Jump to content


Photo
- - - - -

Downforce ideal angle.


  • Please log in to reply
35 replies to this topic

#1 gruntguru

gruntguru
  • Member

  • 6,712 posts
  • Joined: January 09

Posted 10 April 2019 - 22:35

Inspired by the Ducati aero thread. . . . .

 

On the topic of downforce being normal to the road. The optimum angle for aero downforce (and this applies to cars and bikes) is not 90*. It is more effective for the DF vector to lean somewhat towards the centre of the corner radius. Leaning the vector slightly (this lean is in the same direction as body roll in a car but the opposite direction to MC lean) has a small cost to normal force and therefore tyre grip, but a useful contribution to cornering force.

 

I haven't solved the equations for determining the optimum angle but it will depend on the coefficient of friction and probably the ratio of aero DF to vehicle mass. 

 

Fascinating topic - suggests that extra body roll might be useful in high DF cars.



Advertisement

#2 Greg Locock

Greg Locock
  • Member

  • 5,643 posts
  • Joined: March 03

Posted 11 April 2019 - 08:20

In the limiting case you'd get say 1.5 g from the tires and 4g from the body, which has been rolled by 90 degrees. 5.5g

 

If the tire load sensitivity is 0.7 then conventionally you'd get 5g^.7*1.5=4.6g

 

So I think there may be some merit in your idea.

 

Of course when the car rolls the aero lift coefficient will drop.



#3 gruntguru

gruntguru
  • Member

  • 6,712 posts
  • Joined: January 09

Posted 11 April 2019 - 23:58

The concept doesn't rely on low tyre sensitivity - still works if sensitivity = 1 eg

 

Using your mu and DF with tyre load sensitivity = 1.

 

Conventional. 

Normal force = 1 + 4 = 5g

Latacc = 5 x 1.5 = 7.5g

 

Wing canted 10* 

Normal force = 1 + (4 x cos10*) = 1 + (4 x 0.984) = 4.94

Latacc = 4.94 x 1.5 + (4 x sin10*) = 7.41 + 0.694 = 8.1g

 

No idea what the optimum angle is - 10* was a random selection. Intuitively the optimum angle will increase as mu decreases:

 <45* for mu>1.0 (since Latacc increase is greater than Normal force increase)

 >45* for mu<1.0 (conversely)

 and therefore 45* for mu=1.0



#4 imaginesix

imaginesix
  • Member

  • 7,525 posts
  • Joined: March 01

Posted 12 April 2019 - 03:29

Jumping ahead based on that, it would seem the optimal angle is always tan-1(mu(x)). That ignores the effect of body roll on ground effects.

 

If x=1 then the example of 0.7 mu gives us an optimal wing angle of 35 deg, or 55 deg of body roll!

 

It also suggests the probability that I'm wrong is 99% :)



#5 gruntguru

gruntguru
  • Member

  • 6,712 posts
  • Joined: January 09

Posted 12 April 2019 - 04:39

Your maths is better than mine.

 

Your probability and statistics not so good (or maybe its just excess self deprecation?)



#6 Fat Boy

Fat Boy
  • Member

  • 2,184 posts
  • Joined: January 04

Posted 19 April 2019 - 19:30

I don't think that looking for side force through aero is really all that beneficial unless you're on an oval or a track which is heavily handed one direction. The NASCAR guys about a decade ago were putting a ton of 'banana' in their car bodies to produce a side load. It's still done to some extent now, but its not nearly as easy to see because of the present body templates. A little bit of side load goes a long way on an oval, especially on cars which have next to no downforce in the first place. A friend that was in that world told me they would hang the bodies on those cars as curved as they possibly could. Any little bit extra was worth time.

 

xzD3VG1.jpg

 

Click on the pic to see it in 'normal' proportions.


Edited by Fat Boy, 19 April 2019 - 19:31.


#7 gruntguru

gruntguru
  • Member

  • 6,712 posts
  • Joined: January 09

Posted 19 April 2019 - 23:17

Interesting.

