Necking Stress-Strain Graph (Materials Guide)
On a ductile metal’s tensile stress–strain graph, necking usually starts at the peak engineering stress, called ultimate tensile strength (UTS), and continues until fracture. Before this point, deformation is broadly uniform. After it, strain concentrates in a narrowing region, so engineering stress no longer describes local stress accurately; local area measurements and suitable analysis are needed.
Why does the curve fall after it reaches its highest point, even as the narrow part of the specimen continues to deform? The answer lies in how stress is calculated. The graph may show engineering stress, based on the specimen’s original size, or true stress, based on its changing size. These measures tell different stories once necking begins.
I’ll walk through how to read the curve, identify necking, and avoid common errors in interpreting data. The focus is a standard tensile test on a ductile metal, such as a metal that can stretch and deform before breaking.
Tensile-Test Overview and Necking on the Stress–Strain Graph
A tensile test pulls a specimen along its length and records force and extension. The resulting stress–strain graph shows how the material responds as it is stretched. Its main stages include elastic deformation, plastic deformation, strain hardening, necking, and fracture. The exact curve depends on the material and test conditions.
In the elastic region, the specimen changes shape but largely returns to its earlier shape when the load is removed. After yielding, it undergoes plastic deformation, meaning some change remains. As many ductile metals continue to deform, they strain-harden: more stress is needed to keep stretching them.
How to locate necking. On a typical engineering stress–strain curve, the highest point is the UTS. For a uniform tensile specimen, localized necking generally begins at or near this maximum. The specimen then narrows in one region, and deformation becomes concentrated there. The force the specimen carries often drops as the neck grows, so engineering stress falls.
The UTS is not the fracture point. The specimen usually continues stretching after the peak and breaks later. A graph that ends at UTS does not show the full failure process.
What the curve looks like.
| Graph stage | What happens in the specimen | What to look for |
|---|---|---|
| Elastic region | The specimen stretches and can largely recover when unloaded | Stress rises with strain |
| Yield and plastic deformation | Permanent shape change begins | Curve leaves its initial elastic trend |
| Strain hardening | Plastic deformation continues; more stress is needed | Engineering stress often continues to rise |
| UTS and necking onset | Deformation starts to localize in a narrowing region | Peak engineering stress |
| Post-necking | The neck narrows and strain becomes highly localized | Engineering stress often falls |
| Fracture | The specimen separates | End of the test record |
The graph is a summary, not a picture of the specimen. If possible, compare it with the test record and observations of the specimen. A visible neck helps confirm what the curve suggests.
Stress–Strain Definitions, Test Standards, and Interpretation Requirements
Stress and strain describe load and shape change in ways that allow tests on different-sized specimens to be compared. Engineering values use the specimen’s original dimensions. True values use its changing dimensions. Knowing which definitions a graph uses is essential, especially when interpreting the region after necking.
Engineering stress and strain. Engineering stress is the applied force divided by the original cross-sectional area:
[ \sigma_\mathrm{eng}=\frac{F}{A_0} ]
Here, (F) is force and (A_0) is the specimen’s original cross-sectional area. Engineering strain is the change in gauge length divided by the original gauge length:
[ e_\mathrm{eng}=\frac{L-L_0}{L_0} ]
Here, (L_0) is the original gauge length and (L) is its length during the test. Stress is commonly reported in megapascals (MPa); strain is a ratio and has no unit, though it may be shown as a percentage.
True stress and strain. True stress uses the cross-sectional area at that moment, (A), rather than (A_0). True strain adds small length changes over the stretch, so it is written as (\varepsilon_\mathrm{true}=\ln(L/L_0)). Before necking, when deformation is uniform, the common conversions are (\sigma_\mathrm{true}=\sigma_\mathrm{eng}(1+e_\mathrm{eng})) and (\varepsilon_\mathrm{true}=\ln(1+e_\mathrm{eng})).
Those conversions rely on uniform deformation and, in common use, an assumption of constant volume during plastic flow. Once deformation localizes in a neck, one overall gauge-length change cannot describe the local stretch there. Applying these formulas to post-necking engineering data alone does not give a reliable local true stress–strain curve.
Standard and test details. ASTM E8/E8M, Standard Test Methods for Tension Testing of Metallic Materials, sets out methods for tensile testing of metallic materials. Use the edition and specimen requirements that apply to the material and laboratory. Specimen geometry, gauge length, test setup, and measurement method matter when comparing results.
A useful check before interpreting a curve is to confirm the material, specimen dimensions, test method, and strain measurement basis. Check whether the graph reports engineering or true values, and whether strain came from an extensometer or another method. Do not assume two curves are directly comparable just because both are labeled “stress–strain.”
Reading the Curve: Necking, Measurement Limits, and Common Pitfalls
Reading a curve means matching each change in its shape to the test and the specimen. The peak engineering stress is a practical marker for necking onset in a typical uniform tensile test. It is not a complete description of what happens in the neck afterward, where local shape and stress can differ from the rest of the specimen.
The Considère criterion. For uniform plastic deformation, the Considère criterion states that necking begins when the rate of increase in true stress with true strain equals the true stress:
[ \frac{d\sigma_\mathrm{true}}{d\varepsilon_\mathrm{true}}=\sigma_\mathrm{true} ]
In plain language, necking starts when strain hardening can no longer keep the deformation uniform. This criterion explains why the engineering curve reaches a maximum at the onset of localized necking under the usual assumptions. It is a mechanics condition, not a replacement for checking the actual specimen and test data.
What changes after the peak. The measured force divided by the original area can decline while the local true stress in the neck does not. The neck’s area is shrinking, and local deformation is rising. This is why a falling engineering curve should not be read as proof that the material inside the neck is experiencing less stress.
Accurate post-necking analysis needs local information, including neck geometry or area, plus a suitable method for accounting for the stress state. The neck is not in a simple uniform tension state; stress can vary across it. Do not infer local post-necking properties from engineering stress and strain alone.
Common interpretation errors.
| Avoid this conclusion | Why it is unreliable | Better practice |
|---|---|---|
| “UTS is where the specimen breaks.” | UTS is the peak engineering stress; fracture usually comes later. | Mark UTS and fracture as separate events. |
| “The curve falls, so local stress falls.” | Engineering stress uses the original area, not the shrinking neck area. | State which stress measure the graph uses. |
| “I can convert the entire curve to true stress with one formula.” | Uniform-deformation conversions generally fail after localization. | Use local geometry and an appropriate analysis for the neck. |
| “All tensile curves can be compared directly.” | Material, specimen, and test details can affect results. | Check the applicable test method and reporting basis. |
When reading a specific graph, first identify its axes and units. Then find the peak, check the test notes, and determine whether the record includes observations of necking and fracture. This step-by-step check helps separate what the graph directly shows from what requires further analysis.
Applications in Materials Selection, Forming, and Failure Analysis
Necking behavior helps engineers compare how materials stretch and how they fail under tension. But no single point on the curve answers every selection question. A useful comparison states the test basis and the metric being used, and it treats post-necking data with care.
Materials selection. UTS describes the maximum engineering stress recorded in a tensile test. It can be useful when comparing tensile strength, but it does not by itself describe how much a material can deform before breaking. Ductility is often assessed with measures such as elongation or reduction in area; the method and specimen basis should be reported.
Forming assessment. Stretching a metal during forming can create localized thinning. A tensile curve can show how the material hardens before necking, but it does not alone predict performance in every forming process. Forming behavior also depends on the process, material direction, and other test or design conditions.
Failure analysis. A specimen that fractured after a neck formed has a different failure sequence from one that broke without clear localized thinning. The curve, specimen shape, and test record can be considered together. Avoid treating a tensile curve alone as proof of the cause of a component failure.
Applied reading scenarios. In a materials comparison, suppose two engineering curves reach different UTS values. That tells you which test recorded the higher maximum engineering stress, not automatically which material is more ductile or performs better in a forming operation. Compare the same reported measures and test basis, then consider the application’s requirements.
In a failure review, suppose the curve peaks and then declines, and the specimen shows a narrow region before breaking. That pattern is consistent with necking followed by fracture. It does not provide the local true stress in the neck unless local geometry and suitable analysis are available.
A clear report should identify the stress and strain definitions, test standard and applicable edition, specimen details, and relevant test conditions. If post-necking values are included, explain how local measurements and analysis were obtained. These details make comparisons more meaningful and limit claims to what the evidence supports.
Conclusion and FAQ
Necking is a shift from broadly uniform stretching to deformation concentrated in a narrowing region. On a typical engineering tensile curve, its onset is associated with the UTS peak, while fracture occurs later. The central interpretation rule is simple: after necking, engineering stress does not give the local stress in the neck. Use local measurements and suitable analysis to study that region.
What is necking on a stress–strain graph?
Necking is localized narrowing during a tensile test. Instead of stretching evenly, one region of the specimen takes on more deformation. On a typical engineering curve for a ductile metal, necking begins at or near the maximum engineering stress.
Does necking start at UTS?
For a uniform tensile specimen, localized necking generally begins at or near UTS, the peak engineering stress. The Considère criterion describes this onset during uniform plastic deformation. Actual interpretation should consider the specimen and test record.
Is UTS the same as fracture strength?
No. UTS is the highest engineering stress recorded during the test. The specimen usually continues deforming after that peak and fractures later.
Why does engineering stress fall after necking?
Engineering stress divides force by the original cross-sectional area. After necking begins, the force may drop as the specimen deforms, so the calculated engineering stress can fall. This value does not show the local stress in the narrowing region.
How do I convert engineering stress to true stress?
Before necking, under uniform deformation, use (\sigma_\mathrm{true}=\sigma_\mathrm{eng}(1+e_\mathrm{eng})). After necking, this conversion is generally not valid for local stress because deformation is no longer uniform.
What is the Considère criterion?
It is a condition for the onset of necking during uniform plastic deformation: (d\sigma_\mathrm{true}/d\varepsilon_\mathrm{true}=\sigma_\mathrm{true}). It links the material’s strain hardening to the start of localized deformation.
Can I calculate post-necking true stress from the engineering curve alone?
Generally, no. A reliable local estimate needs measurements of neck geometry or area and a suitable analysis. Engineering stress and strain alone do not capture the local shape change and stress state.
Which standard covers a metal tensile test?
ASTM E8/E8M is titled Standard Test Methods for Tension Testing of Metallic Materials. Check the edition and specimen requirements that apply to the material and laboratory.
What should I check before comparing two tensile curves?
Check that the material, specimen details, test method, units, and stress–strain definitions are clear. Also confirm whether strain is engineering or true and whether the curves cover the same part of deformation.
(This article was written by one of our staff writers, Alan Westbrook. Visit our Meet the Team page.)
