Maximum G-force actually achieved in corners

[hollus]

This year the red bulls famously took turn 8 at Istambul flat. I wondered about the G-force involved, the actual lateral acceleration. The most obvious answer is the G-force graphics we often see, but I am not sure of how accurate are those? They seem to stop between 4.5 and 5 Gs, and if we would believe them, braking goes over 5G all the time.

So, can we actually measure lateral acceleration? Turn 8 is a long aero corner, and this is my attempt at pulling out some very rough numbers.

I used this video from Webber's pole lap this year
http://www.mototube.pl/film/4655/flvide ... 521544.flv
to measure the time it takes. 7.4 seconds from turning in of the wheel to the wheel going straight again.

Then I used www.mapmyrun.com to make a rough trace through the racing line in a satellite picture: 530 meters of racing line, using the rubber marks to estimate turn in and turn out points.
I also used that satellite picture to measure the total turning angle: 218 degrees, but I suspect my computer stretched the image a tiny bit.

Let's take those three values at face value and assume that the corner is a perfect circumference arc, which it is not. I can work out these numbers:

Total circumference: 875.23 m
Corner radius: 139.30 m
Car velocity: 71.6 m/s (257.8 Km/h)
Fc = m v2 / r (centrifugal force)
Ac = v2 /r = 36.66 m/s2 (Centrifugal acceleration)
36.66 m/s2 = 3.74G

So I get 3.74G average. The corner is tighter through the first 3 apexes and opens up at the end, so this probably works out to very close to 4G for around 4 seconds at the beginning of the corner.

There, my puny attempt at actually measuring things with all the wrong tools. Anybody with better tools wants to improve on this? What is the (actual) maximum G we get through corners this year? Remember that Silverstone is coming, might it be there? Were there any higher (actual) cornering G forces in the past?

[Scotracer]

Copse generally has the highest cornering G force of the year.

[hollus]

Care to provide any (actual) numbers? Please!

[mep]

Nice work.
I think maximum g-forces will always be achieved during braking, simply because the ratio between downforce and car-mass will mainly affect how big your acceleration is. The highest ratio you have when you have highest speed. So during braking from high speed you get max. deceleration (g-force) for a initial moment. During cornering you get lower g forces but for a much longer time.

[speedsense]

Nice work.
I think maximum g-forces will always be achieved during braking, simply because the ratio between downforce and car-mass will mainly affect how big your acceleration is. The highest ratio you have when you have highest speed. So during braking from high speed you get max. deceleration (g-force) for a initial moment. During cornering you get lower g forces but for a much longer time.

Generally they are just about equal, the limitation is the tire and track surface. The same down force exists at initial braking as does cornering at the same speed.
Though cornering with a hill or banking or combination of both will generally exceed the braking capability on most "aero dependent" cars, where vertical G adds into the aero number.
On the acceleration end, with road racing cars, is almost always the lesser of the three... The traction circle is more of an oval shape with a "flat" side on the acceleration side.

[mep]

Generally they are just about equal, the limitation is the tire and track surface. The same down force exists at initial braking as does cornering at the same speed.

speedsense on the first glance you are right but you missed one major component wich is drag. Aerodynamic drag is big enough to help slowing the car down without the need to transmit forces through the tire. So braking will win unless you are on a roller coaster.

Though cornering with a hill or banking or combination of both will generally exceed the braking capability on most "aero dependent" cars, where vertical G adds into the aero number

It seems like you search for exceptions from the rule.
Ok you corner trough a banked corner, I will brake uphill.
What you can do is mounting vertical wings who give just horizontal forces without the need to transmit them trough the tires.

On the acceleration end, with road racing cars, is almost always the lesser of the three... The traction circle is more of an oval shape with a "flat" side on the acceleration side.

Isn't that due to the fact that you usually accelerate just with either front or rear tires? The effect should not be there on a four wheel driven car, because for the tire it should not be a big difference if you brake or drive.

[WhiteBlue]

Copse generally has the highest cornering G force of the year.

I saw them doing 5.0 there and in reality it could have been more as the FOM scale isn't any higher. Huge downforce this year.

[ISLAMATRON]

I was wondering if the scale maxed out at 5

[hollus]

That was partly my point. The scale maxes up at 5.0G... and often the measurement is there for a while... what does not correspond to reality!

Look at my analysis of turn 8. I get much less than the meter in the screen. Also, if you look carefully, you will see the G-force meter marking 0.3G to the left (average) in the middle of the straights, what again makes no sense.

I suspect those G-force indicators are exxagerated and extremely noisy, drivers would be complaining a bit more if they has more than 5 negative Gs a couple of hundreds of times per hour, I would think.

In silverstone, Webbers lap in the simulator went like 3.9G, 4.4G, 4.2G, 3.5G and so on, in the screen, 4.5, 5.0, 4.5, 5.0...

