Worked example
Example: Wills Example 3.3: lead ore, 25 mm crusher product. The inputs, in the units selected above:
| Input | Value |
|---|---|
| Top size (95 % passing) | 25.000 mm |
| Lower size (5 % passing) | 1.0000 mm |
| Liberation size | 150.00 µm |
| Mineral content (mass fraction) | 5.8000 % |
| Density of the valuable mineral | 7.6000 g/cm³ |
| Density of the gangue | 2.6500 g/cm³ |
| Particle shape | Most ores (f = 0.5) |
| Required relative standard deviation | 1.0000 % |
| Mass of the lot sampled | 1,000.0 t |
Applying the formula below gives:
| Result | Value |
|---|---|
| Size distribution factor | 0.25000 |
| Liberation factor | 0.077460 |
| Mineralogical factor | 118.77 g/cm³ |
| Sampling constant | 1.1500 g/cm³ |
| Minimum sample mass | 179.66 kg |
| Sample mass with Wills's allowance for other errors | 359.31 kg |
Open this example in the calculator and change any input.
Formula
\(\sigma^2 = \left(\frac{1}{m_s} - \frac{1}{M}\right) C d^3, \quad C = f g l m, \quad l = \sqrt{L/d}, \quad m = \frac{1 - a}{a}\left[(1 - a)\rho_r + a\rho_t\right]\)
Variables
| Symbol | Meaning | SI unit |
|---|---|---|
| \(d_{95}\) | Top size (95 % passing) | m |
| \(d_5\) | Lower size (5 % passing) | m |
| \(L\) | Liberation size | m |
| \(a\) | Mineral content (mass fraction) | - |
| \(\rho_r\) | Density of the valuable mineral | kg/m³ |
| \(\rho_t\) | Density of the gangue | kg/m³ |
| \(\text{ore}\) | Particle shape | - |
| \(\sigma\) | Required relative standard deviation | - |
| \(M\) | Mass of the lot sampled | kg |
| \(g\) | Size distribution factor | - |
| \(l\) | Liberation factor | - |
| \(m\) | Mineralogical factor | kg/m³ |
| \(C\) | Sampling constant | kg/m³ |
| \(m_s\) | Minimum sample mass | kg |
| \(2m_s\) | Sample mass with Wills's allowance for other errors | kg |
Method and standards
Gy's fundamental sampling error.
Where it is used
- Designing sampling protocols for plant feed and concentrates.
- Deciding where to crush before splitting.
Common mistakes
- Using d50 instead of the top size.
- Forgetting the other sources of sampling error.
Assumptions and limits
- Two-component ore; Gy's factor values.
References
- Wills and Finch
- Wills' Mineral Processing Technology
- 8th ed. (Butterworth-Heinemann, 2016)
- ch. 3
- eqs. 3.4-3.6 (Gy, 1967, 1982) and Example 3.3
Questions and answers
Why does the top size matter so much?
The error varies with the cube of the particle size; crushing a sample before splitting it reduces the mass needed by orders of magnitude.
What precision should I enter?
The relative standard deviation: for ±0.1 % Pb at 95 % confidence on a 5 % Pb ore, 2σ = 0.02, so σ = 1 %.
What do the defaults represent?
Wills's Example 3.3: 5 % Pb ore (5.8 % galena), crushed to 25 mm, galena liberated at 150 µm, σ = 1 %; Wills gets 177 kg with a rounded factor m.
Method and article reviewed by FireCAD engineering on 17 November 2026. Every result is checked against an independent calculation on each release: how we verify our calculators.