Tension vs. Compression in Trusses — Interpret tension vs compression in truss members - Statics

SECTION 1

Historical Context & Motivation

The ability to determine whether a structural member carries tension or compression is one of the oldest and most consequential problems in civil and mechanical engineering. Ancient civilizations built remarkable timber and stone frameworks—Roman bridges, Chinese pagodas, and medieval cathedral roofs—by relying on intuition and empirical rules about how forces flow through connected bars. However, a rigorous analytical framework did not emerge until the Age of Enlightenment, when mathematicians and engineers began formalizing the equilibrium conditions that govern statically determinate trusses. Understanding this history illuminates why distinguishing tension from compression remains a foundational skill in structural analysis.

1638 • Galileo's Two New Sciences

Galileo Galilei published his analysis of beams under load, establishing one of the first quantitative treatments of structural member failure. Although he focused on bending rather than pure axial forces, his work laid the groundwork for distinguishing how members resist loads.

1826 • Navier's Elastic Theory

Claude-Louis Navier published his treatise on the mechanics of structures, formalizing the concept of internal forces in structural members and introducing the idea that bars in frameworks carry purely axial loads when pin-connected at joints.

1847 • Squire Whipple's Truss Analysis

American engineer Squire Whipple published the first systematic analytical method for determining forces in truss members, enabling engineers to calculate tension and compression in each bar of iron-and-timber bridge trusses.

1864 • Maxwell's Method of Reciprocal Diagrams

James Clerk Maxwell introduced graphical methods for truss analysis, including the reciprocal diagram technique. These graphical tools enabled rapid identification of tension and compression members without extensive algebraic computation.

1890s • Cremona Diagrams & Steel Trusses

Luigi Cremona refined Maxwell's graphical approach into the Cremona diagram, which became standard practice for bridge and roof truss design during the steel construction boom that defined the modern built environment.

Today, even though computational tools such as finite element software handle complex structural systems, the ability to manually interpret whether a truss member is in tension or compression remains essential. It provides the physical insight needed to validate computational results, perform preliminary design, and understand failure modes. The central question this lesson addresses is: given a loaded truss, how do we systematically determine which members pull (tension) and which push (compression)?

SECTION 2

Core Principles & Definitions

Before analyzing any truss, you must internalize several foundational principles that govern how forces distribute through a system of interconnected two-force members. A truss is an assembly of slender bars (members) joined at their endpoints by frictionless pins (joints or nodes), loaded only at the joints, and arranged to form a rigid framework. Under these idealizing assumptions, each member carries only an axial force—either tension (pulling the member apart) or compression (pushing the member together). No bending moments or shear forces develop in an ideal truss member because all forces pass through the pin connections at each end.

Two-Force Member Principle

Each truss member is a two-force member: forces act only at the two pin joints. For equilibrium, these forces must be equal in magnitude, opposite in direction, and collinear along the member's axis. This guarantees purely axial loading.

Tension Convention

A member is in tension (T) when the internal axial force tends to elongate it. In a free-body diagram of a joint, a tension member pulls the joint toward itself—arrows point away from the joint along the member.

Compression Convention

A member is in compression (C) when the internal axial force tends to shorten it. In a free-body diagram of a joint, a compression member pushes the joint away from itself—arrows point toward the joint along the member.

Method of Joints

Isolate each joint as a concurrent-force system. Apply ΣF_x = 0 and ΣF_y = 0. Assume all unknown member forces are tensile; a negative result indicates compression.

Method of Sections

Cut through the truss to isolate a section exposing no more than three unknown member forces. Apply ΣF_x = 0, ΣF_y = 0, and ΣM = 0 to solve for member forces directly.

✦

KEY TAKEAWAY

Think of each truss member as a rope or a stick. A rope can only pull—that's tension. A rigid stick can push back when compressed—that's compression. In the method of joints, you initially assume every member is a rope pulling on the joint. If the math returns a negative sign, it means the member is actually a stick pushing the joint—it is in compression. This sign convention is the single most important bookkeeping rule in truss analysis.

SECTION 3

Visual Explanation — Tension & Compression in a Simple Truss

A symmetric Warren truss loaded at joint D with force P. The bottom chord members (AC, CE, EG) and the diagonal members sloping upward toward the center (BC, DE, EF) carry tension, while the top chord members (BD, DF) and the diagonals sloping downward from the supports (AB, CD, EF) carry compression. The load at D is shown in amber.

