Davis triaxiality factor (Davis & Connelly, 1959) was developed to correlate the available plastic ductility of a stressed point against the uniaxial tensile-test ductility. It uses the ratio of the first stress invariant (sum of principals = 3 × mean stress) to the von Mises equivalent stress.
Strain margin theory: a tensile-test elongation ε_true_ult is measured at TFD = 1. At any other stress state the available ductility scales by 2
(1-TFD) — a heuristic Davis fit to test data across a range of triaxialities. The margin check ensures the FEA point strain at the design factor of safety remains within the available (TFD-corrected) ductility envelope.
Wierzbicki triaxiality η = TFD / 3 = (mean stress) / (von Mises) is the more modern form used in damage / fracture models (Bao-Wierzbicki, Johnson-Cook, GTN). The two are direct multiples; the Davis form survives in legacy aerospace stress reports.
Lode parameter μ distinguishes between stress states with the same triaxiality but different shear character (uniaxial tension μ = -1, pure shear μ = 0, uniaxial compression μ = +1).
References:
- Davis, E.A. & Connelly, F.M. (1959), "Stress distribution and plastic deformation in rotating cylinders of strain-hardening material"
- Wierzbicki, T. & Bao, Y. (2005), Int. J. Mech. Sci. 47
- Bridgman, P.W. (1944), "Stress distribution at the neck of a tension specimen"
- NX Simcenter SOL 401 documentation (stress output post-processing)
Limitations: the Davis 2
(1-TFD) ductility correction is empirical and was originally fit for steels and aluminum alloys. It may overcorrect for materials with limited inherent ductility (CFRP, brittle alloys) or undercorrect for highly anisotropic materials. For fatigue / fracture, use a true damage model (Johnson-Cook, GTN) calibrated to the specific material. The TFD itself is sensitive to von Mises in the denominator — near pure hydrostatic stress (σ_VM → 0) it explodes to infinity, so use the dial classification rather than the raw number for nearly-hydrostatic states.