Centripetal Force: Dynamics of Circular Motion

Swing a bucket of water in a fast vertical circle and the water stays put — at the top of the arc it hangs upside down over your head and does not fall. Ask a crowd why, and most will say something pushes the water outward into the bucket. That answer feels right, is repeated in gyms and playgrounds everywhere, and is exactly backwards. Nothing pushes outward. The water is falling the whole time — and the circle is what falling looks like when you are also moving sideways fast enough. Circular motion is the one place where Newton's second law gets misread by instinct, so this article slows the logic down. A body moving in a circle at speed $v$ is accelerating toward the centre at $v^2/r$ — that much is kinematics, settled in the previous module. Newton's second law then presents an invoice: some real, nameable force must supply $mv^2/r$ , pointed at the centre, every instant, or the circle dies. "Centripetal force" is the invoice, not a new force of nature. The physics of every circular-motion problem is identifying who pays — tension for a whirling stone, friction for a turning car, gravity for the Moon, a normal force for the wall-of-death rider. This identification skill is tested from CBSE and NEET (pick the force, compute $mv^2/r$ ) up through JEE-Advanced (payers that split the bill, payers that max out and fail). It also feeds four archetypes waiting downstream — the flat turn, the banked curve, the vertical circle, the conical pendulum. Master the invoice logic here and those become bookkeeping.

Key takeaways

Uniform circular motion is accelerated motion: $a_c = v^2/r = \omega^2 r$ , pointing at the centre — so Newton's second law demands a net inward force $F_{\text{net}} = mv^2/r$ .
Centripetal force is not a new force. It is the role played by real forces — tension, friction, gravity, normal force, or their combination. Never draw a separate " $F_c$ " arrow on an FBD.
The exam skill is naming the payer: string → tension; flat road → friction; banked road → normal + friction; orbit → gravity; rotor wall → normal force.
Payers have limits: friction maxes at $\mu N$ , strings at breaking tension. "Maximum speed" questions are payer-limit questions: $v_{\max} = \sqrt{\mu g r}$ on a flat road.
No outward "centrifugal force" exists in the ground frame — the outward feeling is your inertia, and the outward force is legal only as a pseudo-force in the rotating frame.
Key idea: a circle at speed $v$ is a standing order for $mv^2/r$ of inward net force — every circular-motion problem is "which real forces pay, and when do they max out?"

The core insight

For a body of mass $m$ moving on a circle of radius $r$ at speed $v$ , Newton's second law requires the real forces to supply a net inward (centripetal) component $mv^2/r = m\omega^2 r$ ; identifying which physical force plays that role — and where it reaches its limit — solves the problem.

Where this lives today

The centripetal invoice sets the size and cost of some of the largest machines ever built. The LHC's 27-kilometre circumference is not a stylistic choice: bending near-light-speed protons into a circle demands a magnetic centripetal force, and even 8-tesla superconducting magnets can bend only so sharply — higher energy means a bigger circle or stronger payers, which is why the proposed successor ring is 91 km around. Medicine runs the same equation smaller and faster: an ultracentrifuge spinning at 100,000 rpm subjects samples to about a million times $g$ — $\omega^2 r$ with a vengeance — separating proteins by how they sediment against it. Your washing machine's spin cycle is the payer-failure case used deliberately: the drum wall pays the invoice for the clothes but the water in the fabric escapes through the holes tangentially, exactly as the cut-string logic predicts. And every satellite in the sky is the one case where the invoice is paid perfectly, forever, for free — gravity supplies precisely $mv^2/r$ , which is what an orbit is.

The physical picture

Start with the stone on a string, whirled in a horizontal circle on smooth ice. At any instant its velocity points along the tangent. Left alone — cut the string — it would keep that velocity and leave along the tangent in a straight line (Newton's first law; not spiral outward, a favourite exam distractor). The string's job is to keep bending the path: every moment, the tension yanks the velocity vector a little toward the centre, turning it without changing its length. Uniform circular motion is perpetual course correction — an acceleration of magnitude $v^2/r$ aimed at the centre, forever. Newton's second law converts that kinematic fact into a bill that must be paid:

Figure

One invoice, three payers — which real force supplies mv squared over r

Hold on to three readings of this picture, because they are the marks.

The force is a role, not a species. "Centripetal" describes direction of duty — centre-seeking — the way "breadwinner" describes a role in a household, not a person. On a free-body diagram you draw only real forces: tension, weight, normal, friction. Then you demand that their net inward component equal $mv^2/r$ . Drawing an extra arrow labelled $F_c$ alongside the real forces is the single most common FBD error in this chapter, and it double-counts the physics.