In the diagram above, observe the general pattern that emerges for a simply supported truss under gravity loading: the bottom chord members are in tension because they resist the tendency of the truss to sag and spread apart at the supports, while the top chord members are in compression because they resist the closing of the truss profile under load. Diagonal web members alternate between tension and compression depending on the direction of the shear they must resist. This pattern is analogous to a simply supported beam: the bottom flange is in tension, the top flange is in compression, and the web resists shear—a truss simply discretizes these functions into individual bars.

ℹ

Sign Convention Reminder

When drawing the free-body diagram of a joint, assume all unknown member forces act away from the joint (i.e., assume tension). If the computed force is positive, the member is in tension. If negative, the member is in compression. This convention is used universally in the method of joints and is the approach adopted throughout this lesson.

SECTION 4

Mathematical Framework — Equilibrium at Joints & Sections

The mathematical foundation for determining tension and compression in truss members rests on the equations of static equilibrium. For a two-dimensional truss in the x–y plane, three independent equilibrium equations govern the entire structure, and two equations govern each isolated joint. Supplementing these with strategic moment equations (method of sections) provides a complete toolkit for solving any statically determinate truss.

GLOBAL EQUILIBRIUM

ΣFₓ = 0 ; ΣFᵧ = 0 ; ΣM₀ = 0

Applied to the entire truss free-body diagram to find support reactions. Fₓ and Fᵧ are force components; M₀ is the moment about any point O. Always solve for reactions first.

METHOD OF JOINTS — JOINT EQUILIBRIUM

ΣFₓ = 0 ; ΣFᵧ = 0 (at each joint)

Each joint is a concurrent force system with no moment equation. Resolve all member forces and external loads into x and y components. Start at a joint with at most two unknowns and progress through the truss sequentially.

METHOD OF SECTIONS — MOMENT EQUATION

ΣM_P = 0

Cut the truss through three members. Sum moments about a point P where two of the three unknown forces intersect. This isolates a single unknown member force directly, significantly reducing computation.

DETERMINACY CHECK

m + r = 2j

A planar truss is statically determinate when the number of members m plus the number of reaction components r equals twice the number of joints j. If m + r < 2j, the truss is unstable; if m + r > 2j, it is statically indeterminate.

The interplay between these equations is straightforward. Begin by checking determinacy. Then compute support reactions using global equilibrium. Finally, apply either the method of joints or the method of sections—or a combination—to find individual member forces. The sign of each result, under the tension-positive convention, directly tells you whether the member is in tension (+) or compression (−).

SECTION 5

Detailed Breakdown — Free-Body Diagram of a Joint

The method of joints is best understood by examining the free-body diagram (FBD) of an individual joint in detail. The following diagram isolates joint B of a simple Pratt truss and shows how forces are resolved into components, how the tension-positive assumption manifests as arrows pointing away from the joint, and how equilibrium yields the member force magnitudes and signs.

Left: a simple three-bar truss with joints A, B, C, and D. Joint B (highlighted in amber) is isolated. Right: the free-body diagram of joint B showing all member forces assumed in tension (arrows pointing away from B). The applied load P acts downward. Angle θ is measured from the vertical to members AB and BC.

From the FBD on the right, we write the equilibrium equations for joint B. By symmetry, the truss geometry makes members AB and BC equal in length and inclined at angle θ from the vertical. Setting up the coordinate system with x horizontal (positive right) and y vertical (positive up):

ΣFₓ AT JOINT B

−F_AB sin θ + F_BC sin θ = 0 → F_AB = F_BC

Horizontal components of the symmetric diagonal members cancel, confirming that for a symmetric truss under a symmetric load, the forces in symmetric members are equal.

ΣFᵧ AT JOINT B

−F_AB cos θ − F_BC cos θ − F_BD − P = 0

Substituting F_AB = F_BC and solving: F_BD = −P − 2F_AB cos θ. If both terms are negative, F_BD < 0, confirming that the vertical member BD is in compression when all members at B point downward.

This procedural approach—draw the FBD, assume tension, write ΣFₓ = 0 and ΣFᵧ = 0, interpret signs—forms the backbone of the method of joints. The power of the approach lies in its systematic nature: you never need to guess whether a member is in tension or compression before solving; the mathematics reveals the answer through the sign of the result.

SECTION 6

Worked Example — Method of Joints on a Pratt Truss

Consider a Pratt truss with three panels, pinned at joint A and on a roller at joint F. The bottom chord joints are A, C, E, F (left to right, spaced 4 m apart), and the top chord joints are B and D at a height of 3 m above the bottom chord, with B directly above C and D directly above E. A vertical load of 12 kN acts downward at joint D. Determine the force in members BD, CD, and CE, and identify each as tension or compression.

