A cantilever bridge is a span built from rigid horizontal arms that project outward from supporting piers and carry load without falling back on a tied arch or suspension cable. Sir Benjamin Baker engineered the most famous example, the Forth Bridge in Scotland, opened in 1890. Each pier carries two balanced arms — an anchor arm tied to the abutment and a cantilever arm projecting toward midspan — and a suspended span often drops in between. The result is a stiff, deep-truss structure that crosses 200–500 m gaps without falsework in the river below.
Inside the Cantilever Bridge
The mechanism is moment-balance, not tension. Each pier supports two arms that swing in opposite directions — the anchor arm reaches back over the abutment and is held down by either dead weight or a tension tie, and the cantilever arm reaches out toward the next pier. Load on the cantilever arm tries to rotate the whole assembly about the pier, and the anchor arm resists that rotation. If you draw the bending moment diagram, you get negative moment over the pier (top fibres in tension) and zero moment at the tip of the cantilever — which is exactly why these bridges look the way they do, deep at the pier and shallow at the tip.
Most long cantilever bridges then drop a suspended span between the two cantilever tips. That suspended span is a simple beam sitting on bearing seats — often called a Gerber beam after Heinrich Gerber, who patented the form in 1866. The suspended span lets each half of the bridge thermally expand and rotate independently without cracking the deck, which matters when you have a 521 m main span like the Quebec Bridge. The hinge geometry has to be tight — bearing seat tolerance under 5 mm vertical, otherwise you get impact loading every time a train rolls onto the suspended span.
Get the balance wrong during construction and the bridge fails before it carries traffic. The 1907 Quebec Bridge collapse killed 75 workers because the compression chord buckling capacity had been miscalculated, and the dead load alone overwhelmed the anchor arm. Modern balanced cantilever construction in post-tensioned concrete uses form travellers that add segments on each side of the pier within roughly 1 segment of each other — typically the unbalanced moment must stay under 3-5% of the balanced capacity, or you tip the pier.
Key Components
- Anchor Arm: The arm extending from the pier back toward the abutment. It carries an upward reaction at the pier and a downward tie-down at the abutment, resisting the overturning moment generated by the cantilever arm. On the Forth Bridge each anchor arm is 207 m long and is weighted with granite-filled counterweights at its end.
- Cantilever Arm: The arm projecting outward from the pier toward midspan. It carries its own weight, the suspended span reaction, and live load — all as a deep cantilever in negative bending. Section depth typically tapers from full structural depth at the pier to roughly 30-40% of that depth at the tip.
- Pier (Main Tower): Carries the full vertical load of both arms plus any live load imbalance. On a steel cantilever like the Forth Bridge the pier is a braced steel tower; on concrete cantilevers it's a hollow box pier post-tensioned vertically. Pier base reactions on a 300 m main-span bridge run 60-120 MN per pier.
- Suspended Span: A simply-supported beam or truss dropped between the two cantilever tips. It sits on bearing seats with elastomeric or roller bearings allowing thermal movement of ±150 mm on a long span. Length is typically 25-40% of the main span.
- Bearing Seats / Hinges: The pin or bearing connection between the cantilever arm tip and the suspended span. These must transfer vertical shear and accommodate rotation but carry zero bending moment — that's the whole point of the Gerber hinge. Bearing surface flatness tolerance is held to 0.5 mm over 1 m to prevent edge-loading.
- Anchorage / Tie-Down: At the back of the anchor arm, either a mass-concrete counterweight or a set of post-tensioned tie-down rods anchored into rock holds the arm against uplift. Uplift forces in service can run 15-30% of the dead-load reaction at the pier, and any slip in the anchorage shows up as a measurable downward deflection at the cantilever tip.
Real-World Applications of the Cantilever Bridge
Cantilever bridges are the right answer when you need to cross a deep gorge, a navigable shipping channel, or a wide river where falsework in the water is impossible — and where a suspension bridge's anchorage cost isn't justified. They dominate the 200-550 m span range, and they're how engineers crossed the world's hardest crossings before cable-stayed construction matured in the 1980s. You see them in rail crossings, highway crossings, and modern segmental concrete viaducts where each segment is cast or lifted into place from the pier outward.
- Railway Infrastructure: The Forth Bridge in Scotland — a 2,529 m steel cantilever rail bridge with two 521 m main spans, opened 1890, still carrying ScotRail traffic.
- Highway Bridges: The Quebec Bridge in Canada — a 549 m main-span steel cantilever, the longest cantilever span ever built, completed 1917, carrying Route 175 and CN Rail.
- Segmental Concrete Viaducts: The Confederation Bridge between PEI and New Brunswick — 12.9 km of balanced cantilever post-tensioned concrete spans of 250 m each, opened 1997.
- Mountain Crossings: The Storms River Bridge on the N2 in South Africa, built by free cantilever method using form travellers from each pier.
- Urban Rail Transit: The Howrah Bridge in Kolkata, India — a 457 m cantilever truss carrying tram, road, and pedestrian traffic across the Hooghly River since 1943.