The payer can max out. Friction can supply at most $\mu N$ ; a string at most its breaking tension; a normal force cannot pull. Every "maximum speed," "minimum coefficient," "when does the string snap" question is asking where the payer's budget fails the invoice: the invoice grows as $v^2$ , the budget doesn't.

Nothing pushes outward — in the ground frame. Your body feels flung outward on a merry-go-round because your inertia wants the tangent while the ride drags you into the circle; the "fling" is the seat pushing you inward, misread. An outward force is legal bookkeeping only in the rotating frame, where it is the centrifugal pseudo-force $m\omega^2 r$ — the frame toll from the pseudo-forces module, never a ground-frame arrow.

The math: the requirement and the payer audit

The magnitude of the required net inward force follows directly from the centripetal acceleration:

F_{\text{net, inward}} = \frac{mv^2}{r} = m\,\omega^2 r \qquad (1)

where: $v$ — speed along the circle; $\omega = v/r$ — angular speed in $\mathrm{rad/s}$ ; $r$ — radius. The two forms are one fact ( $v = \omega r$ ), but choose deliberately: the $v$ -form when speed is given or asked (cars, stones), the $\omega$ -form when everything rotates together (turntables, rotating frames — points at different radii share $\omega$ , not $v$ ). Feel the scalings: at fixed radius, doubling the speed quadruples the invoice — the deep reason speed limits on curves matter so much; at fixed $\omega$ , the invoice grows linearly with radius — the outermost child on the merry-go-round needs the most grip. The solving procedure is an audit, and it never varies. Identify the circle (centre and radius — half the battle in disguised problems). Draw the FBD with real forces only. Resolve along two directions: toward the centre, and perpendicular to the motion's plane. Then write Newton's second law twice:

\text{toward centre:}\;\; \sum F_{\text{inward}} = \frac{mv^2}{r}, \qquad \text{perpendicular:}\;\; \sum F = 0 \;\;\text{(if no motion that way)} \qquad (2)

where: the first equation is the invoice; the second pins down unknowns like the normal force. For a stone on a string on a smooth table: inward, $T = mv^2/r$ ; vertical, $N = mg$ . Done. Every archetype downstream — flat turn, banked curve, vertical circle, conical pendulum — is this same audit with a different payer roster.

Derivation AThe whirling stone and the breaking string

We run the audit on the cleanest payer — a stone of mass $m$ tied to a string of length $r$ , whirled in a horizontal circle on a smooth table at speed $v$ — and then push it to the payer's limit, the breaking tension $T_{\max}$ , to find the maximum safe speed. Strategy: FBD, two directions, then invert the invoice. Watch that the string, not some "circular force," is the only horizontal actor.

Step 1. FBD of the stone: weight $mg$ down, normal $N$ up, tension $T$ along the string toward the centre. Nothing else touches the stone — so nothing else can appear.

Step 2. Perpendicular (vertical) direction — no vertical motion:

N = mg

Step 3. Toward the centre, Newton's second law with the centripetal acceleration:

T = \frac{mv^2}{r}

Why: the tension is not "equal to the centripetal force" as a balance of two things — it IS the net inward force, and the equation assigns its required size. If the stone moved faster, the same string would simply have to pull harder.

Step 4. Push to the limit. The string can supply at most $T_{\max}$ ; the invoice grows as $v^2$ :

\frac{mv_{\max}^2}{r} = T_{\max} \;\;\Rightarrow\;\; v_{\max} = \sqrt{\frac{T_{\max}\, r}{m}}

Beyond $v_{\max}$ the payer defaults: the string snaps, the invoice goes unpaid, and the stone departs along the tangent at whatever velocity it had — straight line, constant speed, per Newton's first law (plus projectile fall if gravity gets a say). Verified: limits — $T_{\max} \to \infty$ allows any speed; $r \to \infty$ at fixed $v$ needs vanishing tension (a straight line is a circle of infinite radius). Numbers: $m = 0.5$ kg, $r = 1.2$ m, $T_{\max} = 60$ N: $v_{\max} = \sqrt{60 \times 1.2/0.5} = \sqrt{144} = 12$ m/s ✓.

