Foundation Engineering / Pile capacity in clay — the α-method
How much α matters
A factor of three, sitting on the term that carries the pile.
Lesson review: expert verification remains open
This lesson has received a full-context editorial review and a check of its technical explanations. Selected arithmetic has been checked separately. This is not independent engineering sign-off.
Corrections checked in this pass
- Aligned the adhesion-curve marker with the actual stress-ratio axis.
Checks needed for project use
- A geotechnical reviewer must confirm drainage and stress basis, bearing-factor formulation, correlations and settlement influence factors. Partial factors and project-specific ground conditions are outside these teaching checks.
- Select adhesion method for soil, installation, effective stress and loading history; the curve is an empirical comparison.
These are educational models and exercises. Check their assumptions before using the results. The full technical content and every diagram have not yet been independently verified.
Concrete and steel calculations retain their stated first-generation Eurocode teaching basis. Use the standards and amendments specified for your project; do not mix generations without an appropriate design basis.
Read the lesson register and outstanding review notesWhy this matters
Because α multiplies the shaft-resistance term, its selected range can strongly affect the result. This lesson tests that sensitivity and considers what additional evidence would support a choice.
Different authorities quote α for a stiff clay anywhere between about 0.3 and 1.0. That is not scatter around a value; it is disagreement about the value.
Applied to the worked pile's lower clay over the full 18 m, α = 0.3 gives 916 kN of shaft resistance and α = 1.0 gives 3054 kN — a factor of 3.3. This sensitivity warrants attention alongside the ground model, base condition and other controlling assumptions.
The disagreement is not incompetence. α depends on installation method, on time since installation, on whether the clay is fissured, on the concrete mix and on how the hole was supported. Correlations fitted to driven piles in one clay genuinely do not transfer to bored piles in another.
The proper response is a load test. A preliminary pile taken to failure on the actual site, with the actual rig and the actual clay, back-figures an α that no correlation can match. On a scheme with many piles it pays for itself several times over, because it usually allows shorter piles than a cautious correlation would.
Where a preliminary test is not possible, the responses in descending order of strength are: a correlation calibrated on comparable clay with the same installation method, cited; the same with a reduced value to reflect the transfer; and a value chosen at the conservative end of the published range with the sensitivity stated. What is never acceptable is a single mid-range α presented without its provenance, because it looks identical on the page to a value that was actually established.
Check yourself
Why does the adhesion factor α fall as the ratio su/σ′v increases?
Check yourself
A scheme needs 400 piles in a stiff clay for which published α values range from 0.3 to 1.0. What is the strongest response?
Try it
The adhesion factor
α falls as the clay stiffens relative to its stress level. Move both and watch which way it goes.
Why it falls
- Installation remoulds a thin annulus, and the shaft never meets the intact clay.
- A heavily overconsolidated clay loses proportionally more strength on remoulding.
- So α is correlated against su/σ′v, not against su.
- Strength ratio su/σ′_v
- 0.75
- Adhesion factor α
- 0.577
- Unit shaft resistance
- 52.0 kPa
- Shaft resistance Qs
- 1763 kN
- Base resistance Qb
- 229 kN
- Shaft share
- 89 %
Shaft resistance across the quoted range α = 0.3 to 1.0
Across the α values different authorities publish for a stiff clay, this shaft is worth anywhere between 916 and 3054 kN — a factor of 3.33. That is the single largest uncertainty in a pile design in clay, and a site load test is the only thing that removes it.
Summary
- α is below 1 because installation leaves a remoulded, softened annulus against the shaft
- α falls as su/σ′v rises — so it was higher in the stronger, deeper clay, not lower
- Apply it layer by layer; averaging su discards the profile and errs unsafely
- Nc = 9 for a deep confined base, against 5.14 at the surface
- α between 0.3 and 1.0 moved shaft resistance by a factor of 3.3 — buy a load test