Figure 1: This G-G plot shows longitudinal acceleration vs. lateral acceleration, and is one way to see how well the rider is using all the available traction. The ideal G-G plot would have all its data points clustered about the perimeter, indicating the rider is always accelerating, braking or cornering as much as possible.
Earlier we discussed the G-G plot (an X-Y plot of lateral and longitudinal acceleration) and the traction circle and how it combines the two acceleration values into an easy-to-read format. The G-G plot, shown here in figure 1, graphs the two forces combined for every point on the racetrack and is useful for a quick evaluation of the overall lap. Ideally, the plot would consists of points only on the outer edge of the traction circle, with no points filling the graph – in other words, the rider is always using the maximum traction available by accelerating, braking or turning at any given point on the racetrack. On the G-G plot, this measure of how much traction is being utilized – usually called Total G – is viewed as how far the point is from the center of the plot and can be calculated by summing the lateral and longitudinal accelerations using vector addition and the Pythagorean theorem:
Figure 1: Data from Mazda Raceway Laguna Seca, with GPS speed in black. In the middle, the lateral acceleration channel is shown in red. At the bottom and in yellow, the lateral acceleration channel is modified to show all positive values for cornering.
Just as it is easier to view both acceleration G and braking G as separate, positive values rather than looking at the raw longitudinal G data for both, it is often useful to modify the lateral acceleration channel in a similar manner so that all values are positive. This requires a simple absolute-value modification of the lateral G channel, and we’ll call the new channel corner G:
Where abs () is the absolute value function. Your software package most likely includes the abs function, but some creative addition and subtraction can yield the same results. This channel is shown in Figure 1, using the same data as used in the previous braking G and acceleration G examples. Again, it is easier to pick out various characteristics using the Corner G channel such as maximum lateral acceleration in each turn as well as how quickly the trace ramps up to a maximum or trails off at the exit of each corner.
A lap of Mazda Raceway Laguna Seca, showing speed (black), acceleration G (green) and braking G (red). The acceleration and braking G channels are derived from the longitudinal acceleration data (blue).
Two channels can be generated by dissecting the longitudinal acceleration channel into its positive and negative components. The first channel, which we will call Acceleration G, shows only the values for positive longitudinal acceleration – when the bike is what we commonly refer to as accelerating. The second channel we’ll call Braking G, and shows only the values for negative acceleration – when the bike is decelerating, or braking. Here, the braking forces will displayed as positive values rather than negative. The real value of these new channels is that it is much easier to visualize braking and acceleration as separate, positive channels, rather than looking only at longitudinal G. Additionally, separating them into two distinct channels makes it easier to use the data in other math channels.
Studying how a parameter changes over time is a large part of analyzing data, and can reveal plenty of information about the rider’s inputs along with the machine setup. For example, we are concerned not only with how hard the rider can brake, but also how quickly (or slowly) the rider applies the brakes at the beginning of the braking zone. A derivative channel of brake pressure or braking G will show that information quickly and easily.
Figure 1: GPS speed (black) and the gear position math channel (red) are displayed here. The lowest level of the gear position trace represents first gear here; each successive upward step is one higher gear, to a maximum of fourth on this particular lap.
While most motorcycles are equipped with a gear position sensor and this can be fed directly to the data acquisition system to determine gear, using a math channel to calculate gear position has a number of advantages. The ratio of rpm and rear wheel speed is directly proportional to the gear selected, and we can create a math channel as such. A gear ratio can be calculated as follows: