Perpendicular Axis Theorem

Take a flat circular disc — a carrom coin, a CD, a chapati rolled flat on a counter. You know its moment of inertia when it spins flat on a table like a top: I_z = \tfrac{1}{2}MR^2, the standard result for rotation about an axis through the centre, perpendicular to the face.

Now flip the coin and spin it end-over-end, the way you flip a coin before a cricket match. You are spinning it about a diameter — a completely different axis. Finding this new moment of inertia seems like it should require setting up a fresh integral over the disc, working through the geometry from scratch.

It does not. One line of algebra gives the answer: the moment of inertia about any diameter is exactly \tfrac{1}{4}MR^2 — half the value you already knew. That factor of one-half is not a numerical coincidence. It falls out of a clean geometric fact about flat bodies and perpendicular axes: the perpendicular axis theorem.

The trick works for every flat shape — not just discs. Rings, rectangular plates, triangular sheets, the flat blade of a ceiling fan — any body whose mass lies entirely in a single plane. Once you have the theorem, a large class of moment-of-inertia problems that look like they need integration reduce to a line of arithmetic.

What the theorem says — the geometry

Picture a flat object — a metal plate, a playing card, a chapati — lying in the xy-plane. Pick any point in the plane and draw three mutually perpendicular axes through it: the x-axis and y-axis both lie in the plane of the body, and the z-axis points straight out of the plane, perpendicular to the body.

Look at the mass element dm in the figure. Its distance from the z-axis is r_z = \sqrt{x^2 + y^2}. Its distance from the x-axis is just |y| — there is no z-component because the body is flat. Its distance from the y-axis is just |x|. The Pythagorean theorem connects these three distances: r_z^2 = x^2 + y^2 = r_y^2 + r_x^2. Since the moment of inertia is \int r^2 \, dm, this means I_z must equal I_y + I_x. The formal proof below simply writes out this argument as an integral.

The physical intuition is Pythagorean. For any mass element at position (x, y) in the plane, its squared distance from the z-axis is x^2 + y^2 — the Pythagorean theorem in the xy-plane. But y^2 is the squared distance from the x-axis, and x^2 is the squared distance from the y-axis. Since moment of inertia is the integral of (squared distance) \times dm, the z-integral is automatically the sum of the x-integral and the y-integral. The Pythagorean theorem at the level of individual mass elements becomes an additive rule at the level of the entire body.

The proof

The proof is five lines, and every line follows from one fact: the body is flat, so every mass element has z = 0.

Setup: The body lies entirely in the xy-plane. Every mass element dm sits at a position (x, y, 0). The three axes — x, y, and z — are mutually perpendicular and pass through the same point (the origin).

Step 1. Write the moment of inertia about the z-axis.

where r_z is the perpendicular distance of each mass element from the z-axis. For a point at (x, y, 0), that distance is r_z = \sqrt{x^2 + y^2}, so:

Why: the z-axis passes through the origin along the z-direction. The distance from any point (x, y, 0) to this axis is the distance in the xy-plane, which is \sqrt{x^2 + y^2}. Squaring removes the square root.

Step 2. Write the moment of inertia about the x-axis.

The distance of (x, y, 0) from the x-axis is r_x = \sqrt{y^2 + z^2} = \sqrt{y^2 + 0} = |y|. So:

Why: the x-axis lies along the x-direction. The perpendicular distance from the x-axis is measured in the yz-plane. But z = 0 for every mass element (the body is flat), so the distance reduces to |y|. This is the key step — flatness eliminates the z^2 term.

Step 3. Write the moment of inertia about the y-axis.

By the same reasoning, r_y = \sqrt{x^2 + z^2} = |x|. So:

Why: identical argument to Step 2, but now the distance from the y-axis is measured in the xz-plane. With z = 0, that distance is |x|.

Step 4. Add equations (2) and (3).

Why: the integral of a sum is the sum of the integrals. Both integrals run over the same mass distribution, so they combine into a single integral over (x^2 + y^2).

Step 5. Compare equation (4) with equation (1).

Why: the right-hand side of equation (4) is identical to equation (1). That is the entire proof — the flatness condition (z = 0) made r_x^2 = y^2 and r_y^2 = x^2, so adding them gives x^2 + y^2 = r_z^2. The Pythagorean theorem, integrated over the body.

The proof is clean because the geometry is clean. For a flat body, the perpendicular distances to the in-plane axes involve only one coordinate each (y for the x-axis, x for the y-axis), while the distance to the perpendicular axis involves both (\sqrt{x^2 + y^2}). The Pythagorean theorem does the rest.

Applications — finding moments the easy way

The theorem's power shows up when you already know I_z (or can look it up from the standard table in the moment of inertia article) and want I_x or I_y.

Disc about a diameter

A uniform disc of mass M and radius R has I_z = \tfrac{1}{2}MR^2 about the axis through its centre, perpendicular to its face — the standard result from integrating r^2 \, dm over the disc area.

Now pick any two perpendicular diameters as the x-axis and y-axis. By the symmetry of the disc, every diameter is equivalent — a disc looks the same from every direction in its plane. So I_x = I_y.

Apply the perpendicular axis theorem:

Without the perpendicular axis theorem, finding I_{\text{diameter}} requires evaluating a double integral with careful limits. The theorem gives the same answer in two lines.

Ring about a diameter

A uniform ring (all mass concentrated at radius R) has I_z = MR^2 about the axis through its centre, perpendicular to its plane. Every mass element sits at the same distance R from the centre, so the integral is trivial: \int r^2 \, dm = R^2 \int dm = MR^2.

