32 replies to this topic

[wegmann]

Does anyone have current info for the contribution of drag to an F1 car's deceleration? Specifically, I'd like to know how much drag force there is at Monza vs. Monaco.

Thanks in advance for any helpful data.

[dc21]

I might be wrong but I thought that since peak downforce at each track is the same, peak drag would also be the same. If you are talking about drag at a specific speed then ??????

Dave

[dolomite]

Figures from an interview with Gary Anderson, quoted in the latest issue of Race Tech:

Monaco 1200kg downforce at 150 mph, lift/drag ratio 2.85-2.9:1
Montreal lift/drag ratio 3.1:1
Monza 1000kg downforce at 150mph, 1200kg at 220mph

[dosco]

Check out Gorgio Piola's books (Formula 1 Technical Analysis).....he has some interesting numbers in there.

Dolomite's numbers look about right, from my recollection, though.

dosco

[Bladrian]

Those downforce figures point out every thing that is wrong with current racing car configurations. With over a ton of downforce on a car, there can be very little, if any, suspension compliance - which knocks any theories regarding relevance to passenger car development into a cocked hat.

[John Galt]

Originally posted by dolomite
Figures from an interview with Gary Anderson, quoted in the latest issue of Race Tech:

Monaco 1200kg downforce at 150 mph, lift/drag ratio 2.85-2.9:1
Montreal lift/drag ratio 3.1:1
Monza 1000kg downforce at 150mph, 1200kg at 220mph

I thought that downforce was one of those figures that increased exponentially in relation to velocity. Those Monza figures seem to belie that. Can someone clarify?

[munks]

dc21 - i don't think its true that peak downforce is the same at every track ... although i imagine its not too far off since tracks with low-downforce configurations tend to have higher top speeds.

most of those numbers look right, but i agree with Mr. Galt (ahh, Atlas Shrugged on Atlas F1) that the Monza #s don't seem to make sense ... only 200kg more for another 70mph? the equation for downforce and drag uses velocity squared (although this ignores the effects of the lower ride-height at the end of the straight ... shouldn't matter *THAT* much).

anyway, i have one of piola's books and it gives the coefficient of lift and drag, but doesn't give the frontal area which would be needed to compute the actual force. adding to the confusion, the frontal area would be different at each track, mostly depending on the rear wing size. does anybody know the frontal areas of an f1 car?

[desmo]

Firstly let make me make clear that I am the furthest thing from an aero expert, but it is my shaky understanding that aero effects (lift, drag) being a function of velocity squared really only approximates the relationship in complex real world conditions with things like small and large scale turbulence, boundary layer seperation areas moving about and ground effects. I would expect aeromaps would show significant deviations from the simple calculated values due to these non-linear effects and probably others I know nothing about, otherwise aeromapping would be as simple as running the car in a given aero trim at one speed and then extrapolating the downforce at all other velocities. I don't think it is done that way.

[wawawa]

Somebody please correct me if I'm wrong, but my understanding is that downforce is really necessary only at the corners - having an additional 1200kg of mass generating downforce on a straight is a tremendous waste.

I know that the rules prohibit movement of aero surfaces, but does anyone know if an attempt was made to get around this problem by designing a fixed aero surface that induces a "non-uniform" airflow? By this, I mean that the aero is designed so that as the speed of the car increases, the airflow around the car suffers a kind of "discontinuity", so that instead of generating more and more downforce, the aerodynamics behave in a totally different way (above a certain threshold), reducing downforce.

I believe that some kind of "discontinuity" occurs at supersonic speeds (just hear me out ) - the Bell XS-1 project had to deal with this while trying to break through the sound barrier. I wonder if it is at all possible to design a surface that deliberately induces some kind of rupture at lower speeds in order to avoid wasting enormous amounts of power lugging an extra-heavy car down the straights.

[TheD2JBug]

wouldn't the drag depend on how good your aero package is ?. For example the 1994 Tyrrell was regarded as having a really neat aero package so they ran less wing = less drag for the appropiate level of downforce. Thus Katayama could hang with the Renaults despite being at a power disadvantage

[dolomite]

I perhaps over-simplified the data supplied in the original article, which is somewhat more ambiguous if you examine it more closely. Here is an extract from the actual words:

"An efficient downforce setup is around the Montreal level, where you are running reasonable wing on the car - nothing barn door-ish. That is where you might get 3.1:1. You might drop down to 2.85/2.9:1 at Monaco."
How much downforce would you see at 150mph at Monaco?
"That is where you would see (the) 1200kgs."
At Monza?
"What really happens is that at Monaco you do 280kph (175mph) top speed and at Monza you do 350kph (220mph) top speed. It is the drag that sets that. You end up close to the same sort of downforce at 350kph at Monza as you see at 280kph at Monaco. At Monza at 150mph you would be lucky to see 1000 kilos."

