Intuition The ONE core idea
A spinning solid body stores its "resistance to being turned" in a 3 × 3 grid of numbers called the inertia tensor — it takes the arrow that says how you spin (ω ) and outputs the arrow that says how much rotation you carry (L ). This page builds, from nothing, every symbol needed to understand that sentence: vectors, cross products, mass sums, matrices, and what "principal axis" even means.
This is a prerequisites page. Before you can read the parent note , you need to own every symbol it throws at you. We go one at a time, each anchored to a picture, each building on the last.
A rigid body is a solid object whose particles never change their distances from one another — a spinning top, a book, a brick. It cannot squash or bend. When it rotates, every particle sweeps around the same axis by the same angle in the same time.
Picture a brick glued together from millions of tiny mass-dots. If the brick turns 10° , every dot turns 10° . That single fact is what makes the whole theory possible — see Rigid body rotation .
The figure below draws this: faint dots are the "before" brick, bright dots the "after". Look at the yellow and green arrows — a corner dot and an inner dot swing through the same angle around the fixed red origin. Nothing stretches; only orientation changes.
Definition Position vector
r = ( x , y , z ) is an arrow from a chosen origin (a fixed point in the body, often the centre) to a particle. The three numbers x , y , z are how far along the three perpendicular axes you must walk to reach it.
The little arrow on top always means "this is a vector — it has a direction , not just a size."
Definition Standard basis vectors
i ^ , j ^ , k ^
Along the x , y , z axes we place three unit arrows — arrows of length exactly 1 — called i ^ , j ^ , k ^ . The little hat ^ is a promise: "this arrow has length 1 ; it only carries direction, not size." Any vector is a sum of these:
r = x i ^ + y j ^ + z k ^ .
So "the x -component of r " just means "how many copies of i ^ you need." We will use the same hat later for unit direction arrows like e ^ .
Definition Length (magnitude)
r
r = ∣ r ∣ = x 2 + y 2 + z 2 is the plain distance from origin to the particle — a single positive number, no direction. That is why it has no arrow and no hat.
Why the topic needs this: the inertia tensor is built from where each particle sits, i.e. from its components x , y , z . No positions, no tensor.
m a and ∑ a
Label each particle with an index a = 1 , 2 , 3 , … . Then m a is the mass of particle number a , and the symbol ∑ a (capital Greek "sigma") means "add up the following expression over every particle a ."
∑ a m a = m 1 + m 2 + m 3 + ⋯ = total mass.
For a smooth solid body there are infinitely many infinitely-small chunks. Adding infinitely many tiny pieces is exactly what an integral does. To make sense of the mass of an "infinitely small chunk," we need one more idea: density .
ρ ( r ) and d m = ρ d V
ρ ( r ) (Greek "rho") is the mass packed into each unit of volume at the point r — kilograms per cubic metre. A tiny box of volume d V sitting at r therefore holds mass
d m = ρ ( r ) d V .
So the smooth-body sum becomes a volume integral :
∫ ( ⋯ ) d m = ∫ body ( ⋯ ) ρ ( r ) d V .
For a uniform body ρ is a constant and can be pulled outside the integral.
∑ ↔ ∫
Beads on a string → use ∑ a m a . Continuous rope → use ∫ ρ d V . Same "add it all up" idea, different notation.
Why the topic needs this: the entries of I are mass-weighted sums of coordinate products. For a solid body you compute them as volume integrals over ρ d V — exactly how the parent's cube example gets 6 1 M L 2 .
Definition Angular velocity vector
ω (Greek "omega") is one arrow that packages everything about a spin: it points along the axis of rotation , and its length is how fast you turn (radians per second). The right-hand rule fixes the direction: curl your right fingers the way the body turns, your thumb points along ω .
The figure shows the disc spinning (green circulation arrows) while the single yellow ω arrow stands straight up the axis — its direction is the axis, its length is the rate.
Intuition Why an arrow for a spin?
A spin has two facts: which axis and how fast/which way . An axis is a direction → arrow direction. Speed+sense → arrow length + right-hand rule. One arrow, both facts. This is why we can write ω = ( ω x , ω y , ω z ) = ω x i ^ + ω y j ^ + ω z k ^ just like a position.
Why the topic needs this: ω is the input to the inertia tensor. L = I ω reads "feed the spin arrow in, get the momentum arrow out."
