1.2 Differential & Integral Calculus Fundamentals in Engineering Analysis

1.2 Differential & Integral Calculus Fundamentals in Engineering Analysis

Quick Answer: Differential calculus models rates of change in welding engineering—most critically the cooling rate derivative $dT/dt$, which governs austenite transformation products in the HAZ, and the constitutive derivative $d\sigma/d\varepsilon$, which defines elastic ($E$) and tangent ($E_t$) moduli. Plastic necking occurs at Considère's condition: $d\sigma_{\text{true}}/d\varepsilon_{\text{true}} = \sigma_{\text{true}}$. Integral calculus determines true arc energy in pulsed welding through $\int v(t) i(t) dt$ and predicts structural fatigue life by integrating the Paris-Erdogan law $da/dN = C(\Delta K)^m$.

---

Differential Rates of Change in Heat Flow & HAZ Cooling Kinetics

During fusion welding, heat flows dynamically from the localized molten weld pool into the surrounding base metal. The transient temperature field $T(x,y,z,t)$ establishes steep thermal gradients. The primary physical variable governing the metallurgical microstructure, hardness, and cracking susceptibility of the heat-affected zone (HAZ) is the instantaneous rate of temperature change with respect to time: the cooling rate derivative $dT/dt$.

Rosenthal's Moving Heat Source Differential Equation

Classical weld thermal analysis originates from Rosenthal's solution to the three-dimensional differential heat conduction equation with a moving coordinate system ($x = \xi - v t$):

$$\frac{\partial T}{\partial t} = \alpha \left( \frac{\partial^2 T}{\partial x^2} + \frac{\partial^2 T}{\partial y^2} + \frac{\partial^2 T}{\partial z^2} \right) - v \frac{\partial T}{\partial x}$$

where $\alpha = k / (\rho c_p)$ is thermal diffusivity, $k$ is thermal conductivity, $\rho$ is density, $c_p$ is specific heat capacity, and $v$ is arc travel speed.

The Cooling Rate Derivative ($dT/dt$)

In transformation metallurgy, the cooling rate between $800^\circ\text{C}$ and $500^\circ\text{C}$ (expressed as $\Delta t_{8/5}$ or the cooling rate at $540^\circ\text{C}$) determines whether austenite decomposes into benign polygonal ferrite and pearlite, tough acicular ferrite, upper/lower bainite, or brittle untempered martensite.

  Temperature (°C)
       |
  1500 +--- Fusion Boundary
       |    \
   800 +-----\-----------------------
       |      \   Austenite Transformation Window (Δt_8/5)
       |       \  Cooling Rate = |dT/dt|
   500 +--------\--------------------
       |         \___
     0 +----------------------------- Time (t)

By differentiating Rosenthal's temperature field with respect to time $t$ along the weld centerline at temperature $T_c$:

1. Thick Plate (3D Conduction, heat dissipating in three dimensions):

   $$\left. \frac{dT}{dt} \right|_{T_c} = -2\pi k \left( \frac{T_c - T_0}{H_{\text{net}}} \right)^2$$

   where $T_0$ is the initial plate preheat temperature, and $H_{\text{net}} = \eta (V I / v)$ is the net linear heat input.

2. Thin Sheet (2D Conduction, heat confined to sheet plane of thickness $t$):

   $$\left. \frac{dT}{dt} \right|_{T_c} = -2\pi k \rho c_p \left( \frac{t}{H_{\text{net}}} \right)^2 (T_c - T_0)^3$$

Metallurgical Implications of the Derivative

• High Cooling Rate ($|dT/dt| \gg 0$): Occurs with low preheat ($T_0$), thin welds on thick plates (3D heat sink), or low heat input. Rapid cooling bypasses the ferrite/pearlite "noses" on the Continuous Cooling Transformation (CCT) diagram, forming high-carbon martensite. In the coarse-grained HAZ (CGHAZ), this drives hardness above $350\text{ HV}$, creating severe susceptibility to Hydrogen-Induced Cracking (HIC).
• Low Cooling Rate ($|dT/dt| \to 0$): Occurs with high preheat and high heat input (e.g., multi-wire SAW). Extremely slow cooling coarsens austenite grains and forms coarse upper bainite and grain-boundary ferrite sideplates, degrading Charpy V-notch impact toughness.

---

Derivatives on Stress-Strain Curves: Elastic, Tangent, and Secant Moduli

The tensile test curve $\sigma = f(\varepsilon)$ embodies fundamental derivatives that dictate structural stability, residual stress development, and weldment ductility.

   Stress (σ)
       |
   UTS +-------------.._ (Necking: dσ_true/dε_true = σ_true)
       |           .    `---
       |         /         Plastic Region: Tangent Modulus E_t = dσ/dε
  σ_ys +--------/ 
       |       / 
       |      / Elastic Region: Young's Modulus E = dσ/dε = const
       |     /|
       |    / |
     0 +---+--+------------------- Strain (ε)

Moduli Formulations

• Young's Modulus ($E$): The first derivative of stress with respect to strain in the linear-elastic regime:

  $$E = \left. \frac{d\sigma}{d\varepsilon} \right|_{\varepsilon \le \varepsilon_y}$$

  For structural steel, $E \approx 200\text{ GPa}$ ($29 \times 10^6\text{ psi}$). It is an inherent atomic lattice property independent of heat treatment.

• Tangent Modulus ($E_t$): The instantaneous local derivative at any specified point in the plastic regime:

  $$E_t = \frac{d\sigma}{d\varepsilon}$$

  In inelastic structural column buckling (Shanley's theory) and non-linear finite element modeling of welded joints, $E_t$ replaces $E$ to govern plastic buckling capacity.

• Secant Modulus ($E_s$): The slope of the chord connecting the origin to a designated point on the stress-strain curve: $E_s = \sigma / \varepsilon$.

Derivation of Considère's Necking Criterion

A central derivation on the CWEng exam is Considère's Criterion, which defines the exact onset of localized necking (plastic instability) during a uniaxial tensile test.

Let tensile load be $P$, instantaneous cross-sectional area be $A$, and true stress be $\sigma_{\text{true}}$. The tensile force is:

$$P = \sigma_{\text{true}} \cdot A$$

At the Ultimate Tensile Strength (UTS), maximum load is achieved, meaning the differential change in load is zero ($dP = 0$):

$$dP = d(\sigma_{\text{true}} \cdot A) = A \, d\sigma_{\text{true}} + \sigma_{\text{true}} \, dA = 0$$

Dividing through by $A \cdot \sigma_{\text{true}}$:

$$\frac{d\sigma_{\text{true}}}{\sigma_{\text{true}}} = -\frac{dA}{A}$$

Assuming plastic deformation is volume-conservative ($V = A \cdot L = \text{constant}$):

$$dV = A \, dL + L \, dA = 0 \implies -\frac{dA}{A} = \frac{dL}{L}$$

By definition, incremental true strain is $d\varepsilon_{\text{true}} = dL / L$. Therefore:

$$d\varepsilon_{\text{true}} = -\frac{dA}{A}$$

Substituting $d\varepsilon_{\text{true}}$ into the differential force equation yields Considère's Criterion:

$$\frac{d\sigma_{\text{true}}}{d\varepsilon_{\text{true}}} = \sigma_{\text{true}}$$

Application to Hollomon's Power-Law Equation

Many engineering metals follow Hollomon's constitutive power-law equation in the plastic regime: $\sigma_{\text{true}} = K \varepsilon_{\text{true}}^n$, where $K$ is the strength coefficient and $n$ is the strain-hardening exponent. Taking the derivative:

$$\frac{d\sigma_{\text{true}}}{d\varepsilon_{\text{true}}} = n K \varepsilon_{\text{true}}^{n-1} = \frac{n (K \varepsilon_{\text{true}}^n)}{\varepsilon_{\text{true}}} = \frac{n \sigma_{\text{true}}}{\varepsilon_{\text{true}}}$$