 

Of course if the car is already fitted with a wing, the question is "should the wing provide pure downforce or is it better to angle the wing (eg with excessive body roll) to utilise some of the lift as side-force?" The answer is clearly "yes".



#8 Fat Boy

Fat Boy
  • Member

  • 2,184 posts
  • Joined: January 04

Posted 20 April 2019 - 01:24

Interesting.

 

Of course if the car is already fitted with a wing, the question is "should the wing provide pure downforce or is it better to angle the wing (eg with excessive body roll) to utilise some of the lift as side-force?" The answer is clearly "yes".

 

"Clearly 'yes'."?  I don't think it's clear in the slightest.

 

Keep in mind that someone actually has to drive the damned thing and anything over a couple degrees of body roll is big and sluggish. Also remember that as soon as you roll the car you start shedding downforce (particularly from under the nose of a GT/prototype). I rarely set a car up completely symmetric, so a bit of asymmetric downforce is not an uncommon thing, but that's to address the particularities of each track, not an attempt to make the car turn aerodynamically. If you're turning both directions, I think you'll find a lot of weaknesses before you find an advantage. Having said that, it's an interesting thought and I've been trying to think of a good way to incorporate the concept (if not the approach). I think it'd be an advantage on an oval, but a road course would be a lot more difficult to make work.



#9 mariner

mariner
  • Member

  • 1,809 posts
  • Joined: January 07

Posted 20 April 2019 - 05:32

Sprint cars which are purely oval track have their huge wings set up with the outer wing endplate high and the inner one low so as to impart a inward roll couple from the   aero to laod the inside rear tyre.

 

As the wing is high up the roll moment is pretty high.

 

Like many things in Sprint cars it is crude but effective



#10 PayasYouRace

PayasYouRace
  • RC Forum Host

  • 19,157 posts
  • Joined: January 10

Posted 20 April 2019 - 10:24

Interesting.
 
Of course if the car is already fitted with a wing, the question is "should the wing provide pure downforce or is it better to angle the wing (eg with excessive body roll) to utilise some of the lift as side-force?" The answer is clearly "yes".


The answer is only yes when the side force of the wing exceeds the grip otherwise provided by the tyres. The intersection of two grip curves, one from the tyres against normal force applied, and one of aerodynamic steering force.

Of course, the ultimate extreme would be a hovercraft. Zero grip and entirely aerodynamically steered.

#11 Regazzoni

Regazzoni
  • Member

  • 2,034 posts
  • Joined: January 10

Posted 20 April 2019 - 18:03

Maurice Phillippe tried the concept on the Parnelli VPJ-1 with its (extreme) dihedral wings. It didn't go far.

 

Interesting the front upper wishbone fairing in line with the side wings as to prepare the flow direction over them.

1972-Dihedral-wing-Indy-car-Parnelli-Jonvpj-1.jpg



#12 gruntguru

gruntguru
  • Member

  • 6,712 posts
  • Joined: January 09

Posted 20 April 2019 - 23:42

The answer is only yes when the side force of the wing exceeds the grip otherwise provided by the tyres. The intersection of two grip curves, one from the tyres against normal force applied, and one of aerodynamic steering force.

 

The answer is always yes. Imaginesix nailed it in post #4 complete with formula. Regardless of the grip level of the tyres the optimum wing angle during cornering is never horizontal (nor vertical) - there is always a benefit to sharing the lift between normal force for the tyres and pure lateral force.


Edited by gruntguru, 20 April 2019 - 23:49.


#13 gruntguru

gruntguru
  • Member

  • 6,712 posts
  • Joined: January 09

Posted 20 April 2019 - 23:49

"Clearly 'yes'."?  I don't think it's clear in the slightest.

 

Keep in mind that someone actually has to drive the damned thing and anything over a couple degrees of body roll is big and sluggish. Also remember that as soon as you roll the car you start shedding downforce (particularly from under the nose of a GT/prototype). I rarely set a car up completely symmetric, so a bit of asymmetric downforce is not an uncommon thing, but that's to address the particularities of each track, not an attempt to make the car turn aerodynamically. If you're turning both directions, I think you'll find a lot of weaknesses before you find an advantage. Having said that, it's an interesting thought and I've been trying to think of a good way to incorporate the concept (if not the approach). I think it'd be an advantage on an oval, but a road course would be a lot more difficult to make work.

