Materials Science · Units, loads, stress and strain

Loading direction, geometry and conditions

Follow forces through a body and its supports, explain why a ruler bends differently after rotation, and distinguish compression instability from material yielding. State direction, duration and environment before transferring a test result.

Useful background: Materials, questions and levels of evidence. Each worked case explains the quantities it uses.

Back to the chapter

After reading, you can

Identify axial tension/compression, shear, bending and torsion from how forces or torques act.
Explain the tension and compression regions in a simple bending case.
Separate cross-sectional area from its arrangement in a ruler-orientation comparison.
Explain qualitatively why compression can cause buckling and why restraints matter.
Select a relevant loading model and identify conditions missing from a material-performance claim.

Turning a ruler changes the test

Hold one end of a rectangular ruler and gently deflect the other. It bends readily when its broad face is horizontal, but much less readily when it stands on its narrow edge. The amount of material is the same. Its arrangement relative to the bending direction has changed.

To understand the comparison, name the action as well as the material. A load is the applied force or turning action; a restraint is something that limits motion, such as a grip, pin or clamp. The load path describes how those actions are transmitted through the body to its supports. A tensile-test number is useful only when it answers the relevant question along that path.

What motion does the loading encourage?

Axial tension pulls along a member’s centreline and tends to lengthen it. Axial compression pushes inward and tends to shorten it. In a straight uniform member, away from the load-introduction regions and before any instability, average axial engineering stress is F/A, using the original cross-sectional area A. Turning a transverse force into an axial one changes the model, even if its magnitude F is unchanged.

Shear stress acts tangentially to a surface, rather than normally through it. It can make neighboring layers shift sideways relative to one another. In the ideal local shear sketch, arrows on all four faces balance forces and turning effects; it is not a drawing of an entire joint. A real pin or adhesive joint needs its actual contact and load distribution.

Bending makes the member’s longitudinal line curve. In the simply supported, downward-loaded sketch, the upper longitudinal region shortens and is in compression, while the lower region lengthens and is in tension. Between them is a neutral surface with no longitudinal extension from bending in this simple homogeneous model. Reverse the bending direction and these roles reverse. Transverse shear is also present in a beam loaded this way.

Torsion twists one cross-section relative to another about the member’s long axis. A hand turning a screwdriver applies a torque, resisted through the shaft and the engaged tip. A circular shaft transmits that torque through shear stresses; its twist is not the axial extension FL/(EA). In the end-view sketch, opposite torque senses are shown from the same viewing direction so that the arrow convention is unambiguous.

A component can combine these actions. An off-centre pull, for example, can introduce bending as well as axial tension. Naming the overall motion is a starting model; it does not mean every point has one uniform stress.

Schematic loading and local-stress directions, not measured shapes or scaled forces. A: equal outward axial pulls on a straight bar. B: local shear element with top-right, bottom-left, left-down and right-up tangential arrows, balancing forces and moments. C: downward midpoint load on a simply supported beam, with a pin and roller, upper longitudinal compression and lower tension in the sagging region; the drawn deflection is exaggerated and the beam also transmits transverse shear. D: opposite torques in two shaft end views, both viewed in the same direction; they describe torsion rather than axial pulling.

Why the ruler’s depth matters

In the simple small-bending model, material farther from the neutral surface undergoes more longitudinal stretching or shortening for the same curvature. Those longitudinal stresses also act at larger distances from the centre of the section. Both effects make the arrangement of the cross-section important, rather than just its total area.

For a uniform rectangular section, a geometric measure of this arrangement is the second moment of area I = bh³/12. Here h is the depth in the up/down bending direction and b is the perpendicular width. I has units of length to the fourth power, such as mm⁴; it is not a mass moment of inertia. For the same homogeneous material, the bending stiffness in this model is proportional to EI, where E is Young’s modulus.

