Anonymous old friend wrote:As an engineer, my first reaction to your post (nice post!) would be: "I don't believe in what I see, much less in what I hear". Do you want to know if those techniques work for you? Easy: take a clock and a kart next weekend and try them.
Now, seriously:
Diagonal in the straights its the opposite of simple. This is going to be a VERY long explanation, so
I will answer only this point in depth. If I have time I will continue later (but I don't think so, this is going to be a very hard week for me), so the "shortening the curve" argument will be... very short.
The simple answer is not that simple:
a. In circuits with transition curves, Mr. Rob is wrong.
b. In circuits without transition curves, Mr. Rob is wrong.
That's why I love to write about racing, btw: to write things like the previous two paragraphs. :propeller:
What I do is to forget about scrubbing the tyres on the exit, think about weight transfer, take care of my throttle and
follow the rubber. I also check for the water traces (there will be dust there), that is, where the water runs when it rains and where is the transition at the exit. For that you need to develop a SERIOUS feeling for the rotation of the road. It is much more important than an illusory shortcut because of a diagonal. As you might know, the tangent of an angle close to 90 degrees is almost 1. THERE IS NO APPRECIABLE DIFFERENCE ON THE LENGTH. As for the scrubbing, my god: you're talking about a minimal twist of your wrist COMPARED with the "invisible twists" the road has and that (you'll be sorry) I'm going to explain in a gigantic email.
The reasons I do that are different from Mr. Rob's, and if he doesn't understand the
reasons, I doubt very much he is right. I might be wrong, but I'm never wrong. Once I thought I was wrong but... it was a mistake: I was right.
Bad jokes apart, there are straights and straights (a. and b.). I find this VERY hard to explain, most people is unaware of this even if
their bodies know (as I have written many times, it's like meeting a chick for the first time: your body
knows). I hope this email does something to improve your understanding of a track, that's why I'm going to spend a long time on it.
Let's take this step by step:
A road is made of three kind of "curves". First, second and third degree ones. By that I mean that the equation of the axis can be:
1. first degree (y = ax + b), that is a straight (duh)
2. second degree (y = (ax + c)^1/2), that is, usually, a circular curve (re-duh)
Nothing new here, isn't it?
Straights and circles. But then you have
3. third degree (y = ax^2 + bx + c), that is, a logarithmic spiral
or a cubic parabola.
NOT all roads (nor circuits) have third degree curves. Old roads were made ONLY of straights and circles. "New" roads (after 1970 or so, and NOT all) have logarithmic spirals between straights and circles.
Many old tracks do not have spirals. For example, Catalunya is made of straights and circles. It has one lonely cubic parabola, but it is "for the show". As it is very evident (as Monza's Parabolica) I have the picture at hand, so there you go:
Repsol curve: the straights edges are green, the circular curve edges are red and the cubic parabola edges are yellow.
To confuse drivers, the new sections at Catalunya, built four or five years ago DO have spirals. However, they are so subtle that is hard for regular drivers to pinpoint them on a picture like the previous one. They're hard to pinpoint even for the people that builds them! If you're into mechanics an example could be clarifying: it takes a couple of years of experience to detect
on sight how the ramps on a cam are made. It's the same with spirals.
So, why spirals? This is also going to be long and (I'm afraid) there is a 50/50 chance that I will leave you confused and exhausted.
Well, allow me to explain that, in a car it is IMPOSSIBLE to take the axis of a straight and then take a perfect circle because of Newton laws.
For that to happen you should keep your steering wheel completely straight and then, at the entrance of the curve (top left of the previous image, where the green edges change into red edges), you should move your steering wheel INSTANTLY, exerting an INFINITE lateral force on the car for it to come to a circular curve (actually, infinite power, but, hey...).
You should go from zero lateral acceleration to a constant lateral acceleration in NO TIME, in one point. Hence, you need infinite force (on the steering wheel) and infinite power (on the tyres) for such a trajectory.
So, the assumption Mr. Rob seems to make, that you
simply go straight and then you change direction is false.
Even
going straight on a straight is impossible. Allow me to explain this fine point, it will improve your racing (in the real world: simulators do not take this in account, that I know).
Most people assumes that when you're in a straight, your car is straight and your steering wheel is straight, isn't it?
For that to happen, the road should be FLAT and straight. If the road had a lateral slope, then you should move your steering wheel slightly to counteract that inclination.
Well, is a road flat?
No.
If you build such a road, when it rains that road would become a giant puddle because the water would not run because the road is flat. Like the old joke goes: "It's raining dogs and cats", "Well, you are right because I'm standing in a poodle!".
So, in any real road you're going slightly sideways. The road is not built flat ON PURPOSE. It has the shape of a roof. That is called "pumping" in Spanish (bombeo) and "crown" in English: the road is inclined, 2% or so, sideways, for water to leave the road when it rains.
Now, you know (if you didn't knew already) that a road is made of straights that have a slight lateral inclination. Moreover, in most modern circuits (but not all, Catalunya is a typical example) you have:
1. straights
2. logarithmic spirals (that is, curves that go from zero curvature to a definite one, like the trajectory of a coin that you roll on the floor) and
3. circular curves.
Having said all this you have to ask yourself: which is the LATERAL PROFILE of such a road? I mean, if you take a knife and cut the road lengthwise, what do you see?
1. In straights, you see the center axis and the two edges below it.
2. In logarithmic spirals you get a transition of the superelevation. You rotate the road around its axis to change the sideslope from a crown to a full superelevation at the entrance of the circular curve. So, one edge rises and the other sinks.
3. In circular curves you have a constant superelevation: the outer edge is over the axis and the inner edge is below it at a constant height along the curve.
So, a superelevation diagram of a road with spirals is something like this:
The green line is the axis on the straight viewed from the side, the yellow line is the spiral axis and the red line is the circular curve axis.
The dotted lines are the Edges of Traffic Lanes (ETL). On the straight (Points On Tangent or POTs)both edges are at the same height, hence you see only one dotted line. Where the Transition Spiral (TS) begins, the outer edge is flat (at the same height as the axis) but the inner edge have kept its two percent inclination. Where both edges have reached two percent is the Point On Spiral (POS). Where both edges have reached full superelevation is the Point on Curve (POC).
The idea of a well taken curve is to take advantage of this rotation as much as possible. You also want to "shorten" the curve, that is, to minimize the time you spend on the curve, where you, presumably, are going slow. Hence, you take the curve at the last possible moment: what's called a late apex. This means that you have an abrupt change in speed, when you're
slowest. Then, at the exit, you will have a loooooong, very looong acceleration. This means that the car will oversteer. So, you correct: that explains Mr. Maldonado reaction. Many drivers, like Schumacher, tap the brakes at this moment, to diminish weight transfer. Hence, more oversteer. So you "loose" your wrists for a moment, to compensate the huge oversteering you're inflicting on your rear tyres. Then, the exit: very smooth movements are required to eliminate understeer (remember, you're accelerating with all your might, so weight transfers
to the back, hence understeering). You dominate all this and you're on your way to victory, probably one second ahead of guys that do not understand all this. Either you
understand it or you
feel it. Preferably, both, at least down here in Latin America, where both ways are appreciated. It is a MUST to go around the track in a bicycle, on foot you do not feel the slopes, much less in a car. One of my favorite tricks is to include a slight longitudinal slope on the braking area and also to use a crown that goes 2 percent but not as a roof but uniform between curves to the same side (yeah , I know, it would be better if we had two beers, two bicycles and took a ride around a track, it would be 10 minutes for me to explain). End of story.
I truly, truly, have no time to explain more of this but there you go: the problem with circuits without spirals is that you rotate the road in the straight. That is, as a designer, when you reach the curve (green to red edges) you MUST have "developed" the full superelevation. HENCE, you rotate the road IN THE STRAIGHT. Here is the diagram. Study it: it means that IN THE STRAIGHT you have superelevation, so you, intuitively move your steering wheel IN THE OPPOSITE DIRECTION OF THE CURVE AT ITS ENTRANCE. That is why modern roads use spirals (and railroads since immemorial times: if you didn't the rail car would "jolt" at every curve entrance, as you probably understand by now).
If you don't believe me, find a road with circular curves (no transitions) and check yourself while you drive, you will notice that (keeping your lane its more evident that cutting the apex) you
counter-steer before the entrance, and only then you take full curve, right where the superelevation begins. Enlightenment will ensue.
I have to go to work, see you.
As you can see the thing is subtle. ALL formula one drivers understand that. Almost ZERO fans do it. Mr. Rob doesn't. If you compare the two percent gain you would have for keeping both wheels at the same height (right on the centerline) with the 30 or 40% difference in traction between rubberized and non-rubberized track AND the same amount between weight transferred/non transferrred (with a tap of the brakes) or the six percent influence of the superelevation on the straight that tracks without spirals bring to you at the entrance of each curve, you will quickly see the guy is pure baloney.
I think.
Finally, go and call venerable to your grandfather...
There is NO AUTHORITY in good engineering. :hihi:
