SOM-07-04 Chapter 07 · Deflection of Beams

Strain Energy & Castigliano's Theorem for Beam Deflections

Subject
Strength of Materials (Solid Mechanics)
Chapter
Deflection of Beams
Standard Exam Coverage
GATE CE, UPSC ESE, SSC JE, State AE/JE, RRB JE
In this chapter (4 topics)
On this page

1. Engineering Context & Overview

While geometric methods (Double Integration, Macaulay's Method, Moment-Area, and Conjugate Beam) rely directly on the differential relationship between curvature and bending moment (EId2ydx2=MEI \frac{d^2 y}{dx^2} = M), energy methods operate on the fundamental principle of Conservation of Energy.

For structures undergoing non-uniform flexure, internal strain energy (UU) is stored in the elastic material fibers. Alberto Castigliano (1847–1884) demonstrated that partial derivatives of this internal energy with respect to external generalized loads yield the corresponding generalized displacements directly.

In modern structural engineering and competitive examinations (GATE CE, UPSC ESE), Castigliano's Second Theorem and its corollary, the Dummy (Fictitious) Load Method, serve as the most potent tools for solving:

  1. Complex non-prismatic members and bent cantilevers (orthogonal frames).
  2. Deflections and rotations in curved beams, arches, and circular rings where classical geometric curvature integration becomes analytically prohibitive.
  3. Statically indeterminate beams where internal redundant reactions can be solved by setting the partial derivative of strain energy with respect to the redundant force to zero (Principle of Least Work).

2. Internal Work & Strain Energy in Pure Bending

Consider an elastic beam subject to transverse loading. At any cross-section at distance xx, an internal bending moment M(x)M(x) acts. Over an infinitesimal element of length dxdx, the relative angular rotation of the two opposite transverse faces is given by:

dθ=κ dx=M(x)EI dxd\theta = \kappa \, dx = \frac{M(x)}{EI} \, dx

Under gradually applied static loads within the proportional limit (E=constantE = \text{constant}), the relationship between bending moment and relative rotation is strictly linear. The incremental internal strain energy dUdU stored in this infinitesimal segment is the area under the linear M−θM - \theta diagram:

dU=12M(x) dθ=12M(x)(M(x)EI dx)=M(x)22EI dxdU = \frac{1}{2} M(x) \, d\theta = \frac{1}{2} M(x) \left( \frac{M(x)}{EI} \, dx \right) = \frac{M(x)^2}{2 EI} \, dx

Integrating over the entire beam length LL (or summing over all distinct structural segments):

U=∫0LM(x)22EI dxU = \int_0^L \frac{M(x)^2}{2 EI} \, dx
   Face A           Element dx           Face B
     │                                     │
     │◄─────────────── dx ───────────────►│
     │                                     │
 ───┼─────────────────────────────────────┼─── Neutral Axis
     │                                     │
     ▼ M(x) (Sagging)                      ▲ M(x) + dM
    (                                     )
    Rel. Rotation dθ = [M(x) / EI] dx
    Incremental Strain Energy: dU = ½ M(x) dθ = [M(x)² / (2EI)] dx

2.1 Axial and Shear Energy Contributions

In general structural members, strain energy is also produced by axial force N(x)N(x), shear force V(x)V(x), and torque T(x)T(x):

Utotal=∫0LN(x)22EA dx+∫0LM(x)22EI dx+∫0LfsV(x)22GA dx+∫0LT(x)22GJ dxU_{\text{total}} = \int_0^L \frac{N(x)^2}{2 EA} \, dx + \int_0^L \frac{M(x)^2}{2 EI} \, dx + \int_0^L \frac{f_s V(x)^2}{2 GA} \, dx + \int_0^L \frac{T(x)^2}{2 GJ} \, dx

where fsf_s is the shear form factor (fs=1.2f_s = 1.2 for rectangular sections, fs≈1.11f_s \approx 1.11 for circular sections).

Important Engineering Observation: For slender beams where the span-to-depth ratio exceeds 10 (L/d>10L / d > 10), flexural strain energy accounts for more than 98% to 99% of the total stored internal energy. Consequently, in all standard competitive examinations (GATE CE, ESE), shear and axial strain energy contributions in beam deflection are routinely neglected unless shear deformation is explicitly requested (Timoshenko beam theory).