Pratt Truss — Determine Forces in BD, CD, and CE

1
Step 1 — Find Support Reactions

The truss spans 12 m (three 4-m panels). Joint A has a pin support providing reactions Aₓ and Aᵧ. Joint F has a roller providing only Fᵧ. The 12 kN load acts downward at D, which is located 8 m from A horizontally. Taking moments about A: ΣM_A = 0 → Fᵧ × 12 − 12 × 8 = 0, so Fᵧ = 8 kN ↑. From ΣFᵧ = 0: Aᵧ + 8 − 12 = 0, so Aᵧ = 4 kN ↑. From ΣFₓ = 0: Aₓ = 0.

Aₓ = 0, Aᵧ = 4 kN ↑, Fᵧ = 8 kN ↑
2
Step 2 — Method of Sections: Cut Through BD, CD, CE

Pass a vertical section through members BD, CD, and CE, cutting the truss into left and right portions. Analyze the left portion, which includes joints A, B, and C. The only external force acting on this portion is the reaction Aᵧ = 4 kN ↑ (Aₓ = 0). Assume all three cut member forces are tensile—that is, pointing away from the left portion toward the right portion.
3
Step 3 — Solve F_BD (Moment about C)

Take moments about joint C(4, 0). Member CD begins at C and member CE also begins at C, so both F_CD and F_CE pass directly through C and contribute zero moment about that point. Only F_BD—which acts horizontally along the top chord through B(4, 3), a perpendicular distance of 3 m above C—produces a nonzero moment about C. The reaction Aᵧ acts 4 m to the left of C. ΣM_C = 0: −Aᵧ(4) − F_BD(3) = 0 → −(4)(4) − 3F_BD = 0 → F_BD = −16/3 = −5.33 kN.

F_BD = −5.33 kN → since the tension-positive assumption produced a negative value, member BD is in Compression (5.33 kN C), consistent with the general pattern that top-chord members carry compression under gravity loading.
4
Step 4 — Solve F_CE (Moment about D)

Take moments about joint D(8, 3). Member BD ends at D and member CD also ends at D, so both F_BD and F_CD pass directly through D and contribute zero moment about that point. Only F_CE—which acts horizontally along the bottom chord, a perpendicular distance of 3 m below D—produces a nonzero moment about D. The reaction Aᵧ acts 8 m to the left of D, measured horizontally. ΣM_D = 0: −Aᵧ(8) + F_CE(3) = 0 → −(4)(8) + 3F_CE = 0 → F_CE = 32/3 = 10.67 kN.

F_CE = +10.67 kN → positive, so member CE is in Tension (10.67 kN T), consistent with the general pattern that bottom-chord members carry tension under gravity loading.
5
Step 5 — Solve F_CD (Force Equilibrium)

With F_BD and F_CE both horizontal, only F_CD contributes a vertical force component to the left portion besides the reaction. Member CD rises from C(4, 0) to D(8, 3)—a 3-4-5 right triangle—so its vertical component equals (3/5)F_CD. ΣFᵧ = 0: Aᵧ + (3/5)F_CD = 0 → 4 + 0.6F_CD = 0 → F_CD = −6.67 kN.

F_CD = −6.67 kN → negative, so member CD is in Compression (6.67 kN C).

⚠

Interpreting Signs

The sign interpretation always follows the same rule established earlier in this lesson: assume all unknowns as tension (pulling away from the joint or section), then a positive result confirms tension and a negative result indicates compression. Always state the final answer explicitly as (T) or (C) after the magnitude—never reinterpret a positive result as compression, even when the geometry of a moment equation feels unfamiliar.

SECTION 7

Method of Joints vs. Method of Sections — Strengths & Limitations

Both the method of joints and the method of sections achieve the same goal—determining internal member forces and classifying them as tension or compression—but they differ significantly in efficiency, applicability, and computational effort depending on the problem context. The table below summarizes their key attributes to help you choose the most efficient approach for a given situation.

Attribute	Method of Joints	Method of Sections
Equations per step	2 (ΣFₓ = 0, ΣFᵧ = 0)	3 (ΣFₓ = 0, ΣFᵧ = 0, ΣM = 0)
Max unknowns per step	2	3
Best used when	All member forces are needed; truss is simple	Only specific interior member forces are needed
Efficiency	Requires sequential joint-by-joint analysis	Can jump directly to a target member
Common pitfall	Choosing a joint with > 2 unknowns	Cutting through > 3 unknowns
Sign convention clarity	Very intuitive—arrows on joint FBD	Requires careful attention to assumed directions on the cut

Comparison of the two primary analytical methods for truss member forces

⚡

STRATEGIC INSIGHT

Think of the method of joints as reading a book page by page—methodical and complete. The method of sections is like using the table of contents to jump to exactly the chapter you need. In practice, experienced engineers use both methods in tandem: the method of joints to sweep through an entire truss, and the method of sections to quickly verify the force in a critical member identified during preliminary design.