For example, turn a 20 mm by 2 mm section through a right angle while keeping the bending direction fixed. Its area stays 40 mm², but h changes from 2 to 20 mm and b from 20 to 2 mm. I changes from 13.333… to 1333.333… mm⁴, a factor of 100. Under the same transverse force, span and supports, the elementary linear beam model therefore predicts one hundredth of the bending deflection in the edge-up orientation.

This is a controlled geometry comparison: assume a long slender ruler, uniform isotropic material, small elastic bending, negligible shear deflection, and no unwanted sideways twisting or instability. Isotropic means its material response does not depend on direction. A layered or direction-dependent material need not preserve the same effective E when tested differently. The factor of 100 is a prediction under the stated model, not a universal ruler measurement.

The same rectangular cross-section, viewed along the ruler. Bending is up/down in both cases.
Quantity	Broad face horizontal	Narrow edge down
Width b	20 mm	2 mm
Depth h	2 mm	20 mm
Area bh	40 mm²	40 mm²
I = bh³/12	13.333… mm⁴	1333.333… mm⁴
Relative EI, same E	1	100
Relative bending deflection, matched linear-beam case	1	0.01

Two cross-sections of the same 20 mm by 2 mm rectangular ruler, viewed along its length with up/down bending fixed. Both outlines use 5 pixels per millimetre and have equal area 40 mm². Flat orientation has width b 20 mm and depth h 2 mm, I 13.333… mm⁴. Edge-up orientation has width b 2 mm and depth h 20 mm, I 1333.333… mm⁴. Dotted horizontal centre lines mark the neutral surface in the homogeneous simple-bending model. The second I is 100 times the first while the axial area is unchanged.

Describe supports and force paths

A support supplies forces or moments because it restricts motion. An ideal pin permits rotation but stops translation at its location; an ideal roller stops motion normal to its supporting surface while allowing sliding along it. A clamp stops both displacement and rotation in the model. Real fixtures have finite stiffness and may slip, so their behavior must be checked when interpreting a test.

Force position matters through its turning effect, called moment or torque. About a chosen point, its magnitude is M = Fd, where d is the perpendicular distance from that point to the force’s line of action. M has units N·m, while F/A has stress units. A clamp carrying a transverse end load must transmit a force and a bending moment; an axial stress quotient alone does not describe it.

Compression introduces another possibility. If a slender member bows slightly, a compressive force can act with an offset from the deflected section and increase its bending. Bending stiffness opposes the bowing. Above a condition-dependent threshold, the straight state can lose stability: this is buckling. It can occur while the material remains elastic, so it is different from reaching a simple compressive yield or crushing stress.

A shorter unsupported length or a suitable lateral brace can obstruct the long sideways bending shape and raise the ideal elastic buckling threshold. End rotation restraints also change which shapes are possible. These changes do not alter the material’s modulus or yield stress. Initial crookedness, load misalignment, brace stiffness and possible movement in another direction still matter; no numerical buckling load follows from length alone.

Holes and sharp section changes redirect the load path and can increase local stress. A rounded transition can reduce a concentration under a given comparison, but it does not make every local stress equal to the section average. Cracks, repeated loads and the joint itself can introduce additional limits developed in later lessons.

Keep direction and time with the result

Material direction is different from component orientation. Rotating the isotropic rectangular ruler changed its geometry relative to bending. Testing along and across a fiber-reinforced material can also change the material response itself. Record direction relative to fibers, layers or processing, rather than assuming that a single scalar value describes every direction.

Also state how the load is applied: slowly increased, held, cycled or delivered as an impact. Some materials continue to deform during a sustained load; that time-dependent deformation is called creep. A short test therefore need not establish an hours-long deformation limit. Temperature and moisture can change a material’s response or state as well, so comparisons need matching or explicitly varied conditions.

An observation becomes interpretable when these details identify a model. If two rulers differ in deflection, first distinguish a change in section orientation or span from a change in material or conditions. A changed reading is evidence to explain, not automatically evidence that the material’s intrinsic modulus has changed.

