Stress and Materials

The last lesson, Forces and Structures, balanced a worker on a plank and stopped there. Equilibrium said the supports push back with 530 N and 230 N. It said nothing about whether a pine plank 50 mm wide can actually take that, or whether it bends permanently and stays sagged the moment the worker steps off. Whether a member survives its load is a materials question, and it starts with a division.

Stress is force spread over area

Hang 10,000 N, roughly a metric ton, from a steel rod, and the danger depends entirely on how thick the rod is. Stress is the force divided by the area carrying it, and its unit is the pascal: one newton per square meter. A pascal is a tiny amount, so real stresses are quoted in megapascals, abbreviated MPa, each worth a million pascals. Atmospheric pressure is about 101 kPa, which helps calibrate the scale.

A 10 mm diameter rod has a cross-sectional area of pi times 0.005 squared, about 7.85 times 10^-5 m^2. The stress in it is 10,000 / 7.85e-5, which is 1.27 times 10^8 Pa, or 127 MPa. Shrink the rod to 5 mm and the area falls by a factor of four, because area scales with diameter squared, so the stress jumps to about 508 MPa, past what mild steel can carry. Same load, same material, one diameter apart: one rod is fine and the other is in trouble. In a straight pull, area, not length, sets the stress.

A felt comparison helps: 127 MPa means every square millimeter of the rod's cross section carries 127 newtons, the weight of a 13 kg mass pressing on each square millimeter. That is why stress numbers look enormous next to everyday weights. The same division works in compression. A table leg carrying 400 N on a 30 by 30 mm cross section sees 400 / 0.0009, about 444 kPa, trivial for wood, which is why furniture legs stay slim. Slender compression members do have a second failure mode, though: they can bow sideways and collapse, a behavior called buckling, at loads far below the material's crush strength. A thin rod that would laugh at 10,000 N of tension can buckle under a few hundred newtons of compression, which is why the compression members of a truss are fatter than the tension ties doing comparable work.

Strain and stiffness

Stress says how hard a material is being pulled or squeezed. Strain says how much it deforms in response: the change in length divided by the original length, a plain fraction with no unit. Stretch a 2.0 m rod by 1.2 mm and the strain is 0.0012 / 2.0, or 0.0006. Engineers say six hundred microstrain.

Over a wide range of everyday loads, strain is proportional to stress: double the pull, double the stretch. The ratio of stress to strain is Young's modulus, written E, and it is the stiffness of the material. It comes out in pascals like stress, because strain has no unit. Steel's E is about 200 GPa, aluminum about 70 GPa, and wood along the grain about 10 GPa. GPa means gigapascals, a billion pascals.

Run the numbers on the 10 mm steel rod from the last section, 2.0 m long and pulled with 10,000 N. Stress 127 MPa divided by E gives strain: 127 times 10^6 / 200 times 10^9 = 6.4 times 10^-4. Multiply by the 2.0 m length and the rod stretches about 1.3 mm. Make the same rod out of aluminum and the strain is 127 times 10^6 / 70 times 10^9 = 1.8 times 10^-3, so the stretch is about 3.6 mm, more than three times as much. Stiffness and strength are separate properties. Steel and a good aluminum alloy can have similar strength, yet steel is roughly three times stiffer, which is why precision machine frames are steel even when aluminum would carry the load.

Stiffness differences are easy to feel in other materials. Common plastics run between 1 and 3 GPa, twenty to a hundred times floppier than steel, which is why a plastic clip springs back visibly while a steel bracket in the same shape barely moves. Rubber sits at the far bottom near 0.01 GPa, stretching by tens of percent instead of fractions of a percent. A material's modulus decides how much a part moves long before strength enters the conversation.

How materials give up

Below a certain stress, a stretched rod springs back to its original length when the load comes off. That region is elastic deformation, and it is the region everyday design lives in. Past a limit called the yield strength, part of the stretch stays after unloading: plastic deformation. Mild steel yields near 250 MPa. Push further and it keeps stretching, actually getting a little stronger through strain hardening, until it reaches its ultimate strength somewhere between 400 and 550 MPa, then necks: one small region thins while the rest stops stretching, and the break happens there.