Setting this derivative equal to $\sigma_{\text{true}}$ per Considère's criterion:

$$\frac{n \sigma_{\text{true}}}{\varepsilon_{\text{true}}} = \sigma_{\text{true}} \implies \varepsilon_{\text{true}} = n$$

Engineering Principle: The true uniform plastic strain at the onset of necking is identically equal to the strain-hardening exponent $n$ of the material.

---

Definite Integrals in Arc Energy & Weld Bead Geometry

Waveform Power Integration in Advanced Arc Welding

In conventional constant-current (CC) or constant-voltage (CV) processes, welding power is modeled as the simple scalar product $P = V \cdot I$. However, modern power sources (pulsed GMAW, Cold Metal Transfer, AC squarewave GTAW) utilize high-frequency waveform modulation ($100\text{ to }500\text{ Hz}$) where voltage and current fluctuate rapidly out of phase.

   Current (I)
       |
   I_p +-------+             +-------+             Peak Current
       |       |             |       |
       |       |             |       |
   I_b +       +-------------+       +------------- Background Current
       |<--tp->|<----tb----->|
     0 +------------------------------------------- Time (t)
               |<----- T ----->|

The true average arc power is the definite integral of the instantaneous power function over wave period $T$:

$$P_{\text{avg}} = \frac{1}{T} \int_0^T v(t) \cdot i(t) \, dt$$

Critical Mathematical Fact: $\frac{1}{T} \int_0^T v(t) \cdot i(t) , dt \ne V_{\text{avg}} \cdot I_{\text{avg}}$.

Multiplying arithmetic average voltage by average current can underestimate or overestimate true heat input by up to 30%! Consequently, ASME Section IX (Appendix H) and AWS D1.1 mandate the use of instantaneous energy integration for waveform-controlled welding:

$$\text{True Energy (J)} = \int_0^{t_{\text{weld}}} v(t) \cdot i(t) \, dt$$ $$\text{True Linear Heat Input} = \frac{\eta}{v_{\text{travel}} \cdot T} \int_0^T v(t) \cdot i(t) \, dt \quad (\text{J/mm})$$

Bead Profile Integration

When macrographs of deposited weld beads are analyzed, the cross-sectional geometry of the reinforcement crown can be modeled mathematically. For a bead with surface profile defined by parabolic distribution $y(x) = h \left(1 - \frac{4x^2}{W^2}\right)$ over interval $[-W/2, +W/2]$:

$$A = \int_{-W/2}^{W/2} h \left(1 - \frac{4x^2}{W^2}\right) dx = h \left[ x - \frac{4x^3}{3W^2} \right]_{-W/2}^{W/2} = h \left( \frac{W}{2} - \frac{W}{6} - \left( -\frac{W}{2} + \frac{W}{6} \right) \right) = \frac{2}{3} W h$$

---

Definite Integrals in Cumulative Fatigue Damage & Fracture Mechanics

Welded joints subject to cyclic loading fail predominantly via fatigue cracking initiated at geometric stress concentrations (weld toes, roots, undercut). Linear Elastic Fracture Mechanics (LEFM) models subcritical crack propagation using calculus.

The Paris-Erdogan Law

The rate of crack extension per stress cycle ($da/dN$) is related to the stress intensity factor range $\Delta K$:

$$\frac{da}{dN} = C (\Delta K)^m$$

where $\Delta K = Y \Delta \sigma \sqrt{\pi a}$, $a$ is crack depth, $\Delta \sigma$ is cyclic stress range, $Y$ is dimensionless geometric boundary correction factor, and $C, m$ are empirical material constants.

