5.3 Failure Modes, Leak vs. Rupture Morbidities, and Degradation Rate Determination

Key Takeaways

  • API RP 580 defines failure modes based on the physical breach geometry of loss of containment: small pinhole leak, medium hole leak, large hole leak, and catastrophic pressure boundary rupture.
  • The failure mode directly dictates consequence morbidity—pinhole leaks lead to manageable localized release/toxic contours, whereas catastrophic rupture results in immediate mass inventory release, jet fires, or vapor cloud explosions (VCE).
  • Structural failure modes (thinning/corrosion allowance consumption) typically exhibit leak-before-break behavior, whereas localized cracking, brittle fracture, and HTHA can trigger full rupture without prior warning.
  • Degradation rates must be established using a hierarchy of data sources: measured equipment-specific inspection data (highest confidence), historical component data in identical service, published corrosion tables (API RP 581), or expert corrosion engineering modeling (lowest confidence).
  • Equipment degradation rates are dynamic; operational changes, feedstock changes (e.g., processing high-TAN or high-sulfur crude), and IOW exceedances require re-evaluating corrosion rates and updating RBI risk profiles via Management of Change (MOC).
Last updated: August 2026

Physical Failure Modes in API RP 580 Risk Assessments

In API RP 580 and API RP 581, a Failure Mode describes the specific physical geometry and structural manifestation of a loss of containment (LOPC) event. Defining the failure mode is essential because it governs the release rate of hazardous process fluids, directly determining the Consequence of Failure (COF).

API RP 581 classifies loss of containment into four standardized hole size categories:

  • Small Hole ($1/4 \text{ in.}$ / $6.4 \text{ mm}$): Pinhole leaks caused by localized pitting, small packing/gasket weeping, or micro-crack penetration.
  • Medium Hole ($1 \text{ in.}$ / $25 \text{ mm}$): Moderate wall breach resulting from localized corrosion, small nozzle failure, or severe flange leaks.
  • Large Hole ($4 \text{ in.}$ / $102 \text{ mm}$): Major nozzle severance, severe pipe wall rupture, or large localized corrosion blowout.
  • Catastrophic Rupture ($>4 \text{ in.}$ or Full Diameter): Instantaneous structural detachment of vessel shell plate, head separation, or full-bore longitudinal line burst.

Leak-Before-Break (LBB) Dynamics vs. Catastrophic Rupture

The physical failure mode is governed by material toughness, degradation mechanism type, and stress state:

┌─────────────────────────────────────────────────────────────────────────┐
│                     FAILURE MODE MORBIDITY DYNAMICS                     │
└───────────────────────────────────┬─────────────────────────────────────┘
                                    │
        ┌───────────────────────────┴───────────────────────────┐
        ▼                                                       ▼
┌───────────────────────────────────────┐   ┌───────────────────────────────────────┐
│ LEAK-BEFORE-BREAK (LBB) DYNAMICS      │   │ CATASTROPHIC RUPTURE DYNAMICS         │
├───────────────────────────────────────┤   ├───────────────────────────────────────┤
│ • Ductile wall thinning / pitting     │   │ • Environmental Cracking (SCC, SOHIC) │
│ • Small pinhole leak (1/4" hole size) │   │ • HTHA / Creep / Brittle Fracture     │
│ • Low mass release rate (W_n)         │   │ • Full-bore split (>4" hole size)     │
│ • Low consequence area (COF)          │   │ • Instant mass release & VCE blast    │
└───────────────────────────────────────┘   └───────────────────────────────────────┘

Leak-Before-Break (LBB) Concept

In ductile metals (such as carbon steel operating above its ductile-to-brittle transition temperature), general or localized thinning slowly consumes wall thickness. As wall thickness reaches minimum structural required thickness ($t_{\text{min}}$), localized yielding creates a small pinhole leak ($1/4 \text{ in.}$ hole). The small leak relieves internal pressure before circumferential hoop stress reaches the ultimate tensile strength of the full cylinder, preventing gross unstable crack propagation. LBB provides early warning via hydrocarbon detectors or visual inspection, enabling safe unit shutdown.

Catastrophic Rupture Mechanics

Instantaneous rupture occurs when degradation mechanisms bypass the ductile leak stage:

  • Environmental Cracking (SCC, SOHIC): Deep micro-cracks reduce effective cross-sectional toughness. When crack depth reaches critical stress intensity ($K_{IC}$), unstable fast fracture propagates at sonic speed along the entire length of the component.
  • High-Temperature Hydrogen Attack (HTHA): Trapped methane microfissures weaken grain boundaries across large volumetric areas, causing sudden wall blowout under normal operating pressure.
  • Brittle Fracture: Low-temperature operation below Charpy V-notch impact transition temperatures causes sudden cleavage fracture without prior plastic deformation.

Release Rate and Consequence Morbidity

The mass release rate ($W_n$) for a liquid or gas release through a failure hole is modeled by fluid mechanics:

Wn=CdAn2ρΔPW_n = C_d \cdot A_n \cdot \sqrt{2 \rho \Delta P}

Where $C_d$ is the discharge coefficient, $A_n$ is the hole area, $\rho$ is fluid density, and $\Delta P$ is pressure differential across the boundary. A catastrophic rupture ($A_n \gg 12.5 \text{ in.}^2$) generates mass release rates several orders of magnitude greater than a pinhole leak ($A_n = 0.049 \text{ in.}^2$). This results in rapid flash vaporization, massive flammable vapor cloud formation, jet fires, or high-overpressure Vapor Cloud Explosions (VCE), exponentially multiplying the calculated COF.


