The Cone Penetration Test (CPT) is one of the most widely used in-situ geotechnical tests in the world, valued for its continuous depth profile, high repeatability, and speed of execution. Unlike the Standard Penetration Test, which measures resistance at discrete intervals and requires stopping and re-seating at each test depth, the CPT pushes a instrumented cone into the ground at a controlled rate and records data continuously — producing a detailed picture of soil stratigraphy and behavior from the ground surface to the termination depth of the test.

This guide covers the complete CPT workflow for practicing geotechnical engineers: how the test works, what it measures, how to classify soils using the Robertson Soil Behavior Type chart, how to estimate engineering soil parameters from CPT data, and how CPT results are used to calculate bearing capacity and settlement for shallow foundation design.

How the CPT works — equipment and procedure #

The CPT is performed by pushing a standardised cone penetrometer into the ground at a controlled rate using hydraulic rams mounted on a heavily ballasted truck or anchored rig. The cone is pushed continuously — there is no hammer, no borehole, and no interruption for sampling. Measurements are recorded electronically at the cone tip throughout the push, typically at depth intervals of 1–2 cm, producing a near-continuous depth profile of subsurface resistance.

CPT cone geometry and standard specifications #

The standard cone dimensions are defined by ASTM D3441 (mechanical CPT) and ASTM D5778 / ISO 22476-1 (electronic CPT, the modern standard):

Parameter	Standard specification
Cone tip angle	60°
Cone tip area (A_c)	10 cm² (standard) or 15 cm² (less common)
Friction sleeve area (A_s)	150 cm² (for 10 cm² cone)
Push rate	2 cm/s (±0.5 cm/s)
Rod diameter	35.7 mm (for 10 cm² cone)
Applicable standard	ASTM D5778 / ISO 22476-1

CPT procedure #

Set up and level the rig at the test location. The rig must be adequately ballasted or anchored to resist the reaction force during pushing — typically 10–20 tonnes for onshore investigations.
Attach the cone to the push rods and position at the ground surface.
Push continuously at 2 cm/s using the hydraulic ram system. The electronic sensors in the cone transmit data to the surface data acquisition unit in real time.
Add push rods as the cone advances — typically in 1 m increments. Brief pauses during rod addition are noted in the data but do not affect interpretation.
Terminate the test when the target depth is reached, when the cone encounters refusal (thrust exceeds equipment capacity), or when an obstruction is encountered.
Extract the cone and rods and move to the next test location.

A complete CPT to 20 m depth typically takes 30–60 minutes including set-up and extraction — significantly faster than drilling and SPT testing to the same depth.

Different methods of deploying CPT CPTU — Different methods of deploying CPT

What CPT measures — q_c, f_s, R_f #

The electronic CPT cone contains strain gauges that measure two primary resistances continuously throughout the push:

Parameter	Symbol	Definition	Typical units	Typical range
Cone tip resistance	q_c	Force on the cone tip divided by the cone tip area (10 cm²). The primary measure of soil resistance — analogous to the SPT N-value but continuous.	MPa	0.1 (soft clay) to 50+ MPa (dense sand, gravel)
Sleeve friction	f_s	Friction force on the 150 cm² sleeve directly above the cone tip, divided by the sleeve area. Measures the skin friction between the soil and the steel sleeve.	kPa	1–500 kPa
Friction ratio	R_f	R_f = (f_s / q_c) × 100%. A derived parameter that reflects the relative soil type — low R_f indicates clean sand; high R_f indicates clay.	%	0.1% (gravel) to 8%+ (soft clay)

The ratio R_f is the foundation of soil classification from CPT data. Sands and gravels have low friction ratios because the cone tip resistance is high relative to sleeve friction. Clays and silts have high friction ratios because sleeve friction is proportionally larger relative to the lower tip resistance.

CPTu — the piezocone test and pore pressure measurement #

The CPTu (piezocone test) is an instrumented variant of the standard CPT that adds a pore water pressure sensor at a defined location on the cone — typically at the u₂ position, immediately behind the cone shoulder. The additional pore pressure measurement significantly enhances the interpretive value of the test, particularly in fine-grained soils.

Why pore pressure measurement matters #

When a cone is pushed through saturated fine-grained soil, the rapid penetration generates excess pore water pressure that cannot dissipate instantaneously. This excess pore pressure partially supports the applied stress, meaning the raw q_c value underestimates the true resistance. The CPTu corrects for this by computing the corrected total cone resistance:

q_t = q_c + u₂(1 − a)

where a is the cone area ratio (typically 0.70–0.85), determined by laboratory calibration of the specific cone used. In stiff or dense soils where q_c is large, this correction is negligible. In soft clays where u₂ can be very large relative to q_c, the correction is essential.

