Calibrating the brake curve to your own leg

Set the top of the brake scale to the force you can hold for a full stint, not to the hardest single push you can manage. A maximum you can only reach once compresses the whole useful range into the bottom of the pedal, and everything you do after that is a workaround for a calibration mistake.

The maximum is a decision, not a measurement

Every load cell pedal set asks you to define the top of the scale, and most drivers set it by pushing as hard as they can once. That produces a scale whose upper third is unreachable in a race and whose lower third contains all the braking you actually do. The consequence is a pedal that feels vague at exactly the pressures you use most.

The better method is a stint test. Drive ten minutes at racing pace, note the pressure you are actually applying at the heaviest braking zone, and set the maximum a little above that. Published capacities give the range the hardware can cover, 90 kg for the Fanatec ClubSport V3, up to 100 kg for the MOZA SR-P and 0 to 200 kg for the CRP2, but none of those numbers is a target. The number you want is your own sustainable maximum, and it is usually far below the ceiling.

Why the curve shape matters

Most software offers a curve between input force and brake output. A linear curve means the relationship is one to one across the range. A progressive curve gives you more resolution at low pressures and compresses the top, and a degressive curve does the opposite. Neither is correct in general, because the right curve depends on what your leg does badly.

If you lock the fronts on turn in, the top of your range is too easy to reach and a progressive curve helps. If you cannot get the last ten per cent of braking performance, the top of your range is too hard and a slightly degressive curve or a lower maximum helps. The distinction between those two problems is visible in the telemetry as either a spike above your intended pressure or a plateau below it.

Resolution decides how much the curve can do. A 12 bit sensor gives 4096 steps across the range, a 15 bit angle sensor gives 32768, and a curve that stretches the lower half of a coarse scale can make the steps visible as steps. Change the maximum before you change the curve, because the maximum moves the whole scale and the curve only redistributes it.

The arithmetic makes the point concrete. A 12 bit sensor across a 90 kg cell gives 4096 steps, which is about 22 grammes of leg force per step. A 15 bit angle sensor across 100 kg gives 32768 steps, about 3 grammes each. Now set the maximum wrongly: calibrate a 90 kg cell to a 40 kg leg and the useful part of the scale is 1820 steps instead of 4096, so more than half the resolution you paid for sits above anything you will ever press.

A test that takes three laps

Lap one: brake as hard as you comfortably can into the heaviest zone and hold it. Note whether the pedal reaches one hundred per cent and whether the fronts lock. Lap two: brake at what feels like eighty per cent and check the telemetry number. Lap three: repeat lap two and compare. If the two eighty per cent laps differ by more than a few per cent, the problem is the calibration rather than your leg.

Then leave it alone for a week. Every calibration change resets the motor learning that makes braking repeatable, and a driver who adjusts the curve daily never accumulates the consistency that the hardware was bought to provide.