 

I certainly wasn't suggesting "excessive body roll" as the best approach. Where moveable aero is allowed (FSAE being one rare example) a tilting wing (about the x axis) would be a possibility.



#14 PayasYouRace

PayasYouRace
  • RC Forum Host

  • 19,157 posts
  • Joined: January 10

Posted 21 April 2019 - 08:04

The answer is always yes. Imaginesix nailed it in post #4 complete with formula. Regardless of the grip level of the tyres the optimum wing angle during cornering is never horizontal (nor vertical) - there is always a benefit to sharing the lift between normal force for the tyres and pure lateral force.

 

I think I misread the maths the first time. Looks like it checks out.

 

I certainly wasn't suggesting "excessive body roll" as the best approach. Where moveable aero is allowed (FSAE being one rare example) a tilting wing (about the x axis) would be a possibility.

 

Wasn't there a prototype supercar a couple of years ago with a tilting wing?



#15 pierrre

pierrre
  • Member

  • 81 posts
  • Joined: June 17

Posted 22 April 2019 - 22:15

i disagree and think absolute downforce contribute more to any sideforces but this depending on tyre performance. on ice, side forces would be much better than downforce but on stickly slicks or high performance tyres...the same rule does not apply

lets take f1, suzuka, 130-r corner, recorded at more than 5g to 6g yet downforce during that corner is not 4 times greater that the vehicles total mass. now lets assume an f1 car can do 3g on corners without downforce...for side force to achieve those g-forces at suzuka's 130-r, it needs 1500kg, say if the cars total weight is 750kg. so when we take a transient look into this, as the wing tilts towards sideforces from downforce...it will loose downforce...and logically it will lose grip and those advantages....

 

again on ice, however, i would take side forces anyday



#16 gruntguru

gruntguru
  • Member

  • 6,712 posts
  • Joined: January 09

Posted 23 April 2019 - 07:08

If you draw a vector diagram you will see that angling the lift vector slightly results in a very small loss of downforce but a significant increase in side force. The scalar sum of the z and y components are greater than the magnitude of their vector sum.

 

Think of a right-angle triangle - the hypotenuse is always shorter than the sum of the base and height.



#17 imaginesix

imaginesix
  • Member

  • 7,525 posts
  • Joined: March 01

Posted 23 April 2019 - 14:46

i disagree and think absolute downforce contribute more to any sideforces but this depending on tyre performance. on ice, side forces would be much better than downforce but on stickly slicks or high performance tyres...the same rule does not apply
lets take f1, suzuka, 130-r corner, recorded at more than 5g to 6g yet downforce during that corner is not 4 times greater that the vehicles total mass. now lets assume an f1 car can do 3g on corners without downforce...for side force to achieve those g-forces at suzuka's 130-r, it needs 1500kg, say if the cars total weight is 750kg. so when we take a transient look into this, as the wing tilts towards sideforces from downforce...it will loose downforce...and logically it will lose grip and those advantages....

again on ice, however, i would take side forces anyday

All you've demonstrated is that aero force parallel to the road would be less effective than aero force normal to the road within the limits of your scenario. This is true, but it leaves out the whole question of this thread which is to determine the optimal intermediate angle.

Your reasoning would be valid if we knew for sure that the progression in DF from a parallel to normal force vector was linear, but we don't know that. In fact it turns out the progression is sinusoidal, which means it may have a peak in between those two end points. And it does indeed have a different peak.

FWIW this is the same invalid argument that supports libertarian politics; since unfettered capitalism is better than unfettered communism (we imagine - no one has ever tried either) that means we must have unfettered capitalism as is no intermediate point could potentially serve us better (even though every society that esisted always sat somewhere in the middle of the spectrum rather than at any end point)

But that's a topic for a different thread.

Edited by imaginesix, 23 April 2019 - 14:48.