[munks]

aha, so we can do some very rough calculations based on dolomite's extract ... (and, once again, assuming drag/downforce go up with velocity squared, which desmo pointed out is an approximation)

if we're getting 1200kgs @ 150mph at monaco, then we'd get about 1633kgs @ 175mph, so ...

if we're getting 1633kgs @ 220mph at monza, then we'd get about 759kgs @ 150mph ("lucky to see 1000kg" is an understatement, then).

but he also said that its the drag that sets the top speed. since horsepower drained by drag goes up with velocity *cubed*, maybe these calcs aren't very accurate.

[Jordan191]

Originally posted by munks
aha, so we can do some very rough calculations based on dolomite's extract ... (and, once again, assuming drag/downforce go up with velocity squared, which desmo pointed out is an approximation)

if we're getting 1200kgs @ 150mph at monaco, then we'd get about 1633kgs @ 175mph, so ...

if we're getting 1633kgs @ 220mph at monza, then we'd get about 759kgs @ 150mph ("lucky to see 1000kg" is an understatement, then).

but he also said that its the drag that sets the top speed. since horsepower drained by drag goes up with velocity *cubed*, maybe these calcs aren't very accurate.

I think we are talking drag , not downforce. the drag is heavily dependant on how efficent one's aerodynamic package is. I.e a more efficient package uses less drag to get the required downforce , etc.

[Jordan191]

ooops
my mistake the range on a single CAR

[wegmann]

Yes, I'm mostly interested in the drag force for a single, average, 2002 car, and the ratio from high- to low-drag setups. Thanks dolomite for the extract - I can now calculate the drag at Monaco, but there doesn't seem to be enough info to determine the same info for Monza accurately enough.

I once heard that the cars were configurable enough to run roughly half the drag at some tracks, but that was years ago and I don't know how true it is today.

Anyone else seen any data?

Thanks!

[Wuzak]

Desmo is correct in that the drag = constant x velocity² is only an approximation. It cannot possibly take into account local variables.

I'm not sure that lift works in the same way, though.

[Wuzak]

As for drag force, you can get a good approximation if you know the top speed and the power of the engine.

Power = Force x Velocity

If we say the Renault has a power of 825hp, and its top speed at Monaco was 175mph, we get a drag force of

F = 802kgf

Now we can apply the approximation:

F = C x V²

Which gives us a C value of C = 1.285kg/m

Therfore, at 150mph, with Monaco settings, the drag force is:

F = 589kgf


At Monza,

V=220mph

At 220mph, F = 638kgf

Therefore, C = 0.647kg/m

At 150mph, F = 296kgf



For a car with Monaco settings to run 220mph, it would require 1639hp, or nearly double what it has!



No doubt someone will prove me totally wrong!

[munks]

good calcs, Wuzak. like you said, this approach in an approximation. i can think of a couple things that probably make it less accurate:

1) the top speed on Monaco may be limited from running out of track, rather than actually reaching a point where the various drag forces exactly match the thrust forces.

2) the engine's rated horsepower doesn't ALL contribute to thrust. some of it is sucked up by alternators, pumps, intake/exhaust, and other drivetrain losses. i'm not sure which of these things are included when the engines are on the dyno.

[Wuzak]

Good points Munks.

There is also the issue of gearing - does the maximum speed come at the same rpm as maximum power?

[Jacaré]

Originally posted by wawawa
I believe that some kind of "discontinuity" occurs at supersonic speeds (just hear me out ) - the Bell XS-1 project had to deal with this while trying to break through the sound barrier.

The supersonic discontinuity is a shock wave, something which only occurs if the source of pressure waves is travelling faster than the waves. At lower speeds, the air retains its elastic properties making such discontinuities impossible.

The speed of sound (pressure waves) isn't constant: it depends on the pressure and temperature of the air. In higher pressure zones, for example on the front edge of the wings of an aircraft like the Bell XS-1, the speed of sound is higher. As the aircraft nears the speed of sound, it enters a transonic zone where some parts of the plane are travelling supersonic and other parts subsonic, depending on the density of the air surrounding that part of the aircraft. This transonic transition phase causes shock waves to emanate from different parts of the aircraft, and they interact in a way which is most difficult to predict. Breaking through the sound barrier isn't a smooth process. Once all parts of the plane are supersonic the aerodynamics are easier to understand.

[Wuzak]

I believe that the Hawker Typhoon had just such a problem.

The wing roots were quite thick, and the air travelling over them could go supersonic at certain aircraft speeds.

They had some wing failures as a result, I think.

