Study the six listed topics as one connected chain: loads create internal actions (axial force, shear, bending moment, torque), internal actions create stresses, and stresses integrate into deflections. Fix one sign convention, write each formula's assumptions beside it, and verify every diagram and stress state against an independent relationship before moving on.
Sign conventions that silently flip your SFD, BMD, and stress answers
Adopt one convention — sagging bending moment positive, shear positive when it rotates a small element clockwise, upward forces positive — and write it at the top of every solution before drawing any diagram.
The trouble is that each topic's standard treatment uses a slightly different habit: axial problems take tension as positive, torsion uses a right-hand rule, and beam diagrams vary between textbooks. If you switch conventions halfway through a problem, every calculation can be arithmetically correct while the diagram is mirrored. A mirrored bending moment diagram then points your extreme fibre stress to the wrong face and bends your sketched deflection curve the wrong way, so downstream topics inherit the error.
Build a two-part habit. First, before computing anything, sketch a small beam element with your positive internal actions drawn in, so every sign you write has a picture behind it. Second, close each problem with boundary checks that depend on signs: the bending moment at a free end with no applied couple must be zero, the shear force diagram must close against the total applied load, and a fixed support must show a nonzero bending moment even where slope and deflection are held to zero. Signed mistakes show up in these conditions even when the arithmetic looks clean.
- Sagging moment positive; draw the element before calculating.
- Shear closes against total load; moment returns to zero at free ends.
- One convention across axial, beam, and torsion problems — never per-topic habits.
Simple versus compound stresses: when the uniaxial formula stops being safe
Simple stress problems assume one-directional loading at a point; compound states need principal stresses. The moment a point carries bending plus torsion, or axial plus bending with eccentricity, sigma and tau cannot be added directly.
For a prismatic bar under a centred axial load, the simple treatment holds: sigma equals P over A, strain follows Hooke's law as sigma over E, and lateral strain is Poisson's ratio times the axial strain with opposite sign. That chain breaks when the load line misses the centroid, because eccentricity converts part of the load into bending and the stress becomes P over A plus or minus My over I. It breaks in a different way when a point carries both a normal stress and a shear stress: those belong to different stress types, and combining them requires transformation to principal stresses, sigma one and sigma two, obtained from Mohr's circle or the transformation equations.
Separate two operations that students blur together. Adding stresses of the same type at the same point — axial plus bending normal stresses — is legitimate superposition. Combining stresses of different types — a bending normal stress with a torsional shear stress — is not addition but a plane-stress transformation. A quick exercise fixes the distinction: take sigma x equal to 80 MPa, sigma y equal to 0, tau xy equal to 30 MPa, and draw Mohr's circle. Expected observations: the centre sits at 40 MPa, the radius is 50 MPa, the principal stresses are 90 MPa and minus 10 MPa, and the maximum shear stress is 50 MPa. If your circle disagrees with those numbers, the sign or the centre is wrong before any formula is at fault.
Reading SFD and BMD relationships as built-in answer checks
The relationships dV/dx equals minus w and dM/dx equals V make your diagrams self-checking: slopes, jumps, and zero points must match, and any contradiction points at a calculation error, not a judgment call.
Worked scenario. A simply supported beam of span 6 m carries a UDL of 10 kN/m over the whole span plus a 20 kN point load at midspan. Each reaction is 40 kN. On the left half, V falls linearly from 40 kN to 10 kN at midspan, jumps down 20 kN to minus 10 kN, then falls linearly again to minus 40 kN at the right support. A plausible mistake is to hunt for the point of zero shear by extending the first straight line past midspan, solving 40 minus 10x equals 0 to get x equals 4 m — a station that lies on the wrong piece of the diagram. The better decision is to notice that the jump at midspan crosses zero, so the maximum bending moment sits exactly under the point load, at 75 kN·m. Checking the alternative station with the correct piecewise expressions gives only 60 kN·m. The mistake matters because the section modulus would be sized from the wrong moment.
Use the differential relationships as a five-point rubric on every diagram you draw. Under a UDL the shear diagram is straight and the moment diagram is parabolic; at a point load the shear diagram jumps and the moment diagram has a kink; at an applied couple the moment diagram jumps; the moment is maximum or minimum where the shear diagram crosses zero; and between loads the slope of the moment diagram equals the shear value. If any one of these fails, stop and re-derive the reactions rather than adjusting the curve by eye.
- UDL gives straight SFD and parabolic BMD.
- Point loads jump the SFD; applied couples jump the BMD.
- Extreme moment lives where V crosses zero — on the correct segment.
Bending and shear stress peak at opposite places in the cross-section
The flexure formula puts maximum normal stress at the extreme fibres and zero at the neutral axis; shear stress does the opposite, peaking at the neutral axis and vanishing at the surfaces. Check both, usually at different stations.
The bending equation M over I equals sigma over y equals E over R rests on stated assumptions: plane sections remain plane, the material is homogeneous and linearly elastic, the beam is initially straight, and the section is symmetric about the plane of bending — with pure bending as the clean case. The shear distribution tau equals VQ over Ib follows a different geometry: a rectangle shows a parabolic profile peaking at the neutral axis, while an I-section concentrates shear in the web, which is why the web is checked for shear even though the flanges carry most of the bending.
