Transformation of velocity between rotating frames

Stand perfectly still on the equator and you are moving at 465 metres per second — faster than the speed of sound — toward the east. You feel nothing, your coffee doesn't spill, and yet every space agency on the planet takes that number deadly seriously: it is why Europe ships its rockets across the Atlantic to Kourou in French Guiana, five degrees from the equator, rather than launching them from home. A rocket sitting "at rest" on the pad has already banked almost half a kilometre per second of free velocity, courtesy of the turning Earth. The bookkeeping behind that free velocity is the velocity transformation between rotating frames. A particle traces a single trajectory through space, but an inertial observer and a co-rotating observer disagree about how fast it moves, because the rotating observer's measuring axes are themselves sweeping through space. The disagreement is captured by one cross product: the inertial velocity equals the velocity measured in the rotating frame plus the entrainment velocity $\boldsymbol{\omega}\times\mathbf{r}$ — the velocity of the rotating-frame point that the particle happens to be passing through. The relation looks almost too simple to matter, and that is exactly why it is worth internalizing. It is the first application of the rotating-frame derivative identity, the cleanest place to learn the discipline of asking which frame measured this vector, and the kinematic seed of everything downstream: differentiate it once more and the Coriolis, centrifugal, and Euler forces tumble out. Master this one line and the strange weather of rotating-frame mechanics becomes bookkeeping.

Key takeaways

The velocity of a particle measured in an inertial frame equals the velocity measured in a rotating frame plus the entrainment (transport) velocity: $\mathbf{v}_I=\mathbf{v}_R+\boldsymbol{\omega}\times\mathbf{r}$ .
This follows from the master operator identity $\left(\frac{\mathrm{d}}{\mathrm{d}t}\right)_{\!I}=\left(\frac{\mathrm{d}}{\mathrm{d}t}\right)_{\!R}+\boldsymbol{\omega}\times$ applied once to the position vector.
The entrainment term $\boldsymbol{\omega}\times\mathbf{r}$ is the velocity a point rigidly fixed in the rotating frame would have as seen from the inertial frame — $465\ \mathrm{m/s}$ eastward for a point on Earth's equator.
If the frames share an origin, the position vector is the same geometric object in both frames; only its time derivative differs, by the rotation of the axes.
Differentiating once more (and reapplying the operator identity) produces the Coriolis, centrifugal, and Euler terms — the subject of the acceleration-transformation article.
Key idea: For frames sharing an origin, the inertial velocity is the rotating-frame velocity plus the entrainment velocity: $\mathbf{v}_I=\mathbf{v}_R+\boldsymbol{\omega}\times\mathbf{r}$ . The position is the same vector in both frames; only its time derivative differs, by the rotation of the axes.

Where this lives today

Every GNSS receiver on Earth runs equation (4) continuously. Satellite orbits are computed in an Earth-centred inertial frame (ECI), but your position is wanted in the Earth-centred, Earth-fixed rotating frame (ECEF) — and the velocity conversion between them is exactly $\mathbf{v}_I=\mathbf{v}_R+\boldsymbol{\omega}_\oplus\times\mathbf{r}$ , where navigation engineers call the entrainment piece the transport rate. Get it wrong and a GPS fix drifts by hundreds of metres per second of error; aerospace textbooks drill the ECI↔ECEF velocity transformation before anything else in orbit determination. The same line item appears in inertial navigation systems, missile guidance, and every satellite-tracking app that draws a ground track. Launch economics still obey Example 2. Europe's Ariane 6 flies from Kourou at $5.2°$ N; SpaceX chose Boca Chica at $26°$ N; polar and retrograde missions accept the entrainment penalty deliberately. And the transformation quietly runs the machines around you: robot arms mounted on rotating gantries, centrifuge payload analysis, and hard-disk servo control all resolve tool and payload velocities through $\mathbf{v}_R + \boldsymbol{\omega}\times\mathbf{r}$ before any force calculation begins — velocity first, dynamics second, exactly the order this corner of mechanics is taught.

The physical picture

Glue a small flag to a merry-go-round at radius $r$ . A child riding the platform sees the flag as the most boring object in the world: it never moves, $\mathbf{v}_R=0$ . You, standing on the ground, watch the same flag sweep out a circle at speed $\omega r$ , tangent to its orbit. Neither observer is wrong. The flag's entire inertial velocity is the entrainment $\boldsymbol{\omega}\times\mathbf{r}$ — the velocity the platform itself donates to anything pinned at that spot.