Substituting $\Delta K$ into the Paris equation:

$$\frac{da}{dN} = C \left( Y \Delta \sigma \sqrt{\pi a} \right)^m = C (Y \Delta \sigma \sqrt{\pi})^m \cdot a^{m/2}$$

Derivation of Total Fatigue Cycles to Failure ($N_f$)

Separating variables $a$ and $N$, and integrating from an initial detectable flaw size $a_0$ to the critical terminal flaw size $a_f$:

$$\int_0^{N_f} dN = \frac{1}{C (Y \Delta \sigma \sqrt{\pi})^m} \int_{a_0}^{a_f} a^{-m/2} \, da$$

For structural ferritic steels conforming to BS 7910 or AWS D1.1, the crack growth exponent is commonly taken as $m = 3.0$:

$$\int a^{-3/2} \, da = \frac{a^{-3/2 + 1}}{-3/2 + 1} = \frac{a^{-1/2}}{-1/2} = -2 a^{-1/2} = -\frac{2}{\sqrt{a}}$$

Evaluating across definite limits $[a_0, a_f]$:

$$N_f = \frac{1}{C (Y \Delta \sigma \sqrt{\pi})^3} \left[ -\frac{2}{\sqrt{a}} \right]_{a_0}^{a_f} = \frac{2}{C (Y \Delta \sigma \sqrt{\pi})^3} \left( \frac{1}{\sqrt{a_0}} - \frac{1}{\sqrt{a_f}} \right)$$

Key Physical Insight: Because $a_0 \ll a_f$, the term $1/\sqrt{a_0}$ overwhelmingly dominates the bracketed expression. Over 90% of the total fatigue life of a welded joint is consumed while growing the crack from its initial microscopic flaw size $a_0$ to just $2\text{ to }3\text{ mm}$! Finishing techniques that reduce initial toe discontinuities (e.g., burr grinding, TIG dressing, ultrasonic impact treatment) yield tremendous extensions in operating life.

Palmgren-Miner Linear Cumulative Damage Rule

When weldments endure variable-amplitude spectrum loading, the cumulative damage fraction $D$ is the sum (or definite integral across stress spectrum $S$) of individual cycle damage ratios:

$$D = \sum_{i=1}^k \frac{n_i}{N_i} = \int_0^n \frac{dn}{N(S)} \le 1.0$$

where $n_i$ is applied cycles at stress range $\Delta \sigma_i$, and $N_i$ is cycles to failure at that stress range. Fatigue failure is predicted when $D = 1.0$.

---

Comprehensive Worked Numerical Example: LEFM Fatigue Life Integration

Problem Statement

A full-penetration transverse butt weld in a crane runway girder is subject to a constant-amplitude nominal cyclic tensile stress range $\Delta \sigma = 140\text{ MPa}$. Ultrasonic inspection identifies an initial weld toe planar flaw of depth $a_0 = 1.2\text{ mm} = 0.0012\text{ m}$. Engineering critical assessment establishes that catastrophic brittle fracture will occur when the crack reaches critical depth $a_f = 15.0\text{ mm} = 0.0150\text{ m}$.

The joint geometry and material parameters are:

• Geometric boundary factor $Y = 1.12$
• Paris constant $C = 3.0 \times 10^{-13} , \frac{\text{m/cycle}}{(\text{MPa}\sqrt{\text{m}})^3}$
• Paris exponent $m = 3.0$

Calculate the total fatigue life $N_f$ in cycles.

