4.7 Comparative Failure-Mode Synthesis & Integrated Flaw and Creep Assessment

7. Comparative Synthesis of Structural Failure Modes

Table 4.3-1: Mechanical Failure Modes in Welded Construction

---

8. Comprehensive Worked Numerical Example: LEFM Flaw Assessment & Larson-Miller Creep Life Prediction

Problem Statement

An engineering critical assessment (ECA) is performed on a welded energy infrastructure facility involving two independent structural failure evaluations:

Part 1: LEFM Critical Flaw Sizing on a Pressure Vessel A cylindrical pressure vessel shell (inner diameter $D_i = 2,000.0\text{ mm}$, wall thickness $t = 40.0\text{ mm}$) fabricated from quenched and tempered steel ($\sigma_y = 480.0\text{ MPa}$, plane strain fracture toughness $K_{Ic} = 55.0\text{ MPa}\sqrt{\text{m}}$ at MDMT of $-20^\circ\text{C}$) operates at internal pressure $p = 8.0\text{ MPa}$. Transverse welding residual stresses in the un-PWHT joint contribute an additional tensile stress $\sigma_{\text{res}} = 60.0\text{ MPa}$ parallel to the hoop direction. Phased array ultrasonic testing (PAUT) reveals an external longitudinal surface crack with geometric factor $Y = 1.12$.

• Calculate the critical crack depth $a_c$ that would trigger catastrophic brittle fracture at $-20^\circ\text{C}$.
• Verify whether plate thickness satisfies ASTM E399 plane strain validity.

Part 2: Larson-Miller Creep Life Extrapolation A high-energy Grade 91 steam line ($T_m = 1,510^\circ\text{C} = 1,783\text{ K}$) operates at nominal steam temperature $T_1 = 575^\circ\text{C}$ ($T_{K1} = 848.15\text{ K}$) under an effective hoop stress of $85.0\text{ MPa}$. Experimental creep rupture data for Grade 91 cross-weld joints at $85.0\text{ MPa}$ establishes a Larson-Miller parameter $LMP = 25.40$ (using $C = 20$, with $T$ in Kelvin and $LMP$ scaled by $10^{-3}$).

• Calculate the projected creep rupture life $t_{r1}$ in operating hours.
• If a boiler burner tilt malfunction causes a continuous $35^\circ\text{C}$ temperature excursion to $T_2 = 610^\circ\text{C}$ ($T_{K2} = 883.15\text{ K}$), calculate the new rupture life $t_{r2}$ and the percentage loss of operational life.

Step-by-Step Solution

Step 1: Calculate operating hoop stress and total acting stress Internal radius $R_i = D_i / 2 = 1,000.0\text{ mm}$. Thin-wall membrane hoop stress:

$$\sigma_{\text{hoop}} = \frac{p \cdot R_i}{t} = \frac{8.0\text{ MPa} \times 1,000.0\text{ mm}}{40.0\text{ mm}} = 200.0\text{ MPa}$$

Total acting tensile stress:

$$\sigma_{\text{total}} = \sigma_{\text{hoop}} + \sigma_{\text{res}} = 200.0\text{ MPa} + 60.0\text{ MPa} = 260.0\text{ MPa}$$

Step 2: Calculate critical crack depth $a_c$ Setting applied stress intensity equal to fracture toughness ($K_I = K_{Ic}$):

$$K_{Ic} = Y \sigma_{\text{total}} \sqrt{\pi a_c} \implies \sqrt{\pi a_c} = \frac{K_{Ic}}{Y \sigma_{\text{total}}}$$ $$\sqrt{\pi a_c} = \frac{55.0\text{ MPa}\sqrt{\text{m}}}{1.12 \times 260.0\text{ MPa}} = \frac{55.0}{291.2} = 0.18887\text{ m}^{1/2}$$

Squaring both sides:

$$\pi a_c = (0.18887)^2 = 0.035673\text{ m}$$ $$a_c = \frac{0.035673}{\pi} = 0.011355\text{ m} = 11.36\text{ mm}$$

Evaluation: Any surface crack exceeding $11.36\text{ mm}$ depth will trigger catastrophic brittle fracture under service pressure at $-20^\circ\text{C}$.

Step 3: Check ASTM E399 plane strain validity

$$B_{\text{req}} \ge 2.5 \left( \frac{K_{Ic}}{\sigma_y} \right)^2 = 2.5 \left( \frac{55.0}{480.0} \right)^2 = 2.5 \times (0.11458)^2 = 2.5 \times 0.01313 = 0.0328\text{ m} = 32.8\text{ mm}$$

Because actual shell wall thickness $t = 40.0\text{ mm} > 32.8\text{ mm}$, plane strain conditions are fully validated.

