SPT Procedure (ASTM D1586)

The Standard Penetration Test is the most widely used in-situ soil test in North America. Procedure: a 51-mm OD split-spoon sampler is driven into the soil at the bottom of a borehole by a 63.5 kg (140 lb) hammer falling 762 mm (30 in). The N-value is the blow count for the second and third 152 mm (6 in) increments of penetration (the first 152 mm is a seating drive and is discarded). Total penetration per test is 457 mm (18 in).

The split spoon is retrieved and opened to provide a disturbed but representative sample for classification, moisture content, and visual examination. This dual function โ€” in-situ strength indication and sample recovery โ€” explains SPT's enduring popularity despite its acknowledged variability.

Energy Corrections

The raw N-value is meaningless without energy correction because hammer efficiency varies significantly by rig type, hammer type, rod length, and borehole diameter. The reference energy is 60% of theoretical free-fall energy (= 0.60 ร— 63.5 ร— 9.81 ร— 0.762 = 474 J baseline). The corrected blow count is:

N60 = N_measured ร— (CE ร— CB ร— CR ร— CS) / 0.60

  • CE = hammer energy ratio (ER/60); donut hammer โ‰ˆ 0.45โ€“1.0; safety hammer โ‰ˆ 0.72โ€“1.20; automatic trip hammer โ‰ˆ 0.90โ€“1.60. Measure with PDA or calibrate per ASTM D4633.
  • CB = borehole diameter correction; 1.00 for 65โ€“115 mm diameter; 1.05 for 150 mm; 1.15 for 200 mm.
  • CR = rod length correction; 0.75 for < 3 m; 0.85 for 3โ€“4 m; 0.95 for 4โ€“6 m; 1.00 for > 6 m.
  • CS = sampler liner correction; 1.00 with liners (standard); 1.10โ€“1.30 without liners (when liner is omitted, sample is less confined and N is slightly higher).

For overburden normalization to 100 kPa (1 atm) effective stress:

(N1)60 = N60 ร— CN where CN = (Pa / ฯƒ'v)^0.5 and Pa = 100 kPa (atmospheric pressure). CN should not exceed 2.0 (Liao and Whitman, 1986).

Correlations for Sands

In granular soils, (N1)60 correlates to relative density Dr and friction angle ฯ†':

  • Relative density: Drยฒ = (N1)60 / 60 (Meyerhof, 1957); more precisely, Dr = โˆš[(N1)60 / C2] where C2 โ‰ˆ 41โ€“46 depending on gradation (Skempton, 1986). Typical ranges: N60 < 4 โ†’ very loose; 4โ€“10 โ†’ loose; 10โ€“30 โ†’ medium dense; 30โ€“50 โ†’ dense; > 50 โ†’ very dense.
  • Friction angle: ฯ†' โ‰ˆ 27.5 + 9.2ยทlog(N60) (Peck, Hanson, Thornburn) or the Hatanaka and Uchida (1996) correlation: ฯ†' = โˆš(20ยท(N1)60) + 20ยฐ for (N1)60 between 1 and 50.

Correlations for Clays

In fine-grained soils, N correlates to undrained shear strength Su, but with much greater uncertainty:

  • Terzaghi and Peck: Su (kPa) โ‰ˆ 6.25ยทN60 (conservative for stiff clays; may overestimate for soft clays)
  • Stroud (1974): Su = f1ยทN60 where f1 ranges from 4 kPa/blow (plastic clays, PI > 30) to 6 kPa/blow (low plasticity, PI < 15).
  • Consistency descriptors: N < 2 โ†’ very soft; 2โ€“4 โ†’ soft; 4โ€“8 โ†’ medium; 8โ€“15 โ†’ stiff; 15โ€“30 โ†’ very stiff; > 30 โ†’ hard.

These correlations carry ยฑ50% uncertainty in clays. Unconfined compression tests (ASTM D2166) or vane shear (ASTM D2573) should be used when Su is critical to design.

Bearing Capacity from SPT

Meyerhof (1956) and Bowles provide direct SPT-to-bearing-capacity correlations for sands. A widely used approximation for a 25 mm (1 in) settlement allowance:

q_allow (kPa) โ‰ˆ 12ยทN60 for B โ‰ค 1.2 m; q_allow โ‰ˆ 8ยทN60ยท[(B + 0.3)/(B)]ยฒ for B > 1.2 m

These are order-of-magnitude estimates only. AASHTO LRFD Section 10.6 provides more rigorous SPT-based methods with defined resistance factors.

Liquefaction Screening

The simplified Seed-Idriss method uses (N1)60 to determine CRR (Cyclic Resistance Ratio). The threshold (N1)60 for liquefaction triggering at Mw = 7.5 is approximately 15 for clean sands with < 5% fines. Soils with (N1)60 > 30 are generally considered non-liquefiable. Fines content adjustments (ฮ”(N1)60) increase the apparent (N1)60cs for the CRR chart. Full procedure covered in Liquefaction Assessment.

Limitations and When to Use CPT

The SPT has significant limitations that every geotechnical engineer should internalize:

  • High variability between operators, rigs, and borehole conditions โ€” coefficient of variation on N can exceed 25%.
  • Meaningless in gravels (N refusal) and highly unreliable in soft clays (N < 2).
  • Provides a point measurement every 1.5 m; misses thin layers.
  • Disturbed sample โ€” no undisturbed clay sample is recovered.

The Cone Penetration Test (CPT) per ASTM D5778 is preferred when continuous profiling is needed, when thin weak layers could control design, in sensitive clays, or for higher-quality liquefaction assessment. CPT provides continuous qc, fs, and u2 data at ~2 cm intervals. Its repeatability and the absence of borehole disturbance effects make CPT-based correlations significantly more reliable than SPT-based ones for critical applications.

Best practice is to use SPT borings for sample recovery and CPT soundings for continuous profiling on the same site, cross-correlating via Robertson's soil behavior type (SBTn) chart.