- Large-Span Pedestrian Crossings: Industrial conveyor and pedestrian bridges over rail yards, often built as steel half-through cantilever trusses where below-deck clearance is the binding constraint.
The Formula Behind the Cantilever Bridge
The governing number on a cantilever bridge is the negative bending moment at the pier — that's where the structure works hardest and where the section is deepest. You use it to size the top chord (in tension) and the bottom chord (in compression) of the cantilever arm. At the low end of the typical range — short cantilever arms under 80 m — the moment is dominated by live load and the design feels like a heavy plate-girder problem. At the high end, 200 m+ cantilever arms like the Forth Bridge, dead load completely swamps live load, the section depth balloons to 40-60 m at the pier, and the design becomes a self-weight problem first and a traffic problem second. The sweet spot for steel truss cantilevers sits around 150-250 m arm length, where dead and live load contributions are within a factor of 3 of each other.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| Mpier | Negative bending moment at the pier (over the support) | kN·m | kip·ft |
| wDL | Distributed dead load along the cantilever arm (self-weight + deck) | kN/m | kip/ft |
| L | Length of the cantilever arm from pier centreline to tip | m | ft |
| PSS | Reaction from the suspended span applied at the cantilever tip | kN | kip |
| MLL | Live load moment contribution at the pier from the governing traffic case | kN·m | kip·ft |
Cantilever Bridge Interactive Calculator
Vary the arm load, arm length, suspended-span reaction, live-load moment, and truss depth to see pier moment balance and chord force.
Equation Used
The calculator evaluates the negative bending moment at the main pier. Uniform dead load contributes wDLL2/2, the suspended span bearing reaction contributes PSSL, and any traffic or construction live-load effect is added as MLL. The chord force is a first-pass estimate using the truss depth as the internal lever arm.
- Dead load is treated as a uniform load on the cantilever arm.
- Suspended span load is represented by a vertical point reaction at the cantilever tip.
- Live-load effect is entered directly as an equivalent moment at the pier.
- Chord force estimate assumes a simple truss couple with lever arm equal to truss depth.
Worked Example: Cantilever Bridge in a 2-lane segmental concrete cantilever bridge over a flood plain
Sizing the negative moment at the main pier of a 280 m main-span balanced cantilever post-tensioned concrete highway bridge crossing a flood-prone river valley on a regional highway alignment in northern Manitoba. Cantilever arm length L = 110 m on each side of the pier, suspended span length 60 m. Dead load w<sub>DL</sub> = 280 kN/m (box girder self-weight plus barriers and asphalt). Suspended span reaction at tip P<sub>SS</sub> = 4,200 kN. Governing live load moment at pier M<sub>LL</sub> = 95,000 kN·m for a CL-625 truck train across both lanes.
Given
- L = 110 m
- wDL = 280 kN/m
- PSS = 4,200 kN
- MLL = 95,000 kN·m
Solution
Step 1 — at the nominal arm length of 110 m, calculate the dead-load moment at the pier:
Step 2 — add the suspended-span reaction acting at the tip:
Step 3 — sum dead-load, suspended-span, and live-load moments to get the design moment at the pier:
Now look at the operating range. At the low end of typical span — say a 70 m cantilever arm on a smaller crossing — MDL drops to (280 × 702)/2 = 686,000 kN·m, MSS drops to 294,000 kN·m, and total pier moment falls to roughly 1,075,000 kN·m. The pier-table depth needed is about 5-6 m and you can build it with conventional post-tensioning tendons in a single tier. At the high end of the typical balanced-cantilever range — a 160 m arm like Confederation Bridge — MDL alone climbs to 3,584,000 kN·m and total pier moment exceeds 4 million kN·m, forcing pier-table depth to 12-14 m and multiple post-tensioning tendon tiers. Dead load goes from contributing 65% of total moment at L = 70 m to over 85% at L = 160 m, which is exactly why long cantilever bridges are dead-load problems first.
Result
The nominal pier moment is 2,251,000 kN·m, which sets a pier-table depth of roughly 8-9 m for a typical post-tensioned box girder using ULS allowable stress on the bottom slab in compression. That depth is what you actually feel standing under the bridge near the pier — the box looks like a building, then tapers visibly toward midspan. Compared to the 70 m arm case (1.07 million kN·m, 5 m deep) and the 160 m arm case (4+ million kN·m, 12+ m deep), the 110 m arm sits in the steel-or-concrete sweet spot where construction stays manageable. If your measured top-fibre strain at the pier reads 20% higher than predicted during the closure pour, the most likely causes are: (1) an unbalanced cantilever during construction — segment placement got more than 1 segment ahead on one side and the form traveller stayed there overnight; (2) tendon friction losses higher than the 0.20 wobble + 0.20 curvature coefficients assumed, which is common when ducts aren't grouted promptly; or (3) creep redistribution underestimated, especially if the concrete was loaded younger than 7 days at segment closure.
