Chapter 10 — Laboratory Testing Program — Railway Ground Hazard Risk Management

Geotechnical Services Manual for Railroad Clients

A Consultant's Reference for Delivering Geotechnical Engineering Services Across North American Railroad Capital Programs

Apps for this Chapter3

Apps on dirtbert.com that support this chapter. Each one runs in a web browser on a desktop or a phone, with nothing to install. How to use the apps (I.3). Tools marked TOOLS open in the Geotechnical Analysis Suite (Benesch sign-in); the orange card below lists each one by section.

NANCYPhase IV · Laboratory TestingGuidance for planning and running the laboratory phase.
NANCYProject Data · Laboratory ResultsIndex, strength, consolidation and compaction results for each sample.
TOOLSAnalytical Tools (by chapter)Laboratory calculators and workbook guides.

NANCY pages ask you to accept the Use and Disclaimers once in each browser.

Analytical Tools for this Chapter7

Calculators and workbook tools from the Benesch Geotechnical Analysis Suite that support this chapter: 4 interactive calculators and 3 workbook guides. Each one is also placed inside the section it supports, in an orange tools strip.

SP	Atterberg Limits — Plasticity ChartUsed in 10.3 Test Methods — ASTM Standards and English Unit Outputs	Calculator ▸	Guide
PV	Percent Saturation — Subgrade SamplesUsed in 10.3 Test Methods — ASTM Standards and English Unit Outputs	Calculator ▸	Guide
PV	CBR Statistical ValuesUsed in 10.3 Test Methods — ASTM Standards and English Unit Outputs	Calculator ▸	Guide
PV	CBR CorrelationsApplies to the chapter as a whole	Calculator ▸	Guide
GW	Flexible Wall PermeabilityUsed in 10.3 Test Methods — ASTM Standards and English Unit Outputs	guide only	Guide
ST	Consolidation Index CorrelationsUsed in 10.3 Test Methods — ASTM Standards and English Unit Outputs	guide only	Guide
DT	Soil DataApplies to the chapter as a whole	guide only	Guide

Tools open on dirtbert.com and need the Benesch sign-in. Drafts pending Benesch final review, not for use in design.

Reference Graphics

General topic illustrations from Benesch's shared geotechnical graphics library, related to this chapter's subject matter — not the numbered figures called out in the text above. Click any image to view it full size.

Geotechnical laboratory testing for earth dam clay cores

10.0 Executive Summary

Laboratory testing is the scientific foundation of every geotechnical design. Field observations classify soils and rock; laboratory tests quantify the engineering properties that drive bearing capacity, settlement, slope stability, and pavement design calculations. For railroad capital projects, the consequences of poorly selected or inadequately executed lab programs are magnified by the scale of the loading environment — 286,000-lb gross rail load, cyclic dynamic amplification, and serviceability tolerances measured in fractions of an inch.

This chapter defines the laboratory testing program for geotechnical investigations serving all nine client railroads: CSXT, NSR, G&W, UPRR, BNSF, CPKC, Amtrak, and CN. It covers test selection by project type, ASTM standards for every test method, sample handling and chain of custody, laboratory QA/QC, turnaround time expectations by client, and accreditation requirements. All test results are expressed in English units throughout. The consultant must confirm current ASTM edition requirements with each client's engineering standards before mobilizing a laboratory program — clients periodically update their referenced editions.

Empirical Equations — Use and Limitations

The correlations below are used or referenced in this chapter. Each is a statistical fit to a particular data set, so a value from one is an estimate: it applies only within the materials and ranges stated, carries the scatter of the original data, and must be confirmed by site-specific testing or calibration before it is relied on for design.

When an equation is used, note it with the result: give the equation ID and source, state that the soil or rock is within its range of applicability, mark the value as correlated (not tested), and record the limitation that governs. The Empirical Equations & Correlations reference has every entry with its full references.

Referenced in this chapter

RK-10UCS from point load indexσ꜀

σ_c ≈ K · I_s(50), K ≈ 24 (range ~15–50)

I_s(50) = F · I_s; F = (D_e/50)^0.45

Variables. D_e = equivalent core diameter (mm); I_s = P/D_e²

Use. Hard, isotropic rock; quick strength index in field.

Limitations. K varies widely with rock type and anisotropy; calibrate with lab UCS. Unreliable in weak rock (σ_c < ~25 MPa).

Source. Broch & Franklin 1972; ISRM 1985 · In this chapter: 10.2 · Full entry and references

Also applicable to this chapter

IX-01Compression index from liquid limit (undisturbed)C꜀

C_c = 0.009 (LL − 10)

Variables. LL = liquid limit (%)

Use. Normally consolidated clays of low to moderate sensitivity.

Limitations. Reliability about ±30%. Not for organic or highly sensitive clays.

Source. Terzaghi & Peck 1967 · Full entry and references

IX-02Compression index from liquid limit (remolded)C꜀

C_c = 0.007 (LL − 7)

Variables. LL = liquid limit (%)

Use. Remolded clays.

Limitations. Underestimates C_c of structured natural clays.

Source. Skempton 1944 · Full entry and references

IX-03Compression index from void ratioC꜀

C_c = 1.15 (e₀ − 0.35)

Variables. e₀ = in-situ void ratio

Use. All clays (general trend).

Limitations. Broad scatter; preliminary use only.

Source. Nishida 1956 · Full entry and references

IX-04Compression index from plasticity indexC꜀

C_c ≈ 0.5 · G_s · (PI/100) (≈ PI/74 for G_s = 2.7)

Variables. G_s = specific gravity of solids

Use. Remolded / normally consolidated clays; critical-state framework.

Limitations. Underestimates C_c for sensitive, structured clays.

Source. Wroth & Wood 1978 · Full entry and references

IX-05Secondary compression ratioC_α

C_α / C_c ≈ 0.04 ± 0.01 (inorganic clays and silts)

≈ 0.05 ± 0.01 (organic clays and silts); ≈ 0.075 ± 0.01 (peats)

Variables. C_α = Δe / Δlog t

Use. Secondary consolidation of soft soils; the ratio holds at a given effective stress level.

Limitations. Use C_c corresponding to the final stress (C_c varies with stress in structured clays).

Source. Mesri & Godlewski 1977; Mesri & Castro 1987 · Full entry and references

IX-06Normalized undrained strength from plasticitySᵤ/σ'

S_u / σ'_v0 = 0.11 + 0.0037 · PI

Variables. PI (%)

Use. Normally consolidated clays (field vane basis).

Limitations. Field vane values; apply Bjerrum correction (IS-25) for stability. Not for OC clays.

Source. Skempton 1957 · Full entry and references

IX-08Friction angle of NC clay from plasticityφ'

sin φ' = 0.80 − 0.094 · ln(PI)

Variables. PI (%)

Use. Normally consolidated, inorganic clays; peak effective-stress φ'.

Limitations. Standard deviation about ±3°. Not residual or fully softened strength; for slope reactivation use residual/fully-softened correlations (e.g., Stark & Eid charts).

Source. Kulhawy & Mayne 1990 (data of Bjerrum & Simons 1960) · Full entry and references

IX-10Undrained strength from liquidity indexSᵤ (remolded)

S_u,r ≈ 170 · exp(−4.6 · LI) (kPa)

Variables. LI = (w − PL)/PI

Use. Remolded clays; anchored at S_u ≈ 1.7 kPa at LL and ≈ 170 kPa at PL.

Limitations. Remolded strength only; intact strength is higher by the sensitivity. Rough estimate.

Source. Wroth & Wood 1978 · Full entry and references

IX-11Hazen hydraulic conductivityk

k = C · D₁₀² (cm/s, D₁₀ in cm)

C ≈ 100 (commonly 40–150)

Variables. D₁₀ = effective grain size

Use. Loose, clean, uniform sands (C_u < ~5); D₁₀ ≈ 0.1–3 mm.

Limitations. Ignores density and fabric. Errors of an order of magnitude are common in silty or well-graded soils.

Source. Hazen 1911 · Full entry and references

IX-13Small-strain shear modulus (Hardin)G_max

G_max = 625 · F(e) · OCR^k · P_a^0.5 · σ'_m^0.5

F(e) = 1 / (0.3 + 0.7 e²)

Variables. σ'_m = mean effective stress; k = 0 (PI=0) to 0.5 (PI ≥ 100)

Use. Uncemented clays and sands; lab resonant-column basis.

Limitations. Underestimates aged/cemented soils; field V_s (G_max = ρV_s²) is preferred.

Source. Hardin 1978 · Full entry and references

10.1 Background and Railroad Context

The laboratory testing program for a railroad geotechnical investigation must be:

Purposeful. Every test must serve a defined analytical need. Ordering triaxial shear strength tests for a track subgrade investigation without a slope stability analysis to use them wastes budget and delays the schedule. Order only what the analysis matrix demands.
Sufficient. Under-testing is the more common and more dangerous failure mode. A single consolidation test in a soft clay stratum extending over 500 ft is not statistically defensible. A minimum of one test per identified stratum per project segment, plus additional tests where variability is observed, is the standard expected by all Class I client railroads.
QA-verified. All nine client railroads that maintain laboratory accreditation requirements (see Section 10.8) will reject data produced by non-accredited laboratories. Confirm accreditation before awarding the laboratory subcontract.

10.2 Test Program Design by Project Type

Table 10-1. Laboratory Testing Matrix by Railroad Project Type
Test	Siding/Track Ext.	Intermodal Yard	Bridge/Culvert	Mainline Rehab	Emergency
Moisture content (D2216)	All samples	All samples	All samples	All samples	All samples
Atterberg limits (D4318)	Fine-grained	Fine-grained	Fine-grained	Fine-grained	Fine-grained
Grain size (D6913/D7928)	Select	Select	Select	Select	Priority
Specific gravity (D854)	Consolidation set	All triaxial set	All triaxial set	Consolidation set	Priority
Standard Proctor (D698)	Structure backfill	As needed	Structure backfill	As needed	Not typically
Modified Proctor (D1557)	Embankment	Yes	Embankment	Yes	Not typically
CBR (D1883)	Subgrade	Pavement design	Not typically	Subgrade rehab	Not typically
Resilient Modulus (T307)	Subgrade (Amtrak)	Pavement design	Not typically	Subgrade rehab	Not typically
Consolidation (D2435)	Soft clay	Soft clay	Soft clay	Soft clay	Not typically
UU Triaxial (D2850)	Not typically	Slope stability	Slope stability	Slope stability	Quick stability
CU Triaxial (D4767)	Not typically	Critical slopes	Critical slopes	Critical slopes	Not typically
Direct shear (D3080)	Not typically	Residual φ'	Rock joints	Residual φ'	Not typically
Unconfined compression (D2166)	Not typically	Select	Select	Select	Quick s_u
Permeability (D2434/D5084)	Drainage design	Drainage design	Drainage design	Drainage design	Not typically
Point load index (D5731)	Rock cuts	Rock cuts	Rock cuts	Rock cuts	Not typically
Sodium sulfate soundness	Ballast	Ballast	Not typically	Ballast rehab	Not typically
LA Abrasion (C131)	Ballast	Ballast	Not typically	Ballast rehab	Not typically

Reference graphics

Ballast, subballast and fouling test suite

For ballast-source qualification, ballast condition surveys and subballast design, add the following to the soil program:

Ballast source: petrographic analysis; LA Abrasion; Mill Abrasion (wet) and the Abrasion Number, AN = LAA + 5 MA; wet attrition (Deval) for soft stone in wet track; clay lumps and friable particles (ASTM C142); flat and elongated particles; specific gravity and absorption.
Fouled ballast: trench samples taken by zone (crib, shoulder, top, bottom). Report the fraction finer than 3/8 in. as fouling material and the fraction finer than No. 200 as fines. Compute the fouling index, and identify the source of the fines by color, plasticity and, where it matters, SEM or XRD mineralogy.
Subballast: gradation for the filter checks against both ballast and subgrade; LA Abrasion ≤ 50; LL and PI on the minus-No. 40 fraction (≤ 25 and ≤ 6).

After Selig & Waters (1994), chs. 7, 8 and 9. Full reference under Supplemental Sources.

Empirical equation note · this section

Values from these correlations are estimates, not test results. State the equation ID and source when one is used, confirm the soil or rock is within its range, and calibrate against site testing before design.

RK-10 UCS from point load index σ꜀Use: Hard, isotropic rock; quick strength index in field. Limits: K varies widely with rock type and anisotropy; calibrate with lab UCS. Unreliable in weak rock (σ_c < ~25 MPa).

10.3 Test Methods — ASTM Standards and English Unit Outputs

10.3.1 Water Content (ASTM D2216)

Method: Oven-dry at 230 ± 9°F for minimum 16 hours (peat and organic soils: 140°F).

Equation:

w = (M_water / M_dry) × 100%

where M_water = mass of water (mass_wet − mass_dry); M_dry = mass of dry soil.

Output: Water content w (%), dimensionless.

Minimum frequency: Every disturbed and undisturbed sample. Water content is the most fundamental test in the program — never omit it.

Reference graphics

10.3.2 Atterberg Limits (ASTM D4318)

Liquid Limit (LL): Casagrande cup method; flow curve plotted; LL at N = 25 blows.

Plastic Limit (PL): ⅛-in. thread rolling on glass plate.

Plasticity Index: PI = LL − PL

Liquidity Index: LI = (w − PL) / PI

Activity: A = PI / (% < 0.002 mm)

Output: LL, PL, PI (all %); LI and A dimensionless.

Minimum frequency: 1 per fine-grained stratum per boring; 1 per 5 ft depth in highly variable fine-grained profiles.

Reference graphics

Reference graphics · click to zoom

10.3.3 Grain Size Analysis

Sieve analysis (ASTM D6913): Applicable for material with particles retained on the No. 200 sieve. Sieves: 3 in., 2 in., 1½ in., 1 in., ¾ in., ½ in., No. 4, No. 10, No. 20, No. 40, No. 60, No. 100, No. 200.

Hydrometer analysis (ASTM D7928): For material passing No. 200 (clay and silt sizes). Combined with sieve for full gradation curve (ASTM D6913 + D7928).

Key outputs:

D₁₀, D₃₀, D₆₀ (in., or mm for reference): particle diameters at 10%, 30%, 60% passing
C_u = D₆₀ / D₁₀ (uniformity coefficient, dimensionless)
C_c = D₃₀² / (D₆₀ × D₁₀) (coefficient of curvature, dimensionless)
Fines content: % passing No. 200 sieve

Minimum frequency: 1 per stratum per boring for classification; additional tests for filter design and liquefaction susceptibility assessment.

Reference graphics

Reference graphics · click to zoom

10.3.4 Specific Gravity (ASTM D854)

Method: Pycnometer (volumetric flask) method; soil displaced volume in de-aired water.

Equation:

G_s = M_sd / (M_sd + M_pw − M_spw) (dimensionless)

where M_sd = mass of dry soil; M_pw = mass of pycnometer + water; M_spw = mass of pycnometer + soil + water.

Typical values: G_s = 2.60–2.75 (mineral soils); 2.60–2.65 (clean quartz sand); 2.70–2.80 (iron-rich soils).

Minimum frequency: 1 per soil type in each project; required for void ratio and consolidation calculations.

Reference graphics

10.3.5 Compaction Tests

Standard Proctor (ASTM D698):

Method A: material passing No. 4 sieve; 3 layers; 25 blows per layer; 5.5-lb hammer; 12-in. drop
Energy: 12,375 ft·lb/ft³

Modified Proctor (ASTM D1557):

Method A or C: 5 layers; 25 blows per layer; 10-lb hammer; 18-in. drop
Energy: 56,250 ft·lb/ft³

Output: Compaction curve (γ_d pcf vs. w %); γ_d,max (pcf); w_opt (%)

Zero-Air-Voids (ZAV) line: γ_d,ZAV = G_s × γ_w / (1 + G_s × w/100) (pcf)

Minimum frequency: 1 Modified Proctor per borrow source or fill material type (not per sample); repeat if source changes during construction.

Reference graphics

Reference graphics · click to zoom

10.3.6 California Bearing Ratio (ASTM D1883)

Method: Laboratory CBR at Modified Proctor density (95% RC minimum). Specimen soaked 96 hours at Standard loading conditions.

Penetration resistance equation:

CBR = (P_sample / P_standard) × 100%

where:

P_sample = measured force at 0.1 in. or 0.2 in. penetration (lb)
P_standard = standard force at same penetration; 0.1 in.: 3,000 lb; 0.2 in.: 4,500 lb

Use the higher CBR from 0.1 or 0.2 in. penetration (recheck if 0.2 in. governs).

Output: CBR (%); used directly for pavement design (Chapter 18, Section 18.1).

Minimum frequency: 1 per subgrade soil type per project; 3+ tests for intermodal yard pavement design.

Reference graphics

10.3.7 Resilient Modulus (AASHTO T307)

Method: Cyclic triaxial test with conditioning sequence (500 cycles at 4 psi deviator) followed by 15 stress states per AASHTO T307 protocol.

M_r from regression (NCHRP 1-28A model):

M_r = k₁ × P_a × (θ/P_a)^k₂ × (τ_oct/P_a + 1)^k₃ (psi)

where:

P_a = atmospheric pressure = 2,117 psf = 14.7 psi
θ = bulk stress = σ₁ + σ₂ + σ₃ (psi)
τ_oct = octahedral shear stress (psi)
k₁, k₂, k₃ = regression constants from test

Minimum frequency: Required by Amtrak for all NEC pavement subgrade investigations; recommended by BNSF for intermodal yard pavement design; 1 test per subgrade soil type.

Repeated-load testing for track subgrade

Repeated-load triaxial (or simple shear) tests give two results that matter for track:

The resilient modulus, E_r = cyclic deviator stress ÷ recoverable axial strain.
The rate at which plastic strain accumulates.

Tiny plastic strains per cycle add up to large settlement over millions of cycles, so report both.

What to look for.

In drained or unsaturated tests, the plastic increment shrinks with each cycle.
In saturated specimens, or at stresses near failure, the increment can grow. That is the runaway case.
Running a series of stress levels identifies the threshold stress used in Chapter 2, §2.4.2.
E_r typically rises over the first cycles. Granular materials need several thousand cycles to stabilize, and their modulus is strongly dependent on confining pressure (E_r = K₁ θ^K₂, where θ is bulk stress).

Test conditions. Test at the field moisture and density, including a wet-of-optimum case (Chapter 18, §18.2.2). Fine-grained modulus falls steeply on the wet side.

After Selig & Waters (1994), chs. 3, 7 and 10. Full reference under Supplemental Sources.

10.3.8 One-Dimensional Consolidation (ASTM D2435)

Method A: Incremental load; load doubling each 24-hour interval; loads: 125–250–500–1,000–2,000–4,000–8,000 psf (standard); extend to 16,000 psf for deep or heavily preloaded sites.

Method B: Constant rate of strain (CRS); continuous loading at a controlled strain rate.

Key outputs (English units):

Parameter	Symbol	Units	Equation
Compression index	C_c	dimensionless	Slope of e-log σ'_v curve (NC range)
Recompression index	C_r	dimensionless	Slope of e-log σ'_v curve (OC range); ≈ C_c/5 to C_c/10
Preconsolidation pressure	σ'_p	tsf	Casagrande graphical construction
OCR	—	dimensionless	σ'_p / σ'_v0
Coefficient of consolidation	c_v	ft²/day	c_v = T_v × H_dr² / t (see Chapter 12)
Secondary compression index	C_α	dimensionless	Δe / Δ log t (secondary range)

Minimum frequency: 1 per fine-grained stratum per project; minimum 3 tests in any stratum requiring settlement analysis.

Reference graphics

Reference graphics · click to zoom

10.3.9 Unconsolidated-Undrained (UU) Triaxial (ASTM D2850)

Method: Undrained shear test on undisturbed sample; three specimens at different confining pressures (typically 500, 1,000, 2,000 psf); no drainage allowed during shear.

Output: s_u (psf or tsf) from Mohr circle at failure; φ_u ≈ 0° for saturated clay.

Note: For saturated clay, UU test gives s_u directly regardless of confining pressure — the Mohr circle diameter is constant. UU is appropriate for short-term (undrained) stability analysis only.

Minimum frequency: 1 per soft or medium clay stratum per boring at critical cross-sections (slope stability, embankment loading over soft clay).

Reference graphics

Reference graphics · click to zoom

Unconsolidated–Undrained (UU) Triaxial Test

Strength parameters: match the test to the drainage condition

CD tests give c′ ≈ 0 for both loose and dense soils. A small c′ appears only from a curved envelope or heavy overconsolidation. Use effective-stress parameters for drained problems.
CU tests. Dense sand develops negative excess pore pressure, so its undrained strength exceeds the drained strength (up to the limit of cavitation). Loose sand develops positive pore pressure, so its undrained strength is lower. φ′ is the same as from CD tests.
UU tests give φ = 0 and c = s_u, which depends only on the initial state. This is the convenient basis for short-term undrained loading.
Unconfined compression works only on fine soils that stand unsupported. Disturbance and unloading make it underestimate s_u.
Partially saturated subgrade. Pore pressure is rarely known, so analyze in total stress with both c and φ. Soil at low saturation behaves essentially drained in any test.

After Selig & Waters (1994), ch. 3. Full reference under Supplemental Sources.

10.3.10 Consolidated-Undrained (CU) Triaxial with Pore Pressure (ASTM D4767)

Method: Specimens consolidated to two or three effective confining pressures (σ'₃); sheared undrained with pore pressure measurement; effective stress Mohr circles give c' and φ'.

Output: c' (psf), φ' (degrees) — effective strength parameters for long-term (drained) analysis.

Minimum frequency: Required for all critical slope stability analyses (NSR clay shale, CPKC prairie clay, CN sensitive marine clay); 1 set (3 specimens) per stratum per project minimum.

Reference graphics

Reference graphics · click to zoom

10.3.11 Direct Shear (ASTM D3080)

Method: Specimen sheared at constant normal stress; three tests at different normal stresses.

Peak and residual strength:

τ_peak = c' + σ_n × tan φ'_peak (psf)

τ_residual = σ_n × tan φ'_r (psf, c' = 0 for residual)

Output: c' (psf), φ' (degrees), φ'_r (degrees).

Use cases: Residual strength for CPKC clay shale and NSR weathered shale slope stability; rock joint shear strength for cut slope design.

Minimum frequency: 1 per stratum requiring residual strength parameters; rock joints: minimum 3 direct shear tests per joint set per project.

Reference graphics

10.3.12 Unconfined Compression (ASTM D2166)

Method: Axial loading to failure at unconfined conditions (σ₃ = 0). Quick and inexpensive.

Output: q_u (psf or tsf); s_u = q_u / 2.

Limitation: UU triaxial provides more reliable s_u; unconfined compression can underestimate s_u in fissured or brittle clays. Use for screening; not for final design in critical applications.

Minimum frequency: Appropriate for routine s_u checks on Shelby tube samples; supplement with UU triaxial for design-level analysis.

Reference graphics

10.3.13 Hydraulic Conductivity (Permeability)

Constant head (ASTM D2434): For coarse-grained soils (k > 10⁻³ ft/day).

k = (Q × L) / (A × h × t) (ft/day)

where Q = flow volume (ft³); L = specimen length (ft); A = cross-sectional area (ft²); h = head differential (ft); t = test duration (day).

Flexible wall permeameter (ASTM D5084): For fine-grained soils (k < 10⁻³ ft/day).

k = (a × L × ln(h₁/h₂)) / (A × t) (ft/day)

where a = standpipe area (ft²); L = specimen length (ft); h₁, h₂ = head at start and end; t = elapsed time (day).

Minimum frequency: 1 per stratum for drainage design; 3 per stratum for seepage analysis through embankments or dams.

Reference graphics

10.3.14 Point Load Index (ASTM D5731)

Method: Rock core or irregular lump loaded to failure between conical platens.

Point load strength index:

I_s = P / D_e² (psi or psf)

where P = failure load (lb); D_e = equivalent core diameter (in.).

Size-corrected index:

I_s(50) = F × I_s where F = (D_e/1.97 in.)^0.45

Estimated UCS of rock:

q_u(rock) ≈ 24 × I_s(50) (psi) [Broch & Franklin 1972; I_s in psi]

Minimum frequency: Minimum 5 specimens per rock type per project; not a substitute for UCS testing for critical rock foundation designs.

Reference graphics

Supplemental reference · LEVI Library

USACE Q / R / S nomenclature and common triaxial errors

USACE manuals (EM 1110-2-1902 slope stability, EM 1110-2-2504 sheet piles, and others cited in this manual) use letter names for the basic triaxial tests: Q = unconsolidated-undrained (UU); R = consolidated-undrained (CU); S = consolidated-drained (CD), which is also the long-term drained case. The Q test applies only to soils that are not free-draining, and R-test specimens should generally be fully saturated before shear. Errors EM 1110-2-1906 lists for R tests: back-pressure increments too large or applied too fast; chamber and back pressure not held precisely (which can overconsolidate the specimen); shearing before consolidation is complete; too fast a strain rate; temperature drift; and water absorbed from the porous inserts after the test, which biases the final water content. For S tests the main errors are strain rate and volume-change measurement. In pervious soils, water redistributes even in a quick Q test, so measure the water content of the middle third separately from the ends.