Go deeper · Derivation B · 1 minThe flat turn — friction pays until it cannot

Derivation BThe flat turn — friction pays until it cannot

We audit the payer every driver depends on: a car of mass $m$ rounding a flat curve of radius $r$ . No string points at the centre here — the only available inward force is static friction from the road on the tyres. We derive the no-skid speed limit $v_{\max} = \sqrt{\mu_s g r}$ . The subtle point to watch: it is static friction (the tyre's contact patch is not sliding), and it is self-adjusting up to its ceiling $\mu_s N$ — the full personality profile from the static-friction module.

Step 1. FBD of the car: weight $mg$ down, normal $N$ up, static friction $f$ pointing horizontally toward the centre of the curve (friction acts to prevent the relative sliding that would otherwise happen — the car drifting out of the turn).

Step 2. Vertical: $N = mg$ (flat road, no vertical acceleration).

Step 3. Inward invoice:

f = \frac{mv^2}{r}

Friction adjusts: at low speed the demand is small and $f$ is small; the car turns effortlessly.

Step 4. The ceiling. Static friction cannot exceed $\mu_s N = \mu_s mg$ :

\frac{mv_{\max}^2}{r} = \mu_s m g \;\;\Rightarrow\;\; v_{\max} = \sqrt{\mu_s g r}

The mass cancels — a loaded truck and a scooter share the same flat-curve speed limit on the same surface, because both the invoice and the budget scale with $m$ . Why: exceeding $v_{\max}$ does not fling the car "outward" — friction simply can't bend the path sharply enough, so the car follows a gentler curve than the road and runs wide, tangentially. Skids on curves are under-payment, not outward payment. *Verified: $\mu_s = 0.5$ , $r = 45$ m, $g = 10$ : $v_{\max} = \sqrt{225} = 15$ m/s (54 km/h). Wet road at $\mu_s = 0.25$ : $v_{\max}$ drops by $\sqrt{2}$ to about 10.6 m/s — the square root is why "slow down a little" buys a lot of safety. The full flat-turn family (and what banking does to remove the friction dependence) lives in the Turning on a Flat Road and Banked Curves archetypes.*

Figure

The payer-limit picture — the invoice grows as v squared until the budget line kills the circle

The payer roster

setting	who pays the $mv^2/r$ invoice
stone on a string (horizontal circle)	tension
car/cyclist on a flat curve	static friction (ceiling $\mu_s mg$ )
banked curve	normal-force component (+ friction if needed) — see Banked Curves
vertical circle, top / bottom	tension + weight / tension − weight — see Vertical Circle
conical pendulum	horizontal component of tension
satellite / Moon in orbit	gravity, exactly
rotor / well of death rider	normal force from the wall
electron in a magnetic field (later)	magnetic force

Limiting cases

case	result	meaning
$v \to 2v$ at fixed $r$	invoice ×4	$v^2$ scaling: speed is expensive
$r \to \infty$ at fixed $v$	$F \to 0$	straight line = circle with no bill
$\mu_s \to 0$ (ice)	$v_{\max} \to 0$	no payer, no turning — cars go straight on ice
fixed $\omega$ , growing $r$	$F = m\omega^2 r$ grows	outer riders on a turntable need the most grip
string cut / payer fails	tangential straight-line exit	inertia resumes; no outward spiral
$m \to 2m$ on flat curve	$v_{\max}$ unchanged	both bill and budget scale with $m$

Common trap"The centripetal force balances the centrifugal force, so the body moves in a circle"

This sentence — seductively symmetric — fails twice in one breath. First, in the ground frame there is no centrifugal force to balance: the only outward thing is your intuition. Second, and more fatal: if the inward force were balanced by anything, the net force would be zero and the body would move in a straight line, not a circle. Circular motion exists precisely because the forces do not balance — there is a net inward force, and the body is perpetually accelerating toward the centre it never reaches. The corrected bookkeeping: (real forces' inward components) $= mv^2/r$ , an equation of demand, not balance. The word "centrifugal" becomes legal only in the rotating frame, where the body is at rest and the pseudo-force $m\omega^2 r$ genuinely balances the payers — the frame-toll story told in the Non-Inertial Frames & Pseudo-Forces module. Declare your frame; then only one of the two words belongs in it.

Worked examples

1 · Board level — stone on a string (CBSE/NEET pattern, 3 marks)

A $0.50$ kg stone tied to a $1.2$ m string is whirled in a horizontal circle on a frictionless table at $2$ revolutions per second. Find the tension. If the string breaks at $60$ N, what is the maximum speed the stone can have?

Step 1. Angular speed: $\omega = 2\pi \times 2 = 4\pi\ \mathrm{rad/s} \approx 12.57\ \mathrm{rad/s}$ .

