Jump to content
  • Advertisement

Archived

This topic is now archived and is closed to further replies.

szinkopa

car aerodynamics

This topic is 5211 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Does somebody know any formula with which I could calculate the force acting on the car body? I know the 1/2*density*area*Cw*speed^2, which acts opposite the velocity. But I think there is some force acting downwards, which pushes the car to the ground, increasing the traction. Or maybe there is also some force acting upwards. Let''s assume, there is no wind. Of course the amount of the force depends on the shape of the car. For example if I have a wing on the car with a certain size and angle, how much force does it create at a certain speed? Thanks in advance

Share this post


Link to post
Share on other sites
Advertisement
the drag and downforce can be modeled quite similarly (same equation, different coefficients). It''s crudely explained in the PHOR (Physics Of Racing Series). It''s available in the article and resource section. Basically, the more the downforce (which gives better traction), the more the drag force (which slows down the car).

Share this post


Link to post
Share on other sites
I dont think there are any true equations for this, as to get an accurate model of the airs behavior over different shaped surfaces you would probably need a wind tunnel and many hours to tinker with.
If it helps at all, most street cars actually generate lift at high speeds. Also the bulk of downforce on racecars comes from the front air dam and under the car, not the huge wing on the back.
In my limited knowledge, I beleive what it does it prevent alot of air from going under the car. So as you drive along the air under the car escapes out the back, but cannot be replaced with more air from the front, creating a low pressure area under the car.
Also I''m not sure if the downforce increases exponentially with velocity as the drag does. I think it may be linear.

I''m not sure if this helps, but it looks like it may have some graphs of drag vs downforce and such:
http://aerodyn.org/Annexes/Racing/racing.html

Share this post


Link to post
Share on other sites
One idea I had was to have a very simple virtual wind tunnel, working using basically ray-tracing technology. Have an imaginary grid with spacing somewhere in the range 1mm to 10cm. Place this grid pointing at the front of the car. Fire a ray from each point on the grid and do the collision detection to see what angle it bounces off the car. Then do something with the results to get drag and lift characteristics.

Obviously pretty crude, but it should give at least a reasonable guess for one shape compared to another; maybe as good as looking at a car and trying to tell how good it is.

I think you''d better account for the behaviour at edges if the rays bouncing off could be deflected by other rays to go more in the same direction. Then air coming off the top of the windsscreen would be ''bent'' along the top of the car somewhat in a more realistic fashion. Or you could let particles/rays ''stick'' to the surface if bouncing off at small angles so as to flow over the body.

What do those who know aerodynamics properly think of my idea? I think it''s pretty simple to implement.

Share this post


Link to post
Share on other sites
Thanks d000hg for the idea. It sounds quite good

I has on idea similar to this. This is also a precalculation. It is also based on the angle of the collision. From this I can calculate a force of an air particle, because it''s perpendicular to the body surface at the collision point. Considering the density of air, and speed of the particles, which means impulse here, and the entire area of the collisions I could create a force vector depending on speed. The simple formula I wrote earlier is also based surely on this.

But the problem is that at high speed, when the car is slipping, drifting, sliding, then this force is different, the air is acting on the side of the car too. And as oliii mentioned earlier, that the twitchiness of my car at high speed could be reduced with aerodynamics. So I also should have formula with a roteted car, whose side is also fired with particles.
A solution could be to do the same force calculation with some discrete rotations, like 10 degree, 20 degree, ... , and at runtime changing, perhaps interpolating between adjecent forces to calculate at, say 13 degree the force of air.
This model however takes in consideration neither the flow of the air, nor the whirl of it, and nor the resistance of the air, which all also have effect on the body.

Share this post


Link to post
Share on other sites
Also you may need to take into account that not all air resistance is due to the air physically hitting a surface. Some of it is viscous drag, as the particles slide across surfaces.

As for the downforce, you could use the lift of an airfoil, only upside down. (for a wing or spoiler on a car).
http://wright.nasa.gov/airplane/lifteq.html
has some equations you can use. It also has info on drag, so that may be of some use.

[edited by - CombatWombat on May 12, 2004 8:46:01 PM]

Share this post


Link to post
Share on other sites
The small wing on the back of street cars is called a ''spoiler''. It''s function is to break-up, or spoil the flow of air as the air moves over the top of the car. A car is similar to that of an aeroplane wing in that the air traveling over the top of the car moves faster then the air traveling over the bottom of the car. The faster moving air creates a low pressure which tends to want to suck the car up - this is called ''lift''. The ''spoiler'' breaks the free flow of air as it moves over the surface of the car, thus counteracting this upwards force. As far as I know these small spoilers don''t actively push the car down (unlike the really large wings on Formulae 1 race cars, or the silly looking wings on some modified street cars) but rather stop the car being ''sucked-up''.

Share this post


Link to post
Share on other sites
Getting drag+lift values at different angles sounds a good idea. If you want a more accurate model that takes particles sticking to the surfaces etc then you''ll get different results at different speeds. You could have a 2D array of angle increments at different speeds. Say 16/32 angle settings (~20/10 degrees each) and the same for speed settings (up to about 50m/s for a normal car or 100m/s for a racing car). Thats like 4/16K of memory, simple linear interpolation would be fine.

Getting air ''particles'' to stick to the car''s surface should be easy enough though. In theory other particles should not hit the car but flow over the car at a distance from the surface. That sounds more like hard work though. You''re not going to get real values - fudging will be crucial - but it should be pretty good. If you have a realistic engine and physics set up, accelerate the car and fudge the values to get correct 0-60 and max speed values.

Share this post


Link to post
Share on other sites

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!