Steel Tensile Testing: Yield Strength, Ultimate Strength and Elongation

Structural steel has five key tensile properties, each with specific meaning for design: (1) Yield strength—the stress at which permanent deformation begins. Below yield, steel deforms elastically—stress is removed and the steel returns to original shape. At yield, steel begins plastic deformation—the material doesn't return to original shape when stress is removed. Designers specify maximum stress equal to yield strength (sometimes with a safety factor below yield) to ensure permanent deformation doesn't occur in normal operation. Exceeding yield in service is generally unacceptable. (2) Ultimate tensile strength (UTS)—the maximum stress the steel can withstand. Beyond UTS, stress actually decreases as the material develops a neck (localized deformation) and approaches failure. (3) Elongation—how much the material stretches before breaking, measured as percentage increase in gauge length. This is a ductility measure—how much the material deforms plastically. High elongation indicates ductility (the material deforms significantly before breaking); low elongation indicates brittleness (sudden failure with little warning). (4) Modulus of elasticity—the slope of the elastic region of the stress-strain curve, indicating material stiffness. Steel has a modulus around 200 GPa (206-211 GPa is typical). Stiffer materials have higher modulus. (5) Poisson's ratio—when material is stretched in one direction, it shrinks in perpendicular directions. For steel, this ratio is approximately 0.30. These properties are not independent—stronger steel often has lower elongation; material changes that increase yield strength often decrease ductility. Steel specifications balance these properties to provide both strength and ductility.

The stress-strain curve shows complete behavior from initial loading to failure. The initial portion (from zero stress to yield) is nearly linear—stress and strain are proportional. This is the elastic region. The slope of this region is the modulus of elasticity. The material behaves as a spring—if stress is removed, strain returns to zero. At yield strength, the curve's slope changes—strain increases without proportional stress increase. This is the plastic region. The material begins permanent deformation. Stress-strain curves show this transition subtly (no dramatic kink for most steels) or more abruptly (some steels show a clear yield point). Beyond yield, strain continues with increasing stress (strain hardening) until the curve reaches its peak at ultimate strength. The strain hardening reflects the material strengthening as it deforms—dislocations in the crystal structure are moving, hardening the material against further deformation. Beyond ultimate strength, stress decreases while strain continues. This decrease reflects necking—the material develops a localized thin section. The necking concentrates stress in the thin section, causing rapid additional strain in that region. Finally, the material breaks at maximum strain. The area under the stress-strain curve (stress multiplied by strain, integrated across the full range) represents the energy the material can absorb before breaking—this is energy toughness. A material with large stress-strain curve area has good toughness (it absorbs significant energy). A material with small area (quickly reaching failure) has poor toughness.

Identifying the yield point from a stress-strain curve is sometimes straightforward and sometimes challenging. For many steels, a discontinuity appears in the curve—stress actually drops at yield, then continues. This is the upper yield point and lower yield point. Standards typically use the lower yield point (the lower stress value) for design. For steels without a distinct yield point discontinuity, yield is identified using the offset method: a line parallel to the elastic region's slope is drawn offset by 0.2% strain, and where this line intersects the stress-strain curve is defined as yield strength. This 0.2% offset represents acceptable permanent deformation—the material has 0.2% residual strain after stress removal, but this is considered acceptable. Modern testing machines automatically identify yield using these criteria. Visual inspection of the curve confirms the automatic identification is correct. If yield identification is ambiguous, the test might be repeated or the material might be rejected for not meeting standard clarity. Yield strength is the most important property for design—structural members are designed so maximum service stresses don't exceed yield. Using yield strength with appropriate safety factors ensures structures operate in the elastic region with acceptable margins of safety.

After yield, stress continues increasing as strain increases (strain hardening). The material becomes stronger as it deforms. This is counterintuitive—one might expect continued deformation would weaken the material, but actually deformation rearranges the crystal structure, creating barriers to further deformation, strengthening the material. This strengthening is limited—continued deformation eventually causes necking (localized diameter reduction in the specimen). At necking, stress concentrates in the thin section. Although the true stress (force divided by actual cross-sectional area, which is decreasing) might still increase, the engineering stress (force divided by original cross-sectional area) decreases because the cross-sectional area is decreasing. The test machine records engineering stress, which is why the curve shows decreasing stress beyond necking. The ultimate strength is the peak of the engineering stress curve—the highest stress the machine records. Beyond this point, deformation is increasingly localized in the neck, and the material breaks. Elongation is measured as the percentage increase in gauge length at breaking compared to original gauge length. High-strength steel might have yield strength of 400 MPa and ultimate strength of 500 MPa with 20% elongation (ductile). Very high-strength steel might have yield strength of 700 MPa and ultimate strength of 800 MPa with only 10% elongation (less ductile but stronger). For structural applications, especially in seismic regions or blast-resistant design, high ductility is essential—the structure must deform without sudden catastrophic failure. For other applications, strength might be prioritized over ductility.

Tensile test results are meaningful only if the specimen and procedure follow standards. Standard specimens have specified diameter (commonly 12.5mm), gauge length (typically 50mm for diameter of 12.5mm), and overall length. Specimens are typically machined from a material sample or taken directly from the product (for wire or rod). The gauge length is marked with witness marks before testing. During testing, the distance between marks is measured to determine elongation at breaking. The specimen must be centered and aligned in the testing machine—eccentric loading produces misleading results. The machine applies force monotonically (continuously increasing without reversals) at a controlled loading rate. Slow loading rates might produce slightly different results than fast rates for some materials. Standards specify acceptable loading rates—commonly 2-50 MPa/second. Quality control procedures ensure testing machines are calibrated regularly and that procedures follow standards precisely. The machine records load and displacement continuously as testing proceeds, generating the load-displacement curve. This curve is converted to stress-strain curve by dividing load by original cross-sectional area (to get stress) and dividing displacement by gauge length (to get strain). Humidity, temperature, and other factors can affect properties—testing is often done at standard conditions (21°C, 65% humidity). Material certified as meeting a certain grade must meet specific minimum properties across a range of samples validated through quality assurance protocols.

Structural steel is specified by grade, which defines required properties. Common grades include S235 (European system, meaning 235 MPa minimum yield), S275, S355, S450, and S460. American system uses grades like Grade 50 (50 ksi or approximately 345 MPa yield) or Grade 60 (60 ksi or approximately 414 MPa yield). Each grade specifies minimum yield strength, minimum ultimate strength, minimum elongation, and sometimes other properties. For example, S355 might require: minimum yield 355 MPa, minimum ultimate 470 MPa, minimum elongation 22%. Samples are tested from the production batch—if all properties meet minimums, the steel is certified. If any property fails (e.g., elongation is only 20% instead of minimum 22%), the entire batch might be rejected unless investigation determines the failure is acceptable. Quality assurance procedures ensure testing is performed by accredited laboratories following standardized methods. Material properties vary somewhat—not all samples have identical properties. Standards specify that properties are the guaranteed minimums—characteristic values (typically the 5th percentile of test results) are often used in design, ensuring even poor samples meet specifications. Construction management and A&E teams verify that material certifications are reviewed and retained before steel is incorporated into the structure.

Structural engineers select steel grade based on design requirements. Stronger steel (higher grade) reduces member size and weight, often reducing cost despite higher material cost. Weaker steel (lower grade) reduces material cost but might increase member size, adding weight and potentially increasing fabrication complexity. The optimal choice balances these factors. For applications requiring ductility (seismic regions, dynamic loads), engineers ensure selected material has adequate elongation—some high-strength steels have unacceptably low elongation. For fatigue applications (cyclic loading), material selection is more complex—fatigue strength depends on yield strength but also on other factors like notch sensitivity and surface finish. Decarburization (carbon loss from steel surface during processing) can reduce surface strength—materials for fatigue service must avoid decarburization. Testing verifies material selection decisions are correct—samples from the delivered material are tested to confirm properties match specifications.

Material properties aren't perfectly uniform—samples show natural variation. Steel from one heat (one batch from the furnace) varies slightly; steel from different heats varies more; steel from different mills varies more still. Testing programs account for this variation. For large orders, testing frequency might be one sample per heat (large batch from furnace), or per mill lot. For smaller orders, perhaps one sample per shipment. Statistical concepts ensure tested samples are representative. A property value is reported with description of how representative it is—for example, 'characteristic value 355 MPa (5th percentile of population)' or 'mean value 360 MPa with standard deviation 5 MPa.' This characterization enables engineers to apply appropriate safety factors. Material certification documents test results—this documentation provides evidence the material met specifications and enables traceability if problems develop later.