CPTu additional parameters #

Parameter	Symbol	Use in interpretation
Pore pressure	u₂	Correction of q_c to q_t; identification of clay layers; dissipation testing for permeability
Pore pressure ratio	B_q = (u₂ − u₀) / (q_t − σ_v0)	Third axis on the Robertson (1990) classification chart; distinguishes sensitive clays, silts, and structured soils
Corrected tip resistance	q_t	Used in all modern CPT correlations in place of q_c for fine-grained soils

CPTu dissipation tests — where the cone is held stationary and the decay of excess pore pressure is monitored over time — can be used to estimate the coefficient of consolidation (c_h) and hydraulic conductivity of fine-grained soils, providing data that is otherwise expensive to obtain from laboratory consolidation testing.

CPT vs SPT — differences and when to use each #

The CPT and SPT are the two dominant in-situ tests in geotechnical practice, but they have fundamentally different strengths. Most site investigation programmes benefit from understanding when each is appropriate — and many quality investigations use both in combination.

Feature	CPT	SPT
Depth profile	Continuous (data every 1–2 cm)	Discrete (every 1.0–1.5 m)
Soil sample	None	Yes — disturbed split-spoon sample
Repeatability	High — automated push, no operator variability	Moderate — influenced by hammer energy and driller practice
Speed	Fast — 20 m in under 1 hour	Slower — borehole advance between tests
Thin layer detection	Excellent — resolves layers as thin as 50–100 mm	Poor — 1.5 m test spacing can miss thin layers
Gravels and cobbles	Refusal likely — CPT cannot penetrate coarse gravel or cobbles	Can penetrate denser material with appropriate drilling method
Pore pressure	Measured directly (CPTu)	Not measured
Equipment availability	Less universal — specialist rig required	Near-universal — available globally
Cost per metre	Lower per metre of investigation depth	Higher per metre (borehole advance time included)

When to choose CPT #

Sites with soft to medium soils (clays, silts, loose to medium-dense sands) where the cone will not encounter refusal
When a continuous stratigraphic profile is important — for detecting thin weak or permeable layers
When speed and cost of investigation are a priority
Offshore and nearshore investigations where boreholes are impractical
Liquefaction screening — CPT provides a more reliable and continuous dataset for liquefaction analysis than SPT

When to choose SPT (or combine both) #

Sites with gravels, cobbles, or fill with rubble where CPT refusal is likely
When soil samples are required for visual classification, laboratory testing, or contamination assessment
In regions where CPT equipment is unavailable or uneconomical for the investigation scale
As part of a combined programme — CPT for rapid spatial profiling, SPT boreholes at select locations for sample recovery and visual verification

For a detailed head-to-head comparison, see CPT vs SPT — which test is right for your site investigation?

Soil Behavior Type (SBT) classification — Robertson chart #

Because the CPT retrieves no physical soil sample, soil identification from CPT data relies on classification using measured resistance parameters. The universally adopted system is the Soil Behavior Type (SBT) framework developed by Robertson (1990), updated Robertson (2016).

The term “Soil Behavior Type” is deliberate — the CPT classifies soils by how they behave mechanically during penetration, not by their literal grain size or mineralogy. An organic silt and a soft clay may have similar grain sizes but very different CPT responses; conversely, a heavily overconsolidated clay may respond similarly to a dense sand in terms of tip resistance. SBT is a behavior-based classification that must be cross-referenced against borehole samples where material type is critical to design.

The Robertson (1990) SBT chart #

The Robertson SBT chart plots the normalised cone resistance (Q_tn) on the vertical axis against the friction ratio (F_r) on the horizontal axis. Both axes are on logarithmic scales. The chart is divided into nine SBT zones:

SBT zone	Soil behavior type	Typical Q_tn	Typical F_r (%)
1	Sensitive fine-grained soil	<12	<1
2	Organic soils — peats	<12	>4
3	Clay — silty clay	<12 to 70	3–8
4	Silt mixture — clayey silt	12–70	2–4
5	Sand mixture — silty sand	12–70	1–2
6	Sand — clean to silty	70–350	0.4–1
7	Gravelly sand to dense sand	>350	<0.4
8	Very dense sand to gravelly sand	>350	0.4–1
9	Very stiff fine-grained soil (heavily overconsolidated or cemented)	Varies	>4