Figure

One flag, two velocity readings — at rest on the platform, circling in the lab

Now let a bug crawl radially outward across the turntable at speed $u$ . The rotating observer reports a purely radial velocity. The ground observer sees that radial crawl plus the tangential carry $\omega r$ of the spot the bug currently occupies, added tip-to-tail as vectors. The bug's inertial velocity is the hypotenuse — and because the carry grows with $r$ , the ground-frame path is a widening spiral even though the bug swears it is walking a straight line.

Figure

Velocity addition on the turntable — radial crawl plus tangential carry gives the inertial velocity

The deep point is the distinction between a vector and its rate of change as measured by somebody. With coincident origins the position $\mathbf{r}$ is one geometric arrow — both observers agree on it completely. But the rotating observer differentiates while holding the platform's basis vectors fixed, while the inertial observer also feels those basis vectors whirl at rate $\boldsymbol{\omega}$ . The gap between their two readings is unavoidably $\boldsymbol{\omega}\times\mathbf{r}$ : the same arrow, two honest but different derivatives. Because the relation is linear in $\mathbf{r}$ and $\mathbf{v}_R$ , it composes beautifully: velocities of extended systems transform piece by piece, and stacked rotating frames simply add their angular velocities — the reason it is the workhorse of rigid-body kinematics and spacecraft attitude dynamics.

The core equations

Everything in this article is one application of the master rule proved in the time-derivatives article: differentiation in the inertial frame equals differentiation in the rotating frame plus one cross product. We restate it because every step below leans on it — and because knowing which frame took the derivative is the entire game:

\left(\frac{\mathrm{d}\mathbf{Q}}{\mathrm{d}t}\right)_{\!I}=\left(\frac{\mathrm{d}\mathbf{Q}}{\mathrm{d}t}\right)_{\!R}+\boldsymbol{\omega}\times\mathbf{Q} \qquad (1)

where: $\mathbf{Q}$ — any vector quantity (position, velocity, etc.); $(\cdot)_I$ — time derivative computed in the inertial frame; $(\cdot)_R$ — time derivative computed in the rotating frame (basis held fixed); $\boldsymbol{\omega}$ — angular velocity of the rotating frame relative to the inertial frame. The extra term is pure geometry — the rotation of the measuring axes, not any physical change in $\mathbf{Q}$ . Feed the identity the position vector of a particle, with both frames sharing an origin on the rotation axis, and it hands back the headline result of this article. The rotating-frame derivative of $\mathbf{r}$ is by definition the velocity the co-rotating observer measures, so:

\mathbf{v}_I=\mathbf{v}_R+\boldsymbol{\omega}\times\mathbf{r} \qquad (2)

where: $\mathbf{v}_I$ — velocity measured in the inertial frame; $\mathbf{v}_R\equiv(\mathrm{d}\mathbf{r}/\mathrm{d}t)_R$ — velocity measured in the rotating frame; $\mathbf{r}$ — position vector (same in both frames since origins coincide); $\boldsymbol{\omega}\times\mathbf{r}$ — entrainment (transport) velocity. Read it as a fair exchange rate between observers: what the lab sees is what the platform rider sees, plus what the platform itself is doing at the particle's location. The entrainment term deserves its own line, because it is the piece students meet everywhere from launch pads to star trails. It is the inertial velocity of the rotating-frame grid point the particle is instantaneously passing:

\mathbf{v}_{\text{entrain}}=\boldsymbol{\omega}\times\mathbf{r},\qquad |\mathbf{v}_{\text{entrain}}|=\omega\,r_\perp \qquad (3)

where: $\omega$ — magnitude of angular velocity; $r_\perp$ — perpendicular distance from the rotation axis, $r_\perp=r\sin\theta$ ; $\theta$ — angle between $\mathbf{r}$ and $\boldsymbol{\omega}$ . Only the distance from the axis matters: on Earth that is $R\cos\lambda$ at latitude $\lambda$ , which is why the equator gets the full $465\ \mathrm{m/s}$ and the poles get nothing. If the rotating frame's origin itself translates — a turntable bolted to a moving truck, or the Earth's centre orbiting the Sun — the origin's inertial velocity simply rides along additively. The general statement is:

