ME 219: Mechanics of Deformable Solids 1
Naveen Chandrashekar
Estimated study time: 1 hr 34 min
Table of contents
Sources and References
Primary textbook — R.C. Hibbeler, Mechanics of Materials, 11th ed., Pearson, 2014.
Supplementary — R.C. Hibbeler, Engineering Mechanics: Statics, 14th ed., Pearson, 2015.
Online resources — MIT OpenCourseWare 2.001 Mechanics and Materials I (Fall 2006); Stanford University ME 180 Mechanics of Solids lecture notes; David J. Morin, Introduction to Classical Mechanics, Cambridge University Press (statics chapters).
Chapter 1: Statics Review — Forces, Moments, and Equilibrium
Statics is the branch of mechanics that studies bodies at rest or moving with constant velocity — bodies for which the net force and net moment are both zero. Before any material can be analyzed for internal stress and deformation, the external forces acting on a structure must be fully resolved. This chapter reviews the essential tools: vector decomposition, equilibrium conditions, moments, couples, distributed loads, and free-body diagrams.
1.1 Force Vectors in Two and Three Dimensions
A force is a vector quantity possessing both magnitude and direction. In two dimensions, a force \( \mathbf{F} \) of magnitude \( F \) acting at an angle \( \theta \) measured counterclockwise from the positive \( x \)-axis has Cartesian components:
\[ F_x = F \cos\theta, \qquad F_y = F \sin\theta \]The resultant of a system of concurrent forces is obtained by summing each Cartesian component independently:
\[ R_x = \sum F_x, \qquad R_y = \sum F_y, \qquad R = \sqrt{R_x^2 + R_y^2} \]In three dimensions, a force is expressed in Cartesian form as:
\[ \mathbf{F} = F_x \hat{i} + F_y \hat{j} + F_z \hat{k} \]with magnitude \( F = \sqrt{F_x^2 + F_y^2 + F_z^2} \). The direction cosines \( \cos\alpha, \cos\beta, \cos\gamma \) are the direction angles that \( \mathbf{F} \) makes with the \( x \)-, \( y \)-, and \( z \)-axes respectively:
\[ \cos\alpha = \frac{F_x}{F}, \qquad \cos\beta = \frac{F_y}{F}, \qquad \cos\gamma = \frac{F_z}{F} \]and they satisfy \( \cos^2\alpha + \cos^2\beta + \cos^2\gamma = 1 \).
1.1.1 Position Vectors and Force Along a Line
When a force acts along the line connecting two points \( A(x_A, y_A, z_A) \) and \( B(x_B, y_B, z_B) \), the unit vector along \( AB \) is:
\[ \hat{u}_{AB} = \frac{\mathbf{r}_{AB}}{|\mathbf{r}_{AB}|}, \qquad \mathbf{r}_{AB} = (x_B - x_A)\hat{i} + (y_B - y_A)\hat{j} + (z_B - z_A)\hat{k} \]so the force is simply \( \mathbf{F} = F\hat{u}_{AB} \). This is the standard technique for cable and truss member forces.
Three forces act at a common point: \( \mathbf{F}_1 = 400 \) N at \( 30^\circ \), \( \mathbf{F}_2 = 600 \) N at \( 120^\circ \), \( \mathbf{F}_3 = 300 \) N at \( 270^\circ \) (all measured from positive \( x \)-axis counterclockwise).
Components:
\[ F_{1x} = 400\cos 30^\circ = 346.4 \text{ N}, \quad F_{1y} = 400\sin 30^\circ = 200.0 \text{ N} \]\[ F_{2x} = 600\cos 120^\circ = -300.0 \text{ N}, \quad F_{2y} = 600\sin 120^\circ = 519.6 \text{ N} \]\[ F_{3x} = 300\cos 270^\circ = 0 \text{ N}, \quad F_{3y} = 300\sin 270^\circ = -300.0 \text{ N} \]Resultant:
\[ R_x = 346.4 - 300.0 + 0 = 46.4 \text{ N} \]\[ R_y = 200.0 + 519.6 - 300.0 = 419.6 \text{ N} \]\[ R = \sqrt{46.4^2 + 419.6^2} = 421.2 \text{ N}, \quad \theta = \arctan\!\left(\frac{419.6}{46.4}\right) = 83.7^\circ \]1.2 Equilibrium of a Particle
A particle is in equilibrium when the vector sum of all forces acting on it is zero. In 2D and 3D respectively:
\[ \sum F_x = 0, \quad \sum F_y = 0 \qquad \text{(2D)} \]\[ \sum F_x = 0, \quad \sum F_y = 0, \quad \sum F_z = 0 \qquad \text{(3D)} \]The procedure is always: (1) isolate the particle, (2) draw the free-body diagram (FBD) with all external forces, (3) resolve into components, (4) solve the resulting algebraic equations.
1.2.1 Trusses
A truss is an assemblage of two-force members connected at frictionless pins. Each member carries only axial force — tension or compression. Two classical methods of analysis are the method of joints and the method of sections.
In the method of joints, equilibrium is written for each joint (treated as a particle) in turn, yielding two scalar equations per joint in 2D. In the method of sections, a cut is passed through at most three members whose forces are unknown, and moment equilibrium about a clever point can isolate a single unknown.
1.3 Moment of a Force
The tendency of a force to rotate a body about a point or axis is measured by its moment (torque). For a 2D planar problem, the moment of force \( \mathbf{F} \) about point \( O \) has magnitude:
\[ M_O = F \cdot d \]where \( d \) is the perpendicular distance (moment arm) from \( O \) to the line of action of \( \mathbf{F} \). The sign is positive counterclockwise by convention.
In 3D, with position vector \( \mathbf{r} \) from \( O \) to any point on the line of action:
\[ \mathbf{M}_O = \mathbf{r} \times \mathbf{F} \]In three dimensions, the cross product gives:
\[ \mathbf{M}_O = \begin{vmatrix} \hat{i} & \hat{j} & \hat{k} \\ r_x & r_y & r_z \\ F_x & F_y & F_z \end{vmatrix} = (r_y F_z - r_z F_y)\hat{i} - (r_x F_z - r_z F_x)\hat{j} + (r_x F_y - r_y F_x)\hat{k} \]The scalar moment about a specific axis \( a \) (say, the \( z \)-axis) is the component of \( \mathbf{M}_O \) along that axis.
