Crane gantries and runway beams

An overhead travelling crane is the most demanding thing most industrial buildings ever carry — not because the loads are large, but because they move, they are dynamic, and they repeat millions of times.

That combination makes a runway beam one of the few parts of an ordinary building where fatigue, not strength, sets the design.

An overhead travelling crane in a heavy engineering workshop, the bridge spanning between two runway beams carried on brackets off the building columns, the crane rail and its clips visible along the top flange of the near runway beam, with the crab, hook block and machine tools on the shop floor below. — The runway beam, its bracket off the column, and the rail clipped to the top flange. The load comes down through that rail onto the one detail on the beam that fatigue cares about most.

What the crane actually does to the building

A cross section through a workshop with an overhead travelling crane, showing the actions it applies. Vertical wheel loads on each runway beam with the dynamic factor noted. Lateral surge from the crab accelerating across the bridge, applied at rail level. Longitudinal thrust from the crane travelling along the runway. A skewing couple from the crane running out of square with the guide rollers bearing on the rail sides. The eccentricity of the runway beam bracket from the column centreline is marked, with the resulting moment carried down the column into the foundation. — Four actions, three directions, and every one of them repeats. The lateral and eccentric effects are usually what govern the column, not the vertical load.

Action	Where it comes from	Applied
Vertical wheel loads	Crane self weight + hoist load, × dynamic factor	Down through each wheel
Surge (lateral)	Crab accelerating or braking across the bridge	Sideways, at rail level
Thrust (longitudinal)	Crane starting and stopping along the runway	Along the building
Skewing	The crane not running square	A lateral couple, every traverse

The fatigue problem

An office floor beam sees its full design load rarely. A runway beam is loaded and unloaded every time the crane passes — millions of cycles in a busy workshop.

Fatigue life depends on:

  the stress RANGE Δσ   (not the peak stress)
  the number of cycles N
  the DETAIL CATEGORY of the weld or connection

Halving the stress range increases life by roughly a factor of eight.

The top flange to web junction is the critical detail

The wheel load lands on the rail, is spread over a short length of the top flange, and is forced into the web. That produces high local stress right at the flange-to-web weld — and the whole thing repeats with every wheel passage.

Two consequences:

Never weld anything unnecessary near the top flange. A bracket, a stiffener, a cleat or a handrail post welded in that zone puts a stress concentration into an already fatigue-critical detail. Where attachments are unavoidable they get a deliberate detail category and a fatigue check.
Bearing stiffeners and their welds need the same attention. They sit in the same load path.

This is also why the runway beam is inspected, not just designed — cracks initiate at these welds and propagate under continued use.

Lateral load is the other half of the structural problem. An I-section is very poor about its weak axis, so on anything but a small crane the surge is taken by a surge girder — a horizontal truss or a channel welded to the top flange, spanning between columns in plan. Small cranes get away without one. Large ones cannot.

Rails and fixings

The rail is a wearing part. It gets replaced, several times, over the life of the building.

Clipped, not welded. Clips allow longitudinal movement under thermal and braking effects and can be released for replacement. Welding the rail down would also lay a continuous weld along the most highly stressed, most fatigue-loaded fibre of the beam — precisely where one should not be.
Resilient pad under the rail spreads the wheel load and damps impact.
Rail joints are where the damage happens.

Alignment

The crane bridge is effectively a rigid measuring stick spanning between the two rails. If the geometry is wrong, the crane cannot run square — and it will tell you.

Check	Why
Span between rail centres	Varying span forces the wheels to crab
Straightness of each rail in plan	Lateral deviation drives the flanges into the rail
Level of each rail and relative level between them	Cross-fall makes the crane run downhill and skew
Rail joint step and gap	Impact
Rail to beam centreline eccentricity	Puts torsion into the beam

Where this connects

The shed it lives in is portal frames and pre-engineered buildings, the detail classification and crack behaviour is fatigue and fracture in steel, the connections are steel connections and erection, and finding the cracks is structural NDT and inspection access.

What to take away

A crane applies four actions — vertical wheel loads, lateral surge, longitudinal thrust and a skewing couple — and all of them repeat.
Dynamic factors are the load, derived from the crane’s hoisting class, not a safety margin.
The bracket eccentricity plus surge is what governs the column and the foundation, which is why retrofitting a crane fails there.
Runway beams are fatigue-governed; stress range and detail category decide the section.
Do not weld attachments near the top flange — that is the critical detail.
Rails are clipped, not welded, because they are a wearing part and a weld there is in the worst possible place.
A step at a rail joint is an impact on every pass, damaging beam and crane alike.
Skew wear is a survey problem; replacing wheels without checking alignment buys new worn wheels.