Picking t₅₀ and t₉₀ from each load increment

Section 10.3.8 gives c_v = T_vH_dr²/t but not how t is found. The two standard curve-fitting methods (Das lab manual) are:

Square-root-of-time (Taylor, 1942): plot dial reading against √t, extend the initial straight line, draw a second line from the same origin with abscissas 1.15 times larger, and read t₉₀ where it cuts the curve. Then c_v = 0.848 H_dr²/t₉₀.
Log-of-time (Casagrande and Fadum, 1940): d₁₀₀ is the intersection of the primary and secondary straight lines. For d₀, pick t₁ and t₂ = 4t₁ in the early curved part and step up by their difference. t₅₀ is at d₅₀ = (d₀ + d₁₀₀)/2. Then c_v = 0.197 H_dr²/t₅₀, and C_α comes from the slope after d₁₀₀.

With double drainage, H_dr is half the average specimen height during the increment. Report which method was used. The two methods commonly give different c_v values for the same increment, and the report should say which one the settlement-time analysis in Chapter 12 used.

Source: U.S. Army Corps of Engineers (1970, with changes). Laboratory Soils Testing. EM 1110-2-1906 · Das, B. M. (2002). Soil Mechanics Laboratory Manual, 6th ed. Oxford University Press

Supplemental reference · LEVI Library

Tests for shale and mudstone: preparation effects and slake durability

Railroad cuts in Appalachian and prairie shales (Chapter 13) need durability as well as strength data, and Atterberg limits of clay shale depend on how the sample was processed. EM 1110-2-1906 therefore standardizes three preparations: blenderized (the primary reference — dried below 122°F at < 30% RH, soaked at least 48 hr, then blended as a slurry above 300% water content or twice the LL); undried (shaved directly into distilled water); and air-dried (one drying–soaking cycle). The spread among the three indicates the strength of interparticle bonds. Report which method produced the limits used for design.

Slake durability (Franklin and Chandra, 1972; ASTM D4644): oven-dried rock lumps are rotated for 10 minutes in a sieve-mesh drum half immersed in 68°F water. I_d is the percentage of the initial dry mass retained. Nagaraj reports reproducibility within 5% at 95% confidence, and the test discriminates well between mudstones that otherwise look alike.

I_d (%)	0–25	25–50	50–75	75–90	90–95	95–100
Durability	Very low	Low	Medium	High	Very high	Extremely high

After Franklin and Chandra (1972). As printed in Nagaraj’s Table 9.7, the 75–90 and 90–95 labels are transposed; the order shown here is the original.

For clay shales, Nagaraj also reproduces the Morgenstern and Eigenbrod (1974) classification, which combines the amount of slaking (by the liquid limit of the slaked material: < 20 very low to > 140 very high) with the rate (change in liquidity index in the first cycle: < 0.75 slow, 0.75–1.25 fast, > 1.25 very fast). Add slake durability to the rock column of Table 10-1 for every cut in shale, mudstone or claystone and for any rock proposed as riprap or fill.

Source: U.S. Army Corps of Engineers (1970, with changes). Laboratory Soils Testing. EM 1110-2-1906 · Nagaraj, T. S. (1993). Principles of Testing Soils, Rocks and Concrete. Elsevier (Developments in Geotechnical Engineering)

Supplemental figures · LEVI Library

**Supplemental Figure 10-S1.** Schematic of fixed-ring and floating-ring consolidometers (EM 1110-2-1906). Apparatus compression in either type must be calibrated out for stiff soils. *Source: U.S. Army Corps of Engineers (1970, with changes). *Laboratory Soils Testing*. EM 1110-2-1906, p. VIII-2.*

**Supplemental Figure 10-S2.** Details of a typical triaxial compression chamber for 1.4-in.-diameter specimens (EM 1110-2-1906). *Source: U.S. Army Corps of Engineers (1970, with changes). *Laboratory Soils Testing*. EM 1110-2-1906, p. X-4.*

Ballast and subballast abrasion tests (Deval, Los Angeles, Micro-Deval)

Table 10-7. Abrasion tests for railway aggregates as described by Profillidis
Test	Equipment and procedure	Result
Deval (1896)	11-lb sample in cylinders 8 in. dia. × 13 in., inclined 30°; 10,000 revolutions at 2,000 rev/h (about 5 h); dry or wet	Deval coefficient from % attrition. Typical minimums: > 12 (limestone), > 14 (hard rock), some railways > 16
Los Angeles (1926)	Drum 28 in. dia. × 20 in.; 11-lb sample with 12 steel balls of 15 oz each; 500 revolutions at 30–33 rpm (about 15 min). EN version: 22-lb sample, 1,000 revolutions	% passing No. 12 (French) or No. 8 (British) sieve. Many specifications: LA < 25. EN reference test
Micro-Deval	1.1 lb of ⅜–½ in. gravel, one 11-lb steel ball and 2.6 qt of water in an 8-in. cylinder; 12,000 revolutions at 100 rpm. EN version: 22 lb, 14,000 revolutions	MDE = % finer than No. 12. Used mainly for subballast; subballast MDE < 15–20

North American ballast specifications (AREMA Chapter 1) use the ASTM Los Angeles test for large coarse aggregate (ASTM C535). Results from the European methods are not interchangeable with ASTM results without correlation.

Cyclic behavior. Ballast takes a large permanent set on first loading as stones rearrange, then accumulates strain slowly: ε_N = ε₁(1 + c·log N), with c ≈ 0.2 (European ORE and British Rail research) or 0.25–0.40 (American research). On that basis it takes 100,000–300,000 cycles to double the first-cycle plastic strain. Ballast modulus roughly doubles over the first 1,000 cycles and then stays nearly constant; gravel subballast modulus does not change with cycles. Repeated-load triaxial programs should run long enough to get past this initial bedding-in.

Figures · click to zoom

Figure 10-11. Deval attrition machine. Profillidis Fig. 12.4. — **Figure 10-11.** Deval attrition machine. Profillidis Fig. 12.4.

Figure 10-12. Elastic deformation of ballast vs. number of loading cycles N. Profillidis Fig. 12.3. — **Figure 10-12.** Elastic deformation of ballast vs. number of loading cycles N. Profillidis Fig. 12.3.

After Profillidis (2014), §12.3–12.4, Eq. 12.1, Figs. 12.3–12.4. Full reference under Supplemental Sources.

Standard index tests for subgrade samples

Table 10-8. Standard laboratory battery for subgrade samples (course summary)
Test	Standard	What it gives
Atterberg limits	ASTM D4318	Liquid limit, plastic limit, plasticity index; classifies fines
Particle-size distribution	ASTM D422	Sieve gradation, % finer than No. 200, hydrometer split of silt and clay
Moisture content	ASTM D4643	Water content by drying
Unconfined compressive strength	ASTM D2166	Direct strength of cohesive samples
Proctor compaction	ASTM D698 / D1557	Target density and optimum moisture for fill and subgrade

Compaction can roughly double a soil’s strength. The fines content of granular fill controls its drainage, frost susceptibility and strength after rain, so specify and test fines on ballast and sub-ballast deliveries rather than accepting material that only looks good dry.

Figures · click to zoom

Figure 10-13. Proctor compaction device. — **Figure 10-13.** Proctor compaction device.

Figure 10-14. Bank, loose and compacted states of soil during earthwork. — **Figure 10-14.** Bank, loose and compacted states of soil during earthwork.

After Geotechnology for Railroaders Program, Module 2; §09–10. Full reference under Supplemental Sources.

Case study: 10.16.1 Curing time changes compaction and CBR results (Look 2023), in §10.16 Case Studies.

10.4 Sample Handling and Chain of Custody

10.4.1 Shelby Tube Handling

Shelby tubes (ASTM D1587 thin-wall samples) must be:

Sealed at both ends with wax within 2 hours of extraction
Stored vertically (upright) in a padded core box or individual foam holders
Transported at ambient temperature; never frozen (unless specifically permitted for permafrost)
Extruded at laboratory in a horizontal extruder within 30 days of collection
Never impact-loaded, dropped, or inverted during transport

Sample Quality Designation (Lunne et al. 1997) — referenced in Chapter 6:

Δe/e₀	Quality
< 0.04	Very good to excellent
0.04–0.07	Good to fair
0.07–0.14	Poor
> 0.14	Very poor

Report SQD for all consolidation and triaxial test specimens. UPRR and BNSF Engineering Instructions require that samples with SQD = Poor or Very Poor be noted in the report with a corresponding caveat on the reliability of strength and compressibility parameters derived from them.

Reference graphics

10.4.2 Chain of Custody

A chain of custody (COC) form must accompany every sample shipment from field to laboratory. Required fields:

Project name and number; client railroad
Boring ID, sample ID, sample type, and depth interval
Date and time of collection; collector's name
Date and time of departure from field; shipper's name
Date and time of receipt at laboratory; receiver's name and signature
Condition of samples at receipt (sealed, intact, disturbed)
Tests assigned (referenced to lab test request form)
Custody transfer: any intermediate handler must sign and date

CSXT and Amtrak require COC forms to be scanned and submitted with the laboratory report as a QC attachment.

10.5 Laboratory QA/QC

10.5.1 Duplicate Testing

A minimum of 10% of all tests (by test type) must be run in duplicate as a QC check. Duplicate tests are run on split halves of the same sample by the same method. Acceptable precision limits:

Table 10-2. Laboratory Test Precision Requirements
Test	Acceptable Duplicate Difference	Action if Exceeded
Water content (D2216)	≤ 1.0%	Repeat test on third portion
Atterberg limits (D4318)	LL ≤ 2%, PL ≤ 2%	Repeat; average of three if still outside
Modified Proctor (D1557)	γ_d,max ≤ 1.0 pcf; w_opt ≤ 1.0%	Rerun entire compaction curve
Consolidation C_c (D2435)	≤ 0.05	Flag and report; use higher value for design
UU triaxial s_u (D2850)	≤ 10% of mean	Investigate specimen preparation
CBR (D1883)	≤ 5 percentage points	Average of three specimens

10.5.2 Reference Standards

The laboratory must maintain certified reference standards for:

Atterberg limits: reference LL material (typical LL ≈ 30%; verified against liquid limit standard)
Proctor: reference soil with known γ_d,max ± 0.5 pcf and w_opt ± 0.5% (checked semiannually)
Sieve analysis: NIST-traceable calibrated sieves (checked annually)
Consolidation: calibrated load frames and displacement transducers (calibrated annually)
Triaxial: calibrated load cells (ASTM E4), pressure transducers, and displacement gauges

10.5.3 Laboratory Report Requirements

Each laboratory report must contain:

Project name, boring ID, sample ID, test date
Test method reference (ASTM standard edition)
Technician and supervisor names
All raw data (loads, deformations, masses, dimensions)
Calculated results with units
Any deviations from standard method noted
Supervisor review signature

Supplemental reference · LEVI Library

Equipment checks that most often go wrong (USACE EM 1110-2-1909)

Temperature: keep rooms used for consolidation, permeability, triaxial (especially with pore pressure) and direct shear within ± 5°F; otherwise results become erratic. EM 1110-2-1906 adds that a temperature rise during undrained shear lowers effective stress — a problem that grows with test duration.
Mechanical compaction rammers: calibrate before first use, after every 500 mold fillings, after any repair, whenever results are questionable, and after any 6-month period without calibration — separately for standard and modified effort and for each foot type. Check drop height at operating speed, not in slow operation.
Consolidometer deformation: for stiff clays and dense sands, a significant part of the measured compression can be in the apparatus and porous stones. Load a dummy setup (wet stones with filter paper, or a metal disk) through the test sequence until readings repeat within about 0.0005 in. between cycles (usually three cycles) and subtract the apparatus correction at each load. In the USACE example, apparatus deformation at 4 tsf was about 0.008 in. — not negligible for an OC clay specimen. Recalibrate each test until the correction proves stable, then semiannually or annually.
Liquid limit device: check new devices on arrival, since many do not meet specification. Check base hardness yearly by dropping a 5/16-in. steel ball from 10 in.: rebound must be 7.3–9.0 in.
Records: keep the date, who calibrated, and the devices and methods used for every calibration. Lab audits for Section 10.8 accreditation will ask for them.

Specimen selection matters more than test count. EM 1110-2-1906 stresses that strength, compressibility and permeability can vary severalfold within a few inches. Specimen selection therefore needs personal inspection of the samples by the laboratory and the design engineer together, not selection from boring logs alone. Its guiding principle: “one good test is not only far better than many poor tests, but is also less expensive.” Undisturbed samples should not be stored for long periods, because physical and chemical changes eventually invalidate them.

Source: U.S. Army Corps of Engineers (1970, with changes through 1986). Calibration of Laboratory Soils Testing Equipment. EM 1110-2-1909 · U.S. Army Corps of Engineers (1970, with changes). Laboratory Soils Testing. EM 1110-2-1906

Case study: 10.16.2 Laboratory variation in Cooroy clay compaction curves (Look 2023), in §10.16 Case Studies.

10.6 Turnaround Time by Client

Table 10-3. Laboratory Test Turnaround Time Requirements by Railroad Client
Test	CSXT	NSR	UPRR	BNSF	CPKC	Amtrak	CN	G&W
Water content, Atterberg, grain size	5 business days	5 business days	5 business days	5 business days	5 business days	5 business days	5 business days	5 business days
Modified Proctor	7 business days	7 business days	7 business days	7 business days	7 business days	7 business days	7 business days	7 business days
CBR (soaked)	10 business days	10 business days	10 business days	10 business days	10 business days	10 business days	10 business days	10 business days
UU Triaxial	10 business days	10 business days	10 business days	10 business days	10 business days	10 business days	10 business days	10 business days
CU Triaxial	15 business days	15 business days	15 business days	15 business days	15 business days	15 business days	15 business days	15 business days
Consolidation (D2435)	14 business days	14 business days	14 business days	14 business days	14 business days	14 business days	14 business days	14 business days
Resilient modulus (T307)	Not required	Not required	Not required	15 business days	15 business days	15 business days	15 business days	Not required

Rush processing (1.5–2× standard fee): available from most accredited labs at 50–70% of standard turnaround times. Confirm availability at project initiation for emergency and fast-track projects.

10.7 Sample Retention Requirements

Table 10-4. Sample Retention Policy by Railroad Client
Railroad	Untested disturbed samples	Tested Shelby tubes	Rock cores	Ballast samples	Governing Std.
CSXT	90 days post-report	1 year post-report	1 year post-report	90 days	CSXT Eng. Std.
NSR	90 days post-report	1 year post-report	1 year post-report	90 days	NSR Eng. Std.
UPRR	6 months post-report	2 years post-report	2 years post-report	6 months	UPRR Eng. Instr. Manual
BNSF	6 months post-report	2 years post-report	2 years post-report	6 months	BNSF Eng. Instructions
CPKC	6 months post-report	2 years post-report	2 years post-report	6 months	CPKC Track Standards
Amtrak	1 year post-report	3 years post-report	3 years post-report	1 year	Amtrak Eng. Std.
CN	6 months post-report	2 years post-report	2 years post-report	6 months	CN Eng. Standards
G&W	90 days post-report	1 year post-report	1 year post-report	90 days	G&W Eng. Bulletin

All nine client railroads reserve the right to request sample re-testing or third-party verification testing within the retention period.

10.8 Laboratory Accreditation Requirements

Table 10-5. Laboratory Accreditation Requirements by Railroad Client
Railroad	Required Accreditation	Scope Verification	Submission Required
CSXT	AASHTO Accreditation Program (AAP) or CCRL	Per-method scope	Yes — with lab report
NSR	AAP or CCRL	Per-method scope	Yes
UPRR	AAP or A2LA	Per-method scope	Yes
BNSF	AAP or A2LA	Per-method scope	Yes
CPKC	AAP (US territory); SCC (Canadian territory)	Per-method scope	Yes
Amtrak	AAP or A2LA; CCRL for NEC	Per-method scope	Yes
CN	AAP (US territory); SCC or BNQ (Canadian territory)	Per-method scope	Yes
G&W	AAP preferred; CCRL accepted	Not required	On request

Key accreditation bodies:

AAP (AASHTO Accreditation Program): Covers all geotechnical and pavement lab tests; scope is per-method
A2LA: ISO/IEC 17025 accreditation; broader scope; accepted by all Class I railroads
CCRL (Cement and Concrete Reference Laboratory): NIST-managed; covers aggregate and concrete tests including LA abrasion and soundness
SCC (Standards Council of Canada): Canadian national accreditation body

10.9 Worked Calculation Examples

Example 10-1: CSXT Bridge — Consolidation Test Analysis

Project context: CSXT Georgia Division bridge approach embankment. Consolidation test on Shelby tube sample from B-3 at 18–20 ft depth (soft gray kaolin clay, CH).