1.3.1 Moment of a Couple
A couple is a pair of forces equal in magnitude, opposite in direction, and separated by a distance \( d \). The resultant force of a couple is zero, but the net moment is non-zero:
\[ M = F \cdot d \]1.4 Equivalent Force Systems and Resultants
Any system of forces and couples can be reduced to a single resultant force \( \mathbf{F}_R \) plus a resultant moment \( \mathbf{M}_{R,O} \) about a chosen point \( O \):
\[ \mathbf{F}_R = \sum \mathbf{F}_i, \qquad \mathbf{M}_{R,O} = \sum \mathbf{M}_O + \sum \mathbf{M}_{\text{couple},i} \]If the resultant force is non-zero, the system can often be further reduced to a single force acting at a shifted point — the location where \( \mathbf{M}_{R} = 0 \).
1.4.1 Distributed Loads
A distributed load (load per unit length, \( w(x) \) in N/m) acting along a beam is statically equivalent to a single concentrated force equal to the area under the load diagram, acting at the centroid of that area.
For a uniform distributed load of intensity \( w_0 \) over length \( L \):
\[ F_R = w_0 L, \qquad \bar{x} = \frac{L}{2} \]For a triangular load varying from 0 at one end to \( w_0 \) at the other end over length \( L \):
\[ F_R = \frac{1}{2} w_0 L, \qquad \bar{x} = \frac{2L}{3} \text{ from the zero end} \]1.5 Equilibrium of a Rigid Body
A rigid body in static equilibrium satisfies:
\[ \sum \mathbf{F} = \mathbf{0}, \qquad \sum \mathbf{M}_O = \mathbf{0} \]for any point \( O \). In 2D this gives three independent scalar equations: \( \sum F_x = 0 \), \( \sum F_y = 0 \), \( \sum M_O = 0 \). With three equations, at most three unknowns can be determined — a structure with exactly three unknowns is statically determinate.
Supports introduce reaction forces and moments. A roller provides one force (normal to surface), a pin provides two force components, and a fixed (cantilever) support provides two force components plus one moment.
A simply supported beam of length \( L = 6 \) m carries a uniform distributed load \( w_0 = 10 \) kN/m over its full length and a point load \( P = 30 \) kN at \( x = 4 \) m from the left pin support \( A \). Find the reactions at \( A \) (pin) and \( B \) (roller at \( x = 6 \) m).
Replace distributed load with resultant:
\[ F_w = 10 \times 6 = 60 \text{ kN at } x = 3 \text{ m} \]Sum moments about A:
\[ \sum M_A = 0: \quad B_y(6) - 60(3) - 30(4) = 0 \implies B_y = \frac{180 + 120}{6} = 50 \text{ kN} \]Sum forces vertically:
\[ \sum F_y = 0: \quad A_y + 50 - 60 - 30 = 0 \implies A_y = 40 \text{ kN} \]Sum forces horizontally: \( A_x = 0 \).
The reactions are \( A_y = 40 \) kN (up), \( A_x = 0 \), and \( B_y = 50 \) kN (up).
Chapter 2: Internal Forces and Cross-Section Properties
Once external reactions are known, internal forces at any cross-section can be determined by the method of sections. Cross-section geometric properties — centroids and moments of inertia — govern how internal forces produce stress.
2.1 Internal Forces and Free-Body Diagrams
To find the internal forces at a cross-section, make an imaginary cut, retain one portion, and apply equilibrium. The cut exposes an internal normal force \( N \), shear force \( V \), and bending moment \( M \) (in a 2D beam).
- Normal force \( N \): positive when tensile (pointing away from the cross-section face).
- Shear force \( V \): positive when the left-hand portion acts upward on the right face (or equivalently, when the portion to the left tends to slide upward relative to the right).
- Bending moment \( M \): positive when it causes sagging (concave up), i.e., tension in the bottom fiber.
The procedure: (1) compute support reactions, (2) cut at the desired location \( x \), (3) draw FBD of the left (or right) portion, (4) write \( \sum F_y = 0 \) and \( \sum M_{\text{cut}} = 0 \) to find \( V \) and \( M \).
2.2 Shear Force and Bending Moment Diagrams
The distributions \( V(x) \) and \( M(x) \) along the beam length are the shear and moment diagrams. Two differential relations connect the load, shear, and moment:
Integrating between two sections \( A \) and \( B \):
\[ V_B - V_A = -\int_A^B w(x)\,dx, \qquad M_B - M_A = \int_A^B V(x)\,dx \]Vertical equilibrium:
\[ V - (V + dV) - w(x)\,dx = 0 \implies \frac{dV}{dx} = -w(x) \]Moment equilibrium about the right face (dropping second-order term \( w(x)(dx)^2/2 \)):
\[ -M + (M + dM) - V\,dx = 0 \implies \frac{dM}{dx} = V \]Key features to remember:
- Where a concentrated force \( P \) acts, \( V \) jumps by \( P \).
- Where a concentrated couple \( M_0 \) acts, \( M \) jumps by \( M_0 \).
- \( M \) is maximum where \( V = 0 \) (provided no point couples exist there).
- Under a uniform load, \( V \) is linear and \( M \) is parabolic.
For the beam from Example 1.2 (\( w_0 = 10 \) kN/m, \( P = 30 \) kN at \( x = 4 \) m, \( A_y = 40 \) kN, \( B_y = 50 \) kN):
Region \( 0 \le x \le 4 \) m (left of point load):
\[ V(x) = 40 - 10x \quad \text{(kN)} \]\[ M(x) = 40x - \frac{10x^2}{2} = 40x - 5x^2 \quad \text{(kN\(\cdot\)m)} \]At \( x = 4 \): \( V(4^-) = 40 - 40 = 0 \) kN, \( M(4) = 160 - 80 = 80 \) kN\(\cdot\)m.
