Tires: load sensitivity, and why grip isn't a coefficient

Press a tire harder and it grips harder — but not proportionally. That single non-linearity is why load transfer costs lap time, why balance tuning works at all, and why four evenly loaded tires beat two heavily loaded ones.

School physics says friction force is μ times normal load, with μ a constant property of the two materials. Tires did not attend that lesson.

For a tire, μ falls as load rises. Press it harder and it does grip harder — but less than proportionally. That one fact underpins everything in this module, and once you’ve seen the curve you can derive most of vehicle dynamics from it.

The curve

Peak lateral μ from the app’s tire model, front compound:

Quadruple the load and μ falls by about 13%. The tire still makes more force — 6093 N against 1750 N — but each extra newton of load buys progressively less grip.

Why this makes weight transfer expensive

Now put chapter 18’s result together with this curve.

Cornering transfers load from the inside tires to the outside ones. The axle’s total load doesn’t change — one wheel gains exactly what the other loses. But because μ falls with load, the heavily loaded tire gains less grip than the lightly loaded one loses. Total grip drops.

Measured on this tire, taking an axle carrying 6000 N and splitting it increasingly unevenly:

Two things to take from that table, and the second is the one people miss.

The loss is real but modest at first, and accelerates. Small transfers cost almost nothing; large ones cost properly. It’s a squared-ish effect, not a linear one, so the last increment of transfer is far more expensive than the first.

It is also smaller than folklore suggests. A percent or two of axle grip is not the difference between a good car and a bad one. Anyone claiming load transfer is throwing away 20% of your grip is overstating it — at least for a tire like this one. Softer, more sensitive compounds show a steeper curve, and the effect is stronger on them, but the honest number here is a few percent.

So why does it matter so much?

Because it’s what makes balance tuning work

If μ were constant, transferring load between wheels on an axle would cost nothing, and moving transfer between axles — the whole of chapter 19 — would change nothing at all. A car’s balance would be fixed by its weight distribution and geometry, and anti-roll bars would be pointless.

Load sensitivity is the mechanism by which the axle with more roll stiffness loses more grip. Small effect, enormous consequence: it’s the lever every balance adjustment pulls on.

That’s the honest framing. Load sensitivity costs you a little grip directly, and gives you the entire balance-tuning toolbox indirectly.

Slip angle, briefly

Grip doesn’t appear the instant you turn the wheel. A tire generates lateral force by running at a slip angle — an angle between where it’s pointing and where it’s actually travelling. Force rises with slip angle, peaks somewhere around 5–10° for a road tire, and falls off beyond that.

Two consequences worth carrying forward:

Peak slip angle rises with load. A heavily loaded tire wants a bigger slip angle than a lightly loaded one — which is the argument that overturned Ackermann back in chapter 11.

Past the peak, more steering makes things worse. That’s the physical meaning of “understeering off the road while turning the wheel harder.” The front tires are past their peak, and the correction is less lock, not more.

Go look at it

Open the strut layout and go to the Handling tab.

1. Find Max grip and the per-wheel loads. Note the outer and inner loads at a moderate lateral acceleration.
2. Raise the Corner severity and watch the split widen — outer up, inner down. Watch Max grip as the split grows.
3. Now attack it from chapter 18. Lower the CG height in Car Design, come back, and check Max grip at the same g. Less transfer, more even loads, more total grip. That’s this chapter and chapter 18 in one measurement.
4. Push the split far enough that the inner wheel load approaches zero. Once a wheel lifts, its contribution is gone entirely and the axle is doing all its work on one tire — the far end of this curve.
5. Change the tire compound and repeat. Different compounds have different sensitivity, and the balance you tuned for one may not hold for another.

What it costs you

The design response to load sensitivity is “keep the loads even,” and that pulls against nearly everything else.

Even loads want a low CG, a wide track, soft springs and soft bars — but soft springs and bars cost body control and, past a point, let the geometry wander into parts of its travel where Module 1’s problems live. There is no setting that is optimal for load evenness and everything else at once.

There’s also a measurement trap worth naming. Load sensitivity means you cannot characterize a tire with one number. A μ figure quoted without the load it was measured at is close to meaningless, and comparing two tires at different loads tells you nothing. This is why tire data is expensive and why serious teams buy it rather than estimating.

And the limit of the model: everything above is peak grip in steady state. Real tires also care about temperature, pressure, camber, wear and how long they’ve been at it — and a tire 20°C from its window will make a mockery of any of these numbers.

Rules of thumb

• Peak μ, road tire: 0.9–1.1. Performance tire 1.1–1.3. Slick 1.4–1.8.
• Load sensitivity: expect roughly 5–15% drop in μ per doubling of load. Softer compounds are more sensitive.
• Peak slip angle: 5–10° for road tires, 4–7° for slicks. Higher load pushes it higher.
• Even loading beats uneven loading, always — the question is only what you had to give up to get it.
• Never quote a μ without its load.
• Rally-raid and desert: on loose surfaces the whole model shifts. Grip comes substantially from the tire digging into and shearing the surface rather than from friction against a solid one, so effective μ can rise with load as the tire bites deeper. That inverts some of the reasoning above — one reason desert setups tolerate load transfer that would be unacceptable on tarmac, and why the same car needs a different balance philosophy on sand than on a road section.

Try this

1. Find the lateral acceleration at which the inner wheel load first reaches zero. What is the car’s total grip doing either side of that point?
2. Compare two tire compounds at the same setup. Does the one with more peak grip also have the balance you want?
3. Using the table above, estimate how much grip a car loses at 1.0 g if its CG is 100 mm too high. Then check it in the app.

Next: one number that summarizes the whole of this module — how much extra steering a car needs as it corners harder, and what happens when the linear answer and the limit answer disagree.

Preset values are illustrative starting points, not a measured setup for any particular car.