#18 Fat Boy

Fat Boy
  • Member

  • 2,184 posts
  • Joined: January 04

Posted 23 April 2019 - 19:50

I certainly wasn't suggesting "excessive body roll" as the best approach. Where moveable aero is allowed (FSAE being one rare example) a tilting wing (about the x axis) would be a possibility.

 

Fair enough. I think I remember that Car & Driver had a 'skidpad challenge' many years ago and someone had a small formula car with something like a sail attached which acted laterally. It was a little hokey, but I think it was worth quite a bit of time.



#19 Ross Stonefeld

Ross Stonefeld
  • Member

  • 62,371 posts
  • Joined: August 99

Posted 24 April 2019 - 07:37

Forgive me if I'm misunderstanding what we and you are talking about a little, but how much asymmetric downforce could you get away with? 

 

Take Road America. 7 rights 4 lefts. But your rights lead on to and end the front straight(turn 14, turn 1) and back(Carousel, Canada corner), and start the remaining straight(T3). So you'd be mostly losing speed in areas you'd have a better chance of defending(a suboptimal T5 probably doesnt expose you in T6) and not exposed in a tactical functional sense. Would it be worth a turn of front wing on one side vs the other? Or is my car now squirreling around under braking. 



Advertisement

#20 gruntguru

gruntguru
  • Member

  • 6,712 posts
  • Joined: January 09

Posted 26 April 2019 - 04:02

Forgive me if I'm misunderstanding what we and you are talking about a little, but how much asymmetric downforce could you get away with?  

Not sure if asymmetric is the best word for it - suggests vertical DF but loading the tyres on one side more than the other.

 

"Canted", non-vertical DF is the thread topic i.e. a combination of downforce and sideforce.

 

It is definitely used in a fixed-geometry configuration on ovals. On dirt ovals in particular, the drifting attitude of the car allows the wing assembly (with large end plate) to provide massive sideforce in corners but little on the straight.



#21 Fat Boy

Fat Boy
  • Member

  • 2,184 posts
  • Joined: January 04

Posted 26 April 2019 - 19:01

Forgive me if I'm misunderstanding what we and you are talking about a little, but how much asymmetric downforce could you get away with? 

 

Take Road America. 7 rights 4 lefts. But your rights lead on to and end the front straight(turn 14, turn 1) and back(Carousel, Canada corner), and start the remaining straight(T3). So you'd be mostly losing speed in areas you'd have a better chance of defending(a suboptimal T5 probably doesnt expose you in T6) and not exposed in a tactical functional sense. Would it be worth a turn of front wing on one side vs the other? Or is my car now squirreling around under braking. 

 

Asymmetry is different that what Grunt is talking about. At RA, you can run quite a bit of asymmetry and go well. Cambers, crossweight, split flaps, etc. The only place it hurts is Billy Mitchell and that's usually offset by the gains. At some point, though, braking, and stability in general, does suffer.

 

What the real discussion here is about is putting the wings at a substantial angle (say 45 deg. in front view) to provide a sizeable component of lateral force. We have completely ignored how the car would respond in a straight line, much less on the brakes. It'd likely be awful.



#22 Ross Stonefeld

Ross Stonefeld
  • Member

  • 62,371 posts
  • Joined: August 99

Posted 27 April 2019 - 06:57

I'll load up iRacing, mangle a front wing, and let y'all know how I get on.



#23 Kelpiecross

Kelpiecross
  • Member

  • 1,447 posts
  • Joined: October 10

Posted 29 April 2019 - 10:41

Not sure if asymmetric is the best word for it - suggests vertical DF but loading the tyres on one side more than the other.

 

"Canted", non-vertical DF is the thread topic i.e. a combination of downforce and sideforce.

 

It is definitely used in a fixed-geometry configuration on ovals. On dirt ovals in particular, the drifting attitude of the car allows the wing assembly (with large end plate) to provide massive sideforce in corners but little on the straight.

  Sprint car wings have such low aspect ratios (by aircraft standards) that I think they would have little effect without massive endplates  to minimise "tip losses".  Even so, they run at such steep angles that surely they must be in a stalled  condition.   I read somewhere that the main effect of a sprint car wing is the drag which transfers weight onto the back wheels aiding traction. 