Step-by-Step Solution

Step 1: Compute the constant stress-geometry multiplier

$$Y \Delta \sigma \sqrt{\pi} = (1.12) \cdot (140\text{ MPa}) \cdot \sqrt{\pi} = 156.8 \cdot 1.77245 = 277.92\text{ MPa}\sqrt{\text{m}}$$

Step 2: Cube the stress-geometry term ($m = 3.0$)

$$(Y \Delta \sigma \sqrt{\pi})^3 = (277.92)^3 = 21,466,743 \, (\text{MPa}\sqrt{\text{m}})^3$$

Step 3: Compute the denominator product with Paris constant $C$

$$C \cdot (Y \Delta \sigma \sqrt{\pi})^3 = (3.0 \times 10^{-13}) \cdot (2.14667 \times 10^7) = 6.440 \times 10^{-6} \, \text{m}^{3/2}/\text{cycle}$$

Step 4: Evaluate the crack limit integration bracket

$$\frac{1}{\sqrt{a_0}} = \frac{1}{\sqrt{0.0012\text{ m}}} = \frac{1}{0.034641} = 28.8675\text{ m}^{-1/2}$$ $$\frac{1}{\sqrt{a_f}} = \frac{1}{\sqrt{0.0150\text{ m}}} = \frac{1}{0.122474} = 8.1650\text{ m}^{-1/2}$$ $$\left( \frac{1}{\sqrt{a_0}} - \frac{1}{\sqrt{a_f}} \right) = 28.8675 - 8.1650 = 20.7025\text{ m}^{-1/2}$$

Step 5: Calculate total cycles to failure ($N_f$)

$$N_f = \frac{2}{C (Y \Delta \sigma \sqrt{\pi})^3} \cdot 20.7025 = \frac{2 \cdot 20.7025}{6.440 \times 10^{-6}} = \frac{41.405}{6.440 \times 10^{-6}} = 6,429,348\text{ cycles}$$

The predicted fatigue life is $6.43 \times 10^6$ cycles.

---

Real-World Engineering Scenarios & Exam Pitfalls

Practical Industrial Scenario

During an offshore platform life-extension audit, non-destructive examination revealed toe undercut cracks up to $a_0 = 2.0\text{ mm}$ deep on heavy jacket nodes. The original design calculations had assumed an initial flaw threshold of $0.5\text{ mm}$. The engineering team utilized the Paris Law integral to evaluate the remaining fatigue life. Because the integral scales with $1/\sqrt{a_0}$, increasing $a_0$ from $0.5\text{ mm}$ to $2.0\text{ mm}$ reduced $1/\sqrt{a_0}$ from $44.7\text{ m}^{-1/2}$ to $22.4\text{ m}^{-1/2}$, cutting remaining operating life by exactly 50%. This analysis forced immediate offshore weld toe grinding remediation before cyclonic storm seasons.

Common Exam Traps

Exam Trap 1: Multiplying Average Voltage by Average Current on Pulsed Arcs When asked to calculate heat input for a waveform-controlled power supply (pulsed GMAW), never multiply $V_{\text{avg}} \times I_{\text{avg}}$. The product of the averages is mathematically not equal to the average of the instantaneous products $\frac{1}{T}\int v(t)i(t)dt$. The exam tests whether you recognize that ASME Section IX requires instantaneous energy meters for advanced waveform procedures.

Exam Trap 2: Unit Inconsistencies in Paris Law Integration A major source of calculation error on the CWEng exam is units mixing. The crack depth $a$ is usually given in millimeters ($1.2\text{ mm}$), but the Paris constant $C$ is typically expressed in terms of meters ($\text{m/cycle}$ and $\text{MPa}\sqrt{\text{m}}$). If you substitute $a_0 = 1.2$ instead of $0.0012\text{ m}$, the life prediction will be incorrect by a factor of $(1000)^{1.5} \approx 31,623$!

Exam Trap 3: Engineering vs. True Stress Necking Slope On an engineering stress-strain curve, necking begins at the Ultimate Tensile Strength where the slope is zero ($d\sigma_{\text{eng}}/d\varepsilon_{\text{eng}} = 0$). However, on a true stress-strain curve, the slope at necking is not zero; it is positive and exactly equals the true stress ($d\sigma_{\text{true}}/d\varepsilon_{\text{true}} = \sigma_{\text{true}}$). Confusing engineering with true coordinates is an intentional trap on Part 1 of the CWEng exam.