Normalised parameters — Q_tn and F_r #

Raw q_c and f_s values increase with depth simply because overburden pressure increases. Plotting raw values on the SBT chart would shift points to higher Q at depth, biasing the classification. Normalisation removes this depth dependency:

Q_tn = [(q_t − σ_v0) / P_a] × (P_a / σ’_v0)ⁿ

F_r = [f_s / (q_t − σ_v0)] × 100%

where σ_v0 is total vertical stress, σ’_v0 is effective vertical stress, P_a is atmospheric pressure (100 kPa), and n is a stress exponent that varies between 0.5 (sand) and 1.0 (clay) depending on SBT zone. The Robertson (1990) method uses an iterative approach where n is updated based on the SBT zone identified from the previous iteration, converging to a consistent classification.

Soil Behavior Type Index I_c #

The SBT Index I_c is a single numerical value that describes the position of a data point on the Robertson chart, allowing automated depth profile classification without manual chart plotting:

I_c = [(3.47 − log Q_tn)² + (log F_r + 1.22)²]^0.5

Approximate SBT boundaries from I_c: I_c < 1.31 → gravelly sand to dense sand (Zones 7–8); 1.31–2.05 → sand (Zone 6); 2.05–2.60 → silty sand to sandy silt (Zones 5–4); 2.60–2.95 → silty clay to clay (Zone 3); I_c > 2.95 → clay to organic soil (Zones 2–3).

The I_c value is widely used to automate fines content correction in liquefaction analysis and to filter correlations by soil type across a depth profile.

Estimating soil parameters from CPT #

One of the principal engineering applications of CPT data is the estimation of soil engineering properties through empirical correlations — similar in principle to SPT correlations, but benefiting from the continuous depth profile and higher repeatability of the CPT. All CPT correlations are empirical and carry inherent uncertainty; local calibration against laboratory test data improves reliability significantly.

Soil parameters for coarse-grained soils (sands) #

Soil property	Common methods / references	Key input
Relative density (D_r)	Baldi et al. (1986); Jamiolkowski et al. (2001): D_r = (1/C₂) × ln(Q_tn/C₀). Calibration constants C₀, C₂ depend on sand compressibility.	Q_tn
Friction angle (φ’)	Robertson & Campanella (1983): φ’ = arctan[0.1 + 0.38 × log(q_t/σ’_v0)]; Kulhawy & Mayne (1990)	q_t, σ’_v0
Young’s modulus (E_s)	Lunne & Christoffersen (1983): E_s = α × q_c, where α = 1.5 (loose sand) to 3.0 (dense sand); Robertson (2009) stiffness ratio method	q_c
Small-strain shear modulus (G₀)	Rix & Stokoe (1991); Robertson (2009): G₀/q_t = ratio dependent on Q_tn	q_t
Stress history (OCR)	Mayne (2005): OCR ≈ 0.33 × (q_t − σ_v0)/σ’_v0 for sands (indicative only)	q_t, σ_v0, σ’_v0

Soil parameters for fine-grained soils (clays and silts) #

Soil property	Common methods / references	Key input
Undrained shear strength (s_u)	s_u = (q_t − σ_v0) / N_kt. The cone factor N_kt typically ranges from 10 to 20; calibration against vane shear or triaxial data is strongly recommended. Cabal & Ramirez-Fuerza; Aas et al. (1986).	q_t, σ_v0, N_kt
Overconsolidation ratio (OCR)	Mayne & Kemper (1988): OCR = k × (q_t − σ_v0)/σ’_v0, where k ≈ 0.33 for most clays; Lunne et al. (1997)	q_t, σ_v0, σ’_v0
Elastic modulus (E_u)	E_u = α_u × s_u, where α_u = 100–500 depending on OCR and plasticity; Mayne (2006) direct q_t method	s_u or q_t
Preconsolidation pressure (σ’_p)	Mayne (2005): σ’_p = 0.33(q_t − σ_v0); Demers & Leroueil (2002)	q_t, σ_v0

Bearing capacity and settlement from CPT #

CPT data can be used directly in shallow foundation design calculations — both for bearing capacity (ensuring the foundation will not fail in shear) and for settlement estimation (ensuring deformations remain within acceptable limits). CPT-based methods bypass the need for separate laboratory shear strength and stiffness testing, making them particularly attractive when CPT is the primary investigation tool.

