16-Civ-A4 Geotechnical Materials and Analysis · May 2015
Question 5 of 6: Rankine active earth pressure on a layered c–φ backfill
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper format: National Examination 98-Civ-A4 Geotechnical Materials and Analysis — May 2015. Closed book, 3 hours, 100 marks. Six questions, answer all. Newmark and rectangular m–n influence charts and a formula sheet are provided at the back of the paper.
Reference texts: Das & Sobhan, Principles of Geotechnical Engineering (Cengage); Craig, Soil Mechanics; Holtz, Kovacs & Sheahan, An Introduction to Geotechnical Engineering. Canadian practice: effective-stress, consolidation and Rankine methods as summarised in the Canadian Foundation Engineering Manual (CFEM, 4th ed.).
Question 5: Rankine active earth pressure on a layered c–φ backfill (20 marks)
Given. Two cohesive-frictional backfill layers, a uniform surcharge and a water table 1 m into the lower layer.
Given data
Item
Layer 1 (0–3 m)
Layer 2 (3–5 m)
Unit weight $\gamma$ (moist)
18.0 kN/m³
18.5 kN/m³
Saturated unit weight
—
20.0 kN/m³
Friction angle $\phi'$
30°
25°
Cohesion $c'$
20 kPa
25 kPa
Surcharge $q$
20 kPa (over the full backfill surface)
Water table
depth 4 m; $\gamma_w = 9.81$ kN/m³
Find. (i) the active earth-pressure distribution with depth, and (ii) the active thrust per metre of wall once the tension zone has cracked.
Approach. Build the vertical effective-stress profile (surcharge + soil, buoyant below the water table), apply the Rankine active relation $\sigma_a' = K_a\sigma_v' - 2c'\sqrt{K_a}$ layer by layer, add hydrostatic water pressure below the water table, then integrate only the positive pressure once the tensile (cracked) zone is discounted.
Figure 4 — Rankine active pressure envelope. The upper backfill is in tension (cracked, discounted); positive earth pressure develops in the lower layer, with hydrostatic water pressure added below the water table.
Active coefficients.
$$K_{a1}=\tan^2\!\left(45^\circ-\tfrac{30^\circ}{2}\right)=0.333,\qquad K_{a2}=\tan^2\!\left(45^\circ-\tfrac{25^\circ}{2}\right)=0.406.$$
Vertical effective stress profile. With the surcharge included and buoyant weight below the water table:
$$\sigma_v'(0)=20,\quad \sigma_v'(3)=20+18(3)=74,\quad \sigma_v'(4)=74+18.5(1)=92.5,$$
$$\sigma_v'(5)=92.5+(20-9.81)(1)=102.7\ \text{kPa}.$$
Active earth pressure, upper layer. With $2c_1'\sqrt{K_{a1}} = 2(20)(0.577)=23.1\text{ kPa}$:
$$\sigma_a'(0)=0.333(20)-23.1=-16.4\ \text{kPa (tension)},\qquad \sigma_a'(3^-)=0.333(74)-23.1=+1.6\ \text{kPa}.$$
The pressure is negative down to the tension-crack depth where $\sigma_v' = 2c_1'/\sqrt{K_{a1}}=69.3$ kPa:
$$z_{c1}=\frac{69.3-20}{18}=\boxed{2.74\ \text{m}}.$$
Active earth pressure, lower layer. With $2c_2'\sqrt{K_{a2}} = 2(25)(0.637)=31.9\text{ kPa}$:
$$\sigma_a'(3^+)=0.406(74)-31.9=-1.8,\quad \sigma_a'(4)=0.406(92.5)-31.9=+5.7,\quad \sigma_a'(5)=0.406(102.7)-31.9=+9.8\ \text{kPa}.$$
The lower layer is in tension until $z_{c2}=3.0+\dfrac{2c_2'/\sqrt{K_{a2}}-74}{18.5}=3.24\ \text{m}$.
Water pressure. Hydrostatic below the water table:
$$u(4)=0,\qquad u(5)=9.81(1)=9.8\ \text{kPa}.$$
Active force after cracking (part ii). Once the tensile zones crack they carry no load, so the thrust is the area of the positive pressure blocks plus the water thrust:
$$P_{a1}=\tfrac12(1.6)(3-2.74)=0.2,\quad P_{a2}=\tfrac12(5.7)(4-3.24)+\tfrac12(5.7+9.8)(1)=9.9,$$
$$P_w=\tfrac12(9.8)(1)=4.9\ \text{kN/m}.$$
$$\boxed{P_a = P_{a1}+P_{a2}+P_w \approx 15.0\ \text{kN/m}.}$$
Check: the cohesion given for the backfill (20–25 kPa) is large relative to the 5 m wall height, so most of the retained soil is in tension and contributes nothing once cracked — the active thrust is modest and is dominated by the water pressure. This is the intended teaching point (cohesive backfill sheds most of its active load through tension cracks), not an error. It assumes the open cracks stay dry; if a crack fills with water, an additional hydrostatic thrust over the cracked depth must be added, which would substantially increase $P_a$. Because the two layers do not share the same $c'$, the diagram legitimately shows a small positive sliver at the base of layer 1 followed by renewed tension at the top of layer 2.