3. Castigliano's Theorems (First & Second)

3.1 Strain Energy (UU) vs Complementary Strain Energy (U∗U^*)

For an elastic body loaded by generalized forces P1,P2,…,PnP_1, P_2, \dots, P_n resulting in generalized displacements Δ1,Δ2,…,Δn\Delta_1, \Delta_2, \dots, \Delta_n:

  • Internal Strain Energy (UU): Stored as internal fiber strain: U=∫∑Pi dΔiU = \int \sum P_i \, d\Delta_i
  • Complementary Strain Energy (U∗U^*): Mathematical dual defined with respect to force increments: U∗=∫∑Δi dPiU^* = \int \sum \Delta_i \, dP_i

For non-linear elastic materials, U≠U∗U \ne U^*. However, for linearly elastic materials obeying Hooke's Law, the force-displacement diagram is a straight line through the origin, making:

U=U∗=12∑i=1nPiΔiU = U^* = \frac{1}{2} \sum_{i=1}^n P_i \Delta_i

3.2 Castigliano's First Theorem

Statement: In any elastic system at constant temperature, the first partial derivative of the total internal strain energy (UU) with respect to any generalized displacement (Δi\Delta_i) equals the corresponding generalized external force (PiP_i) acting at that point in the direction of Δi\Delta_i.

Pi=∂U∂ΔiP_i = \frac{\partial U}{\partial \Delta_i}

Scope: Valid for both linear and non-linear elastic structures. Primarily utilized in displacement-based finite element formulations and stiffness matrix derivations.

3.3 Castigliano's Second Theorem

Statement: In any linearly elastic structure at constant temperature with unyielding supports, the first partial derivative of the total internal strain energy (UU) with respect to any generalized external force (PiP_i) equals the generalized displacement (Δi\Delta_i) at the point of application of PiP_i along its line of action.

Δi=∂U∂Pi\Delta_i = \frac{\partial U}{\partial P_i}

Similarly for rotations: If an external moment / couple MiM_i acts at a node, the rotation θi\theta_i at that point in the direction of MiM_i is:

θi=∂U∂Mi\theta_i = \frac{\partial U}{\partial M_i}

Scope: Strictly restricted to linearly elastic systems (Hooke's Law valid, principle of superposition holds, U=U∗U = U^*).


4. Leibniz's Rule: Differentiation Under the Integral Sign

Directly substituting U=∫0LM(x)22EI dxU = \int_0^L \frac{M(x)^2}{2 EI} \, dx and attempting to square the algebraic moment expressions before integrating leads to severe algebraic expansion errors.

By invoking Leibniz's Integral Rule, we differentiate with respect to PiP_i inside the integral sign:

Δi=∂∂Pi[∫0LM(x)22EI dx]=∫0L2M(x)2EI(∂M(x)∂Pi)dx\Delta_i = \frac{\partial}{\partial P_i} \left[ \int_0^L \frac{M(x)^2}{2 EI} \, dx \right] = \int_0^L \frac{2 M(x)}{2 EI} \left( \frac{\partial M(x)}{\partial P_i} \right) dx Δi=∫0LM(x)EI(∂M(x)∂Pi)dx\Delta_i = \int_0^L \frac{M(x)}{EI} \left( \frac{\partial M(x)}{\partial P_i} \right) dx

Similarly, the rotation at point ii is:

θi=∫0LM(x)EI(∂M(x)∂Mi)dx\theta_i = \int_0^L \frac{M(x)}{EI} \left( \frac{\partial M(x)}{\partial M_i} \right) dx

Practical Advantage of Differentiating Under the Integral Sign

  1. Reduces Polynomial Degree: Instead of integrating M2(x)M^2(x) (which could be degree 4 or 6 for UDL/varying loads), we integrate the product of M(x)M(x) and its linear partial derivative ∂M∂Pi\frac{\partial M}{\partial P_i}.
  2. Allows Setting Fictitious Loads to Zero Before Integration: As shown in Section 5, if PiP_i is a fictitious dummy load, it can be set to zero prior to computing the integral, vastly simplifying the algebra.

5. The Fictitious (Dummy) Load Method & Virtual Work Equivalence

5.1 The Problem of Unloaded Points

Castigliano's Second Theorem requires an active external concentrated force (or moment) at the exact point and in the exact direction of the desired deflection (or rotation). If the desired displacement occurs at a point where no load acts (or where only a distributed load acts):

  1. Step 1 (Introduce Fictitious Load): Apply an imaginary fictitious concentrated force QQ (or fictitious moment M0M_0) at the point of interest in the assumed direction of displacement.
  2. Step 2 (Formulate Moment Equation): Write the section bending moment expression as a function of real loads and the fictitious parameter: M(x,Q)M(x, Q).
  3. Step 3 (Differentiate Partially): Evaluate the partial derivative with respect to QQ: m(x)=∂M(x,Q)∂Qm(x) = \frac{\partial M(x, Q)}{\partial Q}
  4. Step 4 (Set Dummy Load to Zero): Evaluate the limit as Q→0Q \to 0 inside the integral: Δ=lim⁡Q→0∫0LM(x,Q)EI(∂M(x,Q)∂Q)dx=∫0LMreal(x)⋅m(x)EI dx\Delta = \lim_{Q \to 0} \int_0^L \frac{M(x, Q)}{EI} \left( \frac{\partial M(x, Q)}{\partial Q} \right) dx = \int_0^L \frac{M_{\text{real}}(x) \cdot m(x)}{EI} \, dx

5.2 Equivalence to the Unit Load Method (Principle of Virtual Work)

Notice that:

m(x)=∂M(x,Q)∂Q∣Q=0m(x) = \left. \frac{\partial M(x, Q)}{\partial Q} \right|_{Q=0}

represents the internal bending moment produced throughout the structure when a single concentrated load of unit magnitude (Q=1Q = 1) is applied at the desired displacement location!