Step 2. Invoice in the $\omega$ -form: $T = m\omega^2 r = 0.50 \times (4\pi)^2 \times 1.2 = 0.50 \times 157.9 \times 1.2 = \boxed{94.8\ \mathrm{N}}$ — this already exceeds 60 N, so 2 rev/s is not sustainable with this string.

Step 3. Maximum speed from the payer limit: $v_{\max} = \sqrt{T_{\max} r/m} = \sqrt{60 \times 1.2/0.50} = \sqrt{144} = \boxed{12\ \mathrm{m/s}}$ .

Step 4. Sanity: at $v_{\max}$ , $\omega = v/r = 10\ \mathrm{rad/s} \approx 1.6$ rev/s — just below the requested 2 rev/s, consistent with Step 2's overdraft.

2 · Board level — the flat curve speed limit (NEET/ISC pattern)

A car rounds a flat curve of radius $45$ m on a road with $\mu_s = 0.50$ . Take $g = 10\ \mathrm{m/s^2}$ . Find the maximum speed for no skidding, and state what force provides the centripetal force.

Step 1. The payer: static friction from road on tyres, pointing toward the centre of the curve. (Name it — that sentence carries a mark.)

Step 2. Vertical equilibrium: $N = mg$ . Friction ceiling: $f_{\max} = \mu_s mg$ .

Step 3. Set invoice = ceiling: $\dfrac{mv_{\max}^2}{r} = \mu_s mg \Rightarrow v_{\max} = \sqrt{\mu_s g r} = \sqrt{0.50 \times 10 \times 45} = \boxed{15\ \mathrm{m/s}}$ (54 km/h).

Step 4. Mass cancelled — the limit is the same for every vehicle on this surface. Faster requires banking the road (the Banked Curves sibling shows how banking shifts the bill to the normal force).

3 · Competitive level — coin on a turntable (JEE-Main pattern, ~2.5 minutes)

A coin rests on a turntable at $r = 10$ cm from the axis, $\mu_s = 0.4$ , $g = 10\ \mathrm{m/s^2}$ . (a) What is the maximum angular speed before the coin slips? (b) If the turntable spins at exactly that rate, where else on the table could the coin rest without slipping?

Step 1. Payer audit: friction supplies the invoice $m\omega^2 r$ ; ceiling $\mu_s mg$ .

Step 2. (a) Slip threshold: $m\omega_{\max}^2 r = \mu_s mg \Rightarrow \omega_{\max} = \sqrt{\dfrac{\mu_s g}{r}} = \sqrt{\dfrac{0.4 \times 10}{0.10}} = \sqrt{40} = \boxed{6.3\ \mathrm{rad/s}}$ (about 1 rev/s).

Step 3. (b) At fixed $\omega$ , the invoice $m\omega^2 r$ grows with radius while the budget $\mu_s mg$ is flat. The coin survives wherever $r \le 0.10$ m: anywhere closer to the axis, at the same or smaller radius — the rim is the danger zone.

Step 4. Concept lock: turntable problems use the $\omega$ -form because all points share $\omega$ ; a point twice as far out needs twice the friction. Timing note: the $\omega$ -vs- $v$ choice is the whole trick — reach for $m\omega^2 r$ the moment "rpm" or "rev/s" appears, converting revolutions to radians first ( $1\ \mathrm{rev} = 2\pi$ rad). Budget 2.5 min; part (b) should take ten seconds once you see the linear-in- $r$ invoice.

4 · Competitive level — JEE-Advanced stretch: the string through the table (~4 minutes)

A small block of mass $m_1 = 0.30$ kg moves in a circle of radius $r = 0.40$ m on a smooth table, tied to a string that passes through a frictionless hole at the centre and supports a hanging mass $m_2 = 0.90$ kg at rest below. Take $g = 10\ \mathrm{m/s^2}$ . (a) Find the block's speed. (b) If $m_2$ is nudged into hanging equilibrium at a smaller radius $r' = 0.10$ m with the block again in a circle, what is the new speed? (c) Why can't this system be in equilibrium with the string slack?

Step 1. (a) Two FBDs, one string, one tension. Hanging mass at rest: $T = m_2 g = 9.0$ N. Block's invoice: $T = \dfrac{m_1 v^2}{r}$ .

Step 2. Solve: $v = \sqrt{\dfrac{T r}{m_1}} = \sqrt{\dfrac{9.0 \times 0.40}{0.30}} = \sqrt{12} = \boxed{3.46\ \mathrm{m/s}}$ .