\mathbf{v}_I=\mathbf{V}_O+\mathbf{v}_R+\boldsymbol{\omega}\times\mathbf{r}' \qquad (4)

where: $\mathbf{V}_O$ — inertial velocity of the rotating frame's origin; $\mathbf{r}'$ — position relative to the moving origin. Set $\mathbf{V}_O=0$ and you recover (2); set $\boldsymbol{\omega}=0$ and you recover Galileo's plain velocity addition $\mathbf{v}_I=\mathbf{V}_O+\mathbf{v}_R$ — equation (2) is Galileo's rule with a position-dependent relative velocity. One statement in this subject is not a transformation at all, and confusing it with one causes most of the trouble. With coincident origins the position vector is the same geometric object for both observers:

\mathbf{r}_I=\mathbf{r}_R\quad(\text{coincident origins}) \qquad (5)

where: $\mathbf{r}_I,\ \mathbf{r}_R$ — position vectors as geometric objects in each frame. The arrow from origin to particle is one thing; the observers merely resolve it onto different bases. Positions agree, derivatives don't — that asymmetry is the whole content of (2).

Which relation answers which question

you want	use	notes
derivative bookkeeping for any vector	eq. (1)	owned by the time-derivatives article
particle velocity between frames	eq. (2)	common origin on the axis
speed of a co-rotating point	eq. (3)	$\mathbf{v}_R=0$ ; speed $\omega r_\perp$
rotating frame whose origin moves	eq. (4)	add $\mathbf{V}_O$ ; nothing else changes
is position shared between frames	eq. (5)	yes for coincident origins — only derivatives differ
accelerations and fictitious forces	acceleration-transformation article	differentiating (2) again gives Coriolis, centrifugal, Euler

Derivation AVelocity transformation from the operator identity

We are deriving equation (2), the velocity transformation $\mathbf{v}_I=\mathbf{v}_R+\boldsymbol{\omega}\times\mathbf{r}$ , from the operator identity (1) — a two-line calculation whose entire value lies in doing it consciously once, because the same move iterated a second time generates all of rotating-frame dynamics. The strategy is simply to feed the identity the position vector and interpret each resulting piece as a velocity somebody actually measures. The only thing to watch is the setup: both frames must share an origin sitting on the rotation axis, so that one arrow $\mathbf{r}$ serves both observers and equation (5) applies. No approximation enters anywhere — the result is exact for any motion and any rotation rate.

Step 1. State the operator identity, established in the time-derivatives article from the turning of the rotating basis vectors ( $\dot{\hat{\mathbf{e}}}_i=\boldsymbol{\omega}\times\hat{\mathbf{e}}_i$ ):

\left(\tfrac{\mathrm{d}}{\mathrm{d}t}\right)_I=\left(\tfrac{\mathrm{d}}{\mathrm{d}t}\right)_R+\boldsymbol{\omega}\times

Why this is legitimate here: the identity holds for any vector whatsoever, so in particular for the position vector of our particle.

Step 2. Apply it to the position vector $\mathbf{r}$ :

\left(\tfrac{\mathrm{d}\mathbf{r}}{\mathrm{d}t}\right)_I=\left(\tfrac{\mathrm{d}\mathbf{r}}{\mathrm{d}t}\right)_R+\boldsymbol{\omega}\times\mathbf{r}

Why the same $\mathbf{r}$ can appear on both sides: with coincident origins the position is one geometric arrow — equation (5) — so both derivatives act on the same object; they differ only in which basis is held fixed while differentiating.

Step 3. Name the two derivatives. The left side is what an inertial observer calls the particle's velocity; the first term on the right is what the co-rotating observer measures with rulers and clocks fixed to the platform:

\mathbf{v}_I\equiv\left(\tfrac{\mathrm{d}\mathbf{r}}{\mathrm{d}t}\right)_I,\quad \mathbf{v}_R\equiv\left(\tfrac{\mathrm{d}\mathbf{r}}{\mathrm{d}t}\right)_R

Step 4. Conclude:

\mathbf{v}_I=\mathbf{v}_R+\boldsymbol{\omega}\times\mathbf{r}

Why the extra term has a clean physical name: for a particle glued to the platform, $\mathbf{v}_R=0$ and the whole inertial velocity is $\boldsymbol{\omega}\times\mathbf{r}$ — the entrainment velocity of the platform point itself. For a moving particle, the platform point it currently occupies still contributes exactly that much. *Verified: Matches Goldstein Eq. (4.86)–(4.87) and Morin Ch. 10; reduces to $\mathbf{v}_I=\mathbf{v}_R$ for $\boldsymbol{\omega}=0$ , and to pure circular motion $|\mathbf{v}_I|=\omega r_\perp$ for $\mathbf{v}_R=0$ .*

Go deeper · Derivation B · 2 minWhat the rotating observer sees — inverting the transformation

Derivation BWhat the rotating observer sees — inverting the transformation

The goal here is the inverse question, which is where the physical surprises live: given simple inertial motion, what velocity does the rotating observer record? The strategy is to solve equation (2) for $\mathbf{v}_R$ and feed it the two simplest inertial motions there are — a point that does not move at all, and a force-free particle coasting in a straight line. Watch the sign: the entrainment term flips, so objects at rest inertially appear to sweep backwards through the rotating frame, and the farther away they sit, the faster they appear to fly.

Step 1. Solve (2) for the rotating-frame velocity:

\mathbf{v}_R=\mathbf{v}_I-\boldsymbol{\omega}\times\mathbf{r}

Why no new derivation is needed: (2) is a vector identity at each instant, so it can be read in either direction.

Step 2. Feed it an object at rest in the inertial frame — a distant star, with $\mathbf{v}_I=0$ :

\mathbf{v}_R=-\,\boldsymbol{\omega}\times\mathbf{r},\qquad |\mathbf{v}_R|=\omega\,r_\perp

Why the sky wheels overhead: in Earth's rotating frame every fixed star sweeps a daily circle at apparent speed $\omega r_\perp$ — the platform's own turning, billed to the rest of the universe with a minus sign.

Step 3. Note how alarming that speed becomes at large distance. The apparent speed reaches the speed of light at

r_\perp=\frac{c}{\omega}=\frac{3.0\times10^{8}}{7.292\times10^{-5}}\ \mathrm{m}\approx 4.1\times10^{12}\ \mathrm{m}\approx 27.5\ \mathrm{AU}

so in the Earth-fixed frame, Neptune (at about 30 AU) traces its daily circle at just over the speed of light. Why nothing is violated: $\mathbf{v}_R$ is a coordinate velocity, not the motion of any object or signal through space; relativity constrains what moves through spacetime, not how fast rotating coordinates repaint their labels.

Step 4. Feed it a force-free particle coasting on a straight line, $\mathbf{r}(t)=\mathbf{r}_0+\mathbf{u}\,t$ with constant $\mathbf{u}$ :

\mathbf{v}_R(t)=\mathbf{u}-\boldsymbol{\omega}\times\mathbf{r}(t)

Why this matters downstream: even with zero force, $\mathbf{v}_R$ changes in time — through $\mathbf{r}(t)$ and through the turning basis onto which the observer resolves it. A changing measured velocity means the rotating observer will report accelerations for a force-free particle; chasing that thought through one more derivative is precisely the acceleration-transformation article, where the Coriolis and centrifugal terms are born. *Verified: Step 2 reproduces the apparent diurnal motion of the celestial sphere (period one sidereal day, 86164 s); the $c/\omega$ radius recomputes to $4.11\times10^{12}\ \mathrm{m}$ with $\omega=7.292\times10^{-5}\ \mathrm{rad/s}$ .*

One preview belongs here, stated but not derived. Differentiating equation (2) once more — applying identity (1) to $\mathbf{v}_I$ itself — yields the acceleration transformation

\mathbf{a}_I=\mathbf{a}_R+2\,\boldsymbol{\omega}\times\mathbf{v}_R+\boldsymbol{\omega}\times(\boldsymbol{\omega}\times\mathbf{r})+\dot{\boldsymbol{\omega}}\times\mathbf{r} \qquad (6)

where: the three extra terms become, on the force side, the Coriolis, centrifugal, and Euler forces. The full derivation — including the bookkeeping that produces the famous factor of 2 — is the centrepiece of the acceleration-transformation article; the sorted physical consequences live in the Coriolis force and centrifugal force articles.