Lab data (at end of each load stage):

σ'_v (tsf)	Void ratio e
0.125	1.820
0.250	1.795
0.500	1.762
1.000	1.698
2.000	1.590
4.000	1.430
8.000	1.235
16.000	1.015

Step 1: Identify σ'_p (Casagrande graphical method)

From the e-log σ'_v plot (described graphically):

Locate point of maximum curvature: approximately σ' = 1.0–1.5 tsf
Casagrande construction: σ'_p ≈ 1.50 tsf

Step 2: Compute C_c (NC range, 2.0–8.0 tsf)

C_c = Δe / Δ(log σ'_v) = (1.590 − 1.235) / (log 8.0 − log 2.0)

= 0.355 / (0.903 − 0.301) = 0.355 / 0.602 = 0.590

Step 3: Compute C_r (OC range, 0.125–1.0 tsf)

C_r = (1.820 − 1.698) / (log 1.000 − log 0.125)

= 0.122 / (0 − (−0.903)) = 0.122 / 0.903 = 0.135

Step 4: OCR

Current effective overburden: σ'_v0 = (γ_sat − γ_w) × z = (115 − 62.4) pcf × 19 ft = 999 psf = 0.50 tsf

OCR = σ'_p / σ'_v0 = 1.50 / 0.50 = 3.0

Conclusion: Moderately OC clay. Use Chapter 12 OC settlement equations for this stratum (Eq. 7-6 below σ'_p; Eq. 7-7 if stress path crosses σ'_p). Report C_c = 0.590, C_r = 0.135, σ'_p = 1.50 tsf, OCR = 3.0 in the engineering properties table (Table 5 of the GDR/FGR).

Example 10-2: BNSF Pavement — CBR and Modified Proctor Correlation

Project context: BNSF Kansas City intermodal yard expansion. Subgrade soil is lean clay (CL); Modified Proctor performed to establish compaction reference, followed by soaked CBR at 95% RC.

Modified Proctor results:

Water Content w (%)	γ_d (pcf)
10.5	111.8
12.0	115.4
14.0	118.2
16.0	117.1
18.0	113.5

Step 1: Identify γ_d,max and w_opt

Peak dry density: γ_d,max = 118.2 pcf at w_opt = 14.0%

Step 2: Target density for CBR specimen

γ_d,target = 0.95 × 118.2 = 112.3 pcf (at 95% RC per BNSF Engineering Instructions)

Step 3: CBR test results at 0.1 in. penetration

Measured load: P_sample = 730 lb; P_standard = 3,000 lb

CBR = (730 / 3,000) × 100% = 24.3%

At 0.2 in. penetration: P_sample = 960 lb; P_standard = 4,500 lb

CBR(0.2) = (960 / 4,500) × 100% = 21.3%

0.1 in. governs: Design CBR = 24.3%

Step 4: Design interpretation

Per Chapter 18, Table 10-1, CBR = 24.3% → Fair subgrade (CBR 20–50 range). BNSF Engineering Instructions require CBR ≥ 20% for intermodal yard pavement subgrade or blanket layer to achieve this. Subgrade marginally adequate; recommend confirming with ≥ 3 CBR tests across the yard footprint.

Example 10-3: NSR Slope — CU Triaxial Effective Parameters

Project context: NSR Appalachian Division cut slope in residual silty clay (CL-ML). CU triaxial test with pore pressure measurement on three specimens.

Test results:

Specimen	σ₃ (psf)	Δu at failure (psf)	q = (σ₁−σ₃)/2 (psf)	p = (σ₁+σ₃)/2 (psf)
A	1,000	400	820	1,820
B	2,000	710	1,540	3,540
C	3,000	1,050	2,290	5,290

Effective stress parameters:

Specimen	σ'₃ = σ₃ − Δu (psf)	σ'₁ = σ₃ + 2q − Δu (psf)
A	600	2,240
B	1,290	4,370
C	1,950	6,530

Step 1: Plot Mohr circles in effective stress space (described):

Circle A: center = (600+2240)/2 = 1,420; radius = (2240−600)/2 = 820 Circle B: center = (1290+4370)/2 = 2,830; radius = (4370−1290)/2 = 1,540 Circle C: center = (1950+6530)/2 = 4,240; radius = (6530−1950)/2 = 2,290

Step 2: Fit Mohr-Coulomb envelope

Using linear regression on (p', q') pairs: slope = sin φ' = (2,290−820)/(4,240−1,420) × (rise/run from regression)

For this dataset (simplified): sin φ' ≈ 0.517 → φ' = 31°

c' intercept ≈ 200 psf (from y-intercept of fitted line)

Step 3: Report

Effective friction angle: φ' = 31°, effective cohesion: c' = 200 psf

Per NSR Engineering Standards Manual, these parameters apply to long-term (drained) slope stability analysis. NSR requires a minimum of two additional specimens for confirmation before finalizing parameters for critical cut slope designs (FS < 1.3 zone).

10.10 Representative Lab Data Table

Table 10-6. Laboratory Test Summary — CSXT Georgia Division Bridge Approach (8 Borings)
Boring	Sample	Depth (ft)	USCS	LL	PL	PI	w (%)	γ_d (pcf)	s_u (tsf)	φ' (°)	c' (psf)	C_c	C_r	σ'_p (tsf)	k (ft/day)
B-1	SS-3	8–10	CL	42	20	22	24.1	—	—	—	—	—	—	—	—
B-1	ST-1	12–14	CH	65	28	37	38.5	78.2	0.24	—	—	0.52	0.11	1.20	—
B-2	SS-4	10–12	ML	35	22	13	18.2	—	—	—	—	—	—	—	—
B-3	ST-1	18–20	CH	71	29	42	42.1	76.5	0.19	—	—	0.59	0.14	1.50	2.1×10⁻⁵
B-3	SS-6	25–27	SC	—	—	4	9.8	—	—	32°	0	—	—	—	—
B-4	ST-2	15–17	CL	48	22	26	29.4	80.1	0.31	—	—	0.44	0.09	1.35	—
B-5	SS-3	8–10	CL	44	21	23	22.8	—	—	—	—	—	—	—	—
B-6	ST-1	12–14	CH	68	27	41	40.2	77.4	0.22	31°	200	0.57	0.12	1.45	—

10.11 Figure Placeholders

Figure 10-1. Consolidation test e-log σ'_v curve for Example 10-1: void ratio vs. log effective vertical stress with C_c, C_r, and σ'_p identified via Casagrande construction.
Figure 10-2. Modified Proctor compaction curve for Example 10-2: γ_d vs. w% with ZAV line, γ_d,max, and w_opt labeled.
Figure 10-3. CBR load-penetration curve for Example 10-2: force vs. penetration (in.) with 0.1-in. and 0.2-in. reference points marked.
Figure 10-4. CU triaxial effective stress Mohr circles for Example 10-3: three circles in effective stress space with Mohr-Coulomb failure envelope.
Figure 10-5. Laboratory test request form: annotated blank showing all required fields (project, sample ID, tests assigned, priority).
Figure 10-6. Chain of custody form: annotated blank showing field origination, transit custody, and laboratory receipt blocks.
Figure 10-7. Sample Quality Designation (SQD) diagram: Δe/e₀ calculation from consolidation test showing how sample disturbance is quantified.
Figure 10-8. Laboratory accreditation scope verification example: AAP scope certificate showing per-method coverage for ASTM D4318, D2435, D2850, D4767.
Figure 10-9. Grain size distribution curves (semilog): four representative samples from a mixed BNSF subgrade profile showing CL, SM, SW, and GP classifications.
Figure 10-10. Laboratory test sequence timeline chart: schedule from sample receipt to final data delivery for a typical 6-boring bridge investigation (Gantt format).

