Pratt Truss: How It Works and Why It Shaped Steel Bridges

The Pratt truss is one of the most widely used truss patterns in the world. It is easy to recognise: verticals rise between the top and bottom chords, and the diagonals in each panel slope down towards the middle of the span. That small geometric choice puts the long diagonals into tension and the short verticals into compression, which is exactly what steel likes. It made the Pratt pattern a staple of railway and road bridges, and it is still common in roofs and footbridges today.

This guide explains how the pattern works, where it came from, its main variants, a Singapore example, and the weak points an engineer looks for when assessing one.

How a Pratt truss carries load

Any truss spanning between two supports behaves like a deep beam with most of its material removed. The top chord is squeezed, the bottom chord is stretched, and the web members between them carry the shear. What distinguishes one truss pattern from another is how the web is arranged, and therefore which web members are pushed and which are pulled. For the underlying principle, see our explainer on what a truss is.

Under gravity load, follow the weight from a panel point on the bottom chord where a floor beam delivers it. The diagonal rising from that point slopes up and outwards towards the nearer support. It is in tension, and it lifts the load to the top chord. The vertical at the top of that diagonal then carries the load back down, in compression, to the next bottom panel point closer to the support, where the next diagonal picks it up. Panel by panel, the load is passed outwards through alternating tension in diagonals and compression in verticals, until the inclined end posts, also in compression, deliver it to the bearings.

That split matters because the diagonals are the longest web members. A long member in compression must be stocky to avoid buckling, but a long member in tension can be slender, since tension straightens rather than bends it. The Pratt pattern puts the long members in tension and the short ones in compression, which saves material.

Origins of the design

The design was patented in the United States on 4 April 1844 by Thomas Pratt and his father Caleb Pratt. In its early form it combined materials: the top chord and verticals, in compression, were timber, and the bottom chord and diagonals, in tension, were wrought iron. Iron in tension and timber in compression was a sensible pairing for the time.

The pattern came four years after the Howe truss, which arranges the diagonals the opposite way. As iron and then steel became cheaper and more reliable, the logic of the Pratt arrangement won out, and it became one of the dominant forms for metal truss bridges in the later nineteenth and early twentieth centuries.

Variants of the Pratt pattern

Several named trusses are adaptations of the pattern for longer spans or particular loading.

Parker truss. A variant with a polygonal top chord, deeper at midspan, so depth follows the bending demand and material is saved near the ends.
Baltimore and Pennsylvania trusses. Subdivided versions with extra half-height members that shorten the panels and reduce the unbraced length of compression members on long spans. The Pennsylvania version also has a polygonal top chord.
Whipple truss. A double-intersection version whose diagonals cross two panels, used for longer spans in the nineteenth century.
Pratt roof trusses. The same pattern in a sloped or flat-topped form for roofs. Our guide to roof truss types compares it with other roof patterns.

Anderson Bridge: a Singapore example

Singapore has a fine example on the Singapore River. Anderson Bridge, completed around 1909 and officially opened in 1910, is described by the National Heritage Board as a steel truss girder bridge built on the Pratt truss principle, spanning about 61 metres between granite-clad abutments with pedestrian archways. It was built because nearby Cavenagh Bridge could no longer carry the growing traffic, and it is protected today with its neighbours as part of the Singapore River bridges. For the wider family of bridge forms, see our guide to the truss bridge.

Stress reversal and counters

The neat tension-diagonal story holds for steady load across the whole span. Moving loads complicate it. When a heavy vehicle stands on one part of the span, the shear in panels near midspan can change direction, and a diagonal designed to work in tension may be pushed into compression. A slender tension member cannot take that.

Historic examples dealt with this by adding counters, lighter diagonals sloping the other way in the middle panels, so that whichever way the shear acts, one diagonal is in tension. Modern designs either check every diagonal for reversal or use members that can take both. When assessing an old bridge, it is important to identify which members are counters and to check that they are still tight and effective.

Weak points and failure modes

Buckling of compression members. The top chord, end posts and verticals are in compression, and if corroded, damaged or undersized they can buckle. See our explainer on buckling.
Connections. Pins, gusset plates, rivets and bolts concentrate stress. Corrosion between plates, worn pins and cracked gussets are classic findings.
Loose or slack counters. Adjustable rods in old trusses can loosen, leaving panels unable to resist reversed shear.
Fatigue. Tension members under repeated load, especially from trains, can crack at rivet holes and welded attachments.
Fracture-critical tension members. In a bridge with only two trusses, the bottom chord and main diagonals carry load with little redundancy, so they need closer inspection.
Added dead load. Heavier decks, surfacing and services added over the years reduce the capacity left for traffic.

Designing a Pratt truss today

Modern steel versions are usually made from hollow sections, rolled angles or I-sections, with welded or bolted gusset connections. The designer’s main decisions are the depth of the truss, the number and length of panels, and how the members meet at each joint. A deeper truss reduces chord forces but lengthens the web members; more panels shorten the compression members but add joints, and every joint costs fabrication time and inspection effort.

Two details deserve particular care. First, the compression chord must be restrained against buckling sideways, either by bracing between trusses or by a stiff deck. Second, members should meet at a common point at each joint, so that their forces balance without bending the members. Where member centrelines are offset for practical reasons, the resulting moments must be designed for. They are also checked for reversed loading, such as wind uplift on a roof, which can turn tension diagonals into compression members.

Frequently asked questions

Are the diagonals in a Pratt truss in tension or compression?

Under normal gravity loading on a simply supported span, the diagonals are in tension and the verticals are in compression. Near midspan, moving loads can reverse the shear and push a diagonal into compression, which is why older trusses of this type often include counter-diagonals in the central panels.

What is the difference between a Pratt truss and a Howe truss?

The diagonals slope in opposite directions. In a Pratt truss the diagonals are in tension and verticals in compression; in a Howe truss the diagonals are in compression and the verticals in tension. Pratt suits steel; Howe suited timber diagonals with iron rods.

Who invented the Pratt truss?

Thomas Pratt and his father Caleb Pratt received a United States patent for the design on 4 April 1844. The original form used timber for the compression members and iron for the tension members.

Where are Pratt trusses used today?

They are still used for steel footbridges, short and medium-span road and rail bridges, industrial roofs, conveyor galleries and link bridges between buildings, wherever an efficient steel truss is needed.

Assessing a steel truss roof, link bridge or historic structure? Talk to our consultancy or get in touch.