Step 4: Calculate creep rupture life at normal design temperature ($575^\circ\text{C}$)

$$LMP = 25.40 = T_{K1} \left[ 20 + \log_{10}(t_{r1}) \right] \times 10^{-3}$$ $$25,400 = 848.15 \times \left[ 20 + \log_{10}(t_{r1}) \right]$$ $$20 + \log_{10}(t_{r1}) = \frac{25,400}{848.15} = 29.9475$$ $$\log_{10}(t_{r1}) = 29.9475 - 20.0 = 9.9475$$ $$t_{r1} = 10^{9.9475} = 8.861 \times 10^9\text{ hours} \quad (\text{Indefinite design life under ideal baseline temperature})$$

Step 5: Calculate creep rupture life under temperature excursion ($610^\circ\text{C}$) Under the elevated temperature $T_{K2} = 883.15\text{ K}$:

$$20 + \log_{10}(t_{r2}) = \frac{25,400}{883.15} = 28.7607$$ $$\log_{10}(t_{r2}) = 28.7607 - 20.0 = 8.7607$$ $$t_{r2} = 10^{8.7607} = 5.764 \times 10^8\text{ hours}$$

Ratio of remaining life:

$$\frac{t_{r2}}{t_{r1}} = \frac{5.764 \times 10^8}{8.861 \times 10^9} = 0.0650 = 6.50\%$$

Engineering Insight: A mere $35^\circ\text{C}$ operating temperature excursion accelerates creep damage by a factor of 15.4, destroying $93.5\%$ of the component's remaining creep rupture life!

---

9. Real-World Engineering Scenarios & Exam Pitfalls

Industrial Scenario: Type IV Creep Rupture in a Grade 91 Heat Recovery Steam Generator (HRSG)

An 8-year-old combined-cycle power plant experienced an explosive rupture of a $400\text{ mm}$ diameter Grade 91 main steam pipe elbow weld operating at $565^\circ\text{C}$ and $18\text{ MPa}$. Metallurgical failure analysis revealed that the fracture occurred circumferentially along the fine-grained heat-affected zone (FGHAZ), roughly $2\text{ mm}$ from the fusion boundary, with zero wall thinning or macro-deformation. Microscopic examination identified extensive cavitation void arrays along prior austenite grain boundaries—classic Type IV cracking. The root cause was traced to improper post-weld heat treatment (PWHT): during field erection, local induction heating coils produced an under-temperature band ($710^\circ\text{C}$ instead of the code-mandated $750\text{--}770^\circ\text{C}$ per ASME B31.1), failing to re-precipitate complex carbonitrides. The softened FGHAZ experienced accelerated tertiary creep, leading to premature catastrophe.

Common Exam Traps

Exam Trap 1: Assuming High-Strength Steel Improves Welded Fatigue Strength CWEng examination questions frequently ask how much fatigue design allowable stress increases if ASTM A36 steel ($\sigma_y = 250\text{ MPa}$) is replaced with ASTM A514 quenched-and-tempered steel ($\sigma_y = 690\text{ MPa}$) for an as-welded bridge girder. The correct answer is $0\%$! Because geometric notch concentrations at weld toes eliminate initiation life, and high tensile residual stresses enforce $R_{\text{eff}} \approx 1.0$, fatigue design categories (AWS D1.1 / IIW) depend strictly on stress range $\Delta\sigma$, regardless of static yield strength.

Exam Trap 2: Using Celsius Instead of Absolute Temperature in Larson-Miller Calculations The Larson-Miller parameter equation is derived from kinetic Arrhenius rate theory and requires absolute temperature (Kelvin or Rankine). Substituting $T = 575^\circ\text{C}$ instead of $T_K = 575 + 273.15 = 848.15\text{ K}$ will produce completely nonsensical life predictions and is an intentional distractor on Part 1 and Part 2 exams.

Exam Trap 3: Embedded Internal Flaw vs. Surface Crack Dimension in LEFM In the stress intensity formula $K_I = Y \sigma \sqrt{\pi a}$, $a$ represents the half-length of an embedded internal crack ($2a$), but the full depth of an edge or surface crack ($a$). If an exam problem states "an internal embedded planar flaw of length $16\text{ mm}$", you must set $a = 8\text{ mm} = 0.008\text{ m}$. Using $a = 16\text{ mm}$ overestimates $K_I$ by a factor of $\sqrt{2} \approx 1.414$.