Hierarchy of Degradation Rate Determination (API RP 580 Section 9)

Determining accurate degradation rates (corrosion rates expressed in $\text{mpy}$ or $\text{mm/yr}$) is essential for calculating the time-dependent Damage Factor $DF(t)$. API RP 580 Section 9 establishes a 4-tier hierarchy of data sources, ranked by data confidence:

Hierarchy LevelData Source DescriptionConfidence LevelAPI RP 581 Damage Factor Uncertainty Impact
1. Measured Equipment DataSequential ultrasonic thickness measurements at dedicated TMLs over known operating intervals.HighestLowest data variance ($\text{Art} \approx 1.0$), minimizing calculated $DF_{\text{thin}}$.
2. Similar Equipment DataInspection history from identical metallurgy and process fluid service within the same facility.HighLow uncertainty scaling multiplier applied to baseline corrosion rate.
3. Published Industry ModelsCorrosion rates from API RP 581 tables, NACE/AMPP literature, or validated modeling software.ModerateModerate uncertainty multiplier applied, requiring safety margin.
4. Expert Engineering JudgementEstimates derived by corrosion specialists when empirical data is unavailable.LowestHighest data variance multiplier applied, artificially elevating calculated $DF_{\text{thin}}$.

Calculating Measured Corrosion Rates

When thickness data is available, API RP 580 utilizes short-term and long-term corrosion rates:

CRshort-term=tprevioustactualΔtrecent\text{CR}_{\text{short-term}} = \frac{t_{\text{previous}} - t_{\text{actual}}}{\Delta t_{\text{recent}}}

CRlong-term=tinitialtactualΔttotal\text{CR}_{\text{long-term}} = \frac{t_{\text{initial}} - t_{\text{actual}}}{\Delta t_{\text{total}}}

If process conditions have remained stable, the long-term rate reflects baseline kinetics. However, if recent process changes or IOW exceedances have occurred, the short-term rate governs.


Dynamic Degradation Rates & Management of Change (MOC)

Degradation rates are not static values. Operational shifts, feedstock changes, and IOW exceedances alter corrosion kinetics. Under API RP 580 Section 15, any change in process operation requires invoking Management of Change (MOC) protocols to update the RBI assessment:

  • Crude Slate Modifications: Processing opportunity crudes with higher Total Acid Number (TAN) or higher sulfur content accelerates sulfidation and naphthenic acid corrosion rates.
  • Wash Water System Interruptions: Loss of wash water in hydrotreater overheads allows ammonium chloride salt accumulation, accelerating localized corrosion rates from $5 \text{ mpy}$ to over $200 \text{ mpy}$.
  • IOW Exceedances (API RP 584): Operating above maximum temperature limits accelerates creep life consumption and HTHA susceptibility.

Technical Worked Example: Corrosion Rate & POF Risk Trajectory

Problem Statement

An 8-inch Schedule 40 Carbon Steel Distillate Line ($D_o = 8.625 \text{ in.}$, nominal thickness $t_{\text{initial}} = 0.322 \text{ in.}$, $t_{\text{min}} = 0.110 \text{ in.}$) was commissioned in 2016. TML inspection history:

  • 2016 Baseline: $t_{2016} = 0.322 \text{ in.}$
  • 2021 Inspection: $t_{2021} = 0.262 \text{ in.}$
  • 2026 Inspection: $t_{2026} = 0.172 \text{ in.}$

Step 1: Corrosion Rate Analysis

CRlong-term=0.3220.17210 years=0.015 in/yr=15 mpy\text{CR}_{\text{long-term}} = \frac{0.322 - 0.172}{10 \text{ years}} = 0.015 \text{ in/yr} = 15 \text{ mpy}

CRshort-term=0.2620.1725 years=0.018 in/yr=18 mpy\text{CR}_{\text{short-term}} = \frac{0.262 - 0.172}{5 \text{ years}} = 0.018 \text{ in/yr} = 18 \text{ mpy}

An MOC audit reveals that unit throughput increased in 2021, raising fluid velocity. The governing corrosion rate is set to the higher short-term rate: $\text{CR}_{\text{gov}} = 18 \text{ mpy}$ ($0.018 \text{ in/yr}$)....

Step 2: Remaining Life & Inspection Interval Calculation

Remaining Life (RL)=t2026tminCRgov=0.1720.1100.018=3.44 years\text{Remaining Life (RL)} = \frac{t_{2026} - t_{\text{min}}}{\text{CR}_{\text{gov}}} = \frac{0.172 - 0.110}{0.018} = 3.44 \text{ years}

Under API RP 580 principles, as remaining life drops below 4 years, the Damage Factor $DF_{\text{thin}}$ escalates exponentially. To maintain equipment risk within the As Low As Reasonably Practicable (ALARP) zone, the RBI plan targets a Category A ultrasonic grid inspection within 1.72 years (half of remaining life), or mandates pipe spool replacement during the upcoming turnaround.

Loading diagram...
Degradation Rate Determination, Failure Modes, and Release Morbidity
Test Your Knowledge

According to API RP 580, which failure mode is most characteristic of ductile materials experiencing localized pitting wall loss under a Leak-Before-Break scenario?

A
B
C
D
Test Your Knowledge

Which source of degradation rate data provides the highest level of confidence and results in the lowest uncertainty factor in API RP 580/581 Damage Factor calculations?

A
B
C
D
Test Your Knowledge

How should an RBI assessment team respond under API RP 580 Management of Change (MOC) guidelines when a process unit permanently increases feedstock sulfur and naphthenic acid content?

A
B
C
D