The same symmetry argument applies — all diameters of a ring are equivalent, so I_x = I_y:

The ring's moment about a diameter is half its moment about the perpendicular axis — the same factor of two, for the same reason.

These two results — \tfrac{1}{4}MR^2 for the disc and \tfrac{1}{2}MR^2 for the ring — illustrate why the ring has a larger moment about a diameter relative to its mass. In the disc, mass near the centre contributes little to I_{\text{diameter}}, while in the ring all mass is at the maximum radius R.

The pattern is the same every time you use the perpendicular axis theorem: start with the known moment about the perpendicular axis, use symmetry to establish that the two in-plane moments are equal, and divide by two. The theorem is at its most powerful when the body has rotational symmetry — discs, rings, and regular polygons — because symmetry gives you I_x = I_y for free.

Why only flat bodies?

The perpendicular axis theorem has a strict limitation: the body must be flat. It must be a two-dimensional mass distribution — all mass lying in a single plane. The theorem fails for three-dimensional bodies like spheres, cylinders, or cubes.

Here is exactly where the proof breaks. Go back to Step 2. For a three-dimensional body, a mass element sits at (x, y, z) with z \neq 0. Its distance from the x-axis is no longer just |y| — it is \sqrt{y^2 + z^2}. So:

Adding them:

The extra term 2\int z^2 \, dm makes I_x + I_y > I_z for any body with mass away from the plane. The thicker the body, the larger this surplus. Only when z = 0 everywhere — a perfectly flat body — does the extra term vanish and the theorem hold.

Think of it this way: when you rotate a flat disc about the z-axis, every mass element swings around in the xy-plane at a distance \sqrt{x^2 + y^2} from the axis. That same element's contribution to I_x comes from its distance y in one direction, and its contribution to I_y comes from its distance x in the perpendicular direction. The Pythagorean theorem ensures these add up to the z-contribution. But if the body has thickness — mass above and below the plane — then rotating about the x-axis involves the distance \sqrt{y^2 + z^2}, which includes the z-component. This extra dimension adds to both I_x and I_y but not to I_z, breaking the balance.

A quick check with a solid cylinder. Take a uniform solid cylinder: mass M = 2 kg, radius R = 0.1 m, height h = 0.3 m, with its symmetry axis along z.

• I_z = \tfrac{1}{2}MR^2 = \tfrac{1}{2} \times 2 \times 0.01 = 0.010 kg·m²
• I_x = I_y = \frac{M(3R^2 + h^2)}{12} = \frac{2(0.03 + 0.09)}{12} = 0.020 kg·m²

If the theorem held: I_x + I_y = 0.040 kg·m². But I_z = 0.010 kg·m² — the sum is four times too large. The difference, 0.030 kg·m², is exactly \frac{Mh^2}{6} = \frac{2 \times 0.09}{6} — the penalty for mass spread across the 30 cm height of the cylinder.

Watch it break — explore the thickness

A rectangular metal plate has length a = 2 m and width b = 1 m, with total mass M = 1 kg. When the plate is perfectly flat (thickness c = 0), the perpendicular axis theorem holds exactly. Drag the thickness upward and watch I_x + I_y pull away from I_z.

Worked examples

Common confusions

• "The theorem works for any three mutually perpendicular axes." Not quite — the z-axis must be perpendicular to the body itself. All three axes pass through the same point, but the body must be a flat lamina lying in the plane of the other two axes. You cannot apply the theorem to a solid sphere by picking three axes through its centre.

• "The axes must pass through the centre of mass." No. The three axes can meet at any common point — the centre, an edge, a corner. The theorem holds as long as the body is flat and the z-axis is perpendicular to it. Of course, if you want standard tabulated values, centre-of-mass axes are the most useful starting point, and you can shift to other parallel axes using the parallel axis theorem.

• "This theorem is the same as the parallel axis theorem." They are different tools for different jobs. The parallel axis theorem relates the moment about a centre-of-mass axis to the moment about any parallel axis at a distance d: I = I_{\text{cm}} + Md^2. The perpendicular axis theorem relates the moment about a perpendicular axis to the moments about two in-plane axes: I_z = I_x + I_y. They solve different problems but combine beautifully — see the going-deeper section below.

• "I_z is always larger than I_x and I_y." This is actually true — and worth noting. Since I_z = I_x + I_y and both I_x and I_y are strictly positive (assuming the body has finite extent in both directions), I_z must be strictly greater than either one alone. A flat body always has its largest moment of inertia about the axis perpendicular to its face.

• "A thin rod is not a flat body, so the theorem does not apply." A thin rod along the x-axis does lie in the xy-plane — it is a degenerate flat body with zero width. The theorem gives: I_x = 0 (all mass sits on the axis), I_y = \frac{1}{12}ML^2, and I_z = I_x + I_y = \frac{1}{12}ML^2. The result is correct — I_y and I_z are equal because the rod is one-dimensional, so the y-axis and z-axis are equivalent by symmetry.

Where this leads next

• Parallel Axis Theorem — how to shift the axis of rotation to a parallel axis through any point, and why the centre-of-mass axis always gives the minimum moment.
• Torque — the rotational analogue of force, and why moment of inertia matters for angular acceleration.
• Moment of Inertia — the full table of standard moments and the integrals that produce them.
• Newton's Second Law for Rotation — \tau = I\alpha, where knowing I about the right axis is the first step.
• Rotational Kinetic Energy — K = \frac{1}{2}I\omega^2, which depends directly on which axis you choose.