SECTION 8

Connection to Advanced Structural Analysis

The idealized truss analysis covered in this lesson—pin-connected joints, loads only at joints, weightless members—provides a powerful starting point, but real structures introduce complications. As you progress through structural analysis, you will encounter statically indeterminate trusses (where m + r > 2j), space trusses in three dimensions, and trusses with rigid (welded) connections that develop bending moments in addition to axial forces. Understanding the tension-compression classification remains central even in these advanced contexts.

Feature	Ideal (Statics) Truss	Advanced Analysis
Connections	Frictionless pins (zero moment)	Welded/bolted joints (transfer moment)
Internal forces	Axial only (T or C)	Axial + shear + bending moment
Determinacy	m + r = 2j (solvable by statics alone)	m + r > 2j (requires compatibility / stiffness methods)
Member failure mode	Yielding (T) or Euler buckling (C)	Combined stress interaction (axial + bending)
Analysis tools	Method of joints / sections (hand calc)	Matrix stiffness method, FEA software

Progression from ideal truss analysis to advanced structural methods

A particularly important connection arises in structural design: compression members are far more susceptible to buckling than tension members. Euler's critical load formula, P_cr = π²EI / (KL)², shows that long, slender compression members can fail at stresses well below the material's yield strength. This is why correctly identifying compression members is not merely an academic exercise—it directly informs member sizing, bracing requirements, and overall structural safety. Tension members, by contrast, are governed primarily by the material's tensile yield or ultimate strength and their net cross-sectional area.

SECTION 9

Practice Problems

Question 1: CONCEPTUAL

In a simply supported Pratt truss under downward gravity loads applied at the top chord joints, explain why the bottom chord members are generally in tension and the top chord members are generally in compression. Relate your explanation to the bending behavior of an equivalent simply supported beam.

Show answer

A simply supported truss under gravity loads behaves analogously to a simply supported beam: the bottom fiber (bottom chord) is in tension and the top fiber (top chord) is in compression. The external loads cause the truss to sag, and the supports resist this by developing upward reactions. The bottom chord resists the tendency of the supports to spread apart—it acts like a tie pulling the supports together, hence tension. The top chord resists the closing of the truss profile under load, compressing together under the applied forces. This chord force pattern directly mirrors the bending stress distribution in a beam where σ = −My/I, with the sign indicating tension below the neutral axis and compression above it when M is positive (sagging).

Question 2: BASIC CALCULATION

A simple triangular truss (three members) has a horizontal bottom chord AC of length 6 m, a vertical member AB of height 4 m at the left end, and a hypotenuse member BC. The truss is pinned at A and on a roller at C. A horizontal load of 10 kN acts to the right at joint B. Determine the force in member AC and state whether it is in tension or compression.

Show answer

First, find support reactions. Since the roller at C provides only a vertical reaction and the 10 kN load is horizontal, ΣFₓ = 0 gives Aₓ = −10 kN (Aₓ acts to the left). Taking Σ

M_A

= 0: Cᵧ × 6 − 10 × 4 = 0, so Cᵧ = 6.67 kN ↑. Then ΣFᵧ = 0: Aᵧ = −6.67 kN, so Aᵧ = 6.67 kN ↓. Isolate joint A. Member AB is vertical and member AC is horizontal. ΣFₓ at A: Aₓ + F_AC = 0 → −10 + F_AC = 0 → F_AC = 10 kN (positive). Since we assumed tension (away from A, pointing to the right), the positive result confirms F_AC = 10 kN (T). Member AC is in tension, which makes physical sense: the horizontal load at B tries to push B rightward, and member AC resists this by pulling joint A toward C.

Question 3: INTERMEDIATE

A symmetric Howe truss has 5 joints on the bottom chord (A, C, E, G, I) spaced 3 m apart, and 4 joints on the top chord (B, D, F, H) at 4 m height. The truss is pinned at A and on a roller at I. A 20 kN downward load is applied at joint E (bottom chord, center). Using the method of sections, determine the force in member DF (top chord between the third and fourth top joints) and classify it as tension or compression.