Bearing capacity from CPT #

All CPT bearing capacity methods produce an allowable bearing pressure for a specified foundation geometry and settlement limit. The most commonly used methods are:

Method	Reference	Basis	Applicable soils
Robertson & Campanella	Robertson & Campanella (1988)	Directly correlates q_c with allowable bearing pressure for shallow foundations, accounting for foundation width and embedment depth	Sands and clays
Meyerhof (CPT equivalent)	Meyerhof (1956), CPT-based version	Converts q_c to equivalent SPT N-value, then applies SPT bearing capacity formula. Used where SPT-based methods are required by specification.	Sands
Eslami & Fellenius	Eslami & Fellenius (1997)	Pile design from CPT; unit shaft and base resistance directly from q_e (effective cone resistance). Included for pile preliminary estimates.	Sands and clays

Settlement estimation from CPT — Schmertmann method #

The Schmertmann (1970, 1978 revision) method is the standard approach for estimating immediate (elastic) settlement of shallow foundations on sand using CPT data. It uses the CPT-derived Young’s modulus (E_s) profile and an influence factor (I_z) that describes how stress from the foundation disperses with depth.

The general form of the Schmertmann settlement equation is:

s_i = C₁ × C₂ × Δq × Σ(I_z / E_s) × Δz

where:

C₁ = embedment correction factor = 1 − 0.5 × (σ’_v0 / Δq)
C₂ = creep correction factor = 1 + 0.2 × log(t / 0.1) for time t in years
Δq = net foundation pressure = applied pressure minus overburden stress at foundation level
I_z = strain influence factor (triangular distribution peaking at B/2 for square footings, 2B for strip footings)
E_s = Young’s modulus in each layer, derived from q_c
Δz = thickness of each sub-layer

The Schmertmann method requires the CPT profile to be available to the depth of influence (typically 2B for square footings, 4B for strip footings, where B is the foundation width). This is one of the practical advantages of the CPT over SPT for settlement calculation — the continuous profile provides the E_s value in every sub-layer without interpolation.

For foundations on clay, consolidation settlement is calculated from CPT-derived preconsolidation pressure and moduli using established one-dimensional consolidation theory. For a detailed treatment of settlement methods, see the foundation design guide.

CPT limitations and best practice #

The CPT is a powerful tool but has well-defined limitations that every practicing engineer should understand before planning an investigation program:

No soil sample: The CPT retrieves no physical soil sample. SBT classification is behavior-based, not mineral or grain-size based. In materials where visual description, laboratory classification, or contamination screening is required, CPT must be supplemented by boreholes.
Refusal in coarse materials: The cone cannot penetrate dense gravel, cobble layers, or cemented soils. A single cobble can cause early refusal and terminate the test far above the intended depth. Sites with known or suspected coarse fill, glacial till, or alluvial gravel sequences may have very limited CPT depth.
Correlations are empirical and soil-specific: All CPT correlations — N_kt for undrained shear strength, the constant n in Q_tn normalisation, D_r calibration constants — are empirical. They were developed for specific soil types and geological environments. In unusual soils (calcareous sands, volcanic soils, structured clays), correlations may produce unreliable results without local calibration.
Specialist equipment: CPT requires a specialised rig and is not universally available, particularly in developing regions or remote locations. Mobilisation cost can make it uneconomical for small investigations.
Interpretation expertise required: The continuous depth profile from CPT contains very rich information, but extracting it correctly requires understanding of normalisation procedures, SBT chart iterative methods, and the limitations of each correlation. Automated software interpretation without engineering review can produce misleading results.

Best practice recommendations:

Use CPTu rather than CPT wherever possible — the pore pressure measurement significantly improves accuracy in fine-grained soils and is essential for correct normalisation.
Calibrate N_kt and other soil-specific constants against laboratory data from the same site or a geologically similar local site.
Combine CPT with boreholes on most projects — use CPT for spatial profiling between boreholes and SPT boreholes at select locations for sample recovery.
Check CPT-derived stratigraphy against any available borehole data before proceeding to parameter estimation and design.
For sites with potential refusal issues, begin with a CPT grid to map the stratigraphy, then plan supplementary SPT boreholes where the CPT could not penetrate.

How DartiGeo handles CPT interpretation #

DartiGeo’s CPT module provides a complete interpretation workflow — from raw data import to bearing capacity and settlement results — within the same integrated platform used for borehole logging, SPT processing, laboratory testing, and foundation design.