Reference graphics

Flexible Pavement Layers — Subgrade to Asphalt

10.12 Practical Recommendations

Write the laboratory test program before the field crew mobilizes. A well-designed test program begins with the analysis matrix (Table 10-1). Know which tests are needed before drilling begins so that samples are collected, preserved, and shipped appropriately. Changing the test program after samples have dried or disturbed is expensive and unreliable.
Confirm laboratory accreditation per-method, not per-lab. An AAP-accredited lab may be accredited for D2216 and D4318 but not for D4767 (CU triaxial). Always verify the accreditation scope certificate for every test method in the program. All Class I clients will reject data produced outside the accreditation scope.
Ship Shelby tubes within 72 hours of collection. Sample quality degrades with time, handling, and temperature cycling. The SQD classification (Section 10.4.1) will reveal disturbance — but by then the sample is consumed. Minimize handling, seal immediately, and ship promptly.
Always run water content on all samples, even those not assigned to formal tests. Water content takes 30 minutes and costs almost nothing. It documents the in-situ condition of every sample and supports stratigraphic correlation, parameter estimation, and QC. A boring with 20 samples and only 5 water contents is professionally incomplete.
Request electronic data delivery in addition to PDF reports. All major geotechnical laboratories can deliver test data in CSV or Excel format. Request this at the time of laboratory subcontract award. Electronic data is essential for efficient entry into calculation spreadsheets and prevents transcription errors from PDF reports.
Verify specific gravity before computing void ratio in consolidation analysis. The void ratio calculation (e = G_s × γ_w / γ_d − 1) is directly affected by G_s. Assuming G_s = 2.70 for a kaolin-rich clay (G_s may be 2.60–2.65) will overestimate e and underestimate C_c by a meaningful margin. Always measure G_s on the same material used for consolidation testing.
For CPKC and CN Canadian territory projects, confirm metric vs. English unit reporting requirements with the laboratory. Canadian laboratories report in SI units by default. Specify English unit output in the lab subcontract, or include a unit conversion table in the laboratory report appendix, and verify conversions before using values in design.

10.13 Laboratory Testing Checklist

Plain text version

Pre-Field

[ ] Analysis matrix completed (Table 10-1 adapted to project type); tests selected match analytical needs
[ ] Laboratory subcontract awarded; accreditation scope verified per-method for all required tests
[ ] COC forms prepared and printed; field team briefed on COC procedure
[ ] Shelby tube size and quantity confirmed (typically 3-in. OD for ASTM D1587)
[ ] Shipping coolers, foam padding, and waterproof labels procured
[ ] Rush processing option confirmed with laboratory (if project schedule requires)
[ ] Sample retention requirements confirmed with client (Table 10-4)

Field

[ ] Water content taken on every sample (no exceptions)
[ ] Shelby tubes sealed within 2 hours of extraction; ends waxed
[ ] COC form completed for each sample; signed by collector
[ ] Samples shipped within 72 hours; COC transmitted to laboratory with shipment
[ ] Condition at laboratory receipt confirmed (email or call from lab supervisor)
[ ] Test assignments transmitted to laboratory with priority schedule

Laboratory Phase

[ ] 10% duplicate testing performed per QC program (Table 10-2)
[ ] SQD computed for all Shelby tube-based tests; poor quality samples flagged
[ ] All raw data retained by laboratory (not just final results)
[ ] Laboratory report includes technician and supervisor signatures
[ ] Electronic data (CSV or Excel) received and checked against PDF report
[ ] Accreditation certificate attached to laboratory report as QC document

Report Integration

[ ] Lab data master table complete (all samples; Table 10-6 format)
[ ] All design parameters sourced from lab data with traceability to sample ID and test date
[ ] SQD flags noted in engineering properties discussion (Section 6 of GDR/FGR)
[ ] Any test anomalies or out-of-specification results explained in report narrative
[ ] Sample retention calendar established; laboratory notified of client retention requirements

10.14 Cross-Reference Table

Topic	See Also
SPT N-value correction using γ_d and G_s	Chapter 6
In situ test correlation to lab parameters	Chapter 7
USCS classification using Atterberg limits and grain size	Chapter 8
Groundwater (effect on consolidation parameters)	Chapter 9
Bearing capacity using c', φ' from triaxial	Chapter 11
Settlement using C_c, C_r, σ'_p from consolidation	Chapter 12
Slope stability using φ' and c' from CU triaxial	Chapter 13
Liquefaction using fines content and FC correction	Chapter 14
CBR and M_r for pavement design	Chapter 18
Ground improvement parameter selection	Chapter 17
Retaining wall backfill φ' from direct shear	Chapter 16
Subgrade s_u from UU triaxial for subgrade FS	Chapter 2
Reference Proctor for compaction acceptance	Chapter 15
Lab data table format in final report	Appendix E
Lab test data table template (blank)	Appendix B

10.15 Practice Questions

Q1. A consolidation test data set shows the following e-log σ'_v values: at σ'_v = 1.0 tsf, e = 1.42; at σ'_v = 4.0 tsf, e = 1.18. The current effective overburden is σ'_v0 = 0.80 tsf and the Casagrande graphical construction gives σ'_p = 1.60 tsf. Calculate: (a) C_c in the NC range, (b) OCR, and (c) which settlement equation from Chapter 12 applies (NC equation, OC below σ'_p, or OC crossing σ'_p)?

Answer:

(a) C_c = Δe / Δlog σ' = (1.42 − 1.18) / (log 4.0 − log 1.0) = 0.24 / 0.602 = 0.399
(b) OCR = σ'_p / σ'_v0 = 1.60 / 0.80 = 2.0
(c) The soil is OC. The design stress (final σ'_vf) determines which equation applies: if the final effective stress from the new embankment load remains below σ'_p = 1.60 tsf, use OC below σ'_p equation (Eq. 7-6, using C_r). If the final stress exceeds 1.60 tsf, use the OC crossing σ'_p equation (Eq. 7-7, combining C_r and C_c).

Q2. A nuclear gauge reads γ_d,field = 117.4 pcf on a fill material. The Modified Proctor gives γ_d,max = 120.5 pcf. (a) What is RC%? (b) Does this pass CSXT's 95% requirement? (c) The moisture content is 17.2% vs. w_opt = 14.0%. Is the moisture within CSXT's ±3% tolerance?

Answer:

(a) RC = (117.4 / 120.5) × 100% = 97.4%
(b) 97.4% > 95% → PASS ✓
(c) Acceptable range: 14.0% ± 3% = 11.0% to 17.0%. Field MC = 17.2% > 17.0% → Outside tolerance (wet side). Rework or allow to dry; retest before placing next lift.

Q3. Two Shelby tube samples from the same stratum undergo consolidation testing. Sample A gives C_c = 0.45 with SQD "Good to Fair" (Δe/e₀ = 0.06). Sample B gives C_c = 0.31 with SQD "Poor" (Δe/e₀ = 0.12). Which value should be used for design and why?

Answer: Use C_c = 0.45 from Sample A. Sample B's SQD of "Poor" indicates significant sample disturbance during sampling, transport, or extrusion. Disturbance generally artificially reduces apparent stiffness in the early part of the consolidation curve, leading to an underestimated C_c — which would underestimate settlement and produce an unconservative design. Sample A, rated "Good to Fair," provides the more reliable result. Document in the report that Sample B's result was rejected due to poor quality (Lunne et al. 1997 SQD criteria) and that C_c = 0.45 governs settlement calculations for this stratum.

Q4. A CU triaxial test on three specimens gives peak deviator stresses (σ₁ − σ₃) at failure of 1,840 psf, 3,280 psf, and 4,920 psf at confining pressures of 1,000, 2,000, and 3,000 psf respectively. Pore pressures at failure are 450, 820, and 1,200 psf. Calculate: (a) σ'₃ for each specimen, (b) σ'₁ for each specimen, and (c) estimate φ' from the ratio of radii to centers.

Answer:

(a) σ'₃: 1,000−450=550; 2,000−820=1,180; 3,000−1,200=1,800 psf
(b) σ'₁: 550+1,840=2,390; 1,180+3,280=4,460; 1,800+4,920=6,720 psf
(c) Radii: (2390−550)/2=920; (4460−1180)/2=1,640; (6720−1800)/2=2,460 Centers (p'): (2390+550)/2=1,470; (4460+1180)/2=2,820; (6720+1800)/2=4,260 sin φ' ≈ rise/run from p'−q' plot: (2,460−920)/(4,260−1,470) = 1,540/2,790 = 0.552 → φ' ≈ 33.5°

Q5. A soaked CBR specimen gives the following load-penetration data: at 0.1 in. — 880 lb; at 0.2 in. — 1,260 lb. The zero-load reading showed 15 lb due to seating. (a) Correct the loads. (b) Calculate CBR at 0.1 in. and 0.2 in. (c) Report the design CBR.

