Bending Stress
A bent beam is squeezed on one face and stretched on the other, and its middle feels nothing.
Before this Lesson
Take a kitchen sponge, draw a few straight lines down its long side from top to bottom, and bend it into a smile. The lines stay straight. They only tilt, crowding together along the top edge and spreading apart along the bottom, and somewhere between is a level where their spacing doesn't change at all. A steel beam does the same thing, just too slightly to see. That one picture tells you where a beam's stress is, how big it gets, and why depth is worth so much more than width.
The lines stay straight
Cut a short slice out of a beam carrying a positive moment. Its two faces started out parallel. Bending tilts them toward each other at the top, but each face stays flat, like the lines on the sponge. That's the one assumption behind this whole Lesson: plane sections stay plane.
Now look at the fibers running between the two faces. The ones along the top got shorter, the ones along the bottom got longer, and one level kept its length. That level is the neutral axis. Because the faces are flat, the change in length grows in a straight line as you move away from the axis: twice as far, twice the change. Every fiber started out the same length, so the strain grows in a straight line too.
From Strain, Hooke's Law, and Poisson's Ratio: strain is stretch per length, and in the elastic range, stress is strain times the material's stiffness, .
Stress is strain times , so it follows the same straight line: zero at the neutral axis, and largest at the fibers farthest from it, the extreme fibers, a distance from the axis. Call the distance to any fiber , and the stress is some constant times that distance, . That leaves two questions: where is the axis, and what is ?
The neutral axis runs through the centroid
Statics locates the axis. A section in bending carries a moment but no net axial force, so all the push above the axis has to cancel all the pull below it. A patch of area carries a force , so:
That integral is the first moment of the area about the axis, and among horizontal lines, it's zero about only one: the one through the centroid. So the neutral axis runs through the centroid of the cross section. For a rectangle or a W-shape, that's mid-depth. The Handbook uses the neutral axis without saying where it is, so this is yours to carry in.
The flexure formula
The push and the pull form a couple, and that couple is the internal moment .
A patch at distance pushes with a force of on a lever arm of , so its moment about the axis is . Add up every patch:
Distance counts twice, once in how hard the patch pushes and once in its arm. So material near the axis barely contributes, and material far out does most of the work, which is why from Moment of Inertia of Areas shows up here. Now , and the stress at any fiber is:
The Handbook prints this under Stresses in Beams as , with the distance from the neutral axis to the fiber. The largest stress is at the extreme fiber, a distance from the axis:
The minus sign keeps tension positive: count upward, and under a positive moment, a fiber above the axis gets a negative stress, which is compression. You can skip the sign. Find the size from the formula and read the direction off the bend. In the Handbook's sign convention, from Internal Shear and Moment in Beams, a positive moment is a smile, and a negative moment is a frown.
The largest stress in a rectangular beam
A typical question has a statics half and a materials half. Statics finds the largest moment, and the flexure formula turns it into stress.
From Maximum Bending Moment: a simple beam under a uniform load peaks at , with the intensity, and under a center point load at . Both peak at midspan.
For a rectangle, the Handbook's area table gives , and is half the depth, . Watch the units on the way in. The moment usually comes in and the section in inches, so multiply by 12 before they meet. In SI, use and mm, and the stress comes out in MPa, since 1 MPa is 1 N/mm².
One number for the whole section
The largest stress depends on the section only through the ratio , so that ratio gets a name, the section modulus:
The Handbook prints both under Stresses in Beams, with a lowercase , as the elastic section modulus. It doesn't print the rectangle's value, but that's one line away:
For a steel shape, you look it up. The Handbook's Table 1-1: W Shapes Dimensions and Properties, in its Civil Engineering chapter, lists it for each shape. A name such as W18X50 means about 18 in. deep and 50 lb/ft.
| Shape | (in.) | (in.) | (in⁴) | (in³) |
|---|---|---|---|---|
| W24X68 | 23.7 | 0.585 | 1830 | 154 |
| W21X44 | 20.7 | 0.450 | 843 | 81.6 |
| W18X50 | 18.0 | 0.570 | 800 | 88.9 |
| W16X36 | 15.9 | 0.430 | 448 | 56.5 |
| W12X40 | 11.9 | 0.515 | 307 | 51.5 |
Adapted from the FE Reference Handbook, Table 1-1: W Shapes Dimensions and Properties
Check a row. The W18X50 has in., so , the printed value. The full table prints and as plain and under Axis X-X, the strong axis, which is how a W-shape normally works, with its web vertical. Its Axis Y-Y columns are for the weak axis and hold no , so don't reach for them. It also prints a beside the , the plastic section modulus, which waits for Steel Beam Flexural Strength.
Working back from an allowable stress
A design question runs the formula backward. It states an allowable bending stress, since the Handbook lists none for timber or steel beams, and the largest stress can't exceed it. That limit answers three different questions:
- Required section modulus: .
- Minimum depth of a rectangle: set equal to and solve for .
- Largest load: find , then run the moment formula backward for the load.
A fiber that isn't at the face
works at any level, as long as is measured from the neutral axis and not from a face. A favorite level is where the web meets the flange. Put 200 , or 2,400 , on a W18X50. The face is in. from the axis, and the flange is 0.57 in. thick, so the junction is in. from the axis:
The face works at 27.0 ksi, so the whole flange runs close to full stress, which is what a flange is for. Plug in the 0.57 in. itself and you'd get 1.71 ksi, the stress at a fiber that close to the axis, not to the top.
Two faces, two distances
A T-section has a wide flange at one end and only a narrow stem at the other, and that pulls the centroid, and the neutral axis with it, toward the flange. Now the two faces sit at different distances from the axis. The Handbook defines only for a symmetrical section, so go back to and give each face its own .
Those distances fix the sizes: the farther face always has the larger stress. The sign of the moment fixes the directions: which face is in tension.
Why depth beats width
In , width counts once and depth counts twice. Here is the side perpendicular to the neutral axis, whichever way the page is turned. Take the 10 in. by 2 in. plank from Moment of Inertia of Areas, flat and then on edge:
Under the same moment, turning the plank on edge cuts the stress to a fifth, and five is the long side over the short side. That Lesson showed growing 25 times, so why not 25 here? Because also grew five times. On edge, the plank curves a twenty-fifth as much, but its extreme fibers sit five times farther out, and for any given curve, farther fibers stretch more. So they stretch a fifth as much, not a twenty-fifth. The plank sags a twenty-fifth as far, but it's only five times as strong.
So a question about changing a rectangle needs no numbers. For the same moment, the stress goes as .