Why don't vehicles have aerofoils to partially take the weight of the car?

[aerohead]

Originally Posted by AeroMcAeroFace

Doing some calculations it seems that aerofoils of over 100 times lift to drag are common, and given a rolling resistance of 0.01 the difference between lifting using wings and using wheels are identical, so theoretically adding wings would have no effect on total drag.

But with greater than 100 times lift to drag then it is potentially beneficial to add wings.
And if the car is already shaped like an aerofoil, such as a solar car, then most of the lift induced drag would already be there in the shape of the car.

Those are in free air and ground effect may change those things.

But the car has to exist, making that shape have lift would give very little extra drag because the shape is already there, already generating drag.

(I do know about the ultra low rolling resistance tyres that they use on solar cars, but you can't buy those and they only last for 200 miles)

I would mention that an airfoil-shaped car is not a guaranteed prerequisite for lift. All wings exhibit a zero-lift angle-of-attack.
And if you think about it, as long as the air continues to follow the streamline contour of the airfoil, it is simultaneously generating higher local pressure.
If the flow remains attached all the way back, then all local barometric pressure is present at the rear stagnation point, minus the skin friction, lost in the turbulent boundary layer (nothing we can do anything about).
The pressure of forward and rearward stagnation essentially cancel any lift due to acceleration at the suction peak of the body.
Static axle loads are sufficient for perfectly acceptable stability.

[AeroMcAeroFace]

Originally Posted by aerohead

I was looking at the 2015 Yukon Denali, tested by CAR and DRIVER.
This is a little larger SUV than the Tahoe, and a curb weight of 6,060-pounds.
EPA test weight would be 6,360-pounds.
For it's frontal area estimate, and using Cd 0.36, aerodynamic road load @ 70-mph is 67.72% of overall load ( 29.798- horsepower, leaving 32.27% for rolling-resistance ( 14.2- horsepower )
Should there be one particular airfoil design, superior to all others, which, say, produced an evenly-distributed 3,180-pounds lift, then rolling resistance could be cut in half, and the lift-drag table for that airfoil would indicate the required power to create that magnitude of lift. Then it would be a matter of comparison, to establish your cost-benefit ratio.
If you drove into a chain-reaction rear-end collision, involving a breached cooling system, and had to maneuver on top of a glycol-wetted road surface, I can't imagine a sensor package which could identify this as a threat, reacting spontaneously, such that it could avoid catastrophe.
Product-liability attorneys would be all over that. A 'rainmaker'.

I agree with what you have said here, and I think it is clear why normal cars currently don't fly.

However, if the car is already "flying" then there would be able to have control surfaces. How quickly could a spoiler come up? Half a second, I would think is possible.

How the hazard is identified, I don't know, but how would a normal car identify that? Wheel slip, yaw and steering angle sensors in traction control systems can react pretty quickly, use the same technology?

But really the question is not about traffic, it is about theoretical possibility. It seems possible to me if the aerofoil has a lift/drag higher than a wheels lift/drag then it would work. see permalink 8

[freebeard]

I am imagining a car, a bit like an ekranoplan, but never leaves the ground so the wheels always drive the car. Steering angle sensors, wheel slip sensors and brake pedal sensors cause the lift to stop and spoilers to create downforce pop up. Turning at speed can be assisted by aerodynamic aids.

You're not the first. Off hand I can think of two predecessors, Jocko Johnson and Buckminster Fuller.

Jocko Johnson's Triple Nickel


Untitled Document

Bucky Fuller's Dymaxion


Both cases are a tadpole trike that fly the one rear wheel.

With the advent of self driving cars, the mechanistic prediction of impending events is actualized.

[Autobahnschleicher]

Originally Posted by redpoint5

I always thought it dumb to increase cornering ability on a Formula 1 car by creating massive downforce, most of which isn't needed at high speed straights where the majority of downforce is created. I'd rather see a rudder that dynamically moves to fly the car around the track.

Active aero is banned in F1, but they do allow drag reduction systems now.
Also a lot of tracks have many highspeed corners and the tires have a grip coefficient significantly greater than 1,0 so downforce is what they use a lot.

They also kind of circumvented the active aero rule with the now obsolete "F-duct"

[aerohead]

Originally Posted by AeroMcAeroFace

I agree with what you have said here, and I think it is clear why normal cars currently don't fly.

However, if the car is already "flying" then there would be able to have control surfaces. How quickly could a spoiler come up? Half a second, I would think is possible.

How the hazard is identified, I don't know, but how would a normal car identify that? Wheel slip, yaw and steering angle sensors in traction control systems can react pretty quickly, use the same technology?

