The radius of gyration is calculated by taking the square root of the moment of inertia divided by the area (or mass). In its simplest form: k = √(I/A), where k is the radius of gyration, I is the moment of inertia, and A is the cross-sectional area. This single value, measured in metres or inches, tells you how far from a given axis the area or mass of an object is effectively concentrated.
The Core Formula
The radius of gyration works along specific axes, so you calculate it separately for each one. For a cross-section analyzed about the x-axis and y-axis:
- About the x-axis: k_x = √(I_x / A)
- About the y-axis: k_y = √(I_y / A)
Here, I_x and I_y are the area moments of inertia about each respective axis, and A is the total cross-sectional area. You can rearrange the formula to express the moment of inertia in terms of the radius of gyration: I_x = k_x² × A. This rearranged form is useful when you already know the radius of gyration (from a steel table, for example) and need the moment of inertia.
The moment of inertia itself comes from integrating the square of the distance from each tiny area element to the axis of rotation across the entire cross-section: I = ∫ R² dA. The radius of gyration essentially collapses that integral into a single representative distance. Think of it as the answer to the question: “If I squeezed all of this area into a thin ring at one distance from the axis, how far out would that ring be to produce the same moment of inertia?”
Area vs. Mass Radius of Gyration
The formula above uses the area moment of inertia, which is the version most common in structural and civil engineering when analyzing beam and column cross-sections. But there’s a parallel version based on mass. For a three-dimensional object rotating about an axis, the mass-based radius of gyration is k = √(I_m / m), where I_m is the mass moment of inertia and m is the total mass. The math is identical in structure. The difference is purely in what you’re distributing: area in a 2D cross-section, or mass in a 3D body.
Both versions produce a result with units of length. In SI, that’s metres (or millimetres in practice). In US customary units, it’s inches or feet.
Step-by-Step Example: Rectangular Cross-Section
A rectangle is one of the most common shapes you’ll calculate, and it demonstrates the process clearly. Say you have a rectangular cross-section with height h and width b.
Step 1: Calculate the area. A = b × h.
Step 2: Find the moment of inertia about the axis you care about. For the horizontal axis passing through the center (the centroidal x-axis), the moment of inertia of a rectangle is I_x = bh³/12. For the vertical centroidal axis, I_y = b³h/12.
Step 3: Divide and take the square root. The radius of gyration about the horizontal centroidal axis simplifies to k_x = h/√12, which is approximately h/3.46. About the vertical axis, k_y = b/√12, or roughly b/3.46.
Notice that the width b drops out of k_x entirely. The radius of gyration about a horizontal axis depends only on the height of the section, and vice versa. This makes intuitive sense: spreading material farther from an axis increases the moment of inertia about that axis, and the radius of gyration captures exactly how far that spread extends.
For a concrete example, take a rectangle 200 mm tall and 100 mm wide. The area is 20,000 mm². The moment of inertia about the horizontal centroidal axis is (100 × 200³)/12 = 66,666,667 mm⁴. The radius of gyration is √(66,666,667 / 20,000) = 57.7 mm, which matches 200/√12.
Why It Matters for Column Buckling
The radius of gyration plays a central role in predicting whether a column will buckle under load. Euler’s buckling formula uses a quantity called the slenderness ratio, defined as L/r, where L is the unsupported length of the column and r is the radius of gyration of its cross-section. A high slenderness ratio means the column is long and thin relative to its cross-section, making it more likely to buckle. Columns classified as “long” based on their slenderness ratio are analyzed using Euler’s formula, which calculates the critical stress at which buckling begins.
The axis that matters most is the one with the smallest radius of gyration. A column will buckle about its weakest axis first, so engineers look for the minimum k value when evaluating stability. For a wide-flange steel beam used as a column, for instance, the weak axis (usually the one running through the thinner dimension) produces the smaller radius of gyration and governs the design. This is why steel tables for structural shapes always list the radius of gyration for both axes.
If you’re selecting a column section, the practical takeaway is straightforward: a larger minimum radius of gyration lets the column carry more load before buckling, or allows a longer unsupported span at the same load.
Radius of Gyration in Molecular Science
The same concept applies at the molecular scale, though the formula looks slightly different. For a protein or polymer chain, the radius of gyration (R_g) measures how compact or spread out the molecule is. It’s defined as the root-mean-square distance of all atoms from the molecule’s geometric center.
In the simplest version, you take each atom’s distance from the center of the molecule, square all those distances, average them, and take the square root. A mass-weighted version does the same thing but accounts for the fact that different atoms have different masses, weighting heavier atoms more in the calculation.
For polymers, the physicist Paul Flory showed that the radius of gyration follows a power-law relationship with chain length: R_g = R_0 × N^ν, where N is the number of monomers (or residues), R_0 is a scaling prefactor, and ν is an exponent that approaches 0.6 for long polymer chains in good solvents. This exponent reflects how the chain swells when it interacts favorably with the surrounding solvent rather than folding back on itself. Molecular R_g values are typically reported in nanometres.
For Composite and Irregular Shapes
When your cross-section isn’t a simple rectangle or circle, the process adds a step but follows the same logic. For composite shapes (an I-beam, a T-section, or any shape built from simpler pieces), you first calculate the total moment of inertia of the composite shape about the axis of interest using the parallel axis theorem to shift each component’s moment of inertia to the common centroidal axis. Then you divide by the total area and take the square root, just as before.
For truly irregular shapes where no neat formula exists, the moment of inertia is computed numerically, either through integration in software or by dividing the shape into small elements and summing their contributions. The radius of gyration calculation itself never changes: it’s always √(I/A) for area-based problems or √(I/m) for mass-based ones. The challenge is always in getting an accurate moment of inertia, not in the final step.