Civil engineering mechanics & structural foundations of Direct Shear Test.
Shear strength is the foundational mechanical property governing the stability of all geotechnical structures, including shallow footings, retaining walls, cut slopes, and earth dams. Unlike cohesive clays, dry cohesionless sands derive their shear resistance purely from inter-particle friction and interlocking dilatancy between mineral grains. Standardized under ASTM D3080, the direct shear test utilizes a horizontally split brass or stainless steel shear box (typically 60 x 60 mm square, 20 to 25 mm depth). A sand specimen reconstituted to target relative density is placed in the box between porous stones and perforated grid plates. A constant normal compressive stress σ_n is applied vertically via a counterbalanced lever arm or pneumatic piston. The lower half of the shear box is driven horizontally at a controlled slow displacement rate (0.2 to 1.0 mm/min) while the upper half is restrained against a calibrated proving ring or load cell that measures shear force T. As horizontal shear displacement proceeds, shear stress τ = T / A increases to a peak failure stress τ_f. According to the classical Mohr–Coulomb failure criterion, peak shear strength relates linearly to normal stress: τ_f = c + σ_n · tan(ϕ), where ϕ is the angle of internal friction and c is cohesion (c ≈ 0 for clean sand). In this laboratory, direct shear tests are performed across normal stresses from 50 to 250 kPa to evaluate the Mohr-Coulomb failure envelope and calibrate friction tangent tan(ϕ).
1. The Mohr–Coulomb Failure Criterion in Geomechanics
The shear strength of particulate soils is classically modeled by the Mohr–Coulomb failure criterion. On any plane within a stressed soil mass, shear stress $\tau$ at failure is limited by normal effective stress $\sigma'$ acting perpendicular to that plane:
\[ \tau_f = c' + \sigma' \tan \phi' \]where $c'$ is effective cohesion and $\phi'$ is the effective angle of internal friction. In dry sand, pore water pressure is zero ($u = 0$), so total normal stress equals effective normal stress ($\sigma_n = \sigma'$).
2. Direct Shear Box Kinematics & Stress State
In a standard square shear box of cross-sectional area $A$ ($60 \times 60\text{ mm} = 3600\text{ mm}^2 = 0.0036\text{ m}^2$), a vertical normal force $N$ applies nominal normal stress:
\[ \sigma_n = \frac{N}{A} \]When the horizontal drive mechanism displaces the lower box half relative to the upper half by shear displacement $\Delta x$, horizontal shear force $T$ is developed along the horizontal split plane. The average shear stress is:
\[ \tau = \frac{T}{A} \]As displacement advances, shear stress reaches a peak failure value $\tau_f$.
3. Granular Interlocking & Dilatancy Component
P. W. Rowe's stress-dilatancy theory establishes that the peak friction angle $\phi_{peak}$ of dense sand comprises two distinct components: inter-particle sliding friction $\phi_\mu$ and dilation angle $\psi$:
\[ \phi_{peak} = \phi_{cv} + 0.8 \psi \]where $\phi_{cv}$ is the constant-volume critical state friction angle, and $\psi = \arctan(-d\Delta z / d\Delta x)$ is the maximum rate of volumetric expansion (dilatancy).
4. Linear Failure Envelope Model
Across the experimental normal stress regime ($\sigma_n = 50\text{ to }250\text{ kPa}$), the peak failure envelope follows the linear equation:
\[ \tau_f = \tan(\phi) \cdot \sigma_n + c \]where $\tan(\phi) = 0.6250$ corresponds to an internal friction angle $\phi = \arctan(0.6250) = 32.00^\circ$, and $c = 5.00\text{ kPa}$ represents minor experimental apparent cohesion.
Viva voce questions & analytical theory
What is the physical meaning of the angle of internal friction (ϕ) in granular soils?
The friction angle ϕ represents the ratio of shear resistance to normal force at failure (tan ϕ = τ_f / σ_n); physically, it combines sliding friction between mineral surfaces and geometric interlocking dilatancy requiring grains to roll over one another.
How does relative density influence peak vs critical state (residual) friction angles in sand?
Dense sand exhibits high peak friction angle (38° to 45°) accompanied by volumetric dilation (expansion); at large shear strains, particles achieve a constant-volume state (critical state) where friction angle drops to the critical state value ϕ_cv (30° to 33°).
Why must the separation screws be turned to create a clearance gap between box halves before shearing?
Without a small clearance gap (0.5 mm), the brass box halves rub directly against each other, adding substantial metal-to-metal friction to measured shear force and falsely inflating apparent soil strength.
Why does clean, dry sand sometimes exhibit a small apparent cohesion intercept (c ≈ 5 kPa)?
Apparent cohesion in clean sand is an experimental artifact caused by minor shear box boundary friction, surface tension from ambient humidity, or linear extrapolation of a curved failure envelope at very low stresses.
What is the primary limitation of the direct shear test compared to triaxial testing?
The direct shear test forces failure along a predetermined horizontal plane rather than the weakest natural slip surface; stress distribution along the shear plane is non-uniform, and pore water pressures cannot be measured.
How is the angle of internal friction used to calculate Rankine's active earth pressure coefficient K_a?
Rankine's active earth pressure coefficient is K_a = (1 - sin ϕ) / (1 + sin ϕ) = tan^2(45° - ϕ/2); a higher friction angle lowers lateral earth pressure against retaining walls.
Experimental precautions and laboratory safety
- Always remove the vertical alignment pins after applying normal load and before turning on the horizontal drive motor.
- Ensure the shear gap clearance screws are retracted after opening the 0.5 mm gap so they do not contact the lower box during shearing.
- Maintain shearing rate slow enough (≤ 0.5 mm/min) to prevent dynamic inertial forces from distorting shear force readings.
- Reconstitute each test specimen with identical initial mass and height to ensure strictly equal initial relative density.
Theoretical assumptions & boundary condition limits
- The civil engineering physical models apply within the specified calibrated operational window (50.0 to 250.0 kPa).
- Assumes homogeneous, isotropic continuum material behavior without unmodeled micro-fracture anomalies.
- Continuous sensor readings reflect ordinary least-squares regression across successive observations.
- Extreme inelastic yielding or cyclic degradation requires non-linear constitutive continuum damage mechanics.