Cantilever Bridge vs Alternatives
The decision is almost always cantilever bridge versus cable-stayed versus suspension, and the deciding factors are span length, foundation conditions, and whether you can put falsework in the gap below. Here's how the three stack up on the dimensions that actually drive the choice.
| Property | Cantilever Bridge | Cable-Stayed Bridge | Suspension Bridge |
|---|---|---|---|
| Economical span range | 200-550 m | 200-1,100 m | 600-2,000+ m |
| Falsework in gap required | No — built outward from piers | No — built outward from pylons | No — but anchorages are massive |
| Stiffness under live load | High — deep truss or box section | Medium — deck deflects under asymmetric load | Low — deck moves several metres under wind |
| Construction time for 300 m main span | 3-5 years (steel truss) | 2-4 years | 4-7 years incl. anchorages |
| Sensitivity to foundation settlement | High — differential settlement loads the cantilever | Medium — cable adjustment compensates | Low — anchorages on rock decouple deck |
| Indicative cost per m<sup>2</sup> of deck | $$ — moderate steel/concrete tonnage | $$ — efficient material use | $$$ — anchorage and cable cost dominate |
| Typical service life | 100+ years (Forth Bridge at 135 yrs) | 75-100 years | 100+ years |
| Best application fit | Deep gorges, navigable channels, rail crossings | Medium-long highway crossings, urban | Very long crossings over deep water |
Frequently Asked Questions About Cantilever Bridge
The dominant cause is creep deflection, not geometry error. A post-tensioned concrete cantilever loaded at 7-14 days of age will creep an additional 1.5-3× the elastic deflection over the first year as the concrete redistributes stress. If you set the camber based on elastic deflection only, you'll measure exactly the kind of low reading you're describing.
The fix lives in the camber diagram, not the field. Most designers add a creep camber of roughly the elastic deflection multiplied by the creep coefficient (φ ≈ 2.0 for normal-strength concrete loaded young). If you're seeing 40 mm low on a cantilever where elastic deflection at the tip was predicted as 60 mm, your creep coefficient assumption was probably 1.5 instead of the 2.2 you actually got because the concrete was loaded at 5 days.
Three tests. First, navigation clearance — if you need a clear deck profile without overhead cables for shipping radar or military aircraft, cantilever wins. Second, foundation conditions — cable-stayed pylons need foundations capable of taking large horizontal cable thrusts; if your bedrock is deep or weak, cantilever piers (vertical load only) get cheaper fast. Third, deflection limits for rail — if the bridge carries heavy rail traffic, cantilever truss stiffness beats cable-stayed deck stiffness for the same span, and you avoid the alignment maintenance headache cable-stayed rail bridges have shown.
If none of those three apply, cable-stayed almost always wins on cost and construction time at 350 m.
You're probably missing the live-load uplift case. The governing uplift on the anchor arm tie-down isn't dead load alone — it's dead load on the anchor arm minus live load on the anchor arm plus live load on the cantilever arm and suspended span. When traffic fills the cantilever side and clears the anchor side, the overturning moment maximises and uplift peaks.
For typical highway loading, design uplift runs 15-30% higher than the dead-load-only calculation. If you sized the tie-down rods to dead-load uplift only, you'll see anchor lift-off under heavy traffic — measurable as a 2-5 mm cyclic vertical movement at the abutment bearing.
Standard practice keeps the unbalanced moment at the pier under 3-5% of the balanced capacity, which in practice means segment placement stays within 1 segment of equal on each side, and the form traveller positions are mirrored at the end of every shift. The pier itself is usually post-tensioned vertically and temporarily braced to the pier-table for the duration of cantilever construction.
Push beyond 5% and you risk pier rotation, which shows up as a tip elevation difference between the two arms that you can't correct after closure. The 1996 Koror-Babeldaob Bridge collapse in Palau is the canonical lesson on what happens when long-term unbalance and creep aren't controlled.
Thermal movement and statical determinacy. A 60 m suspended span on a typical northern climate bridge expands and contracts roughly ±30 mm seasonally. If you make it continuous with both cantilever arms, that movement either cracks the deck or transfers axial force into the cantilevers — neither is acceptable.
The hinge also makes the structure statically determinate, which Heinrich Gerber's 1866 patent specifically targeted. Statically determinate means support settlement at one pier doesn't redistribute moments into the rest of the bridge — a critical property when you can't perfectly predict foundation behaviour over a 100-year life.
Check your sign convention and your load case. A cantilever arm under gravity load has negative moment at the pier — top fibres in tension, bottom fibres in compression. If your model is showing the opposite, the most likely causes are: (1) your load is applied as uplift instead of gravity, often a sign error on a wind or buoyancy case; (2) you're looking at an erection stage where the form traveller weight on the opposite side has temporarily reversed the moment; or (3) the model boundary condition at the pier is fixed rotation when it should be free, which inverts the moment diagram.
Quickest check — load only self-weight and look at the deflected shape. The tip should drop, and the bottom chord at the pier should be in compression. If it isn't, your sign or boundary is wrong before anything else matters.
References & Further Reading
- Wikipedia contributors. Cantilever bridge. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
