Module 2 · 13 min read · Free chapter
Bearing capacity of shallow foundations
Terzaghi's strip footing equation is qu = c Nc + q Nq + 0.5 γ B Nγ, with 0.4 and 0.3 for the γ term in square and circular footings. On clay with φ = 0, qu = 5.7c + γDf. A water table rising to the surface can halve the capacity in sand, and the allowable pressure must also pass the settlement check.
Start from zero: what bearing capacity means
Push your thumb into soft clay and it sinks in. Push it into firm ground and it stops. But push hard enough on any soil and, at some point, the soil under your thumb gives way and squeezes out to the sides. The pressure at which that happens is the soil's bearing capacity.
A footing is just a big thumb. It must press on the soil without making it fail in shear, and without making it sink so much that the building cracks. These are two separate checks:
- Safety against shear failure, worked out from bearing capacity theory with a factor of safety.
- Settlement within limits, worked out in the settlement chapter.
The pressure that passes both checks is the allowable bearing pressure. If the shear check allows 150 kPa but a settlement limit of 25 mm allows only 100 kPa, the design must use 100 kPa.
Why it matters. Bearing capacity sets the size of every footing in every building. Too small and the footing fails or settles; too large and money is poured into the ground.
The family of bearing capacity terms
Exams love these terms because they look alike. Keep them straight with one idea: the soil at founding level already carried the weight of the soil above it, γDf, before the footing was built. Only the extra pressure is new to it.
| Term | Meaning |
|---|---|
| Ultimate bearing capacity, qu | gross pressure at the base at which the soil fails in shear |
| Net ultimate bearing capacity, qnu | qu − γDf, the extra pressure the soil can take beyond what it already carried |
| Net safe bearing capacity, qns | qnu / F, where F is usually 2.5 to 3 for shallow foundations |
| Safe bearing capacity, qs | qns + γDf, the gross pressure that can be applied safely |
| Allowable bearing pressure, qa | the lower of the safe pressure for shear and the pressure for permissible settlement |
The factor of safety is applied to the net value, not the gross, because the overburden γDf is known exactly and needs no margin.
A footing counts as shallow when its depth is small compared with its width, so that the soil above the base acts only as a surcharge and its shear strength is ignored. Most building footings are within 1 to 2 m of the surface.
How soil fails under a footing
Load a footing step by step and watch the soil, and you see one of three patterns.
In general shear, typical of dense sand and stiff clay, a wedge under the footing pushes the soil on both sides outwards along well-defined slip surfaces that reach the ground. The soil next to the footing heaves up, and the load-settlement curve shows a clear peak. Failure is sudden, and the footing may tilt.
In local shear, typical of loose sand and soft clay, the slip surfaces start but do not reach the ground. The soil compresses a lot first, there is only slight heave, and the load-settlement curve keeps rising without a clear peak. The "failure load" is taken where settlement starts to grow fast.
In punching shear, found in very loose soil or under deep footings, the footing sinks straight down, shearing the soil at its edges, with almost no heave. The soil below simply compresses.
Why it matters. The theory below was built for general shear. Loose soils need reduced strength values, and very loose soils are usually governed by settlement long before shear.
Terzaghi's bearing capacity equation
In 1943, Terzaghi gave the first complete theory for a strip footing of width B at depth Df. He assumed a rough, rigid base, a shallow footing, homogeneous soil to a great depth, and the water table far below. He ignored the shear strength of the soil above the base and treated it as a surcharge q = γDf.
At failure, three zones form below the footing:
- Zone I, a triangular elastic wedge stuck to the rough base, which moves down with the footing as one piece.
- Zones II, radial shear zones on each side, bounded by curved slip lines.
- Zones III, Rankine passive zones, triangles pushed up and out, with slip planes at 45° − φ/2 to the horizontal.