  Maybe a bit like the 300SLR of the mid-fifties  which had a simple air brake which Moss thought aided traction and stability in corners.   



#24 Greg Locock

Greg Locock
  • Member

  • 5,643 posts
  • Joined: March 03

Posted 29 April 2019 - 11:52

Most f1 wings run stalled.



#25 Nemo1965

Nemo1965
  • Member

  • 5,524 posts
  • Joined: October 12

Posted 29 April 2019 - 12:07

It would be interesting to hear Nigel Beresford's opinion on this, a valued but not so frequent poster on the Nostalgia Forum. He worked for Tyrrel in the late 80's and early 90's, I have asked him several questions about rake-angles but in less mathematical terms.

In the thread about Tyrrel (here: https://forums.autos...na#entry8657812) he had some interesting remarks about the optimal angle of a car with a difusor. Worth looking at.

And if I am not mistaken he has either worked recently or is currently working in Formula E....

Edited by Nemo1965, 29 April 2019 - 20:59.


#26 Kelpiecross

Kelpiecross
  • Member

  • 1,447 posts
  • Joined: October 10

Posted 29 April 2019 - 14:12

Most f1 wings run stalled.

 

 Would they produce any downforce if stalled?  An aircraft's wings don't produce lift when stalled - that's why they crash.      



#27 PayasYouRace

PayasYouRace
  • RC Forum Host

  • 19,157 posts
  • Joined: January 10

Posted 29 April 2019 - 16:11

Most f1 wings run stalled.

 

Only at the highest speeds, to reduce drag on long straights. They'd be pretty useless if they were stalled in the corners, as that's no more downforce.



#28 ehagar

ehagar
  • Member

  • 6,526 posts
  • Joined: April 01

Posted 29 April 2019 - 18:36

 Would they produce any downforce if stalled?  An aircraft's wings don't produce lift when stalled - that's why they crash.      

 

 

No. From a pilot's perspective a stall is the point in which weight has exceeded lift, meaning loss of altitude. There is still plenty of lift (or downforce in the case of cars), it's just exceeded the aerofoils critical point and any increase in Angle of Attack does not increase lift. In most cases the pilot has no option but to reduce AoA...



#29 Fat Boy

Fat Boy
  • Member

  • 2,184 posts
  • Joined: January 04

Posted 29 April 2019 - 19:59

Only at the highest speeds, to reduce drag on long straights. They'd be pretty useless if they were stalled in the corners, as that's no more downforce.

 

 Would they produce any downforce if stalled?  An aircraft's wings don't produce lift when stalled - that's why they crash.     

 

Most f1 wings run stalled.

 

 

Racecar wings are almost always running at or close to a stalled position, at least when in a high-downforce configuration. All this means is that the air is separating from the lower surface of the wing. You can still make max downforce with a stalled wing, but it's just not particularly efficient. What it does do is add drag without adding downforce. If you're looking to improve braking and/or you can add front, a stalled rear wing (high above the CG) may very well be exactly what you need.

 

This concept, as well as laminar flow, is often misunderstood in race car circles.



#30 Greg Locock

Greg Locock
  • Member

  • 5,643 posts
  • Joined: March 03

Posted 29 April 2019 - 22:22

 Speed has relatively little effect on the the stalled or unstalled status of the airfoil. Beyond the stall point lift can be made to increase - it is not a given, at a great cost in drag. Figure 8 in this interesting paper shows lift increasing up to 45 degrees of angle of attack. Sadly no drag figures.

 

https://people.eng.u...ings/19/100.pdf



#31 PayasYouRace

PayasYouRace
  • RC Forum Host

  • 19,157 posts
  • Joined: January 10

Posted 30 April 2019 - 07:04

Racecar wings are almost always running at or close to a stalled position, at least when in a high-downforce configuration. All this means is that the air is separating from the lower surface of the wing. You can still make max downforce with a stalled wing, but it's just not particularly efficient. What it does do is add drag without adding downforce. If you're looking to improve braking and/or you can add front, a stalled rear wing (high above the CG) may very well be exactly what you need.

