Soil Exploration

A sampler is judged by four geometric ratios (Hvorslev, IS 1892). Let $D_1$ = inside diameter of cutting edge, $D_2$ = inside diameter of sampling tube, $D_3$ = outside diameter of cutting edge, $D_4$ = outside diameter of sampling tube.

1. Area ratio

$$A_r = \frac{D_3^2 - D_1^2}{D_1^2}\times 100\%$$

It is the ratio of soil displaced by the sampler wall to the area of the sample. A high value means a thick wall, more displacement and more disturbance. For undisturbed samples $A_r \le 10\%$ (up to 20% is tolerated for stiff soils; thin-walled Shelby tube is about 10% or less).

2. Inside clearance

$$C_i = \frac{D_2 - D_1}{D_1}\times 100\%$$

The cutting edge is made slightly smaller than the tube so the sample does not grip the tube wall (reduces friction and drag on the sample) and can expand slightly. Recommended: 0.5 to 3%. Short samples in clay can use about 0.5-1.5%; too large a value lets the sample loosen.

3. Outside clearance

$$C_o = \frac{D_3 - D_4}{D_4}\times 100\%$$

The cutting edge is slightly larger than the outside of the tube so the tube moves through the hole with less friction during driving and withdrawal. Recommended: 0 to 2% (more than this increases area ratio and disturbance).

4. Recovery ratio

$$R = \frac{L}{H}$$

$L$ is the length of the sample recovered and $H$ the length of sampler penetration. It shows how much of the pushed length is actually recovered; $R$ close to 1 means little compression or loss. For an undisturbed sample $R$ should be 96-98% or more (about 0.96 to 1.0).

Also a sharp cutting edge (angle about 10-12 degrees, or less) and slow steady pushing are used.

Depth of exploration

The boring must reach the soil that will be significantly stressed by the structure.

• Stress criterion: go to the depth where the added vertical stress $\Delta\sigma$ falls to about 10% of the foundation pressure $q$ (or 20% of the effective overburden stress at that depth).
• Thumb rules (IS 1892): isolated footing: about 1.5 times the width $B$ below the foundation level; closely spaced footings/raft: about 1.5 times the width of the loaded area; piles: 1.5 times the pile group width below the pile tip, or at least 5 m into hard stratum.
• Sowers' rule: $D = 3S^{0.7}$ for light steel/narrow concrete buildings and $D = 6S^{0.7}$ for heavy steel/wide concrete buildings ($S$ = number of storeys, $D$ in m).
• Borings must go through any fill, peat or soft layer to firm stratum, and at least 3 m into rock (or continue until bedrock is verified).
• Check the groundwater level.

Lateral extent (number and spacing of boreholes)

• Minimum 3 boreholes (not in line) for a small site; at least one at each corner and at the centre for large buildings.
• Reduce spacing if the strata change quickly.

Example

For a 5-storey RCC building, 20 m x 30 m plan on uniform clayey soil, with $B$ about 2 m:

• Depth: $D = 3 \times 5^{0.7} \approx 9.3$ m (light/medium building); check with $1.5B$ to $1.5\times$ raft width, and use not less than about 9-10 m.
• Spacing about 15 m, giving a 3 x 2 grid of 6 boreholes; add one extra borehole where the soil is seen to vary, and one deep hole to check the rock/hard layer.

A site investigation report is the final record of the soil investigation. It must be clear enough for a designer who has never visited the site.

Contents

1. Introduction: project name, location, purpose, scope, date, and names of the client and the investigating agency.
2. Site description: topography, existing structures, drainage, access, nearby sources, geology, seismic zone, past land use.
3. Field work: number, location and depth of boreholes/pits, method of boring, sampling methods, field tests (SPT, vane, plate load, etc.).
4. Laboratory tests: grain size, Atterberg limits, moisture content, unit weight, shear strength, consolidation, chemical tests.
5. Subsoil profile: borehole logs, soil profile sections and the groundwater level, with soil classification (IS 1498/USCS).
6. Analysis and findings: design parameters ($c$, $\phi$, $\gamma$, $C_c$, $N$), safe bearing capacity and expected settlement.
7. Recommendations: type and depth of foundation, allowable bearing pressure, ground improvement, dewatering and construction precautions.
8. Appendices: site plan with borehole positions, bore logs, test results, photographs, and a reference list of codes used.

Things to consider while preparing it

• Use clear, brief language and standard symbols and units.
• Separate the facts (observed data) from the interpretation.
• Give the limits of the investigation and note the possible variation between boreholes.
• Present tables and drawings at a suitable scale and check them for consistency.
• Report the groundwater level and its seasonal variation.
• Make recommendations economical and practical for the construction methods.