Limiting cases

The transformation specializes cleanly in several regimes:

Case	Result	Meaning
Non-rotating frames	$\boldsymbol{\omega}=0\ \Rightarrow\ \mathbf{v}_I=\mathbf{v}_R$	With no rotation the entrainment term vanishes and the two observers agree on velocity (Galilean limit for coincident origins).
Particle fixed in rotating frame	$\mathbf{v}_R=0\ \Rightarrow\ \mathbf{v}_I=\boldsymbol{\omega}\times\mathbf{r}$	A point rigidly attached to the rotating frame moves on a circle; its inertial speed is $\omega r_\perp$ , where $r_\perp$ is the distance from the rotation axis.
Motion along the axis	$\mathbf{r}\parallel\boldsymbol{\omega}\ \Rightarrow\ \boldsymbol{\omega}\times\mathbf{r}=0$	A particle on the rotation axis feels no entrainment; the velocity transformation reduces to identity even though $\boldsymbol{\omega}\neq0$ .
Translating origin	$\mathbf{v}_I=\mathbf{V}_O+\mathbf{v}_R+\boldsymbol{\omega}\times\mathbf{r}'$	If the rotating origin moves with inertial velocity $\mathbf{V}_O$ and $\mathbf{r}'$ is the position relative to that origin, add $\mathbf{V}_O$ . The pure form assumes $\mathbf{V}_O=0$ .

Common trap“The rotating observer's velocity is just the inertial velocity in different components”

It is not — and this single confusion causes more lost marks than any algebra slip in the subject. $\mathbf{v}_R$ is the time derivative of position taken while holding the rotating axes fixed: it is what a platform-riding observer genuinely measures with their own rulers and clocks. It is not the inertial velocity re-expressed in rotating components — that would be the same vector in a different basis, and a basis change can never make a moving flag read zero. A particle glued to the turntable has $\mathbf{v}_R=0$ but $\mathbf{v}_I=\boldsymbol{\omega}\times\mathbf{r}\neq0$ : two different vectors, not two descriptions of one. Two further traps: the clean form (2) assumes the frames share an origin on the rotation axis — if the origin translates, use (4); and $r_\perp$ in the entrainment speed is the distance from the axis, not from the origin.

Worked examples

1 · A bug on a spinning vinyl record (confidence builder)

A record spins at $33\tfrac{1}{3}$ rpm. A bug at radius $r=10.0\ \mathrm{cm}$ crawls radially outward at $u=2.0\ \mathrm{cm/s}$ , as measured on the record. Find its rotating-frame speed, its entrainment speed, and its inertial speed and direction.

Step 1. Convert the spin rate: $\omega = 33.333\ \mathrm{rev/min}\times\dfrac{2\pi\ \mathrm{rad/rev}}{60\ \mathrm{s/min}} = 3.49\ \mathrm{rad/s}$ .

Step 2. Rotating-frame velocity: purely radial, $|\mathbf{v}_R| = u = \boxed{0.020\ \mathrm{m/s}}$ .

Step 3. Entrainment velocity from (3): tangential, magnitude $\omega r = 3.49\times0.100 = \boxed{0.349\ \mathrm{m/s}}$ — seventeen times the bug's own crawl.

Step 4. The two pieces are perpendicular (radial ⊥ tangential), so the inertial speed is $|\mathbf{v}_I| = \sqrt{u^2+(\omega r)^2} = \sqrt{0.020^2+0.349^2} = \boxed{0.350\ \mathrm{m/s}}$ , tilted only $\arctan(u/\omega r) = 3.3°$ away from the tangential direction. The lab sees a particle moving almost purely sideways — the record's carry utterly dominates the bug's honest work, and the ground-frame path is a widening spiral.

2 · Why rockets launch from the equator (exam level)

A rocket stands on its pad, at rest in the rotating Earth frame. Using the sidereal rotation rate $\omega=7.292\times10^{-5}\ \mathrm{rad/s}$ and $R=6.371\times10^{6}\ \mathrm{m}$ , find its inertial speed at Kourou ( $\lambda=5.2°\mathrm{N}$ ) and at Baikonur ( $\lambda=45.6°\mathrm{N}$ ). How much launch benefit does Kourou enjoy, as an absolute number and as a fraction of the $\approx7.8\ \mathrm{km/s}$ speed of low Earth orbit?

Step 1. On the pad $\mathbf{v}_R=0$ , so by (2) the whole inertial velocity is entrainment: eastward, magnitude $\omega R\cos\lambda$ from (3), with $r_\perp=R\cos\lambda$ the distance to Earth's axis.