Two arrows can be multiplied in a way that produces a third arrow . We need it twice: for velocity (v = ω × r ) and for angular momentum (L = r × m v ).
Definition Cross product (geometric)
A × B is the arrow that is perpendicular to both A and B , with length ∣ A ∣∣ B ∣ sin θ (where θ is the angle between them), pointing along the direction your right thumb gives when fingers sweep A into B .
In the figure, blue A and green B lie in the page; the yellow shaded parallelogram is their ∣ A ∣∣ B ∣ sin θ area; the red dot-in-a-circle is A × B pointing straight out of the page — perpendicular to both.
Intuition Why cross product for spinning?
A particle at r spinning with ω moves in a circle — its velocity is perpendicular to both the axis and the position arm. "Perpendicular to both" is exactly the job of the cross product, so v = ω × r is the natural (and correct) formula. And because the component form turns cross products into sums of coordinate products, it is what lets the parent note grind r × ( ω × r ) down into the coordinate sums that fill I .
Recall Quick check: cross product of parallel arrows
If A ∥ B then θ = 0 , so sin θ = 0 , so A × B = 0 .
A particle sitting on the spin axis (r ∥ ω ) therefore has v = 0 — it doesn't move. Correct: the axis is stationary.
A ⋅ B = A x B x + A y B y + A z B z = ∣ A ∣∣ B ∣ cos θ — multiply matching components and add. The result is a single number (no arrow). It measures "how much the two arrows point the same way."
Why the topic needs it: the derivation of I produces the term r ( r ⋅ ω ) . The dot r ⋅ ω collapses to one number that then re-scales r .
Common mistake Confusing dot and cross
Why it's tempting: both "multiply two vectors."
Fix: dot (⋅ ) → number , measures alignment. Cross (× ) → arrow , measures perpendicular twist. In r × ( ω × r ) you use both, in that exact order.
Definition Angular momentum
For one particle, L = r × m v . It is the rotational cousin of ordinary momentum m v : a vector saying how much rotational motion the particle carries and about which axis. See Angular momentum .
Why the topic needs it: L is the output of the inertia tensor. The entire topic exists to answer: given the spin ω , what is L ?
The parent note fills a 3 × 3 grid with entries labelled I xx , I x y , I y z , and so on. Here is exactly what each label means, before you ever meet it in the derivation. The two subscripts i and j each stand for one of the axes x , y , z ; I ij is the entry in row i , column j .
Unpacking that one formula gives every entry the parent uses:
Definition Diagonal entries — moments of inertia
When i = j the switch δ ij = 1 , so the r 2 survives and x i x j = x i 2 :
I xx = ∑ m ( y 2 + z 2 ) , I y y = ∑ m ( x 2 + z 2 ) , I z z = ∑ m ( x 2 + y 2 ) .
Each is a sum of squared perpendicular distances to that axis — always ≥ 0 .
Definition Off-diagonal entries — products of inertia
When i = j the switch δ ij = 0 , so only the − x i x j term is left — note the minus sign, built right into the formula:
I x y = I y x = − ∑ m x y , I x z = − ∑ m x z , I y z = − ∑ m y z .
These can be + , − , or 0 . The minus sign is not a convention you tack on later — it is the − x i x j inside I ij .
Intuition Where the pieces come from (foreshadowing the derivation)
The δ ij r 2 part is the "how far from the axis, squared" piece — resistance grows with distance squared (a mass twice as far is four times as hard to spin). The − x i x j part measures how lopsided the mass is across two axes; when it is nonzero, spinning about one axis leaks momentum toward another, which is exactly why L can tilt away from ω .
Common mistake "Products of inertia are
+ ∑ m x y ."
Why it's tempting: the diagonal moments are positive sums, so a plus feels natural.
Fix: the single formula I ij = ∑ m ( δ ij r 2 − x i x j ) carries a minus on the x i x j term. Off-diagonals are − ∑ m x i x j . Wrong sign → wrong eigenvectors.
A matrix is a rectangular grid of numbers. The inertia tensor is 3 × 3 (three rows, three columns). We write matrices in bold upright : I . Its entry in row i , column j is the I ij we just defined.
Definition Matrix times vector
A 3 × 3 matrix acting on a 3 -vector gives a new 3 -vector: take the dot product of each row with the input vector to get each output component.
I ω = I xx I y x I z x I x y I y y I z y I x z I y z I z z ω x ω y ω z = I xx ω x + I x y ω y + I x z ω z I y x ω x + I y y ω y + I y z ω z I z x ω x + I z y ω y + I z z ω z
A matrix is a machine that turns one arrow into another — possibly rotating and stretching it. Because the incoming spin ω and outgoing L can point in different directions, we need exactly such a machine. That machine is I . See Symmetric and orthogonal matrices .
Definition Diagonal matrix
Only the entries with i = j (top-left to bottom-right) are nonzero; every off-diagonal entry is 0 . A diagonal matrix simply stretches each axis by its diagonal number — it never tilts an arrow.
Definition Symmetric matrix
A matrix equal to its own mirror-across-the-diagonal: the entry in row i , column j equals the entry in row j , column i . The inertia tensor is symmetric because the formula I ij = − ∑ m x i x j (off-diagonal) does not care about the order of x i and x j , so I ij = I j i .
Definition Identity matrix
1
The diagonal matrix with 1 s on the diagonal (its entries are exactly δ ij ). It leaves every vector unchanged: 1 v = v . It is the "do nothing" machine.
Intuition Why we crave the diagonal form
If I were diagonal, then I ω would just scale each component — L and ω would line up axis-by-axis and the messy physics vanishes. Finding the frame where I becomes diagonal is the entire punchline "principal axes."
Definition Eigenvector & eigenvalue
An eigenvector of a matrix M is a special arrow that the matrix does not tilt — it only stretches it. Because only its direction matters, we normally scale it to length 1 and write it with a hat, e ^ (the hat means "unit vector," as in §1). The defining equation is
M e ^ = λ e ^ ,
and the stretch factor λ (Greek "lambda") is the eigenvalue . See Eigenvalues and eigenvectors .
exactly a principal axis
A principal axis is defined as a spin direction where L stays parallel to ω . In symbols, I e ^ = λ e ^ — the eigenvector equation, letter for letter, with the unit arrow e ^ being the axis direction and λ the principal moment. So: principal axes = (unit) eigenvectors of I , and principal moments = eigenvalues of I . That is the whole topic in one sentence.
Rigid body: fixed distances
Length r and squares x2 y2 z2
Mass m_a sum and density rho dV
Angular velocity omega spin arrow
Cross product v = omega cross r
Angular momentum L = r cross m v
Tensor components I_ij formula
Matrix times vector L = I omega
Eigenvalues and eigenvectors
Principal axes and principal moments
Self-test: can you answer each before opening the parent note?
What does the arrow on r mean and what is r without an arrow? r is a direction+length (a vector, position); plain
r = x 2 + y 2 + z 2 is just the distance, a single number.
What do the basis vectors i ^ , j ^ , k ^ mean and what does a hat signal? They are unit (length-1) arrows along the x , y , z axes; a hat marks "this vector has length 1, direction only." Any vector is x i ^ + y j ^ + z k ^ .
When do you use ∑ a m a versus ∫ ρ d V ? ∑ for discrete point masses; ∫ ρ d V (with d m = ρ d V ) for a continuous solid body — same "add it all up" idea.
What two facts does the single arrow ω encode? Its direction = the rotation axis (right-hand rule); its length = the spin rate.
What kind of object does the cross product output, and the dot product? Cross → a new vector perpendicular to both (components ( A y B z − A z B y , … ) ); dot → a single number measuring alignment.
Why is v = ω × r perpendicular to the axis? The cross product is always perpendicular to both inputs, and a spinning particle moves in a circle around the axis.
Write the general inertia-tensor component I ij . I ij = ∑ m ( δ ij r 2 − x i x j ) ; diagonal → moments ∑ m ( y 2 + z 2 ) etc., off-diagonal → products − ∑ m x i x j .
What is special about a diagonal matrix acting on a vector? It only stretches each component — it never tilts the arrow.
Why is the inertia tensor symmetric? The off-diagonal − ∑ m x i x j is unchanged if you swap i and j , so I ij = I j i .
State the eigenvector equation, including what the hat and λ mean. I e ^ = λ e ^ ;
e ^ is a unit-length principal-axis direction,
λ the principal moment — the spin direction where
L ∥ ω .
Once every line is easy → go to the parent note .