The CPT interpretation workflow in DartiGeo includes:

Data import from Excel: Raw CPT depth profiles (q_c, f_s, and u₂ for CPTu) can be imported directly from Excel, eliminating manual data re-entry from field data logger output files.
Soil Behavior Type (SBT) classification: DartiGeo automatically calculates Q_tn, F_r, and I_c at each depth increment and plots the depth profile of SBT zones using the Robertson (1990) chart framework. The full Q_tn–F_r chart with all nine SBT zones is plotted with the data points overlaid, providing a visual check of the classification.
Soil parameter estimation: Published correlations for friction angle, undrained shear strength, Young’s modulus, and other parameters are applied automatically across the depth profile, with results displayed as variation-with-depth plots.
Bearing capacity from CPT: Allowable bearing pressure is calculated using CPT-based methods for the specified foundation geometry (width, shape, embedment depth), with full calculation output.
Settlement from CPT: Immediate settlement is estimated with the I_z influence factor distribution and E_s profile shown in the calculation output.
Professional reports: All inputs, SBT classification plots, parameter profiles, and design results are compiled into formatted PDF, Word, and Excel reports at the click of a button.

Because CPT data entered in DartiGeo feeds directly into the bearing capacity and settlement module alongside borehole data, SPT results, and laboratory test output, the engineer works with a single consistent dataset from field data to final report — with no re-entry between modules.

Visit DartiGeo’s Page →

Frequently asked questions #

What does CPT stand for? #

CPT stands for Cone Penetration Test (also referred to as the Cone Penetrometer Test). The CPTu variant adds the letter “u” to indicate that pore water pressure is also measured during the test — the “u” refers to the pore pressure parameter. In some literature, the CPTu is called the piezocone test, and the rig is called a piezocone penetrometer.

What is a good qc value in CPT? #

There is no universal “good” q_c value — the adequacy of a particular cone resistance depends on the soil type, depth, and the engineering application. As a rough guide for preliminary assessment: q_c below 1 MPa typically indicates very soft clay or loose sand that may require ground improvement; q_c of 5–15 MPa indicates medium-dense sand or stiff clay adequate for many conventional shallow foundations; q_c above 20 MPa indicates dense sand or gravel, or very stiff to hard clay. Always use q_t (corrected for pore pressure) rather than raw q_c for quantitative analysis in fine-grained soils.

What is the Robertson SBT chart and how is it used? #

The Robertson SBT (Soil Behavior Type) chart is a classification chart that plots normalised cone resistance (Q_tn) against friction ratio (F_r) on logarithmic axes. It was developed by P.K. Robertson in 1990 and divides the plot space into nine zones, each corresponding to a distinct soil behavior type ranging from sensitive fine-grained soils (Zone 1) through clays, silts, sands to gravelly sands (Zone 7). Engineers plot each depth increment of CPT data on the chart to classify the soil at that depth. Modern software performs this automatically and produces a depth profile of SBT zones alongside the raw CPT data traces.

How deep can CPT reach? #

Standard onshore CPT rigs typically reach 30–60 m depth in soft to medium soils, depending on rig capacity and friction build-up on the push rods. In very soft soils (offshore, deltaic), depths exceeding 100 m have been achieved with specialist equipment. Practical limitations are usually rig thrust capacity (typically 100–200 kN), refusal from dense layers or cobbles, and the physical length of push rods available on the rig. Where greater depth is required, the test is terminated, casing may be installed to a shallower depth to reduce rod friction, and the cone is re-lowered through the casing to continue.

Can CPT replace SPT entirely? #

Not entirely, for two reasons. First, the CPT retrieves no soil sample — there is no physical material to describe, classify in the laboratory, or test for contamination. In most geotechnical investigations, some physical samples are required for project records, client requirements, or regulatory compliance. Second, CPT encounters refusal in gravels, cobbles, and cemented layers that SPT can penetrate. In practice, most quality site investigations use both tests: CPT for spatial profiling and detailed stratigraphy, and SPT boreholes at select locations for sample recovery and visual verification. The two datasets also allow cross-validation of interpreted parameters, improving confidence in the design inputs.

What is the friction ratio in CPT and what does it tell you? #

The friction ratio R_f = (f_s / q_c) × 100% is the ratio of sleeve friction to cone tip resistance, expressed as a percentage. It is one of the primary indicators of soil type in CPT data. Clean sands and gravels have low friction ratios (typically R_f < 0.5–1%) because the cone tip displaces the soil with high resistance while sleeve friction remains low. Clays and organic soils have high friction ratios (typically R_f > 3–5%) because cone tip resistance is low relative to the sleeve friction. The friction ratio forms one axis of the Robertson SBT classification chart.

Cone Penetration Test (CPT) — Complete Guide to Interpretation & Analysis