Stress-Strain Curve Diagram Explained

Understanding Stress-Strain Diagrams — Free Interactive Trainer

A stress-strain diagram plots engineering stress (σ = F/A₀) against engineering strain (ε = ΔL/L₀) from a tensile test. It reveals Young’s modulus (elastic slope), yield strength (onset of plastic deformation), ultimate tensile strength (peak stress), and fracture point. The curve shape determines whether a material is ductile or brittle.

The stress-strain diagram is one of the most fundamental tools in mechanical engineering and materials science. It describes how a material responds to applied forces, revealing critical properties such as stiffness, strength, ductility, and toughness. This free interactive trainer lets you explore every region of the curve, study 15 key concepts with formulas and worked examples, and test your knowledge through practice problems and a quiz.

Hooke's Law & the Elastic Region

In the initial linear portion of the curve, Hooke's Law applies: stress is directly proportional to strain (σ = Eε). The slope of this line equals Young's Modulus (E), which measures the material's stiffness. Steel has E ≈ 200 GPa, while aluminium is about 70 GPa. The material returns to its original shape when the load is removed — this is elastic behaviour.

Yield Stress, UTS & Fracture

Beyond the elastic limit, the material begins to deform permanently. The yield stress (σ_y) marks this transition. For mild steel, a distinct upper and lower yield point is visible. After yielding, strain hardening strengthens the material until it reaches the Ultimate Tensile Stress (UTS) — the peak of the curve. Beyond UTS, necking occurs (a localised reduction in cross-section), and eventually the material fractures.

Ductile vs Brittle Materials

Ductile materials (mild steel, copper, brass, aluminium, titanium) show a long plastic region and a great deal of elongation before fracture. Brittle materials (grey cast iron, glass, ceramics, concrete in tension) fracture suddenly with almost no plastic deformation. Use the vs dropdown to overlay any two of the 23 materials on the same axes: mild steel and grey cast iron are within a factor of two on strength, but the steel absorbs roughly 270 times the energy per unit volume before it breaks. That ratio, not the peak stress, is what "brittle" actually means.

How to Find Young’s Modulus From a Stress-Strain Graph

Young’s modulus is the slope of the straight part of the curve, so E = Δσ / Δε. The practical difficulty is that on an axis running to 25 % strain, that straight part occupies the first 0.125 % — you cannot read a slope off it by eye. Three things make it possible:

1. Zoom into the elastic region. Switch on the Elastic zoom layer; the inset magnifies the first fraction of a per cent by a factor of fifty or more.
2. Take two points well inside the straight run, not one at the origin. The origin carries the grip take-up — the "toe" — and using it drags the slope down.
3. Fit, don’t eyeball. In Plot My Data the simulator locates the straight region by sliding a short chord along your points, keeping the steepest one, and extending over every neighbouring chord still within 1.5 % of it. It then reports the least-squares slope and the R² of that fit.

A fixed rule such as "fit between 10 % and 40 % of the peak stress" looks reasonable and is wrong for any material that yields low against its tensile strength. Annealed copper yields at 69 MPa but peaks at 220, so that window is 22–88 MPa and three quarters of it is already plastic — it returns about 9 GPa for a 117 GPa metal. The straight region has to be found, not assumed.

How to Find Yield Strength and the UTS From a Graph

Yield depends on whether the curve has a knee. Mild steel does: read the upper yield point at the little peak, and the lower yield point on the flat plateau after it — the lower one is the design value. Aluminium, copper, brass and stainless have no knee, so yield is defined by the 0.2 % offset construction described below.

Ultimate tensile strength is the easiest number on the whole diagram: it is simply the highest point of the engineering curve, σ_u = σ_max. It is not an arbitrary peak, though — it sits exactly where Considère’s condition dσ_true/dε_true = σ_true is met, the point at which strain hardening can no longer keep up with the shrinking cross-section. That is why the strain at the UTS is called the uniform elongation: it is the last strain at which the whole gauge length is still deforming together.

Ductility comes off the horizontal axis: percentage elongation is the strain at fracture × 100, or equivalently (L_f − L₀)/L₀ × 100 measured by fitting the two broken halves back together.

What Is the Area Under a Stress-Strain Graph?

The area under the curve is energy per unit volume. Stress in MPa integrated against a dimensionless strain gives MJ/m³ directly — the units work out because N/mm² × mm/mm is N·mm per mm³. Two areas matter:

• Modulus of resilience, U_r — the triangle under the elastic line only, U_r = ½σ_yε_y = σ_y²/2E. This is the energy a part can absorb and give back. A spring wants it large, and because it goes as σ_y squared, strength matters far more than stiffness.
• Toughness, U_t — the whole area to fracture, U_t = ∫σ dε. This needs strength and ductility together.

The gap between them is startling. For mild steel the simulator integrates its own drawn curve and returns U_t ≈ 91 MJ/m³ against U_r = 0.156 MJ/m³ — the elastic triangle is 0.17 % of the total. Almost everything a ductile metal absorbs, it absorbs after it has already yielded. Switch on the Area (toughness) layer to see both shaded at once.

How to Plot a Stress-Strain Graph From Tensile Test Data

If you have a UTM log rather than a textbook curve, use Plot My Data. Paste two columns — load and extension, or stress and strain — give the original area A₀ and gauge length L₀ if you pasted load, and the simulator does the rest:

• σ = F / A₀ and ε = ΔL / L₀ for every row (strain entered as a percentage is detected and converted);
• the curve is plotted with numbered axes and the points that set the modulus highlighted;
• E, the proportional limit, σ_0.2, the UTS, the strain at the UTS, the elongation and the area under the graph are all extracted and shown as cards;
• the whole thing exports as a CSV of the data plus properties, or a labelled PNG for a lab report.

This is the step most spreadsheet tutorials skip. Getting a chart out of Excel is easy; getting the 0.2 % offset intersection out of it is not, because the offset line has to be constructed from a slope you first had to fit. Here that construction is drawn on your data, and Show Calculation prints each step with the numbers filled in.

Key Concepts Covered

Explore mode covers 15 concepts: stress, strain, Young’s modulus, Poisson’s ratio, proportional limit, elastic limit, yield stress, ultimate tensile stress, 0.2 % proof stress, necking, factor of safety, resilience, toughness, ductile materials and brittle materials. Each has a formula, its units, a description and a worked example calculation, with the diagram drawn from the same material model as the main curve.

How to Use This Simulator

Start in Curve mode — a mild-steel curve is already generated — switch on the elastic zoom, and drag the marker from the origin to fracture while watching the specimen above the chart and the permanent-set card below it. Then change material and press Generate Curve — the difference between mild steel’s yield plateau and 6061-T6’s smooth knee is the reason two different definitions of yield exist. Use Compare with to overlay a brittle curve. Move to Explore for the formulas behind what you have just seen, then Practice for numeric problems and the Quiz to check yourself. If you have lab data of your own, Plot My Data will analyse it the same way.

Mechanical Properties of Common Engineering Materials

These are the 23 materials the simulator ships with, and the numbers below are the ones the model is calibrated from — the curve you see is generated from this table, not drawn beside it. Uniform elongation is the strain at maximum load (where necking starts); total elongation is the value on the mill certificate. Yield is the lower yield point where the material has a plateau, and the 0.2 % proof stress otherwise.

Values are typical and vary with heat treatment, composition, section size and test conditions. Four comparisons in this table are worth generating for yourself:

• Grey cast iron against ductile iron. Chemically almost the same material. Round the graphite into nodules instead of leaving it as flakes and the elongation goes from 0.34 % to 12 %, and the toughness from 0.42 to 51 MJ/m³.
• Mild steel, 1045, 4140. One alloy family, carbon and chromium going up: strength rises from 400 to 1020 MPa, ductility falls from 25 % to 17.7 %, and E never moves off 200–205 GPa.
• Inconel 718 against mild steel. 5.5× the strength at exactly the same 200 GPa stiffness. Strength and stiffness are independent properties.
• Soda-lime glass against anything. 0.018 MJ/m³ of toughness — five thousand times less than mild steel — while being perfectly linear and quite strong right up to the instant it fails.

Key Stress-Strain Formulas

Reading a Real Mild-Steel Curve — Five Landmarks That Matter

Pull a tensile specimen of mild steel (low-carbon, ~0.2% C) in the simulator and the curve traces a path no other engineering material draws quite the same way:

1. Origin to A — Proportional limit (σ_pl ≈ 200 MPa). A straight line with slope E (Young’s modulus ≈ 200 GPa). Above this point the stress-strain relation stops being exactly linear, but elastic behaviour persists.
2. A to B — Elastic limit (σ_el ≈ 220 MPa). The largest stress that produces no measurable permanent strain on unloading. In undergraduate practice we treat σ_pl ≈ σ_el; in calibration-grade work the difference is microns of permanent strain.
3. B to C — Upper and lower yield (σ_y ≈ 250 MPa). The curve drops slightly, then plateaus and even oscillates — a characteristic of mild steel called yield plateau. The dislocation pile-up at carbon-atom obstacles releases suddenly. This kink is absent in aluminium alloys and stainless steels, which is why those need the 0.2% offset method below.
4. C to D — Strain hardening (σ rises to UTS ≈ 400 MPa). Dislocations multiply and entangle, the material gets stronger. The curve climbs but at decreasing slope.
5. D to E — Necking and fracture. At UTS, a local cross-section starts to thin (necking). True stress in the neck continues to climb but engineering stress — load divided by original area — falls. The curve descends to fracture at ε_f ≈ 0.25−0.35 strain (25−35% elongation).

The total area under the curve from origin to fracture is toughness — the energy absorbed per unit volume before failure. For mild steel it is roughly 50−100 MJ/m³. A high-carbon steel of the same UTS but smaller elongation has lower toughness even though it has the same strength.

The 0.2% Offset Method — Defining Yield When the Curve Has No Plateau

Aluminium, brass, copper, stainless steel, and most non-ferrous alloys lack the mild-steel yield plateau — the curve flows smoothly from elastic to plastic with no clear knee. To pin a number to yield strength, engineers use the 0.2% offset method from ASTM E8:

1. Draw the elastic line through the origin with slope E (the linear portion of the curve).
2. Translate this line to the right by ε = 0.002 (0.2% strain) so it now passes through (0.002, 0) instead of (0, 0).
3. The intersection of this offset line with the stress−strain curve is the 0.2% proof stress — the conventional yield strength σ_0.2.

For 6061-T6 aluminium, σ_0.2 ≈ 276 MPa against a UTS of 310 MPa. Without the offset convention this material would have no “yield strength” at all; with it, it fits into the same design framework as mild steel. Select Aluminium 6061-T6 from the Material dropdown and switch on the Elastic zoom: the offset line is drawn on the real curve, and the value the simulator reports is measured at the intersection rather than looked up — which is why it comes back as the published 276 MPa. Try it on mild steel too. The offset construction lands on 250 MPa there as well, agreeing with the lower yield point you can read straight off the plateau; that agreement is the reason the convention was adopted in the first place.

Ductile, Brittle, Polymer — Three Curve Shapes Side by Side

The same UTS does not imply the same usability. A brittle cast iron at UTS 250 MPa is a poor structural choice because it fails without warning; a ductile mild steel at the same UTS gives visible warning (sag, neck) before fracture. This is why pressure-vessel codes require ductile materials with minimum elongation thresholds.

Common Lab Errors — Why Your UTM Curve Does Not Match the Book

• Wrong gauge length. Elongation depends on gauge length; the standard ratio is L₀ = 5.65√A₀ (ISO 6892) for proportional specimens. Using L₀ = 50 mm on a non-standard specimen gives an “A” value you cannot compare with published data.
• Strain measured from machine crosshead, not extensometer. Crosshead displacement includes machine compliance and grip slip; the apparent E will be 30−50% low. Always use an attached extensometer for elastic-region work.
• Strain rate not controlled. Strain rate affects yield stress: faster pull → higher σ_y. ASTM E8 specifies 0.015±0.006 mm/mm per minute for the elastic range, with looser control in the plastic range.
• Grip slip mistaken for plastic strain. If the specimen looks loose at the grips after the test, the first part of the curve may be slip not real elongation. Wedge grips with proper bite resolve this.
• Necking ignored in true-stress calculation. Engineering stress falls after UTS, but true stress σ_true = F/A_{instantaneous} continues to rise. For metal forming calculations you must convert.

Permanent Set — What Is Left When You Unload

Unloading does not retrace the curve. It follows a straight line parallel to the original elastic line, because only the elastic part of the strain is recoverable. So from any point (σ, ε) on the curve:

• the strain that springs back is ε_elastic = σ/E;
• what remains is the permanent set, ε_p = ε − σ/E.

Below the elastic limit ε_p is zero and the specimen returns to its original length. Above it, the bar is permanently longer. Switch on the Unload line layer and drag the marker: at the UTS of mild steel the tool reports about 20 % permanent set out of 20.2 % total strain — the elastic part is by then almost nothing. This is also the mechanism behind the 0.2 % offset: that construction is nothing more than asking "at what stress would unloading leave 0.002 of permanent strain?"

It is why a spring must stay below yield to be a spring, why a bent sheet stays bent (with a small springback of exactly σ/E), and why cold working raises the yield stress of the metal you have already deformed.

How Accurate Is This Simulator, and Where Does It Stop?

The curve is not a drawing. Up to the ultimate tensile strength it is generated by a flow law fitted in true stress-strain space and calibrated against three conditions at once: it passes through the published yield point, it passes through the published UTS, and it satisfies Considère’s necking condition exactly there. Both a Swift law (σ_t = K(ε₀ + ε_p)ⁿ) and a Ludwik law (σ_t = σ_a + K(ε_p − ε_a)ⁿ) are available; the readout names which one a given material used. Elastic and plastic strain compose multiplicatively, so the engineering elastic branch is exactly Hooke’s line.

An automated harness re-derives the curve from the shipped source and checks it against more than 5,000 assertions — every material, plus a sweep of 576 user-defined ones. What it verifies:

Where it stops. Three limits are worth stating plainly:

1. The branch after the UTS is empirical. The neck geometry is not modelled. That segment is an interpolation from the peak down to the published fracture stress at the published elongation, with zero slope at the peak so the UTS really is the maximum. Everything before the UTS is the calibrated flow law.
2. No Bridgman correction. Once necking starts the stress state inside the neck is triaxial, and true stress computed as F/A over-reads. The true curve is therefore drawn dashed past necking, where it is indicative rather than quantitative.
3. Rate and temperature are not modelled. Every curve is quasi-static and at room temperature. Yield stress rises with strain rate and falls with temperature; steels have a ductile-to-brittle transition that this diagram cannot show at all.

Grey cast iron and polypropylene have no straight region, so what is labelled E for them is an initial tangent, and the tool says so on the card. The full record is in docs/stress-strain-validation.md in the project repository.

Standards and References

• ASTM E8/E8M-21 — Standard Test Methods for Tension Testing of Metallic Materials. The American baseline for tensile tests.
• ISO 6892-1:2019 — Metallic materials — Tensile testing — Part 1: Method of test at room temperature. The international (and European) equivalent.
• IS 1608:2018 — the Indian standard, widely used in Indian polytechnics and aligned with ISO 6892.
• Callister, W. D. — Materials Science and Engineering: An Introduction, 10th ed., Chapter 6 (Mechanical Properties of Metals).
• Dieter, G. E. — Mechanical Metallurgy, 3rd ed., for the true-stress true-strain treatment and the dislocation theory behind yield phenomena.

Explore Related Simulators

If you found this Stress–Strain Curve simulator helpful, explore our Universal Testing Machine simulator, Hooke’s Law simulator, Impact Testing simulator, Mohr’s Circle simulator, Thermal Expansion Calculator, and the Sheet Metal Bend Radius Calculator, where the same ductility that sets elongation also sets the minimum bend radius.