Designing a front corner from a blank sheet

Every chapter in this course, applied in the order a real designer works — from packaging constraints and a target camber curve to hardpoints you could hand to a fabricator.

You now have every tool this course teaches. This chapter puts them in order.

The order matters more than any individual number. Suspension design is over-constrained — you have four or five hardpoints and a dozen things you care about — so the sequence in which you fix things is what determines whether you converge or chase your tail.

Here is the sequence, with the chapter each step comes from.

Step 0 — Write down what the car is for

Before any geometry, three decisions that determine every number after:

• What surface, at what speed? A car that lives at 0.3 g on public roads wants a different answer than one at 1.2 g on a circuit.
• How much travel? Circuit 50–75 mm, road 150–200, cross-country 350+. This single number reorders every priority, because every per-millimeter quantity gets multiplied by it (chapter 1).
• What’s fixed? Wheelbase, track, tire size, engine position, ride height. These are your constraints, not your variables.

Skip this step and you’ll optimize numbers that don’t matter for your car.

Step 1 — Package it, then measure what you got

Put the wheel, upright, brake and steering rack where they physically fit. Don’t design geometry yet — find out what geometry the packaging allows.

Now read the numbers that fell out: FVSA, roll center, KPI, scrub, caster, motion ratio. This is your starting point, and it’s usually bad. That’s expected.

Step 2 — Camber curve, from roll angle

The first real design decision, and it comes from chapter 3’s equation run backwards.

Estimate your roll angle at your design lateral acceleration — 2–3° per g is a reasonable first guess for a road car — and the outside wheel’s compression at that roll. Then:

Set that equal to your tire’s preferred angle, solve for gain, and convert to a swing arm with $L_{\text{FVSA}} = 57.3 / \text{gain}$.

You’ll usually get a shorter arm than you can package. That’s the trade from chapter 3: you cannot fix a lot of roll with geometry alone. Take what you can get and plan to make up the rest with static camber and by reducing roll.

For long-travel cars, run it from travel instead: total camber swing budget divided into $57.3 \times \text{travel}$.

Step 3 — Roll center, and check its envelope

Now place the roll center (chapter 4), knowing the swing arm you just chose constrains it. Aim 30–100 mm for a road car, near ground for a circuit car.

Then immediately do the thing most people skip: sweep the travel and look at the migration (chapter 5). A static roll center without an envelope is half a specification. Under 50 mm of movement is good; beyond 90 mm the balance will shift with load.

Expect to iterate between steps 2 and 3 several times. They share the same pivots and they fight.

Step 4 — Side view: anti-dive and wheel path

Only now go to the side view (chapter 9). Tilt the inner pivots fore/aft for 20–40% anti-dive at the front.

Do it after the front view, not before, because the front view has more constraints and less freedom. Side-view inclination is comparatively cheap to add.

Step 5 — Steering axis

KPI and scrub radius (chapter 7), then caster and trail (chapter 8).

Two things worth remembering here. Scrub radius wants to be small, and slightly negative on a front-drive car for brake-failure stability. And caster is the cheapest camber you will ever buy — steer camber costs no swing-arm compromise at all, so take it before you shorten the swing arm any further.

Watch that moving the upper ball joint for scrub also changes your camber curve. Back to step 2 if it moved much.

Step 6 — Tie rod, last

Bump steer (chapter 6) is solved last because it’s a consequence of everything above: the tie rod has to aim at the instant center that steps 2 and 3 produced, and match the arc the arms now sweep.

This is why it’s last, and why changing anything earlier sends you back here. Target under 0.1° per 25 mm at the front.

Then Ackermann (chapter 11) via the steering-arm angle — knowing it fights bump steer for the same hardware.

Step 7 — Motion ratio and springs

Place the spring/damper (chapter 10), aiming for the highest motion ratio the packaging allows, since high MR means lower component loads and more consistent damper velocities.

Then size the spring from a target ride frequency (chapter 12), split front to rear for flat ride (chapter 13), and set preload for ride height — remembering they’re separate controls, and checking that the spring can actually support the corner without preload doing most of the work.

Then damping (chapters 14–15): ζ around 0.4 to start, rebound 1.5–2.5× compression, knee near 0.1 m/s.

Step 8 — Balance, at the whole-car level

Only with both axles designed can you look at balance (chapter 19). Set the roll-stiffness split, then check the understeer gradient (chapter 21) in both the linear range and at the limit — a car that’s neutral at 0.4 g and oversteering at the limit is worse than one that’s mildly wrong throughout.

Step 9 — Iterate, and know when to stop

Go round again. Everything you fixed in step 6 was disturbed by step 5.

Stop when the remaining errors are smaller than your manufacturing tolerance. A bracket welded by hand is good to a couple of millimeters, and a couple of millimeters of hardpoint error moves your roll center — chapter 2 measured 20 mm of bushing movement shifting the swing arm by more than a meter. There is no point optimizing past the accuracy you can actually build.

Go do it

Open the simulator and build one.

1. Start from a preset close to your layout, or from the custom vehicle setup with your own wheelbase, track and tire.
2. Work steps 2 through 6 in order, writing down the target for each before you drag anything. Targets first, dragging second — otherwise you’re exploring, not designing.
3. After each step, re-check the previous one. Note which pairs fight hardest: FVSA against roll center, scrub against camber curve, Ackermann against bump steer.
4. Use the Setup Advisor as a checklist rather than an oracle. It catches the common errors; it doesn’t know what your car is for.
5. When the geometry settles, go to Springs/Shocks for step 7, then Handling for step 8.
6. Export the .slab file. It’s text, so put it in version control alongside whatever else you’re building.

What it costs you

The honest closing note of this whole course: every one of these numbers is a compromise with the others, and there is no setting where all of them are right.

Module 1 kept finding the same thing — one set of pivots, four or five consequences, no independent adjustment. Module 2 found the same in time rather than space: a damper serving two jobs at different velocities. Module 3 found that the thing you can adjust freely (the split) matters precisely because the thing you can’t (total transfer) is fixed.

Good suspension design isn’t finding the setting where everything is optimal. It’s knowing which compromises your car can afford, and making them on purpose.

Rules of thumb — the whole course on one card

Try this

1. Design a front corner for a car you actually want to build. Write the step-0 answers down first.
2. Take an existing car, measure it against the table above, and find the one number most out of place. Is it a fault, or a deliberate choice you can now explain?
3. Pick any two rows in that table and describe how they fight. If you can do that for five pairs, you’ve understood the course.

That’s the whole course. You can find an instant center and predict a camber curve from it, construct and criticize a roll center, kill bump steer, read the steering axis in both views, build anti-dive, convert a spring rate to a wheel rate and then to a ride frequency, separate preload from rate, compute a damping ratio, place a damper knee, use a bump stop as a spring, find the wheel-hop resonance, split load transfer to set balance, explain why a tire’s grip isn’t proportional to load, read an understeer gradient in both the linear range and at the limit, follow the force from brake pedal to contact patch, and predict what lowering a car will break before you buy the springs.

Go build something.

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