Step 2. Kourou: $v = 7.292\times10^{-5}\times6.371\times10^{6}\times\cos 5.2° = 464.6\times0.9959 = \boxed{463\ \mathrm{m/s}}$ eastward.

Step 3. Baikonur: $v = 464.6\times\cos 45.6° = 464.6\times0.6997 = \boxed{325\ \mathrm{m/s}}$ eastward.

Step 4. Launching eastward, the rocket keeps its pad velocity as a head start toward orbital speed. Kourou's advantage: $463-325 = \boxed{138\ \mathrm{m/s}}$ — about $1.8\%$ of the $7.8\ \mathrm{km/s}$ needed for low Earth orbit. That sounds small, but the rocket equation is exponential in required $\Delta v$ : for a typical launcher, shaving $138\ \mathrm{m/s}$ off the budget buys several percent more payload to orbit — worth shipping Ariane stages across an ocean. A polar launch site ( $\cos\lambda\to0$ ) would forfeit the entire bonus, and a launch westward would have to pay it back twice over.

Figure

The launch-site map — entrainment speed ωR cos λ shrinks with latitude

Origins · 1 minGalileo Galilei and how this was discovered

Galileo Galilei

1564–1642 · Pisa, Padua, Florence

Galileo owned the idea that makes this article possible: velocity is relative to a frame, and frames in uniform relative motion share the same physics. His 1632 Dialogue argued it with a thought experiment — butterflies, dripping bottles, and thrown balls behave identically below decks on a smoothly sailing ship — and with it he disarmed the strongest objection to Copernicus: that a moving Earth would blow everything off itself. His velocity-addition rule, $\mathbf{v}_I=\mathbf{V}+\mathbf{v}_R$ for a uniformly moving frame, is exactly equation (4) with the rotation switched off; equation (2) is Galileo's rule upgraded to a carry velocity $\boldsymbol{\omega}\times\mathbf{r}$ that varies from point to point. There is an irony in the upgrade: Galileo's one great misfire, his theory of the tides, tried to explain them by compounding Earth's orbital and rotational entrainment velocities — the right transformation, aimed at the wrong phenomenon. The tides needed Newton's Moon; the transformation needed only two more centuries to reveal what its second derivative was hiding.

History

1632

— Galileo's Dialogue makes the ship's-cabin argument: uniform motion is undetectable from inside, and velocities compose by addition.

1687

— Newton's Principia sharpens the stakes with the rotating-bucket argument: rotation, unlike uniform motion, is absolute — rotating frames are genuinely different.

1765

— Euler's rigid-body treatise formalizes the angular velocity vector, giving the entrainment term

\boldsymbol{\omega}\times\mathbf{r}

its modern form.

1835

— Coriolis differentiates the transformation once more and publishes the acceleration bookkeeping that bears his name.

1969

— Apollo 11 launches due east from Cape Kennedy (

28.5°

N), pocketing

408\ \mathrm{m/s}

of entrainment velocity on its way to the Moon.

Going further

References. Morin, Introduction to Classical Mechanics, Ch. 10 (Accelerating Frames of Reference) · Goldstein, Classical Mechanics 3rd ed., §4.9–4.10 · Landau & Lifshitz, Mechanics Vol. 1, §39 · Taylor, Classical Mechanics, Ch. 9

Provenance. Operator identity and velocity/acceleration transformations verified against Goldstein 3rd ed. §4.9–4.10 (Eqs. 4.86–4.91), Morin Ch. 10, Taylor Ch. 9, and Landau–Lifshitz Vol. 1 §39. Numerical values: sidereal rotation rate $\omega=7.292\times10^{-5}\,\text{rad/s}$ from sidereal day 86164 s; Earth radius $R=6.371\times10^{6}\,\text{m}$ ; equatorial entrainment speed $\omega R\approx465\,\text{m/s}$ — all standard textbook/CODATA-consistent values a verifier can recompute directly.

Builds on: Angular velocity vector of a rotating frame · Time derivatives in rotating frames · Non-inertial reference frames Leads to: Transformation of acceleration in rotating frames · Coriolis force · Centrifugal force · Effective gravity on the rotating Earth

Is the rotating-frame velocity just the inertial velocity written in rotating coordinates?

No, and the distinction is the heart of this topic. Re-expressing a vector in a different basis never changes the vector — a flag moving at $\omega r$ in the lab would still be moving at $\omega r$ after a basis change. But $\mathbf{v}_R$ is a different vector: the derivative of position taken while holding the rotating axes fixed, i.e. what a co-rotating observer actually measures. For the glued flag that measurement is exactly zero. The two vectors differ by the entrainment $\boldsymbol{\omega}\times\mathbf{r}$ , which is equation (2).

If I jump straight up, do I lose my 465 m/s and land west of where I started?

No. While you stand on the pad, you and the ground share the same entrainment velocity; when you jump, nothing removes it — you keep the eastward $465\ \mathrm{m/s}$ as free inertial velocity, so you land essentially where you left. This is precisely Galileo's ship argument applied to the rotating Earth. There are tiny corrections, because at the top of your jump you are slightly farther from the axis, where the ground's entrainment differs — but those effects are second-order bookkeeping that belongs to the acceleration transformation and is treated in the deflection-of-falling-bodies article.

Why does the entrainment speed use the sidereal day of 86164 seconds instead of 24 hours?

Because $\boldsymbol{\omega}$ in the transformation is Earth's rotation rate relative to an inertial frame, and the 24-hour solar day is not that. In 24 hours Earth rotates once relative to the Sun — but the Sun itself has moved along Earth's orbit, so the planet has actually turned about $361°$ relative to the stars. One true rotation takes 23 h 56 min 4 s = 86164 s, giving $\omega=2\pi/86164=7.292\times10^{-5}\ \mathrm{rad/s}$ . Using 24 h would overestimate every entrainment speed by about 0.3% — small, but exactly the kind of error precision navigation cannot afford.

Is the entrainment velocity a real velocity or just bookkeeping?

It is a perfectly real inertial velocity — of a specific object: the rotating-frame grid point the particle currently occupies. The flag glued to the turntable really does circle at $\omega r$ in the lab; the launch pad at Kourou really does move east at $463\ \mathrm{m/s}$ relative to the ECI frame. What is bookkeeping is the split of $\mathbf{v}_I$ into "measured" plus "carried": that split depends on which rotating frame you chose. The physics doesn't care how you split it; your instruments do.

Does the formula still work if the particle is far from the origin, or moving fast?

Yes — equation (2) is exact classical kinematics, valid for any position, any speed, and any rotation rate, with no small- $\omega$ or small- $r$ approximation anywhere. The only structural assumptions are that both frames share an origin on the rotation axis (otherwise use equation (4)) and that time and space are Newtonian. At speeds approaching light the whole Newtonian framework needs replacing, and rotating frames in relativity are genuinely subtle — but that is a different theory, not a flaw in (2).

Which ω do I use when a frame rotates inside another rotating frame, like an instrument on a spinning satellite?

Angular velocities add vectorially: if frame 2 rotates at $\boldsymbol{\omega}_{12}$ relative to frame 1, and frame 3 at $\boldsymbol{\omega}_{23}$ relative to frame 2, then frame 3 rotates at $\boldsymbol{\omega}_{13}=\boldsymbol{\omega}_{12}+\boldsymbol{\omega}_{23}$ relative to frame 1 (see the angular-velocity-vector article). For velocity transformations you may either apply equation (2) twice, once per frame, or apply it once with the summed $\boldsymbol{\omega}$ when the origins coincide. The linearity of the entrainment term in $\boldsymbol{\omega}$ is what makes stacked rotations tractable.

How is this related to the Galilean transformation I learned first?

It is the same idea with one upgrade. Galileo relates frames in uniform relative motion: $\mathbf{v}_I=\mathbf{V}+\mathbf{v}_R$ with a single constant $\mathbf{V}$ for all points. A rotating frame also adds a relative velocity — but a position-dependent one, $\mathbf{V}(\mathbf{r})=\boldsymbol{\omega}\times\mathbf{r}$ . That single difference has large consequences: because the added velocity varies from point to point, a second differentiation no longer returns $\mathbf{a}_I=\mathbf{a}_R$ ; it generates the Coriolis, centrifugal, and Euler terms. Uniform motion hides itself; rotation cannot.

Transformation of velocity between rotating frames

Where this lives today

The physical picture

The core equations

Which relation answers which question

Limiting cases

Worked examples

1 · A bug on a spinning vinyl record (confidence builder)

2 · Why rockets launch from the equator (exam level)

Going further

Before this

Where this leads