Worked example
Example: 0.01 m² coupon to 100 m², 6 mm wall. The inputs, in the units selected above:
| Input | Value |
|---|---|
| Deepest pit in the inspected sample | 0.80000 mm |
| Average penetration (from weight loss) | 0.10000 mm |
| Gumbel scale parameter of pit depths | 0.12000 mm |
| Inspected sample area | 0.010000 m² |
| Total structure area | 100.00 m² |
| Wall thickness | 6.0000 mm |
Applying the formula below gives:
| Result | Value |
|---|---|
| Pitting factor | 8.0000 |
| Expected deepest pit over A | 1.9052 mm |
| Remaining wall at the deepest pit | 4.0948 mm |
Open this example in the calculator and change any input.
Formula
\(PF = \frac{p_{max}}{\bar d}, \quad d_A = p_{max} + \alpha\ln\frac{A}{a}\)
Variables
| Symbol | Meaning | SI unit |
|---|---|---|
| \(p_{max}\) | Deepest pit in the inspected sample | m |
| \(\bar d\) | Average penetration (from weight loss) | m |
| \(\alpha\) | Gumbel scale parameter of pit depths | m |
| \(a\) | Inspected sample area | m² |
| \(A\) | Total structure area | m² |
| \(t\) | Wall thickness | m |
| \(PF\) | Pitting factor | - |
| \(d_A\) | Expected deepest pit over A | m |
| \(t - d_A\) | Remaining wall at the deepest pit | m |
Method and standards
Extreme-value statistics.
Where it is used
- Tank floors and pipe inspection.
- Cooling-water exchanger tubes.
Common mistakes
- Using too few sample areas to fit α.
- Treating average loss as the life-limiting value.
Assumptions and limits
- Gumbel distribution of maximum pit depths.
- The deepest pit of one sample stands in for the Gumbel location parameter λ; the standard procedure fits λ and α from the deepest pits of several samples. The result is a best estimate with no confidence level: add a margin, or use the fitted λ, for an inspection decision.
References
- Pitting factor (ASTM G46) and Gumbel extreme-value area scaling d_A = d_a + α ln(A/a) (Shibata; Jones, Principles and Prevention of Corrosion, 2nd ed. (Prentice Hall, 1996))
Questions and answers
What does the pitting factor show?
How much deeper the worst pit is than the average loss: 1 means uniform corrosion, large numbers mean localised attack.
Why extrapolate to the whole area?
The larger the area, the deeper the worst pit is likely to be; samples underestimate the structure's worst pit.
What do the defaults represent?
Pits found on a 0.01 m² coupon extrapolated to 100 m² of 6 mm wall.
Method and article reviewed by FireCAD engineering on 13 November 2026. Every result is checked against an independent calculation on each release: how we verify our calculators.