What if a beam stood on its edge?

Hold a ruler flat across two books. Press the middle. It bends easily!

Now stand the ruler on its thin edge. Press again. It hardly bends at all. Why?

Try it with a plastic ruler. Lay it flat across a gap and press the middle: it bends easily. Stand it up on its thin edge across the same gap and it barely bends. It's the same ruler, with the same amount of plastic. Only the way it faces has changed.

Does the same trick work for a big steel bridge beam? In this lab you can turn a beam from lying flat to standing on its edge and measure the difference.

A beam's resistance to bending depends on its material and on the shape of its cross-section, not simply on how much material it contains. The same rectangle can be loaded flatwise (wide and shallow) or edgewise (narrow and deep).

This lab uses a solid steel beam with a 0.5 m² cross-section on a 20 m span with a 20 t truck at midspan. It starts 1 m wide and 0.5 m deep, then turns to 0.5 m wide and 1 m deep. Same mass, same load: how much do the sag and the stress change?

Olly's steel beam lies flat, like a wide plank. A truck stops in the middle.

The beam bends a little. It is still safe.

In the model, the steel beam lies flat: 1 meter wide and half a meter tall, stretching 20 meters between its supports. A 20-tonne truck stops in the middle.

The beam sags about 54 millimeters, and the model rates it safe with a safety factor of 3.6. That means it is about three and a half times as strong as it needs to be. The beam weighs 78.5 tonnes.

Lying flat (b = 1 m, h = 0.5 m), the beam has I = b·h³ ÷ 12 ≈ 0.0104 m⁴. With self-weight included and 20 t at midspan, the model gives 54.2 mm of sag and a peak bending stress of 69.7 MPa. Against the model steel's 250 MPa yield strength, that is a safety factor of 3.6.

The beam's mass, 78.5 t, depends only on its volume, so it will not change when we turn it.

Now Olly turns the same beam on its edge. It is tall and thin now.

The truck stops in the middle again. This time the beam bends much less!

Now turn the beam on its edge: half a meter wide and 1 meter tall. Nothing else changes, not the steel, the truck or the weight of the beam.

The sag drops from about 54 millimeters to about 14, just a quarter as much. The safety factor doubles from 3.6 to 7.2. Simply standing the beam up made it four times stiffer and twice as strong against the truck.

On edge (b = 0.5 m, h = 1 m), I ≈ 0.0417 m⁴, four times the flat value. The model's sag falls to 13.6 mm, exactly a quarter of 54.2 mm, because every part of the deflection, from the truck and from self-weight, scales with 1 ÷ I.

The peak stress falls from 69.7 to 34.9 MPa, exactly half, because the section modulus S = b·h² ÷ 6 doubles. The safety factor rises from 3.6 to 7.2. Same 78.5 t of steel throughout.

When a beam bends, its top gets squeezed and its bottom gets stretched.

A tall beam keeps its top and bottom far apart. That makes it very hard to bend.

When a beam bends, its top is squeezed and its bottom is stretched, while the middle hardly changes length. The farther apart the top and bottom are, the more the material has to squeeze and stretch for the beam to curve even a little. So a tall beam fights bending much harder.

Engineers measure this with a number that grows with the beam's height times itself three times. Double the height and that number grows eight times; double the width and it only doubles. That's why depth matters so much.

The USDA Wood Handbook gives the moment of inertia of a rectangular beam as I = b·h³ ÷ 12 and its section modulus as S = b·h² ÷ 6, and bending deflection as proportional to W·L³ ÷ (E·I). Swapping b and h multiplies I by (h_new ÷ h_old)² = 4 and S by 2.

Model limits: Euler–Bernoulli beam theory for one solid rectangle, a single point load, and no checks for buckling, twisting, shear or the supports. That last gap matters here, because tall, thin beams can fail sideways in a way this model cannot show.

Wood experts test how stiff boards are when they stand on their edges.

But a very tall, thin beam can tip and twist sideways. So builders hold it steady.

Depth matters in real buildings too. The lumber grades in the USDA Wood Handbook give each grade's bending stiffness for edgewise loading, with the board standing on its narrow edge.

But there is a catch the model can't show. A very tall, thin beam can bend sideways and twist, a failure called lateral-torsional buckling. Floors help stop this: the Wood Handbook says decking fastened along the top of the beams often gives enough restraint, and Britannica describes diagonal braces called bridging set between floor joists.

The Wood Handbook explains that because a bent beam is compressed along its concave edge, it can buckle by a combination of sideways deflection and twist; long, slender beams free to twist and move sideways buckle once the bending stress reaches a critical value. Imperfections such as bow or twist can amplify this, so lateral or torsional restraints may be needed to hold beams in line.

The most common restraint is a deck attached to the beams' top edges, which often prevents lateral-torsional buckling. Britannica's joist entry describes bridging, diagonal braces between joists, as extra strength. None of this is in the model, so its tallest, thinnest beams look better here than they would in real life.