But really the question is not about traffic, it is about theoretical possibility. It seems possible to me if the aerofoil has a lift/drag higher than a wheels lift/drag then it would work. see permalink 8

At higher velocity the event horizon gets shorter.
It's said that a computer can react in 1/10th the time of a human, at peak physiological efficiency.
That would definitely favor letting the ECU do the 'reacting.' If it's getting the appropriate sensor input. That could be an unknown.
Perhaps there are physical anomalies which facial recognition software could interpret as a threat response, ahead of the brain actually sending any signal in response. Something that's been observed in clinical observation.
If so, constant monitoring of the driver's face/ eyes, might pickup this pre-reflexive anomaly, triggering the car to respond before the human has actually got the signal themselves.
Ballistic actuators, modelled on the airbag, could be designed with enough force to mechanically disable the lifting surfaces in time to recover a traction-dependent avoidance maneuver. As a military canopy is separated by explosive fasteners immediately in advance of a forced ejection-seat deployment, programmed to respond to accelerometer signal threshold protocols cataloged into onboard memory. The pilot may experience white-out or blackout, it doesn't matter; the plane will do whatever it takes to protect the pilot, without any participation.

[seifrob]

Disclaimer:
I am neither expert, nor can I provide some numeric data (any pilot here? input is most welcome).

Even at speed, when you turn really hard (think of deer-evasive manouvers), your car turn radius is at magnitude of ten meters (around 50 m, based on local roundabout size).
Planes do not turn so well. What I have been able to find is 134 m turn radius at 95 km/h for typical sailplane (anybody can confirm?) and they turn by rolling, using full wings area to create turning force. I was not able to find turning radius just by yawing but assume hundreds of meters.
I Did not check the math yet, but, IMHO, to substitute wheels for rudder with similar performance will ask for rudder with impractical size.
About "disabling the lift when needed": Even if you make our ECU and spoiler actuators lighting fast, you do not get instant adhesion. First you need to fight your car inertia. Lets say your car is lifted "just to touch". So, to get full ground contact again, the body needs to drop down what is the suspension lift. Lets say 30 cm. It takes 0,25 s to drop 30 cm by free fall. So, from the instant you "disable the lift" it still takes 0,25 s till your brakes or steering begin to work. Not worth the risk.

[aerohead]

Yes, it's these lag times that could be the difference between property damage, injury, and life and death.
In a multi-redundancy, autonomous driving universe, worst-case-scenario separation distances might be continuously maintained, such that there would always be some safety margin present.

[oil pan 4]

Cars don't fly because people can barely drive.

[JulianEdgar]

Originally Posted by AeroMcAeroFace

In the Tahoe hybrid thread the car would only ever get 60mpg(US) due to rolling resistance, in the insight thread half of the drag was rolling resistance.

Theoretically rolling resistance is proportional to force applied, so a lighter car would have lower rolling resistance.

A car that generates significant lift has lower rolling resistance at higher speed.
Low drag shapes in ground effect often have high lift coefficients.

I am imagining a car, a bit like an ekranoplan, but never leaves the ground so the wheels always drive the car. Steering angle sensors, wheel slip sensors and brake pedal sensors cause the lift to stop and spoilers to create downforce pop up. Turning at speed can be assisted by aerodynamic aids.

As long as it is aerodynamically stable, in all directions, can dump lift as necessary I see no reason why this wouldn't reduce rolling resistance.

Obviously this is really complex but looking at cars like the huayra, certainly possible. Is it the case though that by the time significant rolling resistance reduction occurs, the speed is so high that rolling resistance is a tiny fraction of total drag?

It's only friction with the ground that provides ALL of a car's propulsive, braking and cornering forces. You can easily experience what a car that is only just in contact with the ground would be like by driving on a surface with almost no friction (like on ice, or aquaplaning with all four wheels on water). Basically, impossible to control.

Alternatively, if you wish the car to be like an ekranoplan, you then need to devise different propulsive, braking and cornering systems to those used in cars. For example, propellers or jet engines for propulsion. With ekranoplan-style braking and turning systems, no normal roads would be suitable, and no normal vehicle spacing would be possible either.

Finally, if you wish a car to be able to change from being just in contact with the ground (or not at all in contact with the ground) to being a conventional car when friction with the ground is required to turn, brake or accelerate the car, then cars will need to be enormously complex (because they will need to be two different types of vehicles) and roads will need to be huge and traffic spacing equally huge (because of the finite time it would take to change from one mode to the other).

Sounds like a really bad idea to me.

[freebeard]

I am imagining a car, a bit like an ekranoplan, but never leaves the ground so the wheels always drive the car. Steering angle sensors, wheel slip sensors and brake pedal sensors [do stuff].

Further to Permalink #13, the platform to test this would be a tadpole with front-wheel drive, torque vectoring and autonomous control [ready].

The only vehicle I know of that fits the bill would be the Arcimoto FUV or Roadster. Any smoll light plane could provide the empennage. The hard part would be the software development. It would require Segway-like balancing on the two front wheels, plus the predictive AI.