Answer:

(a) Corrected loads: 0.1 in.: 880 − 15 = 865 lb; 0.2 in.: 1,260 − 15 = 1,245 lb
(b) CBR(0.1) = (865/3,000) × 100% = 28.8%; CBR(0.2) = (1,245/4,500) × 100% = 27.7%
(c) Since CBR(0.1) > CBR(0.2), the design CBR = 28.8% (0.1-in. penetration governs, as is typical). If CBR(0.2) had governed, the load-penetration curve would need to be re-examined for a seating error at low penetration.

10.16 Case Studies

Field case studies by Burt G. Look, from Earthworks: Theory to Practice – Design and Construction (CRC Press/Balkema, 2023), chosen for this chapter’s topics. Each is summarized from the book with its key figures and a closing note on railroad application added by the authors of this manual. Credit for the case material belongs to Look; see item S8 under Supplemental Sources.

Case Study 10.16.1 — Curing time changes compaction and CBR results

Case study · Look (2023) §5.8

Samples of Bundamba clay (plasticity index 47%, 97% passing No. 40, 69%–82% fines) were cured for 0, 1, 4 and 7 days before compaction and soaked and unsoaked CBR testing:

Short curing gave a higher maximum dry density, so the field material is more likely to fail its density ratio or need extra compactive effort.
Short curing gave higher CBR swell, which would trigger more remove-and-replace than is needed (typically 10 in. for the swell measured). Whether the sample arrived wet or dry also changed the swell when curing was inadequate.
Short curing gave a higher unsoaked CBR (thinner pavement, more long-term failures) and a lower soaked CBR (over-designed pavement).

For railroad work: plastic clay subgrade samples for CBR, resilient modulus or Proctor testing should be cured to the period the test standard requires, and the curing time should be reported. A test turned around too quickly can push the design either way.

**Figure 10-LK1.** Curing 0–1 days gave a higher MDD than 4–7 days for a high-plasticity clay. *Source: Look (2023), Figure 5.8-1, p. 143.*

**Figure 10-LK2.** Curing time against soaked CBR and swell: short curing lowers soaked CBR and raises swell. *Source: Look (2023), Figure 5.9-1, p. 143.*

Source: Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. CRC Press/Balkema. Case study §5.8, pp. 142. Summarized and paraphrased; figures reproduced for internal reference. Full reference under Supplemental Sources.

Case Study 10.16.2 — Laboratory variation in Cooroy clay compaction curves

Case study · Look (2023) §7.9

Related section: 10.5 Laboratory QA/QC

Eleven standard compaction tests on clay from one project site at Cooroy, Queensland (55 compaction points) gave maximum dry densities from 97–104 pcf, median 1.62, and optimum moisture contents from 18.6% to 24.5%, median 21.5%. The idea of a unique MDD and OMC does not hold for clays. The repeatability allowed in the Australian standard suggests 85%–115% of OMC as a realistic moisture window, yet many specifications ask for ±2% of OMC, which testing variation alone makes unachievable.

The real target is low air voids: much below 85% of OMC, air voids exceed about 10%. Look also warns that the compacted soil will not stay at its placement moisture, so the moisture target should guide field effort and not be a pass/fail criterion. Density is the pass/fail measure. Compaction well above MDD or dry of OMC raises suction, so the “stronger” material is more sensitive to later moisture change.

For railroad work: the ±2% moisture band suits processed granular materials such as subballast, not high-moisture clays in embankment or subgrade. Use a band that reflects the laboratory’s own repeatability.

**Figure 10-LK3.** Individual five-point compaction curves for 11 samples of Cooroy clay: MDD 97–104 pcf, OMC 18.6%–24.5%. *Source: Look (2023), Figure 7.9-1, p. 233.*

Source: Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. CRC Press/Balkema. Case study §7.9, pp. 230–232. Summarized and paraphrased; figures reproduced for internal reference. Full reference under Supplemental Sources.

Author and Sources

Authors: Daniel P. Colwell and Mario Ruel. This chapter was written and compiled by Daniel P. Colwell and Mario Ruel for Railway Ground Hazard Risk Management.

Technical content is drawn from the published sources below, and credit for that material belongs to their authors and publishers. Tables, equations and figures taken from a specific source are also cited where they appear.

Primary sources

USACE EM 1110-2-1906, Laboratory Soils Testing
USACE EM 1110-2-1909, Calibration of Laboratory Soils Testing Equipment
Das, B. M. (2002). Soil Mechanics Laboratory Manual, 6th ed.
ASTM D854, D1557, D1587, D1883, D2166, D2216, D2434, D2435, D2850, D3080, D4318, D4644, D4767, D5084, D5731, D6913, D698, D7928; AASHTO T 307

Supplemental sources (LEVI Library)

U.S. Army Corps of Engineers (1970, with changes through 1986). Calibration of Laboratory Soils Testing Equipment. EM 1110-2-1909
U.S. Army Corps of Engineers (1970, with changes). Laboratory Soils Testing. EM 1110-2-1906
Das, B. M. (2002). Soil Mechanics Laboratory Manual, 6th ed. Oxford University Press
Nagaraj, T. S. (1993). Principles of Testing Soils, Rocks and Concrete. Elsevier (Developments in Geotechnical Engineering)
Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. CRC Press/Balkema. Case studies §5.8, §7.9

File locations for these documents are listed under Supplemental Sources — LEVI Library.

Supplemental Sources — LEVI Library

Material in the green “Supplemental reference” boxes throughout this chapter was added from the following documents in the LEVI geotechnical library. Text is summarized or paraphrased; tables are reproduced with their original source noted; figures are reproduced from the cited document for internal reference use. Consult the original document before relying on a value for design. Blocks marked “After Selig & Waters (1994)” summarize item S7 for railroad clients; tables are condensed from the book and its figures are not reproduced. The Case Studies section summarizes case studies from item S8 by Burt G. Look; its figures are reproduced from the book for internal reference use, and the closing “For railroad work” note in each case study is added by the authors of this manual.

U.S. Army Corps of Engineers (1970, with changes through 1986). Calibration of Laboratory Soils Testing Equipment. EM 1110-2-1909
U.S. Army Corps of Engineers (1970, with changes). Laboratory Soils Testing. EM 1110-2-1906
Das, B. M. (2002). Soil Mechanics Laboratory Manual, 6th ed. Oxford University Press
Nagaraj, T. S. (1993). Principles of Testing Soils, Rocks and Concrete. Elsevier (Developments in Geotechnical Engineering)
Profillidis, V. A. (2014). Railway Management and Engineering, 4th ed. Farnham, UK: Ashgate. Chapters 7–9, 11–13, 16–17, 20 and 22 (track system, mechanical behavior of track, subgrade, sleepers, ballast, transverse effects, maintenance, slab track, electrification foundations, safety)
Benesch. Geotechnology for Railroaders Program, Modules 1–8 (two-day training course reference: risk management, geology and soils, ballast and sub ballast, drainage, slides, slope and subgrade stabilization, scour, rock falls, beaver dams, winter problems, spring thaw, extreme weather, earthquakes). Field photographs and figures from the course; slope stabilization process slides adapted from material by D. Colwell
Selig, E. T. and Waters, J. M. (1994). Track Geotechnology and Substructure Management. London: Thomas Telford. Used on this page: chs. 3, 7, 8, 9 and 10
Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. Leiden: CRC Press/Balkema (Taylor & Francis). ISBN 978-1-032-10470-6. Case studies used on this page: §5.8, §7.9

Referenced in this chapter

Also applicable to this chapter

Ballast, subballast and fouling test suite

Repeated-load testing for track subgrade

Strength parameters: match the test to the drainage condition

USACE Q / R / S nomenclature and common triaxial errors

Picking t50 and t90 from each load increment

Tests for shale and mudstone: preparation effects and slake durability

Ballast and subballast abrasion tests (Deval, Los Angeles, Micro-Deval)

Standard index tests for subgrade samples

Equipment checks that most often go wrong (USACE EM 1110-2-1909)

Laboratory Testing Checklist

Primary sources

Supplemental sources (LEVI Library)

Picking t₅₀ and t₉₀ from each load increment