At \( x = 4^+ \), the shear jumps: \( V(4^+) = 0 - 30 = -30 \) kN.
Region \( 4 \le x \le 6 \) m:
\[ V(x) = 40 - 10x - 30 = 10 - 10x \quad \text{(kN)} \]\[ M(x) = 40x - 5x^2 - 30(x - 4) \quad \text{(kN\(\cdot\)m)} \]At \( x = 6 \): \( V(6^-) = 10 - 60 = -50 \) kN (equals \( -B_y \), consistent). \( M(6) = 240 - 180 - 30(2) = 0 \) (consistent with roller support).
2.3 Centroids of Cross-Sections
The centroid of a plane area is the geometric center, defined by:
\[ \bar{x} = \frac{\int x\,dA}{A}, \qquad \bar{y} = \frac{\int y\,dA}{A} \]For composite areas (built from simple shapes), the centroid is found by a weighted average:
\[ \bar{x} = \frac{\sum \tilde{x}_i A_i}{\sum A_i}, \qquad \bar{y} = \frac{\sum \tilde{y}_i A_i}{\sum A_i} \]where \( \tilde{x}_i, \tilde{y}_i \) are the centroidal coordinates of each sub-area \( A_i \). Holes are treated as negative areas.
2.4 Moment of Inertia and the Parallel-Axis Theorem
The second moment of area (moment of inertia) about the centroidal \( x \)- and \( y \)-axes measures the distribution of area relative to those axes:
\[ \bar{I}_x = \int y^2\,dA, \qquad \bar{I}_y = \int x^2\,dA \]Standard results for simple shapes (centroidal axes):
- Rectangle \( b \times h \): \( \bar{I}_x = \frac{bh^3}{12} \), \( \bar{I}_y = \frac{hb^3}{12} \)
- Circle radius \( r \): \( \bar{I}_x = \bar{I}_y = \frac{\pi r^4}{4} \)
- Triangle (base \( b \), height \( h \)): \( \bar{I}_x = \frac{bh^3}{36} \)
where \( A \) is the total area of the sub-region and \( d \) is the perpendicular distance between the two parallel axes.
The middle term \( \int \bar{y}\,dA = 0 \) by definition of the centroid. Hence \( I = \bar{I} + Ad^2 \).
For composite sections, the total moment of inertia is:
\[ I_{\text{total}} = \sum \left(\bar{I}_i + A_i d_i^2\right) \]A T-section has a flange 100 mm wide \( \times \) 20 mm thick on top, and a web 20 mm wide \( \times \) 80 mm tall below the flange.
Areas and centroids (measured from the bottom of the web):
Flange: \( A_1 = 100 \times 20 = 2000 \) mm\(^2\), \( \tilde{y}_1 = 80 + 10 = 90 \) mm.
Web: \( A_2 = 20 \times 80 = 1600 \) mm\(^2\), \( \tilde{y}_2 = 40 \) mm.
Centroid of section:
\[ \bar{y} = \frac{2000 \times 90 + 1600 \times 40}{2000 + 1600} = \frac{180000 + 64000}{3600} = 67.8 \text{ mm from bottom} \]Distances from centroid:
\( d_1 = 90 - 67.8 = 22.2 \) mm; \( d_2 = 67.8 - 40 = 27.8 \) mm.
Moment of inertia about centroidal axis:
\[ I = \left(\frac{100 \times 20^3}{12} + 2000 \times 22.2^2\right) + \left(\frac{20 \times 80^3}{12} + 1600 \times 27.8^2\right) \]\[ I = \left(66\,667 + 985\,680\right) + \left(853\,333 + 1\,235\,584\right) = 3\,141\,264 \text{ mm}^4 \approx 3.14 \times 10^6 \text{ mm}^4 \]Chapter 3: Stress and Strain
With the internal forces known and the cross-section geometry characterized, we can define the fundamental quantities that describe the mechanical state at a point in a solid body: stress and strain.
3.1 Normal Stress
For a member of cross-sectional area \( A \) subjected to an axial force \( N \) uniformly distributed over the section:
\[ \sigma = \frac{N}{A} \]Units: Pa \( = \) N/m\(^2\), or MPa \( = \) N/mm\(^2\). Positive \( \sigma \) is tensile.
The assumption of uniform normal stress is valid when the load is applied at the centroid of the section and the cross-section is remote from any stress concentrations (Saint-Venant’s principle, discussed in Chapter 4).
3.2 Shear Stress
For a pin or bolt of cross-sectional area \( A \) carrying shear force \( V \):
\[ \tau_{\text{avg}} = \frac{V}{A} \]Shear stress has two subscripts: the first indicates the face normal direction, the second indicates the direction of the stress component. By the complementary property of shear: \( \tau_{xy} = \tau_{yx} \), \( \tau_{xz} = \tau_{zx} \), \( \tau_{yz} = \tau_{zy} \).
3.2.1 Bearing Stress
Bearing stress acts on the contact area between two bodies:
\[ \sigma_b = \frac{P}{A_b} \]where \( A_b = t \cdot d \) is the projected area (thickness times pin diameter) for a pin-in-hole connection.
3.3 Factor of Safety
Typical values range from 1.5 (well-understood, non-critical) to 3 or more (dynamic loads, high uncertainty).
3.4 Normal Strain
where \( \delta \) is the total deformation and \( L \) is the original gauge length. Strain is dimensionless (m/m or mm/mm). Positive strain is elongation.
3.5 Shear Strain
where \( \theta' \) is the angle between the two elements after deformation. For small angles, \( \gamma \approx \tan\gamma \).
3.6 Mechanical Properties of Materials
3.6.1 The Stress–Strain Diagram
The stress–strain diagram is obtained by loading a standard tensile specimen and recording \( \sigma \) versus \( \varepsilon \). For a typical mild steel, the diagram exhibits:
- Proportional limit: up to this stress, \( \sigma \) and \( \varepsilon \) are linearly proportional.
- Elastic limit: the maximum stress for which the material is fully elastic (no permanent deformation).