Definition

The standard penetration test (SPT) is a dynamic in-situ test in which a standard split-spoon sampler is driven into the soil at the bottom of a borehole by a 63.5 kg hammer falling 750 mm. The N-value is the number of blows required for the last 300 mm of a 450 mm drive.

Procedure (IS 2131)

1. Advance a borehole (150 mm or less) to the test depth and clean the bottom.
2. Lower the split-spoon sampler (outer diameter 50.8 mm, inner diameter 35 mm) on rods to the bottom.
3. Drive it with the 63.5 kg hammer dropping freely 750 mm.
4. Count blows for each 150 mm. The first 150 mm is the seating drive and is ignored.
5. The blows for the next two 150 mm intervals added together give $N$. If $N$ > 50 blows for a drive less than 300 mm, record it as refusal with the penetration.
6. Withdraw the sampler, open it and take the disturbed sample.
7. Repeat at about every 1.5 m (or each change of strata).

Uses of the result

• Relative density of sands and consistency of clays.
• Angle of friction $\phi$ and unconfined strength in empirical correlations.
• Bearing capacity and settlement of footings in sand (Terzaghi-Peck, Meyerhof charts).
• Pile capacity, liquefaction assessment and soil identification.

Nature of sample

A disturbed (but representative) sample, not suitable for strength or consolidation tests, but suitable for visual description, grain size and Atterberg limits.

Corrections

1. Overburden correction: $N$ is higher at larger depth for the same density, so it is reduced to a standard 100 kPa effective stress:

$$N' = C_N N, \qquad C_N = 0.77\log_{10}\frac{2000}{\sigma_v'} \quad (\sigma_v' \text{ in kN/m}^2)$$

(Liao and Whitman: $C_N = \sqrt{100/\sigma_v'}$ is also used.) 2. Dilatancy (water table) correction: in saturated fine or silty sand, excess pore pressure raises the apparent $N$ above 15, so

$$N'' = 15 + \tfrac{1}{2}(N' - 15) \quad (N' > 15)$$

3. Other corrections to the energy ratio (hammer, rod length, borehole diameter, sampler liner) are used in $(N_1)_{60}$ in modern practice.

Factors affecting the quality of a sample

• Sampler design: area ratio, inside and outside clearances, cutting edge sharpness.
• Method of boring and advancing (water jets, disturbed bottom of hole, bottom heave).
• Method of sampling: pushing vs hammering, speed and rate of insertion.
• Type of soil: soft sensitive clays and loose sands are easily disturbed.
• Stress release when the sample is taken out, and possible swelling.
• Handling, transport and storage: vibration, moisture loss, rough handling, time delay, extrusion method.

Common field tests

1. Standard penetration test (SPT)
2. Cone penetration tests (static cone, CPT; dynamic cone, DCPT)
3. Vane shear test (for soft clays)
4. Plate load test
5. Pressuremeter test
6. Field permeability (pumping) tests
7. In-situ density tests (core cutter, sand replacement)
8. Geophysical methods such as seismic refraction and electrical resistivity

Corrections to SPT $N$ in sand

1. Overburden pressure correction: since $N$ increases with depth for the same relative density,

$$N' = C_N N, \qquad C_N = 0.77\log_{10}\frac{2000}{\sigma_v'}\ (\sigma_v' \text{ in kN/m}^2,\ \sigma_v' \ge 25)$$

Here $\sigma_v'$ is the effective vertical stress at the test level. The standard reference stress is about 100 kN/m$^2$ ($C_N = 1$). 2. Dilatancy (saturation) correction: applicable for fine or silty sand below the water table, when $N' > 15$:

$$N'' = 15 + \tfrac{1}{2}(N' - 15)$$

This allows for the negative pore pressure (dilation) in dense fine sand that stiffens it during driving, and for the positive pressures in loose sand; the formula corrects only above 15. 3. Energy, rod length, borehole and sampler corrections are applied in current practice to standardise the energy to 60%: $N_{60} = N\,\eta_H\eta_B\eta_S\eta_R/0.60$.

A borehole log is a graphical and written record of the soil encountered in one borehole against depth. It is prepared from field notes and laboratory results.

Information in the header

• Project name, location, client, borehole number and coordinates/reference level.
• Method of boring, diameter, date started and completed, driller/engineer.
• Groundwater level (date of observation).