Thus, Castigliano's fictitious load formulation is mathematically identical to the classic Unit Load Method (Dummy Unit Load):

Δ=∫0LM(x) m(x)EI dx\Delta = \int_0^L \frac{M(x) \, m(x)}{EI} \, dx

where:

  • M(x)M(x) = Bending moment due to actual applied loads.
  • m(x)m(x) = Bending moment due to a fictitious unit load (1 kN1\text{ kN} or 1 kN⋅m1\text{ kN}\cdot\text{m}) applied at the target point.

6. Deflections of Bent Cantilevers, Frames & Curved Members

Energy methods excel where classical Cartesian coordinates break down.

6.1 Orthogonal Bent Frames (L-Shaped Cantilever)

For an L-frame fixed at base AA, with vertical column ABAB (height HH) and horizontal cantilever beam BCBC (length LL):

U=UBC+UAB=∫0LMBC(x)22EIBC dx+∫0HMAB(y)22EIAB dyU = U_{BC} + U_{AB} = \int_0^L \frac{M_{BC}(x)^2}{2 EI_{BC}} \, dx + \int_0^H \frac{M_{AB}(y)^2}{2 EI_{AB}} \, dy

The displacements at tip CC are obtained by integrating over both orthogonal segments independently.

6.2 Curved Beams and Rings in Their Plane

For a curved beam of centroidal radius RR, an arc element has length ds=R dθds = R \, d\theta. Neglecting axial and shear contributions for slender rings (R/h>10R / h > 10):

U=∫M(θ)22EIR dθU = \int \frac{M(\theta)^2}{2 EI} R \, d\theta

For a quarter-circular cantilever ring of radius RR fixed at θ=0\theta = 0 (vertical tangent) and free at θ=π/2\theta = \pi/2, subjected to a vertical downward load WW:

  • Coordinate measured from free end: 0≤θ≤π/20 \le \theta \le \pi/2.
  • Moment at angle θ\theta: M(θ)=−WRsin⁡θM(\theta) = -W R \sin\theta.
  • Derivative with respect to WW: ∂M∂W=−Rsin⁡θ\frac{\partial M}{\partial W} = -R \sin\theta.
  • Vertical deflection: δv=∫0π/2(−WRsin⁡θ)(−Rsin⁡θ)EIR dθ=WR3EI∫0π/2sin⁡2θ dθ=πWR34EI≈0.7854WR3EI\delta_v = \int_0^{\pi/2} \frac{(-W R \sin\theta)(-R \sin\theta)}{EI} R \, d\theta = \frac{W R^3}{EI} \int_0^{\pi/2} \sin^2\theta \, d\theta = \frac{\pi W R^3}{4 EI} \approx 0.7854 \frac{W R^3}{EI}
  • Horizontal deflection (via dummy horizontal force H→0H \to 0): δh=WR3EI∫0π/2sin⁡θ(1−cos⁡θ) dθ=WR32EI=0.5000WR3EI\delta_h = \frac{W R^3}{EI} \int_0^{\pi/2} \sin\theta (1 - \cos\theta) \, d\theta = \frac{W R^3}{2 EI} = 0.5000 \frac{W R^3}{EI}
  • Universal Ratio: δvδh=π/41/2=π2≈1.5708\frac{\delta_v}{\delta_h} = \frac{\pi / 4}{1 / 2} = \frac{\pi}{2} \approx 1.5708

7. Engineering Diagrams & Visual Walkthrough

Figure 1: Castigliano's Theorems & Flexural Strain Energy

The following diagram illustrates the relationship between strain energy UU and complementary strain energy U∗U^*, the physical flexure of an infinitesimal beam element, and Leibniz's rule for direct differentiation.

Castigliano Strain Energy in Bending

Figure 2: Dummy Load Method & Curved Beam Deflection

The following diagram demonstrates the 4-step algorithm for applying fictitious loads to unloaded points, its equivalence to the virtual unit load method, and closed-form deflection kinematics for curved rings.