[DOHC]

Originally posted by Jacaré
As the aircraft nears the speed of sound, it enters a transonic zone where some parts of the plane are travelling supersonic and other parts subsonic, depending on the density of the air surrounding that part of the aircraft.

Today's commercial jet airliners always operate in the transonic regime. There is usually a supersonic "bubble" both above and below the wing. Sometimes, in favorable lighing conditions you can see the shock along the wing, with the naked eye. It is visible as a thin shadow on thewing surface.


Otherwise, I'm with desmo here, the C*v^2 model is an approximation. It will be a good one for a wing, without any disturbances, but an F1 car is rather far from a wing profile. Tracelling at different speeds, you can expect it to produce different levels of turbulence, and in different areas. So I too believe that you need extensive aero mapping to find out what's going on.

Originally posted by wawawa
avoid wasting enormous amounts of power lugging an extra-heavy car down the straights.

I'm not sure if I interpret you correctly here, but the whole point with downforce from wings is that you don't have to lug an extra-heavy car down the straights. Downforce has a drag penalty but no mass penalty. This means that you don't tap any power from the energy for accelerating the car, but you do tap energy to puch it through the air.

Because a wing in plenty-of-flaps configuration has a lift/drag ratio of 4:1 or so, you see that you get four times more downforce than the drag penalty. That extra downforce also means that you get more traction: you can floor on the gas pedal with little or no risk of causing the rear wheels to spin as they have extra grip due to the downforce. And yet, there is no mass involved, no extra mass to accelerate, no mass to lug along the straight. All you need to do is to pay a 25% or so tax for the drag that comes along with the added downforce. That's a good deal, in accelerating, cornering, and braking.

There's almost no disadvantage with downforce at all. Without wings and downforce, you could theoretically increase top speed, but that will only have an effect if straights are very long. And in braking and cornering you lose. So you put wings, more wings on slow circuits, less on fast circuits. But you always use wings. The 4:1 ratio is hard to beat. You pay a quarter, and get the lot. It doesn't even impede acceleration. It's as close to a free lunch you can get. Venturi tunnels are of course better, but they are forbidden, maybe just because they would be superior.

[GillesVilleneuve]

In Piola's Formula 1 Technical Analysis 2000, there is a table on downforce, which I quote:

Monaco 1999
Lift coefficient = 2.661
Drag coefficient = 1.000
Efficiency = 2.66
Velocity = 290 km/hr
Downforce = 1478 kg

Monaco 2000
Lift coefficient = 3.137
Drag Coefficient = 1.084
Efficiency = 2.89
Velocity = 292 km/hr
Downforce = 1694 kg



Monza 1999
Lift coefficient = 1.845
Drag coefficient = 0.680
Efficiency = 2.71
Velocity = 347 km/hr
Downforce = 1435 kg

Monza 2000
Lift coefficient = 2.029
Drag coefficient = 0.682
Efficiency = 2.98
Velocity = 349 km/hr
Downforce = 1513 kg

*Edit, sorry about that mistake guys, I got mixed up taking the values because of the wrong headings*

[DOHC]

Although the post above gives some indication of the range of drag from Monza to Monaco, I think those figures should be taken with (a large) pinch of salt. There are several reasons.

1) Are we to believe that in 1999 the drag coeff is 2.66 at Monaco and 2.71 at Monza? If the drag coeff is up, you run a higher drag config at Monza. I doubt it. Still, if we look at the 2000 figures, a similar spurious figure is proposed: 2.89 at Monaco and 2.98 at Monza. I think these figures are phony.

2) The "efficiency" measure obtained by computing a lift to drag ratio is something you do for wings, not for entire cars. The point is that an F1 car has

- bodywork that only contributes to drag but produces no downforce
- wings that produce downforce at a drag penalty, presumably with a L/D of 4 to 5.
- an underbody including diffusor that produces downforce at very little drag penalty.

Once you have designed your car, you can't do anything about the first item. When it comes to finding the right aero settings, this item therefore has nothing to do with "aerodynamic efficiency." It's a matter of aero efficiency only to the extent that a more slippery body will require less horespower to push it through the air at high speed. It's got nothing to do with downforce. (Consider front wheels, for example. They're on the car, they have specified dimensions, they add to drag but don't produce downforce.)

Then you have the third item. This can vary a little by adjusting ride height. Diffusors are also modified a little bit but this seems to be improvement over the season rather than special designs and setups for different tracks.

That leaves us with the second item, wings, that are indeed configured diferently for different tracks. Wings, I would say, is the only factor that has anything to do with "aerodynamic efficiency" with respect to high or low-speed tracks. The wing configuration and settings affect top speed, braking, cornering, acceleration. A typical trick is to use a little more wing in qualifying, as you then improve cornering. In the race, however, you run a little less wing because you might often not run in clean air (other cars ahead of you) and you might want a slight edge in top speed, as overtaking is often done by outbraking your opponent at the end of a (long) straight.

3) My third objection is that the figures are produced by Enrico Benzing, who is perhaps better known for his engine hp output measurements derived straight from speed and rev data (as I understand it the latter are measured by accoustics alone). Although you can get estimates this way, their accuracy won't be terribly high. My guess is that Benzing's aero data, as reproduced in Piola's book, are inferred in a similar way.

I think the Piola/Benzing figures at best are estimates, but the inconsistencies concerning drag at Monaco and Monza show that they can't be relied upon.

Also, the "efficiency" puts bodywork and tires into the L/D ratio, and I can't see that such a figure makes any sense at all. You always want less drag, and you always want a certain amount of downforce, sometimes more, sometimes less. You pay for the downforce in terms of increased drag, but the bodywork and tire drag is not part of what you pay for the downforce.

[Jacaré]

Originally posted by DOHC
1) Are we to believe that in 1999 the drag coeff is 2.66 at Monaco and 2.71 at Monza? If the drag coeff is up, you run a higher drag config at Monza. I doubt it. Still, if we look at the 2000 figures, a similar spurious figure is proposed: 2.89 at Monaco and 2.98 at Monza. I think these figures are phony.

I don't really follow this but aren't the drag coeffs for Monza higher than Monaco because the car is travelling significantly faster?

[Ben]

Originally posted by Jacaré

I don't really follow this but aren't the drag coeffs for Monza higher than Monaco because the car is travelling significantly faster?

The drag coefficient is calculated by the following equation;

Cd = F / (1/2*a*rho*V^2)

So the effect of speed is considered relative to the force at that speed.

I, like DOHC, am also interesed with the higher number at Monaco. The only thing I can think, is that the frontal area will be higher with monaco wings. This could make the product Cd*a larger than at Monza.

Ben

[wegmann]

Thank you Gilles! ... I may be mistaken but it appears you switched the Drag and the Efficiency. This should explain the discrepancies pointed out.

[DOHC]

Originally posted by wegmann
Thank you Gilles! ... I may be mistaken but it appears you switched the Drag and the Efficiency. This should explain the discrepancies pointed out.

I think that is probably true. Piola's book is unclear on this point, as the tables have incorrect headings. The "efficiency" is listed under "S" for speed, while the headline "Vel.km/h" is used for speed. There is no headline for "efficiency" at all.

But if we change the interpretation to yours, wegmann, the data make a lot more sense: Italy becomes the configuration with the least drag.

[Jacaré]

Originally posted by Ben


The drag coefficient is calculated by the following equation;

Cd = F / (1/2*a*rho*V^2)

So the effect of speed is considered relative to the force at that speed.

thanks for the explanation. I found a nice little table showing some Cd values for typical shapes

[colejk]

Originally posted by munks
good calcs, Wuzak. like you said, this approach in an approximation. I can think of a couple things that probably make it less accurate:

1) the top speed on Monaco may be limited from running out of track, rather than actually reaching a point where the various drag forces exactly match the thrust forces.

2) the engine's rated horsepower doesn't ALL contribute to thrust. some of it is sucked up by alternators, pumps, intake/exhaust, and other drivetrain losses. i'm not sure which of these things are included when the engines are on the dyno.

With regards to 1) Remember that the cars are geared differently for most tracks, so by the end of the straight there they should be close to redline in top gear. F1 cars are mostly geared for acceleration rather than top speed, which is why if you notice on long straights they rarely accelerate any more towards the latter part of the straight. No doubt that even with Monaco settings and different gearing they could reach a higher speed on a longer straight.

[Jacaré]

Originally posted by colejk


With regards to 1) Remember that the cars are geared differently for most tracks, so by the end of the straight there they should be close to redline in top gear. F1 cars are mostly geared for acceleration rather than top speed, which is why if you notice on long straights they rarely accelerate any more towards the latter part of the straight

is the topgear different to the lower gears with regard to redline just in case the driver gets in the slipstream and can reach a higher max revs in top gear?

[AndreasNystrom]

I guess Monza needs different cooling of the car? cause it runs on pretty much high revs all the time, which could produce more drag? (bigger radiator ducts etc)

[Wuzak]

Originally posted by Jacaré

is the topgear different to the lower gears with regard to redline just in case the driver gets in the slipstream and can reach a higher max revs in top gear?

I think that the cars are geared to achieve their maximum speed at the rpm of maximum power. There should be a margin for extra revs when the car gets into another car's slipstream, which essentially reduces the drag force.