Reading Stress-Strain Curves
Stiff, strong, and tough are three different questions about the same curve.
Before this Lesson
A crane hook has four jobs. It mustn't stretch noticeably under its rated load, or bend for good. Overloaded, it should open up where everyone can see it, long before it lets go. And when a load drops onto it, it has to soak up the jolt without cracking. "Strong" answers only one of those. The stress-strain curve answers all four, each with a different feature.
From Strain, Hooke's Law, and Poisson's Ratio: the straight part's slope is the modulus of elasticity, , a measure of stiffness. A ductile metal runs far past its straight part before it breaks, and a brittle one breaks near its end.
From The Tension Test: the curve plots against , with the 0.2% offset yield strength , the ultimate tensile strength at the peak, and fracture.
Slope, height, width, and area
Ask one curve the hook's questions. How much does it stretch under a load? How much stress does it take before it gives? How far does it stretch before it breaks? How much energy does it take to break it? Each looks at a different feature, and a fifth question, how much energy it gives back, rides along with the fourth:
Stiff isn't strong
Slope and height are separate questions. The Handbook's Table 2 - Average Mechanical Properties of Typical Engineering Materials has a pair that shows it: the aluminum alloy 2014-T6 yields at 60 ksi and A36 steel at 36 ksi, so the aluminum is stronger. But its is 10,600 ksi against the steel's 29,000, so at the same stress it stretches almost three times as far.
Height has one catch. Some metals mark their yield with a sharp corner and a nearly flat stretch, as hot-rolled mild steel often does, and the yield strength is the height of the flat. Many others bend over so gradually that there's no corner to read. For those, the 0.2% offset from The Tension Test picks a point by agreement, and the Handbook's Materials Science chapter defines yield strength by that method.
Ductility: how far it stretches
Ductility is the curve's width, how much strain the metal takes before it breaks, and you measure it on the pieces. When the bar snaps, its elastic strain springs back, so fitting the halves together and measuring between the gauge marks gives the permanent stretch. The Handbook's Materials Science chapter defines ductility that way, under Properties of Materials: "Permanent engineering strain after failure," also called percent elongation. For a ductile metal, the springback is a sliver of the width. With the distance between the gauge marks after fracture, and the change taken on the gauge length, not the whole specimen:
The Handbook's Mechanics of Materials chapter prints it under Definitions as .
The neck gives a second measure. Measure its smallest diameter and compare areas, for the percent reduction in area:
The Handbook writes the areas as and . The cancels, but the squares don't: it's a ratio of areas, so the diameters go in squared.
Area is energy
Why should an area on this plot mean anything? Stress times strain is force over area times length over length:
That's work per unit volume. The area under a load-elongation plot is the work done pulling the bar, and dividing the load by and the elongation by divides that work by the bar's volume. So an area under a stress-strain curve is energy per unit volume, in units of stress: a psi is an in.-lb per in³, and an MPa is a megajoule per m³.
Energy you get back: resilience
Load a bar on its straight part and let go, and it gives back all the work, like a spring. That energy is the triangle under the straight part, . The Handbook prints it under Elastic Strain Energy, in its Mechanics of Materials chapter, as the strain energy per unit volume:
The most a metal can store and give back is the triangle up to yield. Past yield, the work goes into permanent slip, and you don't get it back. That triangle is the modulus of resilience, a name the Handbook doesn't use:
It takes the yield strength, not the ultimate. For A36 steel, Table 2 gives and :
Tool steel L2 has the same and yields at 102 ksi. Squared, that's eight times the resilience, which is why springs are made of high-strength steel.
Energy to break it: toughness
Toughness is the area under the whole curve: the energy per unit volume a material absorbs before it breaks. It takes height and width together, so a strong brittle material can be less tough than a weaker ductile one. The Handbook doesn't define it. Its "fracture toughness," , is a different quantity, about cracks.
A question gives an idealized curve made of straight lines, so the area splits into triangles, rectangles, and trapezoids. The most common is elastic-perfectly plastic: straight up to , then flat until it breaks at a strain of .
The triangle is the modulus of resilience. This shape comes back in Steel Beam Flexural Strength, behind the plastic moment.
Several curves at once
On a plot of several materials, each question still looks at its own feature. Toughness trips people, because it needs height and width at once. And the most brittle is the one with the least strain past its straight part.
A sudden blow at a notch
A tension test is gentle: a smooth bar, pulled slowly. Real fractures tend to start at a notch, such as a weld toe or a crack, often under a sudden load, and both push a metal toward brittle behavior. The Charpy test combines them. A small bar with a V-notch lies across two supports, and a heavy pendulum swings down, strikes the face opposite the notch, and breaks it. The pendulum swings on up the other side, but not as high as it started:
is the pendulum's weight, and and are its release and rise heights, both from the bottom of its swing. It's potential energy, as in Work and Energy, with friction ignored. alone is everything the pendulum had, not what the bar took. In SI, a pendulum given by its mass needs , for an answer in joules.
The Handbook's Materials Science chapter names the test under Impact Test, as "used to find energy required to fracture and to identify ductile to brittle transition," but prints no formula.
Cold steel turns brittle
Repeat the test over a range of temperatures, as the Handbook's Impact Test entry describes, and many steels show a cliff. Warm, the bar bends and tears, leaving a dull, fibrous face, and absorbs a lot of energy: the upper shelf. Cold, it snaps, leaving a flat, shiny face, and absorbs little: the lower shelf. The ductile-to-brittle transition temperature sits in the steep middle.
Source: Adapted from FE Reference Handbook, Impact Test
Why? To tear, layers of atoms have to slip, and in steel, cold makes slipping harder, while splitting the crystal along its planes hardly changes. Below the transition, splitting wins, and a crack runs through without the metal yielding ahead of it. Aluminum and the common stainless steels, such as 304, have no such transition.
Many welded cargo ships of World War II cracked in cold seas, and a few broke in two. The investigations blamed hull steel below its transition temperature, with sharp hatch corners acting as notches. So steel for cold-weather service needs a transition temperature below its lowest service temperature, to stay on its upper shelf on the coldest day.
How hard it is to dent
The Handbook defines hardness under Hardenability, in its Materials Science chapter, as "resistance to penetration," measured by denting the material and measuring the dent. The Brinell test presses a hard ball in and measures the dent's diameter: the smaller the dent, the higher the Brinell hardness number, BHN. The Rockwell test reads how deep a diamond cone or a small ball sinks. The Handbook prints neither procedure.
A dent is permanent, so the metal under it has yielded, and a metal that resists flowing there resists it in a tension test too. For plain carbon steels, the Handbook's Relationship Between Hardness and Tensile Strength gives:
That's one rule in two unit systems, since 3.5 MPa is about 508 psi, so match the constant to the units you want. A BHN of 200 means about 100,000 psi, or 700 MPa.
What does the test measure?
Every property here comes from a test. One more is worth naming: an endurance, or fatigue, test cycles the stress up and down many times, and the Handbook's Properties of Materials defines the endurance limit as "the stress below which fatigue failure is unlikely." When a question names a test, ask what the test measures.