Dummy Load Method and Curved Beam

8. Worked Numerical Examples

Example 1: Cantilever with Combined Full UDL and Tip Load

Problem Statement: A cantilever beam ABAB of span L=3.0 mL = 3.0\text{ m} is fixed at AA and free at BB. It carries a uniformly distributed load w=12 kN/mw = 12\text{ kN/m} over its entire length and a concentrated vertical downward load P=20 kNP = 20\text{ kN} at the free tip BB.
Given flexural rigidity EI=9000 kN⋅m2EI = 9000\text{ kN}\cdot\text{m}^2 (E=200 GPaE = 200\text{ GPa}, I=4.5×107 mm4I = 4.5 \times 10^7\text{ mm}^4).
Using Castigliano's Second Theorem:

  1. Determine the vertical downward deflection at tip BB.
  2. Determine the rotation (slope) at tip BB using the fictitious couple technique.
       w = 12 kN/m
  ══════════════════════════▼ P = 20 kN
  │█████████████████████████│
  │A                       B│
  │◄──────── 3.0 m ────────►│
  Fixed                   Free

Solution:

Part 1: Vertical Deflection δB\delta_B
Let origin be at the free tip BB, with coordinate xx measured towards the fixed support (0≤x≤L0 \le x \le L).
At any section xx:

M(x)=−Px−wx22M(x) = -P x - \frac{w x^2}{2}

Since PP acts directly at BB in the direction of the desired deflection, differentiate M(x)M(x) partially with respect to PP:

∂M(x)∂P=−x\frac{\partial M(x)}{\partial P} = -x

Applying Castigliano's Theorem:

δB=∫0LM(x)EI(∂M(x)∂P)dx=1EI∫0L(−Px−wx22)(−x) dx\delta_B = \int_0^L \frac{M(x)}{EI} \left( \frac{\partial M(x)}{\partial P} \right) dx = \frac{1}{EI} \int_0^L \left( -P x - \frac{w x^2}{2} \right) (-x) \, dx δB=1EI∫0L(Px2+wx32)dx=1EI[PL33+wL48]\delta_B = \frac{1}{EI} \int_0^L \left( P x^2 + \frac{w x^3}{2} \right) dx = \frac{1}{EI} \left[ \frac{P L^3}{3} + \frac{w L^4}{8} \right]

Substitute the numerical parameters:

  • PL33=20×(3.0)33=20×273=180 kN⋅m3\frac{P L^3}{3} = \frac{20 \times (3.0)^3}{3} = \frac{20 \times 27}{3} = 180\text{ kN}\cdot\text{m}^3
  • wL48=12×(3.0)48=12×818=121.5 kN⋅m3\frac{w L^4}{8} = \frac{12 \times (3.0)^4}{8} = \frac{12 \times 81}{8} = 121.5\text{ kN}\cdot\text{m}^3
  • Total numerator =180+121.5=301.5 kN⋅m3= 180 + 121.5 = 301.5\text{ kN}\cdot\text{m}^3
δB=301.5 kN⋅m39000 kN⋅m2=0.03350 m=33.50 mm(downward)\delta_B = \frac{301.5\text{ kN}\cdot\text{m}^3}{9000\text{ kN}\cdot\text{m}^2} = 0.03350\text{ m} = \mathbf{33.50\text{ mm}}\quad\text{(downward)}

Part 2: Slope (Rotation) at Tip BB
There is no concentrated moment acting at tip BB. Therefore, apply a fictitious clockwise couple M0M_0 at tip BB.
The bending moment at section xx becomes:

M(x)=−Px−wx22−M0M(x) = -P x - \frac{w x^2}{2} - M_0

Differentiate partially with respect to M0M_0:

∂M(x)∂M0=−1\frac{\partial M(x)}{\partial M_0} = -1

Now set M0→0M_0 \to 0 and integrate:

θB=lim⁡M0→0∫0LM(x)EI(∂M(x)∂M0)dx=1EI∫0L(−Px−wx22)(−1) dx\theta_B = \lim_{M_0 \to 0} \int_0^L \frac{M(x)}{EI} \left( \frac{\partial M(x)}{\partial M_0} \right) dx = \frac{1}{EI} \int_0^L \left( -P x - \frac{w x^2}{2} \right) (-1) \, dx θB=1EI∫0L(Px+wx22)dx=1EI[PL22+wL36]\theta_B = \frac{1}{EI} \int_0^L \left( P x + \frac{w x^2}{2} \right) dx = \frac{1}{EI} \left[ \frac{P L^2}{2} + \frac{w L^3}{6} \right]

Substitute the parameters:

  • PL22=20×9.02=90.0 kN⋅m2\frac{P L^2}{2} = \frac{20 \times 9.0}{2} = 90.0\text{ kN}\cdot\text{m}^2
  • wL36=12×27.06=54.0 kN⋅m2\frac{w L^3}{6} = \frac{12 \times 27.0}{6} = 54.0\text{ kN}\cdot\text{m}^2
  • Total numerator =90.0+54.0=144.0 kN⋅m2= 90.0 + 54.0 = 144.0\text{ kN}\cdot\text{m}^2
θB=144.09000=0.01600 rad=0.01600×180π=0.9167∘(clockwise)\theta_B = \frac{144.0}{9000} = \mathbf{0.01600\text{ rad}} = 0.01600 \times \frac{180}{\pi} = \mathbf{0.9167^\circ}\quad\text{(clockwise)}