Columns of the log

• Depth/RL scale and thickness of each layer.
• Graphic legend (hatch symbols) and soil description with IS classification (colour, consistency/density, grain size).
• Sample type (disturbed D, undisturbed U, SPT) and depths.
• SPT N-value (and optionally core recovery, RQD for rock).
• Water table and any seepage, loss of drilling water, cavity or boulders.
• Remarks such as refusal and end of hole.

Example (BH-1, RL 100.00 m, GWT 2.0 m)

Boring terminated at 12.0 m. The log is used to draw soil profiles and to choose foundation type and depth.

Given: $D_1 = 59$ mm (cutting edge inside), $D_3 = 63$ mm (cutting edge outside), $D_2 = 60$ mm (tube inside), $D_4 = 62$ mm (tube outside).

(i) Area ratio

$$A_r = \frac{D_3^2-D_1^2}{D_1^2}\times100 = \frac{63^2-59^2}{59^2}\times100 = \frac{488}{3481}\times100 = 14.02\%$$

(ii) Inside clearance

$$C_i = \frac{D_2-D_1}{D_1}\times100 = \frac{60-59}{59}\times100 = 1.69\%$$

(iii) Outside clearance

$$C_o = \frac{D_3-D_4}{D_4}\times100 = \frac{63-62}{62}\times100 = 1.61\%$$

Answer: $A_r = 14.02\%$, $C_i = 1.69\%$, $C_o = 1.61\%$. (The clearances are acceptable, but $A_r > 10\%$, so the sample would be slightly disturbed.)

Given: $D_1 = 69$ mm, $D_3 = 73$ mm (cutting edge inside/outside), $D_2 = 70$ mm, $D_4 = 72$ mm (tube inside/outside).

(i) Inside clearance

$$C_i = \frac{D_2-D_1}{D_1}\times100 = \frac{70-69}{69}\times100 = 1.45\%$$

(ii) Outside clearance

$$C_o = \frac{D_3-D_4}{D_4}\times100 = \frac{73-72}{72}\times100 = 1.39\%$$

(iii) Area ratio

$$A_r = \frac{D_3^2-D_1^2}{D_1^2}\times100 = \frac{73^2-69^2}{69^2}\times100 = \frac{568}{4761}\times100 = 11.93\%$$

Answer: $C_i = 1.45\%$, $C_o = 1.39\%$, $A_r = 11.93\%$. Both clearances are within limits; the area ratio is a little over the 10% limit for a good undisturbed sample.

Site investigation (soil exploration) is the process of finding the nature, extent and properties of the soil and rock and the groundwater at a site.

Importance

• To select a safe and economical foundation type and depth.
• To get design parameters for bearing capacity and settlement.
• To find problem soils (soft clay, peat, swelling, collapsible, liquefiable soil) and groundwater.
• To plan the construction method, dewatering, excavation support.
• To find construction material sources and the stability of slopes.
• To avoid failures, delay and cost increases from unforeseen conditions.

Stages

1. Reconnaissance (desk study and site visit): study maps, geology, aerial photos, past records, and walk over the site.
2. Preliminary exploration: a few borings/test pits and simple tests to learn the general strata, groundwater and the likely foundation type.
3. Detailed exploration: more boreholes, undisturbed sampling, field and laboratory tests to get the design parameters.
4. Supplementary exploration (during construction): extra tests where unexpected conditions appear.
5. Report preparation with recommendations.

Undisturbed samples keep the in-situ structure, density, water content and stress history of the soil. They are required because:

• Shear strength tests (unconfined, triaxial, direct shear) must be done on soil with natural structure to get true $c$ and $\phi$.
• Consolidation tests need them for settlement estimates of clays.
• Permeability, unit weight and sensitivity determination.
• Disturbed samples change the properties, so the results would be wrong and the design either unsafe or uneconomical. They are needed mainly for cohesive soils and for final design, while disturbed samples are enough for classification.

During sampling

• Clean the borehole bottom of loose soil and use a sampler with good geometry (small area ratio, correct clearances).
• Push the sampler slowly and smoothly (not hammered) in soft soils; do not rotate it.
• Keep the borehole full of water/drilling mud when below the water table to avoid heave.
• Leave the sampler for a short time and then twist to cut off the base before withdrawing.

Transportation

• Seal the ends of the tube with wax, cap them, and label (project, borehole, depth, top/bottom, date).
• Keep the tubes vertical (top up), pack in sawdust/cushion and avoid vibration, shock and heat.

Storage

• Store in a cool, humid room without freezing or drying, away from direct sun.
• Test as soon as possible (ideally within 2-3 weeks) and keep the extrusion in the direction of sampling.
• Protect disturbed samples in airtight bags or jars to preserve moisture content.