- Yield point: stress at which significant permanent deformation begins (often accompanied by a visible yield plateau in mild steel).
- Ultimate stress \( \sigma_u \): the maximum engineering stress.
- Fracture stress: the stress at which the specimen breaks (lower than \( \sigma_u \) in engineering stress due to necking).
Typical values: steel \( E \approx 200 \) GPa, aluminum \( E \approx 70 \) GPa, concrete \( E \approx 30 \) GPa.
For most structural metals, \( 0.25 \le \nu \le 0.35 \). Thermodynamic stability requires \( -1 \le \nu \le 0.5 \).
where \( G \) is the shear modulus. It is related to \( E \) and \( \nu \) by:
\[ G = \frac{E}{2(1 + \nu)} \]3.6.2 Ductility and Toughness
- Percent elongation: \( \%EL = \dfrac{L_f - L_0}{L_0} \times 100\% \)
- Percent reduction in area: \( \%RA = \dfrac{A_0 - A_f}{A_0} \times 100\% \)
3.6.3 True Stress and True Strain
Engineering stress and strain are computed using original dimensions. True (Cauchy) stress and logarithmic strain use the current (deformed) dimensions:
\[ \sigma_{\text{true}} = \frac{P}{A_{\text{current}}}, \qquad \varepsilon_{\text{true}} = \ln\!\left(\frac{L_f}{L_0}\right) \]True stress is always higher than engineering stress beyond the yield point because the cross-sectional area decreases. True strain is related to engineering strain by \( \varepsilon_{\text{true}} = \ln(1 + \varepsilon_{\text{eng}}) \).
Chapter 4: Axial Loading
Axial loading is the simplest case of internal force application: a prismatic bar subjected to forces along its centroidal axis. This chapter derives the deformation formula, treats statically indeterminate bars, introduces thermal effects, and discusses Saint-Venant’s principle.
4.1 Deformation of an Axially Loaded Bar
Consider a bar of original length \( L \), cross-sectional area \( A \), and elastic modulus \( E \), subjected to axial force \( N \). From Hooke’s Law:
\[ \sigma = \frac{N}{A} = E\varepsilon = E\frac{\delta}{L} \]Solving for deformation:
\[ \delta = \frac{NL}{AE} \]For a bar composed of \( n \) prismatic segments:
\[ \delta = \sum_{i=1}^n \frac{N_i L_i}{A_i E_i} \]The quantity \( AE \) is called the axial stiffness of the member.
A stepped steel bar (\( E = 200 \) GPa) consists of segment AB: \( L_1 = 0.5 \) m, \( A_1 = 400 \) mm\(^2\), carrying \( N_1 = 80 \) kN tension; and segment BC: \( L_2 = 0.8 \) m, \( A_2 = 200 \) mm\(^2\), carrying \( N_2 = -40 \) kN (compression).
\[ \delta_{AB} = \frac{80{,}000 \times 500}{400 \times 200{,}000} = \frac{4 \times 10^7}{8 \times 10^7} = 0.5 \text{ mm (elongation)} \]\[ \delta_{BC} = \frac{-40{,}000 \times 800}{200 \times 200{,}000} = \frac{-3.2 \times 10^7}{4 \times 10^7} = -0.8 \text{ mm (shortening)} \]Total deformation of bar: \( \delta_{\text{total}} = 0.5 - 0.8 = -0.3 \) mm (net shortening).
4.2 Statically Indeterminate Axial Structures
A structure is statically indeterminate when the number of unknown reactions exceeds the number of independent equilibrium equations. Additional equations come from compatibility (geometric) conditions.
The three-step approach for statically indeterminate problems:
- Equilibrium: write \( \sum F = 0 \) involving the unknowns.
- Compatibility: express the geometric constraint (e.g., total deformation is zero for a fixed–fixed bar).
- Force–deformation: substitute \( \delta = NL/(AE) \) to obtain additional equations.
A steel bar (\( E = 200 \) GPa, \( A = 300 \) mm\(^2\)) of total length 1.0 m is fixed at both ends A and B. A force \( P = 90 \) kN is applied at point C, located 0.4 m from A.
Step 1 — Equilibrium:
Let \( R_A \) and \( R_B \) be the reactions (positive = tension on bar from wall).
\[ R_A + R_B = P = 90 \text{ kN} \]Step 2 — Compatibility:
The total elongation of the bar is zero (both ends are fixed):
\[ \delta_{AC} + \delta_{CB} = 0 \]Step 3 — Force–deformation:
Segment AC carries \( N_{AC} = -R_A \) (compressive if \( R_A \) reacts upward) — more carefully: taking rightward positive, \( N_{AC} = -R_A \) and \( N_{CB} = R_B \) if \( P \) acts rightward. Using the standard setup:
\[ \frac{R_A \times 0.4}{AE} + \frac{-R_B \times 0.6}{AE} = 0 \implies R_A(0.4) = R_B(0.6) \]Combined with \( R_A + R_B = 90 \):
\[ R_B = \frac{0.4}{0.4 + 0.6} \times 90 = 36 \text{ kN}, \qquad R_A = 54 \text{ kN} \]4.3 Thermal Strain
A temperature change \( \Delta T \) causes a material to expand or contract. The thermal strain is:
\[ \varepsilon_T = \alpha \Delta T \]where \( \alpha \) is the coefficient of thermal expansion (units: 1/°C or 1/K). For steel, \( \alpha \approx 12 \times 10^{-6} \) /°C; for aluminum, \( \alpha \approx 23 \times 10^{-6} \) /°C.
The thermal deformation of a free bar:
\[ \delta_T = \alpha \Delta T L \]If the bar is constrained, the thermal deformation is suppressed and a thermal stress develops. Compatibility requires the mechanical deformation to equal the negative of the thermal deformation:
\[ \delta_{\text{mech}} + \delta_T = 0 \implies \frac{\sigma L}{E} + \alpha \Delta T L = 0 \implies \sigma = -E\alpha\Delta T \]A temperature increase in a constrained bar produces compressive stress.