Example 2: Orthogonal L-Frame Bent Cantilever

Problem Statement: An L-shaped rigid frame consists of a vertical member ABAB of height H=2.0 mH = 2.0\text{ m} fixed at base AA, and a horizontal member BCBC of length L=3.0 mL = 3.0\text{ m} rigidly joined at 90∘90^\circ at knee BB. At the free tip CC, two concurrent concentrated loads act: a vertical downward load P=15 kNP = 15\text{ kN} and a horizontal load Q=10 kNQ = 10\text{ kN} directed towards the right.
Both members have identical flexural rigidity EI=12,000 kN⋅m2EI = 12,000\text{ kN}\cdot\text{m}^2. Neglecting axial and shear deformation, calculate:

  1. The vertical deflection δv\delta_v at CC.
  2. The horizontal deflection δh\delta_h at CC.
           B ═════════════════════════ C ──► Q = 10 kN
           ║                           │
           ║                           ▼ P = 15 kN
    H = 2m ║
           ║
         ══╩══ A (Fixed Base)
           ◄──────── L = 3m ──────────►

Solution:

We break the structure into two segments: BCBC and ABAB.

1. Member BCBC (Horizontal leg, length L=3.0 mL = 3.0\text{ m}):
Let origin be at free end CC, with coordinate xx directed towards BB (0≤x≤L0 \le x \le L):

M(x)=−PxM(x) = -P x

Notice that the horizontal force QQ acts axially along BCBC and induces zero bending moment in BCBC.

∂MBC∂P=−x,∂MBC∂Q=0\frac{\partial M_{BC}}{\partial P} = -x, \qquad \frac{\partial M_{BC}}{\partial Q} = 0

2. Member ABAB (Vertical leg, height H=2.0 mH = 2.0\text{ m}):
Let origin be at knee BB, with coordinate yy directed downwards towards base AA (0≤y≤H0 \le y \le H).
The vertical force PP acts at lever arm L=3 mL = 3\text{ m}, producing a constant hogging moment −PL-P L.
The horizontal force QQ acts at lever arm yy from knee BB, producing an additional bending moment −Qy-Q y:

M(y)=−PL−QyM(y) = -P L - Q y ∂MAB∂P=−L,∂MAB∂Q=−y\frac{\partial M_{AB}}{\partial P} = -L, \qquad \frac{\partial M_{AB}}{\partial Q} = -y

Vertical Deflection δv\delta_v at CC:

δv=∂U∂P=∫0LMBCEI(∂MBC∂P)dx+∫0HMABEI(∂MAB∂P)dy\delta_v = \frac{\partial U}{\partial P} = \int_0^L \frac{M_{BC}}{EI} \left( \frac{\partial M_{BC}}{\partial P} \right) dx + \int_0^H \frac{M_{AB}}{EI} \left( \frac{\partial M_{AB}}{\partial P} \right) dy δv=1EI[∫0L(−Px)(−x) dx+∫0H(−PL−Qy)(−L) dy]\delta_v = \frac{1}{EI} \left[ \int_0^L (-P x)(-x) \, dx + \int_0^H (-P L - Q y)(-L) \, dy \right] δv=1EI[PL33+PL2H+QLH22]\delta_v = \frac{1}{EI} \left[ \frac{P L^3}{3} + P L^2 H + \frac{Q L H^2}{2} \right]

Substitute the numerical values:

  • PL33=15×273=135.0 kN⋅m3\frac{P L^3}{3} = \frac{15 \times 27}{3} = 135.0\text{ kN}\cdot\text{m}^3
  • PL2H=15×9.0×2.0=270.0 kN⋅m3P L^2 H = 15 \times 9.0 \times 2.0 = 270.0\text{ kN}\cdot\text{m}^3
  • QLH22=10×3.0×4.02=60.0 kN⋅m3\frac{Q L H^2}{2} = \frac{10 \times 3.0 \times 4.0}{2} = 60.0\text{ kN}\cdot\text{m}^3
  • Total numerator =135.0+270.0+60.0=465.0 kN⋅m3= 135.0 + 270.0 + 60.0 = 465.0\text{ kN}\cdot\text{m}^3
δv=465.012000=0.03875 m=38.75 mm(downward)\delta_v = \frac{465.0}{12000} = 0.03875\text{ m} = \mathbf{38.75\text{ mm}}\quad\text{(downward)}

Horizontal Deflection δh\delta_h at CC:

δh=∂U∂Q=∫0LMBCEI(∂MBC∂Q)dx+∫0HMABEI(∂MAB∂Q)dy\delta_h = \frac{\partial U}{\partial Q} = \int_0^L \frac{M_{BC}}{EI} \left( \frac{\partial M_{BC}}{\partial Q} \right) dx + \int_0^H \frac{M_{AB}}{EI} \left( \frac{\partial M_{AB}}{\partial Q} \right) dy

Since ∂MBC∂Q=0\frac{\partial M_{BC}}{\partial Q} = 0, member BCBC contributes zero flexural displacement to δh\delta_h:

δh=1EI[0+∫0H(−PL−Qy)(−y) dy]=1EI∫0H(PLy+Qy2)dy\delta_h = \frac{1}{EI} \left[ 0 + \int_0^H (-P L - Q y)(-y) \, dy \right] = \frac{1}{EI} \int_0^H \left( P L y + Q y^2 \right) dy δh=1EI[PLH22+QH33]\delta_h = \frac{1}{EI} \left[ \frac{P L H^2}{2} + \frac{Q H^3}{3} \right]

Substitute the numerical values:

  • PLH22=15×3.0×4.02=90.0 kN⋅m3\frac{P L H^2}{2} = \frac{15 \times 3.0 \times 4.0}{2} = 90.0\text{ kN}\cdot\text{m}^3
  • QH33=10×8.03=26.6667 kN⋅m3\frac{Q H^3}{3} = \frac{10 \times 8.0}{3} = 26.6667\text{ kN}\cdot\text{m}^3
  • Total numerator =90.0+26.6667=116.6667 kN⋅m3= 90.0 + 26.6667 = 116.6667\text{ kN}\cdot\text{m}^3
δh=116.666712000=0.009722 m=9.72 mm(towards right)\delta_h = \frac{116.6667}{12000} = 0.009722\text{ m} = \mathbf{9.72\text{ mm}}\quad\text{(towards right)}

Example 3: Quarter-Circular Cantilever Curved Bar

Problem Statement: A slender curved bar in the form of a quarter of a circle of radius R=1.2 mR = 1.2\text{ m} lies in a vertical plane. The lower end AA is rigidly built-in (tangent is vertical), while the upper end BB is free (tangent is horizontal). A vertical concentrated downward load W=5.0 kNW = 5.0\text{ kN} is applied at free end BB.
Given flexural rigidity EI=3000 kN⋅m2EI = 3000\text{ kN}\cdot\text{m}^2. Neglecting axial and shear strains:

  1. Calculate the vertical downward deflection δv\delta_v at BB.
  2. Calculate the horizontal deflection δh\delta_h at BB.
  3. Verify the ratio δv/δh\delta_v / \delta_h.

Solution:

Let angular coordinate θ\theta be measured from the free tip BB (0≤θ≤π/20 \le \theta \le \pi/2):

  • Arc differential length: ds=R dθds = R \, d\theta.
  • At section θ\theta, the horizontal distance from BB to the section is x=Rsin⁡θx = R \sin\theta.
  • Therefore, the bending moment due to WW is: M(θ)=−WRsin⁡θM(\theta) = -W R \sin\theta
  • To obtain the horizontal deflection, apply a fictitious horizontal load HH directed to the right at tip BB. The vertical distance from BB down to the section is y=R(1−cos⁡θ)y = R(1 - \cos\theta).
  • The total moment including the dummy load is: M(θ,H)=−WRsin⁡θ−HR(1−cos⁡θ)M(\theta, H) = -W R \sin\theta - H R (1 - \cos\theta)

Taking partial derivatives:

∂M∂W=−Rsin⁡θ,∂M∂H=−R(1−cos⁡θ)\frac{\partial M}{\partial W} = -R \sin\theta, \qquad \frac{\partial M}{\partial H} = -R(1 - \cos\theta)

1. Vertical Deflection δv\delta_v:

δv=∫0π/2M(θ,0)EI(∂M∂W)ds=1EI∫0π/2(−WRsin⁡θ)(−Rsin⁡θ)(R dθ)\delta_v = \int_0^{\pi/2} \frac{M(\theta, 0)}{EI} \left( \frac{\partial M}{\partial W} \right) ds = \frac{1}{EI} \int_0^{\pi/2} (-W R \sin\theta)(-R \sin\theta) (R \, d\theta) δv=WR3EI∫0π/2sin⁡2θ dθ\delta_v = \frac{W R^3}{EI} \int_0^{\pi/2} \sin^2\theta \, d\theta

Since ∫0π/2sin⁡2θ dθ=π4\int_0^{\pi/2} \sin^2\theta \, d\theta = \frac{\pi}{4}:

δv=πWR34EI\delta_v = \frac{\pi W R^3}{4 EI}

Substitute the values (W=5 kNW = 5\text{ kN}, R=1.2 mR = 1.2\text{ m}, EI=3000 kN⋅m2EI = 3000\text{ kN}\cdot\text{m}^2):

R3=(1.2)3=1.728 m3R^3 = (1.2)^3 = 1.728\text{ m}^3 δv=π×5.0×1.7284×3000=27.143412000=0.0022619 m=2.262 mm(downward)\delta_v = \frac{\pi \times 5.0 \times 1.728}{4 \times 3000} = \frac{27.1434}{12000} = 0.0022619\text{ m} = \mathbf{2.262\text{ mm}}\quad\text{(downward)}

2. Horizontal Deflection δh\delta_h: Set H→0H \to 0 in the integrand:

δh=lim⁡H→0∫0π/2M(θ,H)EI(∂M∂H)ds=1EI∫0π/2(−WRsin⁡θ)[−R(1−cos⁡θ)](R dθ)\delta_h = \lim_{H \to 0} \int_0^{\pi/2} \frac{M(\theta, H)}{EI} \left( \frac{\partial M}{\partial H} \right) ds = \frac{1}{EI} \int_0^{\pi/2} (-W R \sin\theta)[-R(1 - \cos\theta)] (R \, d\theta) δh=WR3EI∫0π/2(sin⁡θ−sin⁡θcos⁡θ) dθ\delta_h = \frac{W R^3}{EI} \int_0^{\pi/2} (\sin\theta - \sin\theta \cos\theta) \, d\theta

Evaluating the definite integral:

∫0π/2sin⁡θ dθ=[−cos⁡θ]0π/2=0−(−1)=1\int_0^{\pi/2} \sin\theta \, d\theta = [-\cos\theta]_0^{\pi/2} = 0 - (-1) = 1 ∫0π/2sin⁡θcos⁡θ dθ=[sin⁡2θ2]0π/2=12−0=12\int_0^{\pi/2} \sin\theta \cos\theta \, d\theta = \left[ \frac{\sin^2\theta}{2} \right]_0^{\pi/2} = \frac{1}{2} - 0 = \frac{1}{2}

Integral value =1−12=12= 1 - \frac{1}{2} = \frac{1}{2}.

δh=WR32EI\delta_h = \frac{W R^3}{2 EI} δh=5.0×1.7282×3000=8.646000=0.001440 m=1.440 mm(towards right)\delta_h = \frac{5.0 \times 1.728}{2 \times 3000} = \frac{8.64}{6000} = 0.001440\text{ m} = \mathbf{1.440\text{ mm}}\quad\text{(towards right)}

3. Ratio Verification:

δvδh=(π/4)(WR3/EI)(1/2)(WR3/EI)=π2≈1.5708\frac{\delta_v}{\delta_h} = \frac{(\pi / 4) (W R^3 / EI)}{(1 / 2) (W R^3 / EI)} = \frac{\pi}{2} \approx \mathbf{1.5708} Calculated: 2.2619 mm1.4400 mm=1.5708\text{Calculated: } \frac{2.2619\text{ mm}}{1.4400\text{ mm}} = 1.5708

The analytical prediction is perfectly verified.


9. Comprehensive Beam Deflection Method Selection Matrix

With all four deflection methods now mastered across Chapter 07, civil engineers must select the optimal method based on structural configuration:

Method Best Suited For Critical Limitations Exam Preference
Double Integration Uniform, single-segment prismatic beams (EI=constEI = \text{const}) with continuous loading. Cumbersome for multiple point loads and discontinuous spans due to boundary constants. Textbooks & basic derivations; rarely used in GATE for complex beams.
Macaulay's Method Prismatic simply supported beams with multiple concentrated loads and partial UDLs. Not suited for stepped/variable cross-sections (EI1≠EI2EI_1 \ne EI_2) or curved geometries. Highest priority for SSC JE, State AE, and standard GATE prismatic beams.
Moment-Area (Mohr) Stepped / non-prismatic cantilevers, symmetric spans, and locations of known zero slope. Difficult when neither slope nor tangent point is readily known (e.g., unsymmetric beams). Best for stepped cantilevers and fixed-end moment derivations.
Conjugate Beam Complex support configurations, internal hinges, overhanging spans, non-prismatic sections. Requires converting supports; sign conventions of conjugate shear/moment can trip students. Standard method for GATE / ESE involving internal hinges and overhangs.
Castigliano / Strain Energy Orthogonal bent frames, portal frames, trusses, and curved rings / circular arches. Computationally tedious for long multi-span continuous beams; requires dummy loads for unloaded points. Mandatory method for curved members, frames, and indeterminate reactions.

10. Common Misconceptions & Exam Pitfalls

[!WARNING] Pitfall 1: Squaring the Moment Expression Before Differentiating Expanding (M1+M2+… )2(M_1 + M_2 + \dots)^2 before taking the derivative leads to massive polynomial terms and inevitable algebraic errors. Always differentiate under the integral sign first: ∫MEI∂M∂Pdx\int \frac{M}{EI} \frac{\partial M}{\partial P} dx.

[!CAUTION] Pitfall 2: Forgetting to Set the Dummy Load to Zero Before Integration Students often keep QQ as a variable throughout the integration, resulting in pages of wasted algebra. Once the partial derivative ∂M∂Q\frac{\partial M}{\partial Q} has been computed, set Q=0Q = 0 immediately inside the integrand before computing the integral.

[!WARNING] Pitfall 3: Applying Castigliano's 2nd Theorem to Non-Linear Elastic Systems Castigliano's Second Theorem (Δi=∂U/∂Pi\Delta_i = \partial U / \partial P_i) is derived from the equality of strain energy and complementary energy (U=U∗U = U^*). It is strictly valid only for linear elastic materials. For non-linear elastic systems, ∂U/∂Pi≠Δi\partial U / \partial P_i \ne \Delta_i; instead, Engesser's theorem applies: Δi=∂U∗/∂Pi\Delta_i = \partial U^* / \partial P_i.

[!NOTE] Pitfall 4: Neglecting Member Contributions in Bent Frames In bent cantilevers (such as the L-frame in Example 2), students frequently evaluate the integral only for the member carrying the direct load (BCBC), completely forgetting that the vertical leg (ABAB) undergoes flexure from the transferred moment and contributes significantly to both vertical and horizontal tip displacements.


11. Real-World Engineering Applications & Standard Codes

  1. Deflection Limits in IS 456:2000 (Clause 23.2):
    • The final deflection due to all loads including the effects of temperature, creep, and shrinkage should not normally exceed Span/250\text{Span} / 250.
    • Deflection including effects of temperature, creep, and shrinkage occurring after erection of partitions and application of finishes should not exceed Span/350\text{Span} / 350 or 20 mm20\text{ mm}, whichever is less.
  2. Deflection Limits in IS 800:2007 (Table 6):
    • Industrial building crane gantry girders: δmax⁡≤L/500\delta_{\max} \le L / 500 (electric overhead travelling crane) or L/400L / 400 (manually operated).
    • Roof purlins and building rafters: δmax⁡≤L/150\delta_{\max} \le L / 150 (elastic sheet roofing) or L/180L / 180 (brittle tile roofing).
  3. Curved Structural Elements:
    • Circular arch ribs, tunnel linings, and expansion loops in high-pressure steam and oil piping rely directly on Castigliano's strain energy formulations to absorb thermal expansion displacements without exceeding permissible flexural stresses.

12. Quick Revision Summary & Formula Sheet

Physical Quantity / Condition Formula / Expression Units
Bending Strain Energy (UU) U=∫0LM(x)22EI dxU = \int_0^L \frac{M(x)^2}{2 EI} \, dx N⋅m\text{N}\cdot\text{m} or J\text{J}
Castigliano's 1st Theorem Pi=∂U∂ΔiP_i = \frac{\partial U}{\partial \Delta_i} N\text{N} or kN\text{kN}
Castigliano's 2nd Theorem (Displacement) Δi=∂U∂Pi=∫0LM(x)EI(∂M(x)∂Pi)dx\Delta_i = \frac{\partial U}{\partial P_i} = \int_0^L \frac{M(x)}{EI} \left(\frac{\partial M(x)}{\partial P_i}\right) dx m\text{m} or mm\text{mm}
Castigliano's 2nd Theorem (Rotation) θi=∂U∂Mi=∫0LM(x)EI(∂M(x)∂Mi)dx\theta_i = \frac{\partial U}{\partial M_i} = \int_0^L \frac{M(x)}{EI} \left(\frac{\partial M(x)}{\partial M_i}\right) dx radians\text{radians}
Dummy / Unit Load Method Δ=∫0LM(x) m(x)EI dx\Delta = \int_0^L \frac{M(x) \, m(x)}{EI} \, dx m\text{m} or mm\text{mm}
Quarter-Circular Ring (Vertical Tip Deflection) δv=πWR34EI≈0.7854WR3EI\delta_v = \frac{\pi W R^3}{4 EI} \approx 0.7854 \frac{W R^3}{EI} m\text{m} or mm\text{mm}
Quarter-Circular Ring (Horizontal Tip Deflection) δh=WR32EI=0.5000WR3EI\delta_h = \frac{W R^3}{2 EI} = 0.5000 \frac{W R^3}{EI} m\text{m} or mm\text{mm}
Universal Ring Deflection Ratio δvδh=π2≈1.5708\frac{\delta_v}{\delta_h} = \frac{\pi}{2} \approx 1.5708 Dimensionless
L-Frame Vertical Deflection (Tip Load PP) δv=PL33EI+PL2HEI\delta_v = \frac{P L^3}{3 EI} + \frac{P L^2 H}{EI} m\text{m} or mm\text{mm}