Structural

Beam Bending (Euler-Bernoulli)

Deflection, stress and curvature from EI, L, load and end conditions

Euler-Bernoulli beam theory predicts how a slender beam deflects and develops bending stress when loaded transversely. The governing equation, EI d⁴y/dx⁴ = w(x), reduces every common case to a closed-form formula once the boundary conditions are fixed. Three numbers — flexural rigidity EI, length L, and load type — are enough to predict deflection within a few percent for any beam with length-to-depth ratio above ten. The same theory underlies aircraft wings, bridge girders and the rebar in your house's slab.

  • Governing equationEI d⁴y/dx⁴ = w(x)
  • Cantilever tip deflection (point)δ = FL³ / (3EI)
  • Bending stressσ = M·c / I
  • ValidityL/d > 10 (slender beams)
  • E (steel)200 GPa
  • E (concrete)~25 GPa

Interactive visualization

Press play, or step through manually. The visualization is yours to drive — try it before reading on.

Open visualization fullscreen ↗

Watch the 60-second explainer

A condensed visual walkthrough — narrated, captioned, under a minute.

The Euler-Bernoulli equation

For a slender beam bending under a distributed load w(x), the deflection y(x) satisfies:

EI · d⁴y/dx⁴ = w(x)

where:
  E = Young's modulus  (Pa)
  I = second moment of area  (m⁴)
  w(x) = transverse load per unit length  (N/m)
  y(x) = deflection (m)

Integrating twice gives the bending moment, twice more gives the deflection, with four constants of integration set by boundary conditions. The bending stress at a point is then σ = M·c / I, where c is the distance from the neutral axis.

The theory rests on three assumptions: cross-sections that were plane before bending stay plane after bending, those sections stay perpendicular to the deformed neutral axis (so shear deformation is ignored), and the material is linearly elastic. Together these collapse the beam from a 3D solid into a 1D problem in y(x).

Boundary conditions

CantileverSimply supportedFixed-fixedPropped cantileverOverhangingContinuous (multi-span)
End AFixed (clamped)Pin (no moment)FixedFixedPinPin
End BFreeRollerFixedPin / rollerRoller w/ overhangPin (intermediate)
Static determinacyDeterminateDeterminateIndeterminate (×2)Indeterminate (×1)DeterminateIndeterminate
Tip deflection (point P at midspan)PL³/(3EI) at tipPL³/(48EI)PL³/(192EI)0.0098·PL³/EIvaries by overhangvaries by spans
UDL midspan deflectionwL⁴/(8EI) at tip5wL⁴/(384EI)wL⁴/(384EI)wL⁴/(185EI)variesvaries
Max momentwL²/2 at fixed endwL²/8 at midspanwL²/12 at endswL²/8 (fixed end)variesvaries
Use caseDiving boards, balconiesMost simple beamsSlabs in moment framesContinuous girder over a pierCantilever overhangsMulti-span bridges

Worked example: cantilever tip deflection

A steel cantilever beam of length 2 m supports a 10 kN point load at its tip. The cross-section has second moment of area I = 4 × 10⁻⁵ m⁴. How much does the tip deflect?

δ = FL³ / (3EI)

  F = 10 kN     = 10,000 N
  L = 2 m       → L³ = 8 m³
  E = 200 GPa  = 2 × 10¹¹ Pa
  I = 4 × 10⁻⁵ m⁴

EI = 2 × 10¹¹ × 4 × 10⁻⁵
   = 8 × 10⁶ N·m²

δ = (10,000 × 8) / (3 × 8 × 10⁶)
  = 80,000 / 2.4 × 10⁷
  = 3.33 × 10⁻³ m
  = 3.3 mm

The tip drops 3.3 mm under load. To halve that deflection, doubling I (e.g. choosing the next size up in I-beams) is the cheapest path, because doubling E means switching from steel to a stiffer material — there are very few materials stiffer than steel that aren't exotic.

Worked example: bending stress in the same beam

What's the maximum bending stress, and is the beam safe? Take the cross-section depth as 0.2 m, so c = 0.1 m (distance from neutral axis to extreme fibre):

Maximum moment occurs at the fixed end:
  M_max = F × L = 10,000 × 2 = 20,000 N·m

Maximum bending stress:
  σ_max = M·c / I
        = 20,000 × 0.1 / (4 × 10⁻⁵)
        = 2,000 / 4 × 10⁻⁵
        = 5.0 × 10⁷ Pa
        = 50 MPa

Steel yield (typical S275): 275 MPa
Safety factor: 275 / 50 = 5.5

The beam is well below yield. In practice, code-prescribed safety factors and load combinations would take the design moment to perhaps 1.4 × 20 kN·m = 28 kN·m, still leaving stresses around 70 MPa — comfortably elastic.

Calculating I for common shapes

  • Rectangle (b × h): I = b·h³ / 12. Doubling h gives 8× the I.
  • Solid round (radius r): I = π·r⁴ / 4.
  • Hollow round tube: I = π(r₀⁴ − r_i⁴) / 4. Most material near the outer fibre, so very efficient in bending — bicycle frames, scaffold poles.
  • I-beam (idealised): I ≈ A_f · d² / 2 + I_web, dominated by the flange contribution if flanges are thin.
  • Channel / angle: I depends on orientation; can be 5–10× different about the two axes.