Step 3. (b) The tension requirement is unchanged ( $T = m_2 g = 9.0$ N — the hanging mass still just hangs), so $v' = \sqrt{Tr'/m_1} = \sqrt{9.0 \times 0.10/0.30} = \sqrt{3} = \boxed{1.73\ \mathrm{m/s}}$ . Smaller circle, slower speed for the same payer — the invoice $mv^2/r$ at fixed $T$ forces $v \propto \sqrt{r}$ .

Step 4. (c) Slack string means $T = 0$ : the block then has no inward force (smooth table) and moves in a straight line, while $m_2$ free-falls. The circular state requires the string taut — the constraint and the invoice sustain each other. Timing note: the design pattern is "one tension, two jobs" — vertical equilibrium sets $T$ , the circle spends it. Spot that and (a) is 90 seconds. Part (b) tests whether you re-derive instead of pattern-matching a memorized $v = \sqrt{\mu g r}$ -type formula that doesn't apply here. Budget 4 min.

Exam traps

Drawing $F_c$ as an extra force. The FBD carries real forces only; "centripetal" is the role their inward components must fill. An extra $mv^2/r$ arrow double-counts and typically doubles your tension answers.

"Centripetal force balances centrifugal force." See the pitfall box: balanced forces make straight lines. In the ground frame the net inward force is unbalanced by construction.

Cut-string trajectory. The body leaves along the tangent (then falls as a projectile if airborne) — never radially outward, never in a spiral. MCQ options showing outward-spiral exits are bait.

Using $v$ -form with rpm data. "Rev/s" or "rpm" means all points share $\omega$ : convert to $\mathrm{rad/s}$ ( $\times 2\pi$ ) and use $m\omega^2 r$ . Forgetting the $2\pi$ scales forces by $39.5$ — an error too large to miss if you sanity-check.

Payer-limit inequality direction. No-slip/no-snap conditions read (invoice) ≤ (budget): $mv^2/r \le \mu_s mg$ , $mv^2/r \le T_{\max}$ . Setting equality first and forgetting which side grows leads to inverted "minimum/maximum" answers.

Radius identification. The $r$ in the invoice is the radius of the actual circle traced — for a conical pendulum it is $L\sin\theta$ , not the string length; for a car it is the curve's radius, not the road's width. Disguised-radius setups are JEE's favourite dressing.

Which $g$ . CBSE/ISC 9.8; JEE/NEET as declared (usually 10); AP allows 10; A-Level 9.81. The flat-curve $v_{\max} = \sqrt{\mu g r}$ inherits the choice directly.

Origins · 1 minChristiaan Huygens and how this was discovered

Christiaan Huygens

1629–1695 · The Hague & Paris Academy of Sciences

Christiaan Huygens got to circular motion before Newton. In De Vi Centrifuga (written 1659, published 1673) he derived the quantitative law of the "centre-fleeing tendency" of whirled bodies — the magnitude $v^2/r$ — from pure geometry, decades before force had a proper definition. Newton inverted the viewpoint: not a centre-fleeing force possessed by the body, but a centre-seeking (centripetal — Newton coined the word as a deliberate answer to Huygens') force required from outside, and used it to test universal gravitation against the Moon's orbit. Huygens' fingerprints are everywhere in this chapter regardless: he built the first pendulum clock, solved the compound pendulum, invented the balance-spring watch (fighting Hooke over it), discovered Titan and Saturn's rings with lenses he ground himself, and championed the wave theory of light a century and a half before it won. Few scientists have been so right, so early, about so much.

History

1659

— Huygens derives the

v^2/r

law of circular motion (De Vi Centrifuga, published 1673).

1666

— Newton's "Moon test": terrestrial gravity, diluted by distance squared, pays the Moon's centripetal invoice.

1687

— Principia defines centripetal force and derives Kepler's laws from the inverse square.

1957

— Sputnik: the first artificial payer-is-gravity circle,

v_{\text{orb}} \approx 7.8

km/s.

2008

— the LHC's superconducting magnets begin supplying the centripetal force that bends 7-TeV protons around a 27 km ring.

Going further

Texts. NCERT Physics Class 11, Ch. 4 (§ dynamics of uniform circular motion) · HC Verma, Concepts of Physics Vol. 1, Ch. 7 · Kleppner & Kolenkow Ch. 2–3 (the honest vector treatment).