The practical consequence is that one beam usually needs two checks at two different stations. Maximum bending moment and maximum shear force rarely occur at the same cross-section — a uniformly loaded simply supported beam has its largest moment at midspan and its largest shear at the supports. A deep timber beam near its supports may be governed by the shear check even though the bending check passes at midspan, while a steel I-beam at midspan is usually governed by bending. When comparing candidate sections, compute both the required section modulus from the maximum moment and the web shear from the maximum shear force, and let the larger requirement choose the section rather than defaulting to the bending check.
Torsion of shafts: circular-shaft assumptions and combined loading
The torsion equation T over J equals tau over r equals G theta over L holds for solid or hollow circular shafts in pure torsion with linear elasticity. Combined bending and torsion needs principal stresses or an equivalent moment, not direct addition.
Circular symmetry is what lets plane sections remain plane and radii stay straight under twist, which is why the torsion equation exists at all; a non-circular cross-section warps, and applying T over J to it is invalid regardless of how convenient it looks. Within the valid case, strength uses tau equals 16T over pi d cubed for a solid shaft, and stiffness uses the angle of twist theta equals TL over GJ — two different checks with two different allowables.
Worked scenario. A solid circular shaft of diameter 50 mm carries a bending moment of 0.8 kN·m together with a torque of 1 kN·m. The surface stresses are sigma equal to 32M over pi d cubed, about 65.2 N/mm², and tau equal to 16T over pi d cubed, about 40.7 N/mm². The plausible mistake is adding them, 105.9, or ignoring the bending stress entirely — the first overestimates by roughly a factor of two, the second understates the load on the material. The better decision is a plane-stress transformation: the maximum shear stress is the square root of (sigma over two) squared plus tau squared, about 52.2 N/mm², which you can cross-check through the equivalent torque T e equal to the square root of M squared plus T squared, giving the same 52.2. It matters because shaft allowables are stated in shear, so only the transformed value compares against the material limit.
Choosing a deflection method: Macaulay, moment-area, or superposition
Match the method to the loading: Macaulay's method handles beams with point loads and couple discontinuities in one equation, moment-area suits simple moment diagrams, and superposition assembles standard tabulated cases quickly.
Macaulay's method writes a single expression for EI d²y/dx² using bracketed terms that switch on where loads begin, so point loads and applied couples do not force you to restart the integration at each discontinuity. Apply boundary conditions to sort constants: zero deflection at each simple support, zero deflection and zero slope at a fixed end. Moment-area uses two theorems on the M over EI diagram — its area gives the change in slope between two points, and its moment about a point gives their relative deflection — which is efficient when the M over EI diagram is a simple rectangle or triangle.
Superposition builds the answer from standard cases, such as a simply supported beam under a central point load or a UDL, added load by load. It is fast but conditional: deflections must stay small and the material linear, because each load is assumed to act on the undeformed geometry. A workable rule for the decision itself: one or two point loads plus a UDL on a single span leans toward Macaulay; a cantilever or a span with a triangular or rectangular moment diagram leans toward moment-area; a load set made of recognisable standard cases leans toward superposition. If two methods are both plausible, solve by one and estimate the other to the right order of magnitude as a check.
| Method | Best suited to | Key quantities | Watch out for |
|---|---|---|---|
| Macaulay (double integration) | Single spans with mixed point loads, UDLs, and couples | EI d²y/dx², boundary conditions for constants | Bracket terms must switch on at the right stations |
| Moment-area | Spans whose M/EI diagram is a simple shape | Area and first moment of the M/EI diagram | Signs of area contributions; supports that allow rigid-body movement |
| Superposition | Load sets matching tabulated standard cases | Tabulated end deflections and slopes | Only valid for small deflections and linear behaviour |
| Direct integration | One clean load case with simple boundary conditions | EI d²y/dx² integrated twice | Restarting at every discontinuity costs time |
A six-stage preparation sequence with readiness checks
Prepare in chain order — simple stresses, compound stresses, diagrams, beam stresses, torsion, deflection — because each stage consumes the previous one's outputs, then close with mixed problems that cross two idealizations.
Stage one, simple stresses and strains: axial members, Hooke's law, Poisson's ratio, and a first pass at units. Stage two, compound stresses: Mohr's circle and principal stresses, verified against the exercise values above. Stage three, shear force and bending moment diagrams for cantilevers, simple spans, and overhangs, scored against the five-point rubric. Stage four, bending and shear stresses, deliberately attached to your stage-three diagrams so the maximum-moment and maximum-shear stations come from real diagrams. Stage five, torsion, including the combined bending-plus-torsion scenario. Stage six, deflection by all three methods, then mixed problems that chain two stages in one question.
Use these readiness checks as learning milestones — they indicate study progress, not a predicted exam result. You are ready to move on when: you can state the assumptions behind the flexure and torsion equations without looking; an overhanging-beam SFD and BMD passes all five rubric points on your first attempt; you can compute a combined bending-and-torsion shaft by both the maximum shear stress route and the equivalent torque route and they agree; you can justify your deflection method choice in one sentence; and every numerical answer carries a consistent N–mm–MPa unit audit. Where any check fails, return to that stage's exercises rather than rereading theory.
- Stage 1–2: axial, Hooke's law, Poisson, then Mohr's circle with the 80/0/30 exercise.
- Stage 3–4: diagrams scored on the five-point rubric, then flexure and VQ/Ib checks.
- Stage 5: pure torsion plus the combined-loading scenario by two routes.
- Stage 6: all three deflection methods, then mixed two-idealization problems.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.