Examples of information that changes the question. These are categories to define, not a complete test method.
Information	Why it matters
Force direction and application point	Separates axial action from bending or combined loading
Dimensions and section orientation	Determines area, bending arrangement and unsupported span
Supports, grips and lateral restraints	Determines allowed displacement, rotation and instability shapes
Material direction and state	A value may apply only along one orientation and processing condition
Temperature, moisture and loading history	Short, dry, slow loading may not represent another condition

Worked examples

Tension survival does not rule out compression buckling

A long thin rod carries a tensile force but bows sideways under a smaller compressive force. Explain why this is possible. Predict what shortening its unsupported length or adding an effective midpoint lateral brace could change, without claiming a numerical capacity.

The tensile case pulls along the member; the compression case pushes and can amplify a small sideways offset. They are different physical problems even for the same original cross-section and material.
The rod can lose stability in compression before reaching its compressive material limit. Sideways movement alone does not establish plastic yielding: an elastic buckling mode is possible.
Shortening the unsupported span makes the long bending shape harder to develop. A suitable midpoint brace resists lateral displacement there and changes the allowed shapes. Either can raise the ideal elastic instability threshold under the relevant assumptions.
Neither intervention raises the material’s E or yield stress. The actual threshold still requires dimensions, end conditions, brace effectiveness, direction, imperfections and an appropriate stability model. Check material failure as a separate possibility.

There is no contradiction. Compression can be limited by a geometry-and-restraint instability that the tensile material test did not assess.

Separate axial area from bending arrangement

For the 20 mm by 2 mm isotropic ruler section, calculate the change in I when it is turned edge-up. Does the same rotation multiply its axial stiffness EA/L by 100?

With the broad face horizontal, b = 20 mm and h = 2 mm: I_flat = 20 × 2³/12 = 13.333… mm⁴.
With the narrow edge down, b = 2 mm and h = 20 mm: I_edge = 2 × 20³/12 = 1333.333… mm⁴. Thus I_edge/I_flat = 100. At the same E, the ratio of bending stiffnesses EI is also 100.
For matched transverse load, span and supports in the stated linear beam model, deflection scales inversely with EI, giving δ_edge/δ_flat = 1/100 = 0.01. This is a relative prediction; no absolute deflection has been calculated.
Both areas are 20 × 2 = 40 mm². In an admissible axial tension comparison with the same E and L, EA/L is unchanged. The section’s bending arrangement changed, while its axial area and material modulus did not.

Bending stiffness rises by a factor of 100 in the stated model; axial stiffness remains unchanged. The word “stiffer” must retain the loading mode.

One force magnitude, two different demands

A straight uniform bar has cross-sectional area 100 mm². Compare a centred 50 N axial pull with a 50 N transverse force whose line of action is 200 mm from a clamp. What can be calculated before a bending-response model is supplied?

In the axial case, average engineering stress is 50/100 = 0.50 N/mm² = 0.50 MPa, away from grips or section changes.
In the transverse case, the applied force has moment magnitude M = 50 × 0.200 = 10 N·m about the clamp. The support must resist its turning effect as well as balance the force.
If that perpendicular distance becomes 100 mm at the same force, M becomes 50 × 0.100 = 5 N·m. The force magnitude did not halve, but the moment did.
The 10 N·m is not a stress, and 0.50 MPa does not describe the transverse bending stress distribution. Section shape, bending orientation, material behavior and restraints are needed to predict stresses and deflection there.

The axial average is 0.50 MPa; the transverse force gives a 10 N·m turning demand about the clamp. They are different quantities from different loading models.