Provenance. All results are standard. A verifier should recompute: Example 1's $T = 94.8$ N at $4\pi$ rad/s and $v_{\max} = 12$ m/s; Example 2's $v_{\max} = 15$ m/s; Example 3's $\omega_{\max} = \sqrt{40} = 6.3$ rad/s; Example 4's $v = \sqrt{12} = 3.46$ m/s and $v' = \sqrt{3} = 1.73$ m/s.

Prerequisites. Circular Motion: Centripetal & Tangential Acceleration · Newton's Second Law · Static Friction. Leads to. Turning on a Flat Road · Banked Curves · Motion in a Vertical Circle · The Conical Pendulum. Deep dive (UG). Centripetal Force Dynamics (UG treatment) · the Coriolis pseudo-force story.

Is centripetal force a real force or not? My textbook lists tension, friction, gravity… and then centripetal.

It is real in the way "the winning goal" is a real goal — a role filled by an actual player. The tension IS the centripetal force in the string problem; friction IS it on the flat curve. What it is not is an additional force to draw alongside them. Practical test for your FBD: every arrow must answer "what object exerts this, by what mechanism?" Tension, friction, gravity, normal all pass; a bare " $F_c$ " fails. Then the phrase "acts as the centripetal force" appears in your equation, not your diagram: $T = mv^2/r$ .

If the net force points inward, why doesn't the body fall into the centre?

Because it also has a huge sideways velocity. The inward force spends every instant turning that velocity, not shrinking the radius — the body falls toward the centre and misses, forever. Cut the speed and it really does fall inward (a too-slow satellite descends); kill the force and it flies off straight. A circle is the exact truce between the two. Newton's own picture: a cannonball fired fast enough that the ground curves away exactly as fast as the ball falls.

Why do I feel thrown outward on a ride if no outward force exists?

Your nerves report real forces acting on you — and what acts on you points inward: the seat, the rail, the car door pressing you toward the centre. The "outward" sensation is your inertia: your body attempts the tangent, and the ride keeps intercepting it. You interpret "the wall keeps shoving me inward" as "something presses me into the wall." In the rotating frame that feeling gets an honest name — the centrifugal pseudo-force $m\omega^2 r$ — but that is frame bookkeeping (the Pseudo-Forces module), not a ground-frame interaction.

Does the centripetal force do any work on the body?

In uniform circular motion, none. The force points at the centre; the velocity points along the tangent; they stay perpendicular at every instant, and a force perpendicular to motion does zero work. That is why the speed never changes even though the body accelerates continuously — all the force's effort goes into redirecting, none into speeding up. Consistency check you can quote: kinetic energy is constant, so net work must be zero. (In a vertical circle gravity does do work and the speed varies around the loop — that richer story is the Vertical Circle archetype.)

How do I decide between mv²/r and mω²r in a problem?

By what is shared or given. Data in rev/s, rpm, or "all points of a rigid disc": use $m\omega^2 r$ (convert to rad/s first — multiply rev/s by $2\pi$ ). Data as a speed in m/s (a car, a stone's speed): use $mv^2/r$ . They are the same law through $v = \omega r$ ; picking the natural one just avoids conversion errors. Classic discriminator: on one turntable, the outer coin needs MORE force ( $\omega$ fixed, $F \propto r$ ); two cars at the same speed, the one on the tighter curve needs more ( $v$ fixed, $F \propto 1/r$ ).

For a car turning on a flat road, why is it STATIC friction when the car is obviously moving?

Static and kinetic refer to the contact patch, not the vehicle. A rolling tyre's contact patch is momentarily at rest on the road — no sliding — so the friction available is static, with its higher coefficient and self-adjusting personality. It aims wherever needed to prevent incipient sliding: toward the curve's centre when turning. Only when the demanded force exceeds $\mu_s N$ does the patch actually slide — a skid — downgrading the payer to kinetic friction, which is both weaker and no longer steerable. This is exactly the ABS story from the friction modules, wearing a curve.

Satellites are in free fall, so what pays their centripetal invoice — and why don't they need engines?

Gravity pays, exactly and entirely: $GMm/r^2 = mv^2/r$ defines the orbital speed. No engine is needed because nothing is being resisted — there is no friction to fight in vacuum, and the invoice is redirection, not propulsion (zero work, as above). Engines fire only to change orbits. "Weightlessness" aboard is the free-fall story from the apparent-weight module: astronaut and station both fall toward Earth at the same rate, missing forever. The full orbital economics — speeds, periods, energies — arrives in the Gravitation chapter.