Balancing the forces on the wedge, and splitting the passive resistance into parts from cohesion, surcharge and soil weight, gives:
qu = c Nc + q Nq + 0.5 γ B Nγ
The three terms are the contributions of cohesion, surcharge and the weight of soil below the base. Nc, Nq and Nγ are the bearing capacity factors, which depend only on φ. Commonly used forms are:
Nq = e^(π tan φ) tan²(45° + φ/2), Nc = (Nq − 1) cot φ, Nγ = (Nq − 1) tan(1.4φ)
The factors grow very fast with φ. For φ = 32°, Terzaghi's tables give Nq ≈ 28.5 and Nγ ≈ 26.9. For a saturated clay loaded quickly, φ = 0, and Terzaghi's values are Nc = 5.7, Nq = 1, Nγ = 0, so:
qu = 5.7c + γDf (strip footing on clay)
Terzaghi extended the result to other shapes with simple multipliers:
- square: qu = 1.3 c Nc + q Nq + 0.4 γ B Nγ
- circular (B = diameter): qu = 1.3 c Nc + q Nq + 0.3 γ B Nγ
An easy way to remember the last term: 0.5 for strip, 0.4 for square, 0.3 for circle.
For local shear, he kept the same equation but reduced the strength: c′ = (2/3)c and tan φ′ = (2/3) tan φ, then read the factors for φ′.
A worked case: a strip footing 1 m wide at 0.6 m depth in clay with unconfined strength 140 kPa and γ = 21.8 kN/m³. Then c = 140/2 = 70 kPa and qu = 5.7 × 70 + 21.8 × 0.6 ≈ 399 + 13 = 412 kPa. With F = 3 on the full value, the allowable load is about 412 × 1/3 ≈ 137 kN per metre of wall.
Why it matters. The three-term equation tells a designer where capacity comes from. On sand, depth matters most, because the q Nq term grows with Df. On clay, width hardly helps, because Nγ = 0; only cohesion and depth count.
Effect of the water table
Terzaghi assumed the water table was far below. In real sites it rises in the monsoon, and when it enters the zone that resists failure, the soil there weighs less in effective terms, so it resists less. The rule is simple: wherever a γ appears in the equation, use the effective unit weight of the soil it refers to.
The failure zones reach down to about a depth B below the base. So:
- Water table at depth B or more below the base: no effect; use the normal equation.
- Between the base and depth B: only the 0.5 γ B Nγ term changes. Its γ is interpolated, γ′ + (d/B)(γ − γ′), where d is the depth of the water table below the base.
- At the base: q Nq uses the full γ, but the 0.5 γ B Nγ term uses γ′.
- Between the ground surface and the base: the surcharge becomes q = γ D1 + γ′ D2, where D1 is the depth above the water table and D2 below it, and the γ term uses γ′.
- At the ground surface: both terms use γ′.
A worked case shows the size of the effect. A strip footing 1.8 m wide at 1.5 m depth sits in sand with φ = 32°, γ = 17 kN/m³ above water and γsat = 19 kN/m³, so γ′ ≈ 9 kN/m³:
| Water table | qu (kPa) |
|---|---|
| deep (more than 3.3 m below ground) | about 1138 |
| 2.5 m below ground (1 m below the base) | about 1052 |
| at the base | about 944 |
| 1 m below ground | about 830 |
| at the ground surface | about 603 |
A rise of the water table from deep down to the surface cuts the capacity nearly in half. Design codes therefore use the highest likely water level, not the level seen on the day of the boring.
Why it matters. In the flood plains of Bihar and the North East, the water table can rise to near the surface every monsoon. A footing designed on summer borings can lose almost half its margin each year.
Beyond Terzaghi: general equation and field tests
Terzaghi's equation covers strip, square and circular footings with vertical loads. Later work by Meyerhof, Hansen and Vesic made it general. Each term is multiplied by factors for shape (s), depth (d) and load inclination (i):
qu = c Nc sc dc ic + q Nq sq dq iq + 0.5 γ B Nγ sγ dγ iγ
The Indian code for shallow foundations uses an equation of this kind, with a water table factor on the γ term. Depth factors give credit for the shear strength of the soil above the base, which Terzaghi ignored. Inclination factors reduce capacity when the load leans, as under wind or earth pressure.
