Classical Mechanics
Parallel Axis Theorem
Move the rotation axis a distance d from CM: moment of inertia increases by Md²
The parallel axis theorem (Steiner 1840s, but in earlier Euler form): for a rigid body of mass M, the moment of inertia I about an axis parallel to and a distance d from an axis through the center of mass (with moment I_cm) is I = I_cm + Md². Proof uses the fact that the cross-term ∫r×r_cm dm vanishes when integrated over the body about its CM. Examples: a uniform disc of mass M and radius R has I_cm = MR²/2 about its center; about its rim, I = MR²/2 + MR² = (3/2)MR². The perpendicular axis theorem (for planar bodies): I_z = I_x + I_y. Crucial for physical pendulum analysis (T = 2π√(I_pivot/Mgd_cm)), gyroscopes, and any rotation about a non-CM axis.
- StatementI = I_cm + Md²
- AuthorSteiner 1840s (Huygens prior)
- Disc centerI = MR²/2
- Disc rimI = 3MR²/2
- Perpendicular axisI_z = I_x + I_y (planar)
- Physical pendulumPeriod uses this
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
Why parallel axis matters
- Physical pendulum. The classic application — a swinging compound body has period T = 2π √((I_cm + Md²)/Mgd). Used in clock escapements, seismic pendulums, and tuned mass dampers.
- Rotor balancing. An off-CM rotor has effective inertia I_cm + Md², increasing torque demands and creating cyclic forces on bearings; static and dynamic balancing both rely on parallel-axis bookkeeping.
- Robot arm dynamics. Each link contributes its inertia about the joint axis, computed from the link's I_cm plus its parallel-shifted contribution Md² where d is the joint-to-CM distance.
- Engineering structures. Beam bending uses the second moment of area I, with the same parallel-axis structure: I about a non-centroidal axis is I_centroidal + Ad². Determines stiffness of girders, channels, and I-beams.
- Wheels and rolling kinematics. Rolling KE can be computed as (1/2) I_contact ω² with I_contact = I_cm + MR². Same shape, different roll speeds depending on where mass sits relative to the axle.
- Sports physics. A baseball bat swung about a grip point off the bat's CM has effective inertia I_grip = I_cm + Md²; bat speed and the sweet spot location follow from the formula.
- Tools and instruments. Hammers, axes, and rackets are designed to keep moment-arm geometry favorable; parallel axis quantifies the trade between mass and reach.
Common misconceptions
- "Md² could be smaller than I_cm minus something." No — Md² is non-negative, so I about any parallel axis is always greater than or equal to I_cm. The CM axis minimizes I among parallel axes.
- "Applies to any axis." Only to axes parallel to the CM axis. Tilted axes need the full inertia tensor and a rotation, not just an Md² shift.
- "Inertia adds linearly across axes." Wrong direction — I = I_cm + Md² shifts to a parallel axis. To combine moments about different axes you have to use the tensor, not just add scalars.
- "Theorem applies to mass moments only." The same algebra works for second moments of area in beam theory, plus the tensor generalization in 3D rigid-body dynamics.
- "Distance d is from any reference." d is specifically the perpendicular distance between the two parallel axes. Picking the wrong distance gives the wrong I.
- "Steiner discovered it." Huygens used the equivalent relation in his 17th-century pendulum-clock work; Jakob Steiner formalized it in the 19th century. Common in older European texts as Steiner's parallel axis theorem.
Frequently asked questions
Why does the moment of inertia increase by Md² off-axis?
Define r' = r − d, where d is the offset from the new axis to the CM. Substitute into I = ∫|r|² dm: ∫|r' + d|² dm = ∫|r'|² dm + 2d · ∫r' dm + d² ∫dm. The middle integral ∫r' dm is zero by definition of the center of mass — that is the magic step. The first integral is I_cm; the last is M d². The Md² term reflects that every mass element is now farther from the axis by at least d, so its contribution to I picks up a constant geometric penalty proportional to d².
What is the perpendicular axis theorem?
For a planar lamina (a flat 2D body in the xy-plane) only, I_z = I_x + I_y, where the three axes pass through the same point. Proof: I_x = ∫y² dm, I_y = ∫x² dm, I_z = ∫(x² + y²) dm = I_x + I_y. Useful shortcut for thin plates, discs, and rings: knowing two of the three diametral and axial moments gives you the third for free. Does NOT extend to 3D solids — it is strictly a flat-body theorem.
How is it used in physical pendulum analysis?
A physical pendulum is any rigid body swinging from a pivot above its center of mass. Its small-angle period is T = 2π √(I_pivot / Mgd_cm), where d_cm is the distance from pivot to CM. By the parallel axis theorem, I_pivot = I_cm + M d_cm². So T = 2π √((I_cm + Md_cm²)/(Mgd_cm)). Plug in I_cm for the body shape — for example a uniform rod of length L pivoted at one end has I_cm = ML²/12, d_cm = L/2, giving T = 2π √(2L/3g). Underpins rate gyros, balance wheels, and any pendulum that is not a point mass.
Why does the cross term vanish at CM?
The center of mass of a body, expressed in coordinates whose origin sits at the CM, is at zero by construction: ∫r' dm = 0. So when you expand |r' + d|² = |r'|² + 2 r' · d + d², the middle cross term ∫(2 r' · d) dm = 2 d · ∫r' dm = 0. This is exactly the geometric meaning of CM — it is the unique point about which the first moment of mass vanishes. If you parallel-shift to any other reference point (not the CM), the cross term does not vanish and the simple Md² formula breaks.
What is the analog for tensors?
Yes — the full tensor version is I_O = I_cm + M (d² 𝟙 − d ⊗ d), where d is the displacement vector from the CM to the origin O, 𝟙 is the 3×3 identity, and d ⊗ d is the outer product matrix d_i d_j. Diagonal elements grow by M(d_y² + d_z²) for the x-axis component, etc.; off-diagonals pick up −M d_i d_j products of inertia. Recovers the scalar version when you take the projection along a single axis. Used in robot dynamics and aerospace inertia computations.
Why is rolling kinetic energy K = (1/2)Iω² + (1/2)Mv²?
A rolling body has two contributions: translation of its CM (with energy (1/2)Mv²) and rotation about its CM (with energy (1/2)I_cm ω²). Equivalent restatement using I_contact = I_cm + MR² (parallel axis to the rolling contact point, distance R): K = (1/2) I_contact ω². Both forms agree when v = ωR (rolling without slipping). For a uniform sphere I_cm = (2/5)MR², so K = (7/10)Mv²; for a solid cylinder K = (3/4)Mv²; for a thin hoop K = Mv². Why a sphere beats a hoop down a ramp.