4.4 Saint-Venant’s Principle
This principle justifies the use of \( \sigma = N/A \) for prismatic bars away from connections, holes, or abrupt changes in geometry. At stress concentrations, the local peak stress exceeds the average by a stress concentration factor \( K \):
\[ \sigma_{\max} = K \sigma_{\text{avg}} \]Values of \( K \) depend on the geometry and are tabulated in reference handbooks. For a circular hole in a wide plate under uniaxial tension, \( K = 3 \) at the edges of the hole.
Chapter 5: Torsion
Torsion arises when a structural member is twisted about its longitudinal axis by applied torques. The analysis of circular shafts is exact within linear elasticity; non-circular sections require more advanced methods (not covered in ME 219).
5.1 Shear Stress in Circular Shafts
Consider a solid circular shaft of radius \( c \) and length \( L \), subjected to an internal torque \( T \). The fundamental assumptions are: plane cross-sections remain plane and circular after twisting; radii remain straight; material is linear elastic and homogeneous; the shaft is in pure torsion (no bending or axial load).
Consider a small radial line element at radius \( \rho \) in the cross-section. After the shaft twists by angle \( \phi \) over length \( L \), this element shears by arc \( \rho\phi \). The shear strain at radius \( \rho \) is:
\[ \gamma(\rho) = \frac{\rho\phi}{L} \]By Hooke’s Law, shear stress is linear in \( \rho \):
\[ \tau(\rho) = G\gamma = \frac{G\phi}{L}\rho \]The stress distribution is linear, zero at the center, maximum at the outer radius. The resultant torque must equal the internal torque \( T \):
\[ T = \int_A \rho\,\tau(\rho)\,dA = \int_A \rho \cdot \frac{G\phi}{L}\rho\,dA = \frac{G\phi}{L}\int_A \rho^2\,dA = \frac{G\phi}{L} J \]where \( J = \int_A \rho^2\,dA \) is the polar moment of inertia. Therefore:
\[ \frac{G\phi}{L} = \frac{T}{J} \]Substituting back:
\[ \tau(\rho) = \frac{T\rho}{J} \]The maximum shear stress occurs at the outer surface \( \rho = c \):
\[ \tau_{\max} = \frac{Tc}{J} \]Polar moments of inertia:
- Solid circle radius \( c \): \( J = \dfrac{\pi c^4}{2} \)
- Hollow shaft inner radius \( c_i \), outer radius \( c_o \): \( J = \dfrac{\pi}{2}(c_o^4 - c_i^4) \)
5.2 Angle of Twist
From the derivation above, the angle of twist \( \phi \) over a length \( L \) under torque \( T \) is:
\[ \phi = \frac{TL}{GJ} \]For a shaft with multiple segments or a varying torque distribution:
\[ \phi = \sum_{i=1}^n \frac{T_i L_i}{G_i J_i}, \qquad \text{or} \qquad \phi = \int_0^L \frac{T(x)}{GJ(x)}\,dx \]The sign convention for torque follows the right-hand rule: positive torque on a cross-section face whose outward normal points in the positive \( x \)-direction means the torque vector points in the positive \( x \)-direction.
A solid steel shaft (\( G = 80 \) GPa) consists of segment AB: diameter 60 mm, length 1.2 m, torque 1500 N\(\cdot\)m; and segment BC: diameter 40 mm, length 0.8 m, torque 800 N\(\cdot\)m.
Polar moments:
\[ J_{AB} = \frac{\pi (0.03)^4}{2} = 1.272 \times 10^{-6} \text{ m}^4 \]\[ J_{BC} = \frac{\pi (0.02)^4}{2} = 2.513 \times 10^{-7} \text{ m}^4 \]Maximum shear stresses:
\[ \tau_{AB} = \frac{1500 \times 0.03}{1.272 \times 10^{-6}} = 35.4 \text{ MPa} \]\[ \tau_{BC} = \frac{800 \times 0.02}{2.513 \times 10^{-7}} = 63.7 \text{ MPa} \]Angles of twist:
\[ \phi_{AB} = \frac{1500 \times 1.2}{80 \times 10^9 \times 1.272 \times 10^{-6}} = 0.01769 \text{ rad} = 1.01^\circ \]\[ \phi_{BC} = \frac{800 \times 0.8}{80 \times 10^9 \times 2.513 \times 10^{-7}} = 0.03183 \text{ rad} = 1.82^\circ \]Total twist of C relative to A: \( \phi = 0.01769 + 0.03183 = 0.0495 \) rad \( = 2.84^\circ \).
5.3 Statically Indeterminate Torsion
When a shaft is fixed at both ends, the torque reactions cannot be determined from equilibrium alone. The procedure parallels the axial case:
- Equilibrium: \( T_A + T_B = T_{\text{applied}} \)
- Compatibility: total angle of twist of one end relative to the other is zero.
- Force–deformation: \( \phi = TL/(GJ) \).
A uniform shaft (\( G, J, L \)) is fixed at A and B. A torque \( T_0 \) is applied at C, located \( aL \) from A (\( 0 < a < 1 \)).
Equilibrium: \( T_A + T_B = T_0 \).
Compatibility: \( \phi_{AC} + \phi_{CB} = 0 \) (both ends fixed).
\[ \frac{T_A \cdot aL}{GJ} - \frac{T_B \cdot (1-a)L}{GJ} = 0 \implies T_A a = T_B(1-a) \]Solving: \( T_A = T_0(1-a) \), \( T_B = T_0 a \).
The longer segment carries the smaller torque — again, stiffer paths attract more load.
5.4 Power Transmission
Rotating shafts transmit mechanical power. The relationship between power \( P \), torque \( T \), and angular velocity \( \omega \) (in rad/s) is:
\[ P = T\omega = T(2\pi f) = T(2\pi N/60) \]where \( f \) is frequency in Hz and \( N \) is rotation speed in rpm. Solving for the required torque for a given power:
\[ T = \frac{P}{\omega} \]This torque is then used in the torsion formula to check that \( \tau_{\max} \le \tau_{\text{allow}} \) and that the angle of twist does not exceed the design limit.