For clays, Skempton gave a simple value of Nc that rises with depth and with shape: Nc = 5(1 + 0.2 Df/B)(1 + 0.2 B/L), up to a limit of about 7.5. A surface strip on clay has Nc = 5, close to the theoretical 5.14.
When the load is eccentric, with e from the centre, the footing is treated as if it had a smaller effective width B′ = B − 2e centred under the load, and the capacity is worked out for B′.
Two field methods give bearing capacity directly:
- Plate load test. A steel plate of 300 to 750 mm is placed at founding level in a pit at least five times its width, and loaded in steps. Each step is held until settlement stops. In clay, the ultimate bearing pressure is about the same for the plate and the full footing. In sand, it grows with width, and the settlement of a footing of width B is estimated from the plate settlement Sp: Sf = Sp [B (Bp + 0.3) / (Bp (B + 0.3))]², with widths in metres. The test only tests soil to about twice the plate width, so a weak layer deeper down can be missed.
- SPT and CPT. Charts and formulas link N or qc to the pressure that gives 25 mm settlement in sand, which usually governs design there.
Try it yourself. On ujjwalinfra, the guided lab CBR test loads a plunger in steps and reads a dial gauge at each step, the same routine a plate load test follows in the field on a larger scale.
Descriptive lens: how this topic is asked in written papers
Written papers test bearing capacity in four main ways:
- Define and distinguish the bearing capacity terms and failure modes.
- Explain Terzaghi's theory, its assumptions and failure zones.
- Compute qu, qns and qs for a footing, with shape and water table effects.
- Describe the plate load test and its limitations, or compare it with SPT-based design.
Practice questions (our own, not previous-year questions).
- Distinguish between general, local and punching shear failure with sketches of the load-settlement curves. (10 marks)
- State Terzaghi's assumptions and write his bearing capacity equation for strip, square and circular footings. What changes for a saturated clay under quick loading? (10 marks)
- A 2 m square footing is placed at 1.5 m depth in sand with γ = 18 kN/m³, γsat = 20 kN/m³ and φ = 30° (Nq = 22.5, Nγ = 19.7). Find the net safe bearing capacity with F = 3 when the water table is deep, and again when it rises to the base. Comment on the result. (15 marks)
Answer plan for question 3.
- Write the square form: qu = q Nq + 0.4 γ B Nγ, since c = 0.
- Deep water table: q = 18 × 1.5 = 27 kPa; qu = 27 × 22.5 + 0.4 × 18 × 2 × 19.7 ≈ 608 + 284 = 892 kPa.
- Net: qnu = 892 − 27 = 865 kPa; qns = 865/3 ≈ 288 kPa.
- Water table at base: γ′ ≈ 10.2 kN/m³ in the γ term: qu ≈ 608 + 0.4 × 10.2 × 2 × 19.7 ≈ 608 + 161 = 769 kPa; qns ≈ (769 − 27)/3 ≈ 247 kPa.
- Comment: about 14% drop; design for the highest seasonal water table and also check settlement, which usually governs for footings on sand.
Quick revision
- Allowable bearing pressure = lower of the shear-safe pressure and the settlement-limited pressure.
- qnu = qu − γDf; qns = qnu/F with F = 2.5 to 3; qs = qns + γDf.
- Dense sand and stiff clay: general shear; loose sand and soft clay: local shear; very loose or deep: punching.
- Terzaghi strip: qu = c Nc + q Nq + 0.5 γ B Nγ.
- Square: 1.3 c Nc + q Nq + 0.4 γ B Nγ; circle: 1.3 c Nc + q Nq + 0.3 γ B Nγ.
- φ = 0: Nc = 5.7, Nq = 1, Nγ = 0; qu = 5.7c + γDf.
- Local shear: c′ = 2c/3, tan φ′ = (2/3) tan φ.
- Water table deeper than B below the base: no effect; at the surface: capacity roughly halves in sand.
- Eccentric load: B′ = B − 2e.