Try the reasoning

A report says “the part carries 100 N.” Describe two different loading arrangements consistent with those words and name the missing details needed to interpret deformation.
Show the answer and reasoning
It could be a centred axial pull, or a transverse load on a clamped arm. Required details include force direction and position, shape and dimensions, support/rotation constraints, material response and loading conditions.
The first might admit σ = F/A and δ = FL/(EA); the second involves bending and support moment. The common force magnitude does not select the model or establish a deformation limit.
A notch is made rounder without changing the remote axial force. A learner assumes the stress has become uniform because the sharp corner has gone. Assess that inference.
Show the answer and reasoning
A rounder transition can reduce a concentration, but uniformity does not follow.
Force still has to pass around the remaining shape change, and stress may vary locally. Define the comparison geometry, stress measure and loading before claiming how much a concentration changed. A remote average F/A does not resolve the local field.
Replace the flat 20 mm by 2 mm rectangle by a 4 mm-wide, 10 mm-deep rectangle, keeping the same isotropic material and length. For the same up/down bending model, compare area, I, bending stiffness and relative deflection with the flat section.
Show the answer and reasoning
Both areas are 40 mm². The new I is 4 × 10³/12 = 333.333… mm⁴, 25 times the flat value. EI is 25 times larger and the matched bending deflection is 1/25 = 0.04 of the flat case.
The depth is cubed in I, but the width also changes; using only the depth ratio would give the wrong factor. In the axial model EA/L remains the same because area, E and L are unchanged. The bending conclusion retains the slender, small-deformation and restraint assumptions.
A flexible cable has a high measured tensile strength. A proposal uses it as a straight compression brace between two points. What physical question is missing?
Show the answer and reasoning
Whether the proposed member can remain stable and transmit compression in that configuration.
An ordinary flexible cable goes slack or changes shape when pushed; its tensile strength does not make it a compression strut. A suitable compression member needs a stability and geometry assessment, not just the tensile property.
A 100 N hand force acts perpendicular to a 0.20 m wrench arm. Find its torque about the shaft axis. Does dividing 100 N by the shaft area calculate the torsional shear stress?
Show the answer and reasoning
The torque is 100 × 0.20 = 20 N·m. Dividing the force by shaft area does not establish the torsional shear-stress distribution.
The shaft transmits a turning action, whose relation to local shear depends on its cross-section and an appropriate torsion model. The axial force/area expression answers a different question.
The same polymer strip deflects more in a warm one-hour hold than in a cool ten-second test, using the same section orientation and nominal force. Can the difference be assigned solely to temperature? Propose a more informative comparison.
Show the answer and reasoning
No: temperature and time have both changed. Hold the duration and loading history fixed while comparing temperature, and compare durations at the same temperature, retaining geometry, supports and material state.
Time-dependent deformation and temperature-dependent response can contribute together. Repeat measurements and check fixture motion as well. The existing two readings do not isolate either contribution or establish a universal material trend.

What to carry forward

Load direction, position and restraints determine the force path and admissible model.
Bending combines longitudinal compression and tension; cross-sectional arrangement matters as well as area.
Buckling can limit a compressed component while its material remains elastic.
Keep material direction, time and environment with any claimed property or component response.

References

Teaching resources and research supporting this lesson’s explanations and examples.

The International System of Units (SI)BIPMSI Brochure, 9th edition, updated 2026bipm.orgRead source
University Physics Volume 1 · 12.3 Stress, Strain, and Elastic ModulusSamuel J. Ling, Jeff Sanny and William Moebs · OpenStaxTextbook sectionopenstax.orgRead source
Massachusetts Institute of Technology2 sources
1. Engineering Mechanics for Structures · selected chaptersLouis L. Bucciarelli / MIT OpenCourseWareTextbook chapters, 2002; MIT course 2004Engineering Mechanics for Structures · selected chapters Chapter 2: Force and Moment Chapter 4: Stress Chapter 8: Shafts in Torsion Chapter 9: Beams in Bending Chapter 10: Deflections and Bucklingocw.mit.edu
2. Rate- and Time-Dependent Polymer Deformation · Laboratory Module 2MIT Department of Mechanical Engineering / MIT OpenCourseWareUniversity laboratory notes, Spring 2004 · PDFocw.mit.eduRead source