Load types and their consequences

  • Point load (P): sharp moment peak at the load location; deflection scales as PL³/EI.
  • Uniformly distributed load (UDL, w): smooth parabolic moment diagram; deflection scales as wL⁴/EI.
  • Triangular / varying distributed: integrate w(x) numerically or use influence-line tables.
  • Pure moment at end (M): constant moment along the beam; deflection scales as ML²/EI.
  • Multiple loads: superpose individual cases (linearity holds in elastic theory).
  • Self-weight: always treat as a UDL of the beam's mass per unit length × g.

When Euler-Bernoulli breaks down

  • Short, deep beams (L/d < 10). Shear deformation is no longer negligible. Switch to Timoshenko beam theory, which adds a shear term GA·κ to the stiffness matrix.
  • Large deflections (δ/L > ~0.1). Geometry changes during loading invalidate the small-deflection assumption. Use elastica theory or nonlinear finite element analysis.
  • Plastic hinges form. Past first yield, parts of the cross-section yield and the beam softens; use plastic limit analysis with M_p = σ_y × Z (plastic section modulus).
  • Composite or layered beams. If shear lag between layers matters (sandwich panels), partial-interaction theory replaces the single-EI assumption.
  • Buckling under axial load. A bending beam also under axial compression can buckle laterally before reaching its bending capacity — covered by separate column buckling theory.

Common failure modes

  • Yield at extreme fibre. The most basic case — bending stress σ = M·c/I exceeds yield strength σ_y at the most highly loaded section. Design check: σ_max ≤ σ_y / safety factor.
  • Lateral-torsional buckling (LTB). Compression flange of an I-beam buckles sideways out of plane; happens before yield in long unrestrained beams. Brace the compression flange or use a closed section.
  • Shear yield in the web. For short, heavily-loaded beams, the web can yield in shear before the flanges yield in bending. Add web stiffeners or thicker web plates.
  • Fatigue at stress concentrations. Welded connections, holes for bolts, and notches concentrate stress. Cyclic loading drives crack growth from these initiation sites; design for stress range, not just peak stress.
  • Flange local buckling. Thin compression flanges can wrinkle locally before global LTB or yield. Code limits on flange b/t ratios prevent this.
  • Excessive deflection at service. Even an unyielded beam can fail serviceability if it sags more than allowed (typically L/250 or L/360 for floors). This often controls design over strength.

Frequently asked questions

What does Euler-Bernoulli beam theory assume?

Three core assumptions: (1) plane sections perpendicular to the neutral axis remain plane after bending; (2) those sections remain perpendicular to the deformed neutral axis (no shear deformation); and (3) the material is linearly elastic. The theory works well for slender beams (length-to-depth ratio > 10) but underestimates deflection for short, deep beams where shear matters — those need Timoshenko theory.

What is EI and why does it matter?

EI is the flexural rigidity — Young's modulus E times second moment of area I. It's the single number that determines how stiff a beam is in bending. Doubling E (changing material) doubles stiffness; doubling I (changing cross-section) also doubles stiffness, but I scales as the fourth power of depth, so making a beam twice as deep makes it 16× stiffer in bending. This is why structural beams are I-shaped and oriented with the web vertical.

Why is a fixed-fixed beam stiffer than a simply-supported one?

End restraints prevent rotation at the supports, which forces the beam to develop reaction moments at both ends. These end moments curve the beam upward near the supports, partially cancelling the downward sag in the middle. The result: a fixed-fixed beam under uniform load deflects only 1/5 as much as a simply-supported beam of the same EI and length. The cost is the moment connection at each end, which is more expensive to fabricate.

What's the difference between a propped cantilever and a fixed-fixed beam?

A propped cantilever is fixed (built in) at one end and simply supported at the other — it cannot rotate at the fixed end but can rotate freely at the propped end. A fixed-fixed beam is built in at both ends, restrained against both translation and rotation everywhere. Propped cantilevers split the difference between simple and fully fixed in stiffness and complexity; they're common in continuous bridge girders where intermediate supports prop a long span.

Why do I-beams have the web vertical?

Bending stiffness is governed by the second moment of area I about the neutral axis, and I depends on the cube of distance from neutral axis to the most distant material. Most of the area in an I-beam sits in the flanges (top and bottom). Standing the beam web-vertical puts those flanges far above and below the neutral axis, maximizing I. Lay the same beam on its side (web horizontal) and I drops by typically 10× to 30× — the same beam becomes ten times more flexible.

What is lateral-torsional buckling?

When a beam bends downward, the top flange compresses and the bottom flange stretches. If the compression flange isn't restrained against sideways motion, it can buckle sideways, twisting the entire cross-section out of plane. This is lateral-torsional buckling (LTB) and it's a hidden failure mode for long, slender beams under bending. Prevention: shorten unbraced length with intermediate restraints, or use a section with high torsional stiffness like a closed box.