift occurs because the airflow over the top of a car is faster than across the bottom (fig 69). This occurs to some degree in all cars. As the speed increases, the pressure decreases, according to Bernoulli's theorem. The top of the car therefore has a lower pressure than the bottom, and the result is a lifting force.


    I include a few entries from a table he has generated.


    v(mph) lift force (pounds)
    50 ........ 115.0
    60 ........ 165.6
    70 ........ 225.4
    120 ........ 662.4
    150 ........ 1035.0


    The abovementioned source shows my point and in fact disproves some of your statements. However, I must concede that the source shows only lift force at speeds of 50 mph (80 kmh) and above, while my results only span from 5-30 kmh.

    So I ask this: I believe the source is reliable, and at high speeds, there will be in fact lift. But can anyone shed any light on the lift/downforce a passenger car will experience at low speeds?

    If you are right and that drag force results in downforce, would I then expect a graph of lift force against air velocity to first grow exponentially in the negative region (indicating increasing downforce as my results have shown), and then curve back up and increase exponentially into the positive reason (increasing lift as B. Parker has stated) ??

    Also, yes, my hypothesis agrees with you: that at low speeds, the spoiler is negligible in creating downforce, so I am trying to confirm if the downforce observed actually originates from the car body itself.

    Once again, thank you very much for your responses.
  11. Mar 22, 2007 #10


    User Avatar
    Science Advisor

    At speeds that low, vertical aerodynamic effects are hardly measurable. But regardless, almost all passenger cars develop lift. Engineers work very hard to minimize that lift, and only a handful of exotics manage to generate a small amount of downforce. A car which is pitched downward will eventually generate downforce, but most cars would look ridiculous by the time all of the lift was removed (among other problems).

    It is also not correct to think of a car as a wing. Among other reasons, the ground has a huge effect on the airflow around the car. Even if the body was shaped like an airfoil, it wouldn't act at all the way you'd expect.
  12. Mar 22, 2007 #11


    User Avatar
    Science Advisor

    The flow around a vehicle is far too complicated to analyze using any simple principles like these. At best, you'd get a vague qualitative agreement with reality.

    Still, above a certain speed (which is usually very low), lift tends to increase with the square of the speed. That's how that table you quoted was generated.
    Last edited: Mar 22, 2007
  13. Mar 22, 2007 #12
    Thanks Stingray, do you have any comments on the results I generated? Seems like they completely oppose existing theory
  14. Mar 22, 2007 #13

    Just trying to exclude all possibilities here. But you've only done wind tunnel tests? Perhaps there are varying pressure gradients within the tunnel itself which are affecting your results. If you could have pressure sensors underneath your model and on top, you could measure the relative pressure difference as see if that difference you need for lift is actually there.

    Now it is not my aim to dispute or disprove Bernoulli's principle, however I believe that since your results disagree with Parker, you should look at other reasons why this may be the case. I think that to ONLY consider the car as an airfoil is not valid because the car is firstly a bluff body cutting through the air close to the ground. So you have drag, uneven surfaces and obstructions and also possibly ground effects playing a part. The shape of an airfoil (or car) is not everything which matters: there is also the angle of attack to consider (direction of oncoming airflow) and drag. As I stated previously, even an aircraft flying upside down can generate lift even though the air travels faster over the bottom of the wing than over the top. I said this so you can re-consider looking at the car as an airfoil only, and consider that it is a bluff body which a number of effects possibly affecting your results. Car manufacturers afterall, design cars to stay on the ground, not to fly. If they've done a good job on the Integra maybe this is why you aren't seeing the lift you are expecting. But rule out a problem with your test setup first?
  15. Mar 22, 2007 #14
    I'm afraid my paper is due tomorrow and I don't have the time (and resources) to perform further trials with pressure sensors. But I see your point about angle of attack. I will have to take another look at my results, I didn't really include AOA of the car and of the spoiler as factors because the calculations are quite complex, but admittedly that is a very poor reason to rule out such a factor.

    Thanks for your help.
  16. Mar 22, 2007 #15


    User Avatar
    Science Advisor

    How did you do your measurements? How did you make your model? How large was it? How wide is the wind tunnel? How fast was the airflow? Did you simulate wheel rotation and road motion? Is the effect statistically significant?

    Anyway, you might want to take a look at a book called "Race car aerodynamics" by Joseph Katz. It discusses all of these things without getting into full-on fluid mechanics. It's also filled with interesting graphs (including one of the lift and drag coefficients versus AOA).
  17. Mar 22, 2007 #16

    My experimental setup is simply a 1/24 scale model of the Honda Integra (built by someone else from a kit) placed on an electronic balance in a wind tunnel about a foot wide and a foot high. I used a standing fan to generate the wind so I only achieved low speeds from 0-25 kmh, which I measured with a wind speed meter. No I did not simulate wheel or road motion, I simply measured straight-motion downforce.

    Thanks for the book suggestion, I'll see if I can get my hands on it.
  18. Mar 22, 2007 #17


    User Avatar
    Science Advisor

    You've definitely got a lot of experimental error introduced because of your set up. Just by using a fan with no flow straightening device and insufficient straight length of tunnel prior to the test section, you don't have any way to ensure that you have a smooth, non-rotational flow at the model.

    Honestly, since your report is due, I would say something along the lines of any other experimental report. Here's what I expected to happen, here's what actually happened, here's why I think the two don't match. For your report, I would do a bit of research into the components of a properly designed wind tunnel to see just how complicated they can be.
  19. Mar 22, 2007 #18


    User Avatar

    Staff: Mentor

    The two methods are both correct - they are just two different ways of calculating/measuring the same thing and depending on your situation one may be more suitable than the other.

    Important question: Is the rear of the car body higher than the front? Most cars make such terrible airfoils that ground effects are a very significant fraction of the equation.
  20. Mar 22, 2007 #19


    User Avatar
    Science Advisor

    Besides the issues Fred mentioned, a small model like that tends to be pretty inaccurate. Things like the airflow through the engine compartment can have a very large effect. I heard a story in the design of the 3rd gen Corvette (in the late 60's) that it initially had terrible front-end lift. That was largely cured by adding some slats behind the front wheels to give the air flowing through the radiator an easier path out of the engine compartment.

    Anyway, your small wind speeds are also a problem. In these experiments, things can usually be called "comparable" if something called the Reynolds number is similar. In your case, think of it as velocity times length. So a 1/24 scale model at 25 km/hr acts like a full-size one at ~1 km/hr. As I understand, drag and lift coefficients are usually doing odd things in that range. They're certainly not constant.

    I agree with Fred that you should just explain the problems with your setup. Even though it didn't quite work, it sounds like a fun experiment.
  21. Mar 22, 2007 #20
    Thanks everyone for the advice
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook