Technical Explanation: Failure Theories for Ductile Materials
In engineering design, real components rarely experience the clean uniaxial tension of a coupon test. A shaft sees combined bending and torsion; a pressure vessel wall carries hoop and axial stress simultaneously. To check such a multi-axial stress state against a material's yield strength Sy — measured in a simple 1D tensile test — engineers reduce the full stress state to a single scalar value using a failure theory. For ductile metals under static loading, the two most trusted criteria are the Von Mises and Tresca theories, both implemented by this calculator.
Step 1: Principal Stresses in Plane Stress
The calculator assumes a 2D plane stress state defined by the in-plane normal stresses σx, σy and the shear stress τxy, with the out-of-plane stress σz = 0. The first step is to transform this state into principal stresses — the normal stresses on planes where shear vanishes:
σ₁,₂ = (σx + σy)/2 ± √[((σx − σy)/2)² + τxy²]
The third principal stress is σ₃ = 0. The principal stresses are then ordered (σ₁ ≥ σ₂ ≥ σ₃), because both failure theories depend only on the ordered extremes of the stress state, not on the original coordinate system — a direct consequence of Mohr's circle transformation.
Von Mises (Maximum Distortion Energy Theory)
The Von Mises criterion predicts yielding when the distortion energyper unit volume — the portion of strain energy that changes shape rather than volume — equals the distortion energy at yield in simple tension. For ductile metals under static loading it agrees closely with experiment and is the default output of nearly every FEA post-processor. In terms of principal stresses:
σvm = √[½((σ₁ − σ₂)² + (σ₂ − σ₃)² + (σ₃ − σ₁)²)]
The equivalent stress is then compared directly to yield: yielding is predicted when σvm ≥ Sy. Yielding under this criterion is insensitive to hydrostatic pressure (equal triaxial stress), which matches the observed behavior of ductile materials.
Tresca (Maximum Shear Stress Theory)
The Tresca criterion states that a material yields when the maximum shear stress in the material reaches the maximum shear stress at the yield point in simple tension. Instead of an "equivalent stress," Tresca is more accurately described as computing a stress intensity based on the maximum principal stress difference:
σTresca = max(|σ₁ − σ₂|, |σ₂ − σ₃|, |σ₃ − σ₁|)
Yielding is predicted when σTresca ≥ Sy. The Tresca yield surface is more conservative than Von Mises: its yield boundary (a hexagon) lies entirely inside the Von Mises surface (an ellipse) in principal-stress space. The largest divergence occurs in pure shear, where Tresca is about 15.5% more conservative (σTresca = 2τ versus σvm = √3·τ). It is often preferred where safety margins must be strictly preserved.
Input Parameters
- σx, σy: Normal stresses on the element faces, positive in tension, negative in compression, in consistent units (typically MPa or psi).
- τxy: In-plane shear stress. The principal-stress formula depends on τ², so the sign convention for shear does not affect the result — but the magnitude must be correct.
- Yield Strength (Sy): Taken from material data for the specific alloy, heat treatment, and temperature. Never substitute ultimate tensile strength (Su) when the design intent is to avoid permanent deformation.
Interpreting the Factor of Safety (FoS)
FoS = Sy / σequivalent. An FoS below 1.0 indicates the calculated stress exceeds yield; under the assumptions of the selected criterion, yielding is predicted, meaning plastic (permanent) deformation is expected to begin. That does not automatically imply fracture or collapse — a ductile component may carry load well beyond first yield — but permanent deformation is itself a failure mode for precision components, seals, and fatigue-loaded parts. Typical design targets are FoS = 1.5–2.0 against yield for static loads, higher for impact or uncertain loads. A passing FoS does not replace a complete verification: fatigue, buckling, fracture toughness, creep, and stress concentrations at notches all require separate checks this calculator does not perform.
Assumptions and Limitations
- Applies to ductile materials (steels, aluminum, copper alloys) whose compressive and tensile yield strengths are approximately equal. For brittle materials such as cast iron, glass, or ceramics, use the Maximum Normal Stress (Rankine) or Mohr theory instead.
- Assumes static, monotonic loading at a defined temperature. Neither criterion predicts fatigue failure under cyclic loads.
- Assumes a 2D plane stress condition — appropriate for thin plates, pressure vessel free surfaces, and shaft surfaces, but not for thick components where σz is significant.
- Assumes isotropic, linear-elastic material behavior before yield, with no strain-rate or size effects.
- Input stresses must already include any stress concentration factors; the theory operates on the local stress state at the point of interest.
Practical Engineering Use
These criteria appear throughout mechanical design: shaft design under combined bending and torsion, pressure vessel and piping wall checks, welded and bolted joint evaluation, and as the pass/fail basis in FEA models where every element's Von Mises stress is screened against yield. Comparing both criteria on the same stress state is also good practice — agreement between them confirms a marginally safe design, while divergence (Tresca lower) reveals how much conservatism the stricter theory would demand.