 

This concept, as well as laminar flow, is often misunderstood in race car circles.

 

 

The phrase, "or close to", is the key there. They'll be running close to the stall so that under certain conditions they will benefit from the stall condition, such as on a long straight or under braking. But it'll be useless if running stalled at all times because there'd be no downforce benefit. In a world where L/D is usually key, there's very few situations where you'd want to compromise L with a big helping of D.



#32 Greg Locock

Greg Locock
  • Member

  • 5,643 posts
  • Joined: March 03

Posted 30 April 2019 - 08:39

OK, can we see some data on this? From memory the Ferrari 2000 rear wing was running an L/D of about 3 in its high downforce configuration, implying it was well beyond the linear range.



#33 mariner

mariner
  • Member

  • 1,809 posts
  • Joined: January 07

Posted 30 April 2019 - 08:54

With reference to Kelpiecross's comment on Sprint car wings being a drag deice to transfer weight rearwards I might be wrong but I don't think that is the objective due to the excess rear weight transfer sprint cars get mechanically. They have very short wheelbase (84 inches) and a high CG plus most of the  weight is either the motor well back or fuel way back. They can easily pull wheelie on restart and the drivers try to avoid this so aero load backwards would worsen things. The whole wing can be slid back and forth in the race by hydraulics and this is use as chassis tuning aid as the track condition changes.

o

On anther subject nobody has mentioned pure mechanical tyre grip being well over 1 g these days. I am too dumb to do the maths or see the logical flaw but if a tyre can generate a greater lateral force than the weight on it isn't it  better to apply what aero load you have through the tyre and , in effect, multiply t rather than just using it 1:1 sideways?



#34 Kelpiecross

Kelpiecross
  • Member

  • 1,447 posts
  • Joined: October 10

Posted 30 April 2019 - 10:16

No. From a pilot's perspective a stall is the point in which weight has exceeded lift, meaning loss of altitude. There is still plenty of lift (or downforce in the case of cars), it's just exceeded the aerofoils critical point and any increase in Angle of Attack does not increase lift. In most cases the pilot has no option but to reduce AoA…

 

  I think I see how it works a little better now.   A wing (or even a flat plate) will still deflect air downwards and gain lift from this deflection even when running at a steep angle and stalled.  But a "proper" wing running at the optimum angle also gains lift from the airstream that bends over the top of it and is also deflected downwards  - thus possibly even doubling the lift.     



#35 imaginesix

imaginesix
  • Member

  • 7,525 posts
  • Joined: March 01

Posted 30 April 2019 - 15:51

With reference to Kelpiecross's comment on Sprint car wings being a drag deice to transfer weight rearwards I might be wrong but I don't think that is the objective due to the excess rear weight transfer sprint cars get mechanically. They have very short wheelbase (84 inches) and a high CG plus most of the  weight is either the motor well back or fuel way back. They can easily pull wheelie on restart and the drivers try to avoid this so aero load backwards would worsen things. The whole wing can be slid back and forth in the race by hydraulics and this is use as chassis tuning aid as the track condition changes.

o

On anther subject nobody has mentioned pure mechanical tyre grip being well over 1 g these days. I am too dumb to do the maths or see the logical flaw but if a tyre can generate a greater lateral force than the weight on it isn't it  better to apply what aero load you have through the tyre and , in effect, multiply t rather than just using it 1:1 sideways?

 

It has been mentioned, and the maths has been done for you and it has been explained.

 

You'd be right if the available options were binary; either 90 deg or 0 deg rotation of the wing. But that's not the premise of this thread.


Edited by imaginesix, 30 April 2019 - 15:51.


#36 mariner

mariner
  • Member

  • 1,809 posts
  • Joined: January 07

Posted 30 April 2019 - 19:32

hanks 

 

It has been mentioned, and the maths has been done for you and it has been explained.

 

You'd be right if the available options were binary; either 90 deg or 0 deg rotation of the wing. But that's not the premise of this thread.

Thanks, I first saw the thread on my mobile when travelling and skipped yor  formula as I take time to understand them. Now I've looked properly I can see how it works.