A solid steel shaft (\( G = 80 \) GPa, \( \tau_{\text{allow}} = 75 \) MPa) must transmit 50 kW at 600 rpm. Find the minimum required diameter.
Torque:
\[ T = \frac{P}{\omega} = \frac{50{,}000}{2\pi \times 600/60} = \frac{50{,}000}{62.83} = 795.8 \text{ N\(\cdot\)m} \]Required \( J/c \) from torsion formula:
\[ \frac{J}{c} \ge \frac{T}{\tau_{\text{allow}}} = \frac{795.8}{75 \times 10^6} = 1.061 \times 10^{-5} \text{ m}^3 \]For a solid shaft, \( J/c = \pi c^3/2 \):
\[ c \ge \left(\frac{2 \times 1.061 \times 10^{-5}}{\pi}\right)^{1/3} = \left(6.758 \times 10^{-6}\right)^{1/3} = 0.01884 \text{ m} \]Minimum diameter \( d = 2c = 37.7 \) mm. Round up to next standard size, e.g., 40 mm.
Chapter 6: Bending
Bending is the most common loading mode in structural engineering. When a beam carries transverse loads, internal bending moments develop that produce normal stresses varying across the cross-section. This chapter derives the flexure formula and extends it to composite sections and eccentric axial loading.
6.1 Assumptions in Pure Bending
The Euler–Bernoulli beam theory rests on several assumptions:
- The beam is initially straight and has a uniform cross-section.
- The material is linear elastic, homogeneous, and isotropic.
- The cross-section is symmetric about the plane of bending.
- Plane cross-sections remain plane after bending (the kinematic hypothesis).
- Deformations are small.
6.2 The Flexure Formula
Consider a beam segment bent to a radius of curvature \( \rho \). A fiber at distance \( y \) from the neutral axis (NA) has original length \( ds = \rho\,d\theta \). After bending, its length becomes \( (\rho - y)\,d\theta \) (the negative sign arises because positive \( y \) is measured upward, and a positive moment creates tension below the NA and compression above by convention in some texts — here we follow the convention that \( y \) is measured from the NA, positive upward, and positive moment creates compression above and tension below).
The normal strain at distance \( y \):
\[ \varepsilon = \frac{(\rho - y)\,d\theta - \rho\,d\theta}{\rho\,d\theta} = -\frac{y}{\rho} \]By Hooke’s Law:
\[ \sigma = E\varepsilon = -\frac{Ey}{\rho} \]The strain and stress vary linearly with \( y \). The neutral axis is where \( \sigma = 0 \), i.e., \( y = 0 \).
Locating the neutral axis: The net axial force across the section must be zero (pure bending):
\[ \int_A \sigma\,dA = -\frac{E}{\rho}\int_A y\,dA = 0 \implies \int_A y\,dA = 0 \]This means \( \bar{y} = 0 \) — the neutral axis passes through the centroid of the cross-section.
Relating curvature to moment: The net moment about the NA equals the internal bending moment \( M \):
\[ M = -\int_A y\sigma\,dA = \int_A y\cdot\frac{Ey}{\rho}\,dA = \frac{E}{\rho}\int_A y^2\,dA = \frac{EI}{\rho} \]where \( I = \int_A y^2\,dA \) is the second moment of area about the centroidal axis. Therefore:
\[ \frac{1}{\rho} = \frac{M}{EI} \]Substituting back into the stress expression:
\[ \sigma = -\frac{Ey}{\rho} = -\frac{My}{I} \]With the convention that positive \( M \) causes tension at positive \( y \) (below NA in many structural conventions), it is common to write:
or equivalently (taking absolute value for maximum stress at \( y = c \), the distance to the extreme fiber):
\[ \sigma_{\max} = \frac{Mc}{I} = \frac{M}{S} \]where \( S = I/c \) is the section modulus (units: m\(^3\) or mm\(^3\)).
Positive \( M \) (sagging) produces tension at the bottom fiber (negative \( y \)) and compression at the top fiber (positive \( y \)).
The section modulus \( S \) is a single cross-section property that encapsulates both \( I \) and the extreme-fiber distance \( c \). Maximizing \( S \) for a given cross-sectional area is the design goal: I-beams and wide-flange sections concentrate material far from the neutral axis to maximize \( S \) and hence minimize bending stress.
A simply supported timber beam (rectangular section: 100 mm wide, 200 mm deep; \( E = 12 \) GPa) spans 4 m and carries a midspan point load \( P = 20 \) kN. Find the maximum bending stress.
Maximum moment (at midspan):
\[ M_{\max} = \frac{PL}{4} = \frac{20{,}000 \times 4}{4} = 20{,}000 \text{ N\(\cdot\)m} \]Section modulus:
\[ I = \frac{bh^3}{12} = \frac{100 \times 200^3}{12} = 66.67 \times 10^6 \text{ mm}^4 \]\[ S = \frac{I}{c} = \frac{66.67 \times 10^6}{100} = 666.7 \times 10^3 \text{ mm}^3 \]Maximum bending stress:
\[ \sigma_{\max} = \frac{M}{S} = \frac{20{,}000 \times 10^3 \text{ N\(\cdot\)mm}}{666.7 \times 10^3 \text{ mm}^3} = 30.0 \text{ MPa} \]Tension at the bottom fiber, compression at the top.
The T-section from Example 2.2 has centroid at \( \bar{y} = 67.8 \) mm from the bottom and \( I = 3.14 \times 10^6 \) mm\(^4\). If the section carries a positive bending moment \( M = 25 \) kN\(\cdot\)m:
Distances from neutral axis to extreme fibers:
\( c_{\text{bottom}} = 67.8 \) mm (tension, bottom)
\( c_{\text{top}} = (80 + 20) - 67.8 = 32.2 \) mm (compression, top)
Stresses:
\[ \sigma_{\text{bottom}} = \frac{M \cdot c_{\text{bottom}}}{I} = \frac{25 \times 10^6 \times 67.8}{3.14 \times 10^6} = 539.8 \text{ MPa (tension)} \]\[ \sigma_{\text{top}} = \frac{M \cdot c_{\text{top}}}{I} = \frac{25 \times 10^6 \times 32.2}{3.14 \times 10^6} = 256.4 \text{ MPa (compression)} \]The unsymmetric section means different extreme fiber stresses even for the same \( M \). For efficient design, the material should be placed such that the larger extreme fiber distance corresponds to the material’s stronger direction.
6.3 Composite Beams
A composite beam is made of two or more materials bonded together. The analysis uses the transformed-section method: convert all materials to an equivalent cross-section in a single reference material (usually the stiffer one).
To transform material 2 into the reference material 1, multiply the width of material 2’s cross-section by \( n \). Thicknesses and lengths remain unchanged.
After transforming the section, the flexure formula applies to the transformed section with reference modulus \( E_1 \):
\[ \sigma_1 = -\frac{My}{I_T}, \qquad \sigma_2 = -n\frac{My}{I_T} \]where \( I_T \) is the moment of inertia of the transformed section about its centroidal axis, and \( y \) is measured from the centroid of the transformed section.
The key insight: both materials have the same strain at the same \( y \)-location (perfect bonding, plane sections remain plane), but the stress differs by the factor \( n \) due to the different stiffnesses.
At any \( y \) from the neutral axis, the kinematic assumption gives:
\[ \varepsilon = -\frac{y}{\rho} = -\frac{My}{EI} \]This is the same for all materials bonded together. The stress in each material is:
\[ \sigma_k = E_k \varepsilon = -\frac{E_k M y}{E_1 I_T} \]For material 1: \( \sigma_1 = -My/I_T \). For material 2: \( \sigma_2 = -(E_2/E_1)(My/I_T) = -n(My/I_T) \).
The transformed section replaces material 2’s width \( b_2 \) with \( nb_2 \) so that the integrand \( \int y^2 dA \) for the transformed section automatically accounts for the difference in stiffness:
\[ I_T = \int_{A_1} y^2\,dA + n\int_{A_2} y^2\,dA \]A rectangular wood beam (60 mm wide, 120 mm deep, \( E_w = 10 \) GPa) has a steel plate (60 mm wide, 10 mm thick, \( E_s = 200 \) GPa) bonded to its bottom. The assembly carries a positive bending moment \( M = 8 \) kN\(\cdot\)m. Find the maximum stress in each material.
Modular ratio: \( n = E_s/E_w = 200/10 = 20 \).
Transformed section (reference: wood): Replace the steel plate with a wood-equivalent plate 60 \( \times \) 20 = 1200 mm wide, 10 mm thick.
Transformed areas and centroids (from bottom of steel plate):
Steel (transformed): \( A_T^{(s)} = 1200 \times 10 = 12{,}000 \) mm\(^2\), \( \tilde{y}_s = 5 \) mm.
Wood: \( A_w = 60 \times 120 = 7{,}200 \) mm\(^2\), \( \tilde{y}_w = 10 + 60 = 70 \) mm.
Centroid of transformed section:
\[ \bar{y} = \frac{12{,}000 \times 5 + 7{,}200 \times 70}{12{,}000 + 7{,}200} = \frac{60{,}000 + 504{,}000}{19{,}200} = 29.4 \text{ mm from bottom} \]Moment of inertia of transformed section:
\[ I_T = \left(\frac{1200 \times 10^3}{12} + 12{,}000 \times (29.4 - 5)^2\right) + \left(\frac{60 \times 120^3}{12} + 7{,}200 \times (70 - 29.4)^2\right) \]\[ = \left(100{,}000 + 7{,}112{,}320\right) + \left(8{,}640{,}000 + 11{,}850{,}048\right) = 27{,}702{,}368 \text{ mm}^4 \approx 2.77 \times 10^7 \text{ mm}^4 \]Maximum wood stress (top fiber, \( y = 130 - 29.4 = 100.6 \) mm above NA):
\[ \sigma_w = \frac{M \cdot y_{\text{top}}}{I_T} = \frac{8 \times 10^6 \times 100.6}{2.77 \times 10^7} = 29.1 \text{ MPa (compression)} \]Maximum steel stress (bottom of steel, \( y = 29.4 \) mm below NA):
\[ \sigma_s = n \cdot \frac{M \cdot y_{\text{bottom}}}{I_T} = 20 \times \frac{8 \times 10^6 \times 29.4}{2.77 \times 10^7} = 170.1 \text{ MPa (tension)} \]6.4 Eccentric Axial Loading
When an axial load is applied eccentrically (not at the centroid of the cross-section), it produces both axial normal stress and bending stress. The total normal stress is the superposition:
\[ \sigma = \frac{N}{A} \pm \frac{M y}{I} \]where \( M = N \cdot e \) is the bending moment created by the eccentricity \( e \) (distance from centroid to load line).
where \( M_z = N e_y \) and \( M_y = N e_z \), and \( y, z \) are measured from the centroid.
A short concrete column (\( 300 \) mm \( \times \) \( 300 \) mm square cross-section, \( E = 25 \) GPa) carries a compressive load \( P = 800 \) kN applied at eccentricity \( e = 50 \) mm from the centroidal axis.
Geometric properties:
\[ A = 300 \times 300 = 90{,}000 \text{ mm}^2 \]\[ I = \frac{300^4}{12} = 675 \times 10^6 \text{ mm}^4, \qquad c = 150 \text{ mm} \]Internal forces at any cross-section:
\( N = -800 \) kN (compressive, negative), \( M = N \cdot e = -800{,}000 \times 50 = -40 \times 10^6 \) N\(\cdot\)mm (note: sign depends on sense of eccentricity).
Stresses at extreme fibers:
Axial component: \( \sigma_N = N/A = -800{,}000/90{,}000 = -8.89 \) MPa.
Bending component at \( y = +150 \) mm: \( \sigma_{M,+} = -My/I = -(-40 \times 10^6)(150)/(675 \times 10^6) = +8.89 \) MPa.
At \( y = -150 \) mm: \( \sigma_{M,-} = -(-40 \times 10^6)(-150)/(675 \times 10^6) = -8.89 \) MPa.
