Plane trusses

Look at any steel bridge or the sloping roof of a warehouse and you will see the same skeleton repeating: triangles of slender bars pinned together at their ends. That skeleton is a plane truss, and its genius is a single restriction — every member is loaded only along its own axis, never bent. Pin two straight bars and a load at their shared ends and each bar can only push or pull along its length; there is nothing else it can do. Engineers exploit this ruthlessly: a member loaded purely in tension or compression uses every fibre of its cross-section, so trusses span long distances with remarkably little material. This article picks up where equilibrium of rigid bodies (free-body diagrams, ∑F=0\sum F=0, ∑M=0\sum M=0) left off, and turns those tools on frameworks of pinned bars. You will see how to test whether a truss is properly built before solving it, how to strip it down joint by joint for every member force, and how to shortcut straight to a handful of forces with one well-chosen cut.

The core insight

A plane truss carries every load as pure axial tension or compression in its pin-jointed members, and once you confirm m+r=2jm+r=2j you solve for those forces joint by joint (method of joints) or, for a few targeted members, with one clean cut (method of sections).

The core idea

Idealizing assumptions. A plane truss is modelled as an assembly of straight, weightless, two-force members connected by frictionless pins, with all external loads and support reactions applied only at the joints. Because a two-force member has forces at only its two end pins and no other load along its length, equilibrium forces those two end forces to be equal, opposite, and directed along the line joining the pins — so the member can only push (compression) or pull (tension); it carries no shear and no bending moment. This one fact is what makes truss analysis solvable with nothing but ∑Fx=0\sum F_x=0 and ∑Fy=0\sum F_y=0.

Determinacy and stability. Before solving anything, check whether the truss is even solvable with statics alone. For a plane truss with mm members, rr reaction components, and jj joints:

m+r=2jm + r = 2j

where: mm is the number of two-force members, rr is the number of independent reaction components at the supports, jj is the number of pin joints, and the factor 22 counts the two scalar equilibrium equations (∑Fx=0\sum F_x=0, ∑Fy=0\sum F_y=0) available at each joint. If m+r<2jm+r<2j the frame has too few members for the joints available — it is a mechanism, free to collapse under load. If m+r>2jm+r>2j it is statically indeterminate: more members than the equilibrium equations can resolve, requiring compatibility (deformation) equations beyond this course. Equality is necessary but not sufficient — the members must also be arranged so no panel can flex as a mechanism, which is why every panel of a good truss is triangulated (a triangle is the only polygon that cannot change shape without changing a side length).

Method of joints. Isolate each pin as a free particle. Because a pin transmits no moment, only two equations apply there: ∑Fx=0\sum F_x=0 and ∑Fy=0\sum F_y=0. Start by finding the external reactions from overall equilibrium of the whole truss, then work joint by joint, always choosing a joint with at most two unknown member forces so the two equations solve it directly. Assume every unknown member force is tension (pulling away from the joint); a negative answer simply means the member is in compression.

Method of sections. When you need only a few specific member forces deep inside a large truss, cut an imaginary section through no more than three members with unknown forces, isolate one side as a free body, and apply all three equilibrium equations — ∑Fx=0\sum F_x=0, ∑Fy=0\sum F_y=0, and ∑MO=0\sum M_O=0 about a well-chosen point. Choosing the moment centre where two of the three unknown forces intersect eliminates them, so each unknown often falls out of a single equation, without ever visiting the joints in between.

Figure
A triