In many situations it is desirable to get a single value of power for the currently loaded data, in particular in series where the power measurements are used to regulate the series. To make it as simple as possible to generate a single power figure for a run, there is a special tool included within the performance measurements control under “Power”.
This works by:
- taking the power calculated within the power variable
- finding either the 98% or 95% maximum of power for each lap
- finding the statistical mode of the power measurements from all laps
This has been found to be the most reliable and robust way of calculating power and it’s the recommended way to handle race series regulations. The added advantage of this method is that it not only gives a single summary “Reported Power”, but also a “Quoted Precision” which is a good indication of the spread on the data and is a simple additional check to highlight the reliability of the data.
To use this method of power measurement, you simple load the data and add a lap marker (manually, automatically or from file).
There is only one option to choose on the screen either:
- ignore highest 2% of power readings (equivalent to saying take the 98th percentile maximum) or…
- ignore highest 5% of power readings (equivalent to saying take the 95th percentile maximum) or…
Ignoring the last few percent of measured vehicle power is essential to avoid instantaneous power spikes generated from energy stored in the engine and drivetrain inertia having an undue influence on the power reading. For example, if you rev the engine and quickly enguage the clutch then the vehicle will surge forward, but then the acceleration rate will quickly reduce and then the vehicle will be accelerated by the engines power. Ignoring the last few percent of the power readings allows these spikes to be removed, and report the true power readings.
Typically ignoring the highest 5% of power reading is more robust, but on some tight tracks where maximum power is only infrequenty used, it might be preferable to only ignore the highest 2%. In most cases the different between the 2 results is minimal.
Note that for power reading to be accuracte then the correct procedures for mounting and using the system must be observed. There are some additional notes on this here.