Total stress:
At \( y = +150 \) mm: \( \sigma = -8.89 + 8.89 = 0 \) MPa. (Neutral fiber!)
At \( y = -150 \) mm: \( \sigma = -8.89 - 8.89 = -17.78 \) MPa (compression).
The eccentricity of \( e = c = 150 \) mm places the neutral axis exactly at the far edge of the section; any greater eccentricity would cause tension on one face, which concrete cannot sustain in design.
Supplementary Topics and Integration
Review: Putting It Together — A Complete Structural Analysis
Modern structural analysis in ME 219 proceeds through the following logical chain, which unifies all six chapters:
Statics (Chapters 1–2): Find external reactions using equilibrium of the entire structure. Then cut at the section of interest and draw a free-body diagram to find internal forces \( N \), \( V \), \( M \), \( T \).
Cross-section geometry (Chapter 2): Calculate \( A \), \( \bar{y} \), \( I \), \( J \), \( c \) for the cross-section.
Stress (Chapters 3–6): Apply the appropriate formula:
- Axial: \( \sigma = N/A \)
- Bending: \( \sigma = My/I \)
- Torsion: \( \tau = T\rho/J \)
- Shear (transverse): \( \tau = VQ/(It) \) (advanced, beyond ME 219 scope here)
Deformation (Chapters 4–5): Calculate elongation \( \delta = NL/(AE) \) or twist \( \phi = TL/(GJ) \) to check serviceability.
Design check: Compare computed stresses against allowable values and verify factors of safety.
A steel truss member (\( E = 200 \) GPa, \( \sigma_{\text{allow}} = 150 \) MPa in tension) has a rectangular cross-section 50 mm \( \times \) 25 mm. It carries an axial tensile force of 120 kN applied eccentrically at 5 mm from the centroid of the 50 mm dimension.
Section properties:
\( A = 50 \times 25 = 1{,}250 \) mm\(^2\); \( I = (25)(50^3)/12 = 260{,}417 \) mm\(^4\); \( c = 25 \) mm.
Stresses:
\( \sigma_N = 120{,}000/1{,}250 = 96 \) MPa.
\( M = 120{,}000 \times 5 = 600{,}000 \) N\(\cdot\)mm.
\( \sigma_{M,\max} = 600{,}000 \times 25/260{,}417 = 57.6 \) MPa.
\( \sigma_{\max} = 96 + 57.6 = 153.6 \) MPa \( > 150 \) MPa — slightly overstressed.
Either increase the section size or reduce the eccentricity. If the eccentricity is reduced to 4 mm: \( M = 480{,}000 \), \( \sigma_M = 46.1 \) MPa, \( \sigma_{\max} = 142.1 \) MPa \( < 150 \) MPa — acceptable.
Sign Convention Summary
Consistency in sign conventions is essential. The following conventions are adopted throughout:
| Quantity | Positive |
|---|---|
| Normal stress \( \sigma \) | Tensile |
| Axial force \( N \) | Tensile (away from face) |
| Shear force \( V \) | Left-portion upward on right face |
| Bending moment \( M \) | Sagging (concave up), bottom in tension |
| Torque \( T \) | Right-hand rule along positive \( x \)-axis |
| Normal strain \( \varepsilon \) | Elongation |
| Shear strain \( \gamma \) | Decrease in right angle |
| Coordinate \( y \) (bending) | Upward from neutral axis |
Common Errors and How to Avoid Them
Key Formulas Reference
The following table collects the central formulas from all chapters:
| Topic | Formula | Conditions |
|---|---|---|
| Axial stress | \( \sigma = N/A \) | Uniform, centroidal |
| Axial deformation | \( \delta = NL/(AE) \) | Prismatic, elastic |
| Thermal deformation | \( \delta_T = \alpha \Delta T L \) | Unconstrained |
| Bending stress | \( \sigma = My/I \) | Elastic, homogeneous |
| Max bending stress | \( \sigma_{\max} = Mc/I = M/S \) | At extreme fiber |
| Torsional shear stress | \( \tau = T\rho/J \) | Circular shaft, elastic |
| Max torsional stress | \( \tau_{\max} = Tc/J \) | At outer surface |
| Angle of twist | \( \phi = TL/(GJ) \) | Uniform shaft |
| Parallel-axis theorem | \( I = \bar{I} + Ad^2 \) | Any parallel axis |
| Polar MOI (solid) | \( J = \pi c^4/2 \) | Solid circle |
| Section modulus | \( S = I/c \) | Bending design |
| Shear modulus | \( G = E/\left[2(1+\nu)\right] \) | Isotropic material |
| Poisson’s ratio | \( \nu = -\varepsilon_{\text{lat}}/\varepsilon_{\text{long}} \) | Uniaxial load |
Appendix: Physical Intuition and Conceptual Synthesis
Why the Neutral Axis Passes Through the Centroid
This result follows directly from the requirement that net axial force be zero in pure bending. Mathematically, \( \int_A \sigma\,dA = 0 \) with \( \sigma \propto y \) forces \( \int_A y\,dA = 0 \), which defines the centroid. Physically: if the neutral axis were above the centroid, more area would be in tension than compression, creating a net tensile force that would violate equilibrium in the absence of an applied axial load.
Energy Methods Preview
The elastic strain energy stored per unit volume in a member under normal stress is:
\[ u = \frac{\sigma^2}{2E} \]For a bar under axial load, integrating over the volume:
\[ U = \int_0^L \frac{N^2}{2AE}\,dx \]For a shaft under torsion:
\[ U = \int_0^L \frac{T^2}{2GJ}\,dx \]These expressions underlie Castigliano’s theorem and virtual work methods used in ME 321 and beyond.
The Importance of Linearity
All formulas in ME 219 assume linear elastic behavior: the constitutive law is Hooke’s law, deformations are small (so geometry does not change appreciably under load), and superposition is valid. These are excellent approximations for metals at stresses well below yield. The course lays the essential foundation; nonlinear behavior, large deformations, plasticity, and fracture mechanics are treated in upper-year courses.