Should I change the thread title to "are the accelerometers from the FIA reliable?" or to "G-forces are overrated"?

[timbo]

Look at my analysis of turn 8. I get much less than the meter in the screen.

As you pointed yourself, you calculated average acceleration over the curve, so maximum figures can be quite higher.
In the book about Ferrari F1-2000 car, Peter Wright shows telemetry records from several Schumacher's qualifying laps from 2000, and he reached 4Gs lateral at Imola.
Those cars have much less downforce when modern F1 cars.
I'm certain 5G is possible.

[hollus]

Anyone playing RFactor wants to pitch in? With the right mod, if your times compare well to actual F1 times, probably the tracks are accurate enough to measure this. What G forces are you getting?

[timbo]

Anyone playing RFactor wants to pitch in? With the right mod, if your times compare well to actual F1 times, probably the tracks are accurate enough to measure this. What G forces are you getting?

I play a bit. On BMW Sauber 2007 (considered pretty good model AFAIK) I'm getting pretty close to 5Gs (like 4.8-4.9)in Copse and Maggots/Becketts.

[marcush.]

Sampling rate is also a big filter not to be forgotten .
I ´d believe the FIA screen does not really have a high sampling rate so if it were 5gs we were shown the real gs could be considerably higher but we would not even see them becuse they have not logged when they happened...So these official graphs may not really mean a lot ..
of course I dn´t know their logging rate...someone does??

[Jonsson]

The accelerometers themselves are very reliable, and very accurate. All Formula 1 cars use a Model 4203 Triaxial Accelerometer built by Measurement Specialties. They were originally designed for missile systems and record a tri-axial G-force measurement 20 times per second. It also has a 30 G maximum range.

I think the focus on this discussion should evolve towards how many lateral Gs the body can handle before it passes out. That is what the FIA keeps in mind when coming up with new regulations every year. Humans can withstand about 9 positive (downward) Gs for about 30 seconds before passing out. Lateral Gs are different. The human brain cannot withstand 5 lateral Gs for more than a few seconds without losing consciousness.

One factor to think about is whether there is a slight bank around the corner. This transfers a significant amount of lateral G into positive G, so if the total lateral G force on-screen says 3.75, and there is a bank, the driver is experiencing added positive Gs that could definitely amount to more than 5 total Gs of force in a laterally-downward vector.

[speedsense]

Generally they are just about equal, the limitation is the tire and track surface. The same down force exists at initial braking as does cornering at the same speed.
speedsense on the first glance you are right but you missed one major component wich is drag. Aerodynamic drag is big enough to help slowing the car down without the need to transmit forces through the tire. So braking will win unless you are on a roller coaster.

The same drag is present in cornering at the same speed. As is engine drag if you lift the throttle.
My information comes from years of analyzing data on high down force race cars. (Formula Atlantic and GTP cars,etc.) You can make all the assumptions you want. Data doesn't lie or make mistakes....

Though cornering with a hill or banking or combination of both will generally exceed the braking capability on most "aero dependent" cars, where vertical G adds into the aero number

It seems like you search for exceptions from the rule.
Ok you corner trough a banked corner, I will brake uphill.
What you can do is mounting vertical wings who give just horizontal forces without the need to transmit them trough the tires.

The original OP's question about Max G's achieved, is what I'm addressing...

"Exceptions from the rule"...which rule is that? How about this rule, data acquisition measures what is really happening and leaves little that you need to "search" for....

So we are only talking race tracks that are flat? Not many of those in F1.
The influences of hills, downhills, corner banks etc, will always contort the traction circle. If all the braking was up hill, the traction circle would look like a tear drop... not reality, but the point is there..

Most racetracks have at least one corner, one braking area where the G's get Maxed out, over the rest of the race track, (the OP's question)

On the acceleration end, with road racing cars, is almost always the lesser of the three... The traction circle is more of an oval shape with a "flat" side on the acceleration side.

Isn't that due to the fact that you usually accelerate just with either front or rear tires? The effect should not be there on a four wheel driven car, because for the tire it should not be a big difference if you brake or drive.

part of the reason but...

..when you brake, the majority of the braking force is on the front of the car, even when biased to the rear. And when you corner, the majority of the weight is on the outside loaded wheels. It's all relative...
Also four wheel drive cars, do not change the curve as much as you think it does. Weight transfer still moves to the rear of a car...very observable in DA...

Brakes have way more energy than the Horsepower equivalent energy especially with downforce, the engine has to push that as well and works against acceleration unlike braking where it works for it.
To make the traction circle round again on an F1 car you would need probably around 1200-1400 horsepower with the same DF.... but the tires will always limit the g force and the traction circle, thus keeping the oval flat side acceleration curve.