Show answer

By symmetry of loading and geometry, each support reaction is 10 kN upward (Aᵧ = Iᵧ = 10 kN). Cut a vertical section through members DF, DE (or EF), and EG, exposing three unknowns. Analyze the left portion containing joints A, B, C, D, and E. Take moments about joint E to eliminate F_DE and F_EG (both pass through E). Joint E is at (6, 0) and member DF runs horizontally at height 4 m. The perpendicular distance from E to F_DF is 4 m. Σ

M_E

= 0: Aᵧ × 6 − F_DF × 4 = 0 → 10 × 6 − F_DF × 4 = 0 → F_DF = 15 kN. The sign is positive assuming F_DF acts to the left (away from the section on the left portion, i.e., compression on the top chord). Therefore F_DF = 15 kN (C). This confirms the expected pattern: the top chord of a simply supported truss under gravity loading is in compression.

Question 4: APPLIED

A roof truss for a warehouse is modeled as a Warren truss with a span of 18 m (six panels of 3 m each) and a height of 3 m. The truss is pinned at the left support and on a roller at the right. A uniform snow load is represented by equal downward loads of 5 kN at each of the five interior top-chord joints. Determine the force in the bottom chord member at the center of the truss and the force in one of the diagonal members adjacent to the center. Identify each as tension or compression.

Show answer

Total load = 5 × 5 = 25 kN. By symmetry, each reaction = 12.5 kN. The bottom chord member at the center spans the middle panel. Use the method of sections: cut through the center diagonal, the top chord member, and the bottom chord member at the center panel. Take moments about the top chord joint at the center, which is at height 3 m directly above the center of the bottom chord. The perpendicular distance from this joint to the bottom chord is 3 m. The moment of external forces on the left portion about this point equals Aᵧ × 9 − 5 × 6 − 5 × 3 − 5 × 0 = 12.5 × 9 − 30 − 15 = 112.5 − 45 = 67.5 kN·m. Setting this equal to F_bottom × 3: F_bottom = 22.5 kN (T). For the diagonal, use ΣFᵧ = 0 on the left section. The net vertical force on the left portion (excluding the diagonal) is 12.5 − 5 − 5 − 5 = −2.5 kN. The diagonal makes a 45° angle with the horizontal (3 m rise over 3 m run). Therefore F_diag × sin 45° = 2.5, giving F_diag = 3.54 kN. The sign analysis shows this diagonal is in tension.

Question 5: CRITICAL THINKING

A Pratt truss and a Howe truss have identical spans, heights, and loading conditions. In a Pratt truss, the diagonals slope downward toward the center; in a Howe truss, the diagonals slope upward toward the center. Explain, without performing numerical calculations, why the diagonals in a Pratt truss tend to be in tension while the diagonals in a Howe truss tend to be in compression. Discuss the structural design implications of this difference, particularly regarding member sizing and material efficiency.

Show answer

In both truss types, the vertical shear force is largest near the supports and diminishes toward the center. The diagonals are the primary shear-carrying members. In a Pratt truss, the diagonals slope downward from the top chord to the bottom chord as you move toward the center. The shear force on a section near the support is upward on the left portion, and the diagonal's geometry means it must pull (tension) to provide an upward vertical component on that section. In a Howe truss, the diagonals slope in the opposite direction, so the same shear force requires the diagonal to push (compression) to provide the upward component. The design implication is significant: tension members can be made from slender rods or cables because they are not susceptible to buckling, whereas compression members must be stockier to resist Euler buckling. Consequently, Pratt trusses are generally more material-efficient for long-span structures because their long diagonals are in tension (thin and light) and only the short verticals are in compression. Howe trusses, with compression diagonals, require heavier diagonal members and are less common in modern steel construction, though they were historically favored in timber construction where compression in wood (parallel to grain) is efficient.

SUMMARY

Lesson Summary

Summary

Every member in an ideal truss carries a purely axial internal force that is classified as either tension (elongation, positive by convention) or compression (shortening, negative by convention). The method of joints isolates each joint as a concurrent force system and applies ΣFₓ = 0 and ΣFᵧ = 0, while the method of sections cuts through up to three members and uses ΣFₓ = 0, ΣFᵧ = 0, and ΣM = 0 to solve for specific member forces directly. Both methods rely on the tension-positive sign convention: assume all unknowns as tension, and let the algebraic sign of the result reveal the true sense of the force.

In a typical simply supported truss under gravity loading, bottom chord members are in tension and top chord members are in compression, mirroring the stress distribution in a simply supported beam. Diagonal web members carry the shear and alternate between tension and compression depending on truss type and loading. The distinction between tension and compression has critical design implications: compression members are governed by Euler buckling and require larger cross-sections, while tension members are limited by yielding of the net section and can be designed with slender, efficient profiles. Mastering this classification is the essential first step toward safe and economical structural design.