Mechanical Behavior and Materials Testing

Introduction

Mechanical behavior describes how a material responds when subjected to force, displacement, temperature, time, or repeated loading. Engineers use this knowledge to select materials, establish design allowables, predict service life, diagnose failures, and verify manufacturing quality. A material may deform elastically and recover its shape, plastically and retain permanent deformation, fracture suddenly, or deform gradually under a sustained load. Real materials often display more than one of these behaviors.

Mechanical testing converts these behaviors into measurable properties such as elastic modulus, yield strength, tensile strength, ductility, hardness, fracture energy, creep rate, and fatigue life. Reliable interpretation requires more than reading a test-machine output. Specimen geometry, loading rate, temperature, surface finish, alignment, measurement uncertainty, and the applicable standard all influence the result.

This chapter focuses on tensile, compression, hardness, impact, creep, and fatigue testing, together with standards, uncertainty, and statistically reliable property data.

Learning Objectives

After studying this chapter, you should be able to:

1. Distinguish elastic, plastic, viscoelastic, creep, and fatigue behavior.
2. Interpret engineering stress–strain and true stress–strain data.
3. Explain the principles and limitations of major mechanical tests.
4. Select appropriate standards and specimen conditions.
5. Estimate uncertainty and distinguish random variation from systematic error.
6. Use statistical methods to report reliable engineering properties.
7. Relate laboratory results to design, quality assurance, and failure prevention.

1. Fundamental Mechanical Response

1.1 Stress and strain

For a uniform axial specimen, engineering normal stress is

where σ is engineering stress in pascals (Pa), F is axial force in newtons (N), and A₀ is the original cross-sectional area in square metres (m²). Engineering strain is

where L₀ is the original gauge length and L is the instantaneous gauge length. Strain is dimensionless, although it is commonly reported as mm/mm or percent.

These definitions are convenient for standard tensile tests but do not account for changing area. True stress and true strain are

and, for uniform deformation,

where A is the instantaneous area. The logarithmic strain expression is valid while the gauge section deforms continuously and remains measurable. After necking, local geometry must be measured carefully; the simple uniform-deformation conversion is no longer generally adequate.

1.2 Elastic behavior

Elastic deformation is recoverable when the load is removed. In the initial linear region of many metals,

where E is Young’s modulus, measured in Pa or GPa. The equation assumes small strain, uniaxial loading, homogeneous material, and linear elastic behavior. The slope of the unloading curve may differ if plasticity, damage, or viscoelastic effects have developed.

For isotropic linear elasticity, Young’s modulus, shear modulus G, and Poisson’s ratio ν are related by

This relation applies to homogeneous, isotropic, linear elastic materials. Anisotropic composites require more elastic constants and cannot generally be described by this single relationship.

The elastic strain energy density in uniaxial linear elasticity is

where uₑ has units of J/m³. This is the recoverable energy stored per unit volume.

1.3 Plastic behavior and yielding

Plastic deformation is permanent. Yielding begins when the applied stress reaches a material-dependent yield criterion. Some low-carbon steels show a distinct upper and lower yield point; many metals do not. For materials without a sharply defined yield point, the 0.2% offset method is common: a line parallel to the elastic slope is displaced by ε = 0.002, and its intersection with the stress–strain curve defines the proof stress.

Important tensile properties include:

• Yield strength or proof stress: stress associated with the beginning of significant permanent deformation.
• Ultimate tensile strength (UTS): maximum engineering stress.
• Fracture stress: engineering stress at failure.
• Uniform elongation: elongation before localized necking.
• Percentage elongation: permanent gauge-length increase after fracture.
• Reduction of area: decrease in cross-sectional area at the fracture location.

Ductility is not identical to toughness. Ductility measures plastic deformation; toughness is the energy absorbed before fracture and depends on both strength and deformation capacity.

1.4 Strain hardening and necking

After yielding, many metals strengthen as plastic strain increases. A simplified power law for uniform plastic deformation is

where σₜ is true flow stress, εₚ is true plastic strain, K is the strength coefficient, and n is the strain-hardening exponent. This is an empirical approximation, normally applied over a specified plastic-strain range. It should not be extrapolated blindly to elastic strain, very large strain, or post-fracture behavior.

Necking begins when the material can no longer increase its load-carrying capacity uniformly. Under common assumptions, the Considère condition for the onset of diffuse necking is

The condition assumes monotonic uniaxial tension and sufficiently uniform deformation before necking.

1.5 Viscoelastic and time-dependent behavior

Viscoelastic materials combine elastic energy storage with viscous dissipation. Polymers, adhesives, biological materials, and some composites show strong dependence on loading time and temperature. A simple Maxwell model consists of a spring and dashpot in series; a Kelvin–Voigt model consists of a spring and dashpot in parallel. These are conceptual models rather than universal descriptions.

For a constant stress σ₀, a Maxwell model gives stress relaxation and a characteristic relaxation time

where η is viscosity in Pa·s and E is elastic modulus in Pa. For a constant strain, the stress decreases approximately exponentially in the ideal Maxwell model. Real polymers often require generalized Maxwell or Prony-series models.

Creep is progressive deformation under sustained stress, usually at elevated temperature. Stress relaxation is reduction in stress under sustained strain. Both are time-dependent and temperature-sensitive. The time–temperature superposition principle can sometimes shift data measured at different temperatures along a reduced-time axis, but its validity must be demonstrated for the material and temperature range.

2. Tensile Testing

A tensile test applies controlled axial displacement or force to a standardized specimen until fracture. A testing machine records force and extension; software converts these data to stress and strain using specimen dimensions and gauge length.

2.1 Test procedure

A sound procedure includes specimen identification, dimensional measurement, alignment, extensometer installation, controlled loading rate, environmental control, and post-fracture measurement. Standards such as ASTM E8/E8M or ISO 6892-1 specify specimen geometry, strain-rate methods, temperature, reporting, and definitions. The applicable standard depends on material type and jurisdiction.

The tensile curve usually contains:

1. An initial elastic region.
2. Yielding or proof-stress determination.
3. Plastic flow and strain hardening.
4. Maximum engineering stress.
5. Necking.
6. Fracture.

The area under a complete engineering stress–strain curve approximates the tensile energy per unit original volume, but after necking it is not a direct local fracture-energy measure. Fracture toughness requires a separate fracture-mechanics test.

2.2 Worked example: tensile properties

A cylindrical specimen has diameter 10.00 mm and gauge length 50.00 mm. At a recorded force of 25.0 kN, its gauge length is 50.50 mm. The original area is

The engineering stress is

The engineering strain is

Assuming uniform deformation, true strain is ln(1.0100) = 0.00995. The true stress cannot be calculated reliably from the engineering conversion unless the specimen remains uniform and its current area is inferred consistently; local necking would invalidate that simplification.

3. Compression Testing

Compression tests apply shortening rather than extension. They are useful for brittle materials, foams, powders, concrete, ceramics, polymers, and materials used under compressive loading. Ductile metals may not fracture in compression; instead, they may barrel because friction between specimen ends and platens restrains lateral expansion.

For a specimen of initial height h₀ and area A₀, engineering compressive strain may be reported as

and compressive stress magnitude as σc = F/A₀. Sign conventions vary, so reports must state whether compression is positive or negative.

End friction, platen parallelism, specimen slenderness, alignment, loading rate, and temperature strongly affect results. Lubrication can reduce friction but may be prohibited or prescribed by a standard. Buckling must be avoided in slender specimens. Standards such as ASTM E9 address compression testing of metallic materials, while other materials require dedicated standards.

4. Hardness Testing

Hardness is resistance to localized plastic deformation, usually measured by indentation or penetration. It is rapid, economical, and often suitable for production control, but it is not a fundamental constant independent of test method.

Common methods include:

For a Brinell test, hardness is calculated from force F, ball diameter D, and indentation diameter d by

The formula requires consistent force and length units and applies to the specified Brinell geometry and procedure. Hardness conversions between scales are empirical and should not be treated as exact universal transformations. Surface curvature, thickness, spacing between indents, surface roughness, anisotropy, and work hardening can produce misleading results.

Hardness may correlate approximately with tensile strength for a particular alloy and heat-treatment condition. Such correlations must be validated for the relevant material batch and should not replace tensile testing when design certification is required.

5. Impact Testing

Impact tests measure energy absorbed during rapid fracture. The Charpy V-notch test uses a notched, simply supported specimen struck by a pendulum. The absorbed energy is the difference between pendulum energy before and after fracture, with machine corrections applied according to the standard.

Impact energy is sensitive to temperature, notch geometry, specimen orientation, thickness, loading rate, and microstructure. A ductile-to-brittle transition may occur in some body-centred cubic steels. The transition temperature is not a single universal material constant; it depends on the chosen criterion, such as absorbed energy or fracture appearance.

Impact energy should not be confused with plane-strain fracture toughness K_IC. A Charpy test is valuable for comparative toughness and quality control, whereas fracture toughness testing characterizes crack-growth resistance under specified geometry and constraint.

6. Creep and Stress-Rupture Testing

A creep test maintains a specified stress and temperature while measuring strain as a function of time. The classical creep curve has three stages:

1. Primary creep: decreasing creep rate as the material hardens.
2. Secondary creep: approximately steady creep rate.
3. Tertiary creep: accelerating strain associated with damage, necking, voids, or cracking.

The minimum or steady-state creep rate is often used for design comparisons. A simplified power-law relation is

where ε̇c is creep strain rate, A is a material constant, σ is stress, m is the stress exponent, Q is activation energy, R is the gas constant, and T is absolute temperature in kelvin. This equation applies only within the mechanism and range used to determine its constants. It should not be extrapolated across changes in creep mechanism, phase transformation, oxidation, or microstructure.

Creep rupture tests continue until failure and record rupture time, temperature, stress, and strain. Larson–Miller-type parameters are sometimes used for preliminary long-term extrapolation, but extrapolation over decades requires evidence that the governing mechanism remains unchanged.

7. Fatigue Testing

Fatigue is progressive damage caused by repeated or fluctuating stress, often below the monotonic yield strength. A fatigue test applies controlled cycles and records the number of cycles to crack initiation or final fracture.

For a sinusoidal stress cycle,

where σₘ is mean stress and σₐ is stress amplitude. The stress ratio is

A completely reversed cycle has R = −1 and zero mean stress. An S–N curve relates stress amplitude to cycles to failure. It is strongly affected by surface finish, residual stress, size, corrosion, temperature, frequency, load sequence, and defects.

For long-life design, the modified Goodman relation is sometimes used:

where Se is an endurance-strength parameter and Sut is ultimate tensile strength. This is an engineering approximation, not a universal fatigue law. It assumes the chosen correction factors and failure criterion are appropriate. Other criteria, such as Gerber or Soderberg, may be more suitable for particular designs.

Fatigue results are statistical because small defects and microstructural variations cause scatter. Test reports must identify whether failure means crack initiation, complete fracture, or a run-out at a specified cycle count.

8. Standards, Measurement, and Uncertainty

8.1 Standards and traceability

Standards define specimen dimensions, calibration, loading rate, environmental conditions, data reduction, and reporting. Examples include ASTM, ISO, EN, and national standards. The engineer must verify the current edition, material scope, and any customer or regulatory requirement. A result obtained using a non-equivalent method should not be presented as compliance with a specified standard.

Calibration establishes the relationship between indicated and actual force, displacement, temperature, or time. Traceability links calibration to recognized references through an unbroken documented chain. Machine verification alone does not guarantee a valid test if specimen alignment or preparation is poor.

8.2 Sources of uncertainty

Uncertainty arises from:

• force and displacement calibration;
• dimensional measurement;
• specimen-to-specimen variation;
• temperature and humidity;
• alignment and gripping;
• surface preparation and machining;
• operator and data-reduction choices;
• material anisotropy and microstructural gradients.

Random effects cause scatter; systematic effects shift results in one direction. Repeating a test reduces uncertainty from random effects but does not remove calibration bias or a flawed method.

For independent input quantities xᵢ, a first-order combined standard uncertainty is

where y = f(x₁, x₂, …), u(xᵢ) is the standard uncertainty of input xᵢ, and covariance terms are omitted only when inputs are independent. If inputs are correlated, covariance terms must be included. An expanded uncertainty is commonly reported as

where k is a coverage factor selected according to the reporting objective and degrees of freedom. A statement such as “318 ± 4 MPa, k = 2” is incomplete unless the measurand, confidence interpretation, method, and units are clear.

9. Statistical Reliability of Property Data

A sample mean is

and the sample standard deviation is

The standard error of the mean is s/√n when observations are independent and identically distributed. A confidence interval for a mean should use an appropriate distribution, often the t-distribution for small samples and unknown population standard deviation.

Mechanical properties may be non-normal. Fatigue lives commonly use logarithmic scales, while failure or rupture data may require Weibull analysis. Outliers must not be removed merely because they are inconvenient; removal requires a documented technical cause such as specimen damage, machine malfunction, or confirmed measurement error.

Design values are not automatically equal to sample means. A conservative lower tolerance limit may be needed to ensure that a specified proportion of the material population exceeds a minimum strength. Sample size, lot definition, sampling plan, confidence level, and material traceability must be reported.

10. Illustrative Case: Qualification of an Aluminum Bracket

Background

A manufacturer is qualifying a heat-treated aluminum alloy for an aircraft-support bracket. The bracket experiences static tension, vibration, occasional impact during handling, and sustained moderate temperature.

Problem

Initial tensile tests show an average proof stress of 285 MPa, but one specimen gives 258 MPa. The team must determine whether the low value represents random scatter, a material defect, or a testing problem.

Analysis

The engineers first review specimen orientation, machining marks, heat-treatment records, extensometer verification, gauge dimensions, and fracture locations. The low-strength specimen fractured near a grip mark, outside the valid gauge region. The result is therefore investigated as a potentially invalid test, not automatically discarded. Additional specimens are sampled from different locations in the production lot.

Hardness measurements are taken across the lot to detect heat-treatment gradients. A Charpy-type impact test is not used as a substitute for fracture toughness, but it is used for comparative screening. Fatigue tests are conducted at the bracket’s stress ratio and with representative surface finish. Creep testing is performed at the maximum service temperature because short tensile tests cannot establish long-term deformation.

Approach and outcome

The laboratory repeats the tensile test using the applicable standard, documents the exclusion of the grip-fractured specimen, and reports both individual results and summary statistics. The new data show a lower-bound proof stress above the design requirement, but fatigue performance reveals sensitivity to a sharp machining transition. The design is modified with a larger fillet and controlled surface finishing. Process control adds hardness mapping and periodic tensile verification.

Learning points

A single property value is not sufficient for qualification. Validity depends on specimen failure location, standard compliance, lot sampling, uncertainty, and service-specific loading. Static strength, hardness, fatigue, impact, and creep tests answer different questions and cannot be substituted casually for one another.

11. Best Practices and Limitations

1. Define the service failure mode before selecting a test.
2. Use the current, material-specific standard and record deviations.
3. Verify calibration, alignment, dimensions, temperature, and loading rate.
4. Report individual observations, not only an average.
5. Separate measurement uncertainty from material scatter.
6. Preserve specimens, fracture surfaces, and raw data where failure analysis may be required.
7. Avoid extrapolating hardness correlations, creep laws, or fatigue curves beyond validated ranges.
8. Use representative surface condition, environment, frequency, and stress ratio.
9. Treat safety-critical properties as statistical design data, not isolated measurements.
10. Maintain impartial reporting and never suppress an inconvenient valid result.

Computer vision and machine learning can assist fracture-surface classification, automated extensometry, and anomaly detection. However, training-data bias, poor calibration, lack of explainability, and domain shift can produce unsafe conclusions. AI outputs should support—not replace—validated measurement, engineering judgment, and standards-based acceptance.

Chapter Summary

Mechanical behavior includes recoverable elasticity, permanent plasticity, viscoelastic response, creep, stress relaxation, fracture, and fatigue damage. Tensile tests provide modulus, yield or proof stress, strength, and ductility; compression tests characterize shortening and load-bearing behavior while requiring attention to friction and buckling. Hardness testing is rapid but method-dependent. Impact testing measures absorbed energy under rapid notched fracture and is distinct from fracture toughness. Creep and stress-rupture tests establish time- and temperature-dependent behavior, while fatigue tests characterize failure under repeated loading.

Reliable data require standardized procedures, calibrated equipment, suitable specimens, controlled environments, uncertainty evaluation, and statistical treatment of scatter. Means and standard deviations summarize samples but do not automatically provide safe design values. Engineering decisions must connect the test method and its limitations to the actual service conditions.

Key Terms

Elastic deformation: Recoverable deformation after unloading.

Plastic deformation: Permanent deformation caused by irreversible material rearrangement.

Young’s modulus: Slope of the linear elastic uniaxial stress–strain curve.

Poisson’s ratio: Negative ratio of transverse strain to axial strain in elastic deformation.

Yield strength: Stress defining the onset of specified permanent deformation.

Proof stress: Offset-based yield measure, commonly using 0.2% strain.

Ultimate tensile strength: Maximum engineering tensile stress.

Ductility: Capacity for plastic deformation before fracture.

Toughness: Energy absorbed before fracture.

Strain hardening: Increase in flow stress with plastic strain.

Viscoelasticity: Combined elastic and viscous, time-dependent behavior.

Creep: Progressive deformation under sustained stress.

Stress relaxation: Decrease in stress under sustained strain.

Hardness: Resistance to localized indentation or plastic deformation.

Impact energy: Energy absorbed during rapid notched fracture.

Fatigue: Progressive damage under cyclic loading.

Stress ratio: Minimum stress divided by maximum stress in a fatigue cycle.

Traceability: Documented connection of measurements to recognized references.

Measurement uncertainty: Quantified doubt associated with a reported result.

Tolerance limit: Statistical bound intended to contain a specified proportion of a population.

Review Questions

Short Answer Questions

1. Distinguish engineering stress from true stress.
2. Why is a 0.2% offset method used for many metals?
3. State two factors that influence hardness measurements.
4. What is the difference between creep and stress relaxation?
5. Why should fatigue results be treated statistically?

Long Answer Questions

1. Explain the major regions of an engineering tensile stress–strain curve and the material behavior represented by each region.
2. Compare tensile, compression, hardness, and impact tests in terms of principle, measured property, and limitations.
3. Describe the three stages of a creep curve and explain why temperature control is essential.
4. Explain how specimen preparation, alignment, calibration, and loading rate affect mechanical-test validity.
5. Discuss why Charpy impact energy should not be interpreted as a direct measurement of fracture toughness.

Analytical / Application Questions

1. A 12 mm diameter specimen carries 30 kN while its gauge length increases from 60.0 to 60.72 mm. Calculate engineering stress and strain, stating assumptions.
2. A fatigue specimen is tested between +180 MPa and −60 MPa. Calculate mean stress, stress amplitude, and stress ratio, then explain the likely significance of the nonzero mean stress.
3. A production lot has a low tensile result from a specimen that fractured at a grip mark. Propose a technically defensible investigation and explain when, if ever, the result may be excluded.

Multiple Choice Questions with Answers

Question 1

Which equation defines engineering stress in a tensile test?

A. σ = F/A₀

B. σ = A₀/F

C. σ = ∆L/L₀

D. σ = ln(L/L₀)

Correct Answer: A. σ = F/A₀

Question 2

A specimen has an original gauge length of 50 mm and a loaded length of 50.5 mm. What is its engineering strain?

A. 0.001

B. 0.005

C. 0.010

D. 1.010

Correct Answer: C. 0.010

Question 3

Which property is normally obtained from the slope of the initial linear tensile curve?

A. Hardness

B. Young’s modulus

C. Impact energy

D. Fatigue life

Correct Answer: B. Young’s modulus

Question 4

What is the principal purpose of a hardness test?

A. Measure resistance to localized indentation

B. Measure long-term creep rupture directly

C. Determine fracture toughness exactly

D. Measure thermal conductivity

Correct Answer: A. Measure resistance to localized indentation

Question 5

In a fatigue cycle, stress amplitude is calculated as:

A. (σmax + σmin)/2

B. σmax/σmin

C. (σmax − σmin)/2

D. σmin − σmax

Correct Answer: C. (σmax − σmin)/2

Question 6

Which stage of creep generally shows an approximately constant creep rate?

A. Primary

B. Secondary

C. Tertiary

D. Elastic

Correct Answer: B. Secondary

Question 7

A Charpy test primarily measures:

A. Absorbed energy during rapid notched fracture

B. Young’s modulus

C. Steady-state creep rate

D. Poisson’s ratio

Correct Answer: A. Absorbed energy during rapid notched fracture

Question 8

Repeating a test generally reduces which type of uncertainty most directly?

A. Random uncertainty

B. Known calibration bias

C. Incorrect specimen geometry

D. Systematic alignment error

Correct Answer: A. Random uncertainty

Question 9

For a fatigue cycle with σmax = 180 MPa and σmin = −60 MPa, the mean stress is:

A. −120 MPa

B. −60 MPa

C. 60 MPa

D. 120 MPa

Correct Answer: C. 60 MPa

Question 10

Why should a hardness-to-tensile-strength conversion be used cautiously?

A. Hardness has no units

B. The correlation is empirical and material-condition dependent

C. Tensile strength is always lower than hardness

D. Hardness tests require cyclic loading

Correct Answer: B. The correlation is empirical and material-condition dependent