Materials split into two failure personalities. Ductile materials such as steel, aluminum, and copper stretch visibly, neck down, and warn before they fail. A bent steel nail is ductile behavior on display. Brittle materials such as glass, cast iron, and concrete in tension snap with no warning at all: a crack forms and crosses the whole part in microseconds. Glass fails in practice at perhaps 50 MPa even though its theoretical limit is far higher, because every scratch on its surface concentrates stress at its tip. Wood sits between the extremes, carrying 40 to 80 MPa along the grain depending on species, and splitting cheerfully along that same grain when pulled across it.

Two more ideas finish the failure story. Stress concentrates at any abrupt change in shape: the material flowing around a hole or a sharp inside corner carries several times the average stress right at the edge, which is why cracks start at holes and notches instead of in the calm middle of a plate. Rounded fillets spread that concentration out, and a fillet in the right place is often the cheapest strength upgrade available. Fatigue, the slow growth of cracks under repeated loading, is the other quiet killer: a part that survives one million lifts is not guaranteed to survive two million. Steel has an approximate endurance limit below which it can last indefinitely, while aluminum has none and accumulates damage from any repeated stress, which is why aircraft aluminum lives under an inspection schedule.

The yield strength, not the ultimate, is the number designers plan around for parts that must keep their shape. A bracket that has gone plastic is a bracket that no longer aligns anything, even if it is a long way from snapping.

Factor of safety

No engineer runs a part right at its limit. The factor of safety is the ratio of the strength to the working stress: design a rod to run at 100 MPa in steel that yields at 250 MPa and the factor is 2.5. Why 2 or more? Because loads exceed their estimates whenever anything moves fast or lands hard, because material properties vary batch to batch, because hidden flaws and corrosion eat margin over the years, and because repeated loading can crack steel at stresses well below yield, a subject called fatigue. Each of those alone would justify margin; together they demand it.

The factor is tuned to how well everything is known. Elevator cables work around a factor of 8, because a snapped cable is catastrophic and the loads repeat millions of times. Aircraft structures run factors near 1.5, which sounds reckless until the other half of the deal shows up: certified materials, inspections, and weight so expensive that every kilogram is fought over. A bracket holding a garden hose reel can sit comfortably at 3 or 4, because extra steel is cheaper than analysis there.

Read the factor honestly in both directions. A factor of 1.2 on a carefully measured load is braver than a factor of 4 on a guess, because the factor only multiplies the number that was estimated; if the estimate is off by three, no factor rescues it. That is why good practice pins down the real load first and then picks a modest, justified margin, instead of sprinkling safety everywhere and shipping a heavy, expensive part that still fails where the guess was wrong.

import math

F = 10000.0           # 10,000 N hanging from a steel rod
yield_stress = 250e6  # mild steel yield strength, Pa

for d_mm in (4, 8, 12, 16):
    r = d_mm / 2000.0              # radius in meters
    area = math.pi * r * r
    stress = F / area
    print(f"d = {d_mm:2d} mm  stress = {stress/1e6:6.1f} MPa  "
          f"safety factor = {yield_stress/stress:4.2f}")

The 4 mm rod runs at 795.8 MPa, a factor of 0.31: it yields the moment the load lands and keeps stretching. The 8 mm rod survives elastically at 198.9 MPa but with a factor of 1.26, one overloaded delivery away from a bent hanger. The 12 mm rod at a factor of 2.83 is the sensible pick. The 16 mm rod is safe and heavy, paying for margin nobody asked for. That is the whole design conversation in four lines.

Statics now has both halves: the geometry balances and the material holds. The plank stays a plank. The next lesson, Mechanisms and Motion, starts making parts move on purpose: levers, gears, and pulleys, the trades that turn a small force into a useful one.