Preliminary exploration gives a general picture of the site. It provides:

• Types and sequence of soil strata and their approximate thickness and extent.
• Depth to bedrock or firm stratum and the groundwater level.
• General soil classification and rough strength (from SPT or hand tests).
• Presence of problem soils such as soft clay, fill, peat, or expansive soil.
• Approximate bearing capacity to select the possible type of foundation.
• Information needed to plan the detailed exploration (number, spacing and depth of borings).

In percussion drilling a heavy chisel bit is repeatedly raised and dropped to break the soil or rock, and the cuttings are removed by a bailer.

Merits

• Can penetrate hard soils, boulders, gravel and rock.
• Simple, rugged equipment, cheap and suitable in remote areas.
• Gives a good idea of groundwater and depth to rock; reaches large depths.

Demerits

• Slow in soft soils and gives disturbed, mixed samples.
• Soil below the bit is disturbed, so undisturbed sampling is difficult and strata boundaries are hard to identify.
• Not suitable for soft clays or loose sands below the water table (need casing).
• Cannot give proper records of thin layers.

Auger boring (hand auger or small power auger) is the best method for road sub-grade investigation.

• Sub-grade soil is needed only to a shallow depth (about 1-3 m below formation level), where augers work well.
• It is cheap, fast and light, so many holes at close spacing (about 250-500 m or less) can be made along the alignment.
• It gives disturbed but representative samples that are enough for classification, moisture content, Atterberg limits, compaction (OMC, MDD) and CBR tests.
• It needs little equipment and is easily taken to hilly road sites. Test pits can be used additionally for in-situ density and CBR checks.

The pressuremeter (Menard) test inflates a cylindrical probe inside a borehole and measures the radial pressure against the volume change. It is distinctive because:

• It loads the soil horizontally and in a large volume, giving a true in-situ stress-strain curve (not just a single index like SPT $N$).
• It gives the modulus ($E_p$), limit pressure ($p_l$) and yield (creep) pressure directly, which are used in a design method for bearing capacity and settlement (Menard's method).
• It can be used at any depth, in all soils from soft clay to weak rock, with little disturbance when the probe is placed in a pre-bored or self-boring hole.
• The test is done under controlled stress/strain increments, so the result is rational and repeatable, unlike the dynamic, empirical tests.

Site nature: Kathmandu valley is a lacustrine basin filled with soft to stiff lake clays and silts, fine micaceous sands, thin peat/black clay and gravelly layers, with high groundwater table, low strength at shallow depth and high liquefaction/amplification risk.

Recommended drilling

• Rotary (wash) boring with casing and bentonite mud to the required depth (15-30 m or more), because soft clay and sand below water table would collapse in simple augering; percussion is not recommended for disturbance.
• Use a thin-walled piston (Shelby) sampler in clay/silt for undisturbed samples and split-spoon sampling in sands.

Suitable field tests

Laboratory tests (consolidation, triaxial/UCS, Atterberg limits) on the undisturbed samples complement these. CPTu and SPT give the greatest amount of directly usable design data for bearing capacity, settlement of soft clay, pile capacity and liquefaction.

For a bridge foundation the investigation must give the information needed for the pier, abutment and approach design. It seeks:

1. Subsoil profile at every pier and abutment: boreholes at each foundation location, with soil type, thickness, density/consistency, and depth to rock or hard stratum.
2. Engineering properties: unit weight, $c$, $\phi$, SPT $N$, compressibility and bearing capacity (and rock RQD, strength, core recovery for rock).
3. Groundwater data: level, fluctuation, artesian pressure and chemical content (sulphates, chlorides) for durability.
4. River data: high flood level, low water level, discharge and velocity, bed material and the scour depth (decides foundation level); type of bank, river training and possible shifting.
5. Deep foundation parameters: pile or well capacity, skin friction, end bearing, negative skin friction, and presence of boulders and obstructions that could stop the sinking of wells/piles.
6. Seismic and geology: faults, liquefiable sand layers, seismic zone, slope stability of abutments and approach embankments.
7. Construction aspects: cofferdam or caisson needs, dewatering, access, availability of local materials, and effects on adjacent structures. The depth of boring is at least 1.5 times the width of the foundation below the foundation level (and 3 m or more into rock).

Method: Terzaghi-Peck corrections. First the overburden correction $N' = C_N N$ with $C_N = 0.77\log_{10}(2000/\sigma_v')$, then the dilatancy correction $N'' = 15 + \tfrac12(N'-15)$ for $N' > 15$.

Observed N: the first 150 mm is the seating drive, so

$$N = 21 + 22 = 43$$

Overburden stress: no water table is given, so

$$\sigma_v' = \gamma z = 17 \times 20 = 340\ \text{kN/m}^2$$

Overburden correction:

$$\begin{aligned} C_N &= 0.77\log_{10}\frac{2000}{340} = 0.593 \\ N' &= 0.593\times 43 = 25.48 \end{aligned}$$

Dilatancy correction ($N' > 15$):

$$N'' = 15 + \tfrac12(25.48-15) = 20.24$$

Answer: corrected N-value $\approx 20.2$, say 20. (If the dilatancy correction is applied first, $15+0.5(43-15)=29$ and $29\times0.593=17.2$; the order recommended by Terzaghi-Peck, overburden then dilatancy, is used here.)

Method: overburden correction $N' = C_N N$, $C_N = 0.77\log_{10}(2000/\sigma_v')$, then dilatancy correction (saturated fine sand) $N'' = 15+\tfrac12(N'-15)$ for $N'>15$.

Observed N: ignore the first 150 mm.

$$N = 23 + 17 = 40$$

Effective overburden (fully submerged, $\gamma_w = 9.81\ \text{kN/m}^3$):

$$\sigma_v' = (19 - 9.81)\times 6 = 55.14\ \text{kN/m}^2$$

Overburden correction:

$$\begin{aligned} C_N &= 0.77\log_{10}\frac{2000}{55.14} = 1.201 \\ N' &= 1.201\times 40 = 48.03 \end{aligned}$$

Dilatancy correction ($N'>15$, fine saturated sand):

$$N'' = 15 + \tfrac12(48.03-15) = 31.52$$

Answer: corrected N-value $\approx 31.5$ (about 32).

Formulae ($D_1$ = inside dia. of cutting edge, $D_2$ = inside dia. of tube, $D_3$ = outside dia. of cutting edge, $D_4$ = outside dia. of tube):

$$C_i = \frac{D_2-D_1}{D_1}\times100,\quad C_o = \frac{D_3-D_4}{D_4}\times100,\quad A_r = \frac{D_3^2-D_1^2}{D_1^2}\times100$$

Given: $D_1 = 68$ mm, $D_2 = 70$ mm, $D_3 = 74$ mm, $D_4 = 72$ mm.

$$\begin{aligned} C_i &= \frac{70-68}{68}\times100 = 2.94\% \\ C_o &= \frac{74-72}{72}\times100 = 2.78\% \\ A_r &= \frac{74^2-68^2}{68^2}\times100 = \frac{852}{4624}\times100 = 18.43\% \end{aligned}$$

Classification: limits for an undisturbed sample are $A_r \le 10\%$, $C_i = 0.5$-$3\%$, $C_o = 0$-$2\%$.

• $C_i = 2.94\%$ is within 0.5-3%.
• $C_o = 2.78\%$ exceeds 2%.
• $A_r = 18.43\%$ is much more than 10%.

Answer: $C_i = 2.94\%$, $C_o = 2.78\%$, $A_r = 18.43\%$. Because the area ratio and outside clearance are too large, the sample is disturbed.

Take the 40 mm as the inside diameter $D_1$ and the 42 mm as the outside diameter $D_3$ of the cutting edge (wall 1 mm thick).

$$A_r = \frac{D_3^2-D_1^2}{D_1^2}\times100 = \frac{42^2-40^2}{40^2}\times100 = \frac{164}{1600}\times100 = 10.25\%$$

The limit for an undisturbed sample is $A_r \le 10\%$. Since $A_r = 10.25\%$ is slightly greater than 10%, the sample is classed as disturbed (marginal; it may be accepted as nearly undisturbed in soft or stiff clay if the clearances are right, but by the criterion it is not).

External radius 30 mm gives outside diameter $D_3 = 60$ mm. Wall thickness 2.25 mm gives inside diameter

$$D_1 = 60 - 2\times2.25 = 55.5\ \text{mm}$$ $$A_r = \frac{D_3^2-D_1^2}{D_1^2}\times100 = \frac{60^2-55.5^2}{55.5^2}\times100 = \frac{519.75}{3080.25}\times100 = 16.87\%$$

Answer: $A_r = 16.87\%$.

Recommendation: not recommended for undisturbed sampling. The area ratio is more than 10% (limit 10%, or 20% at most for stiff soil), so the thick wall displaces too much soil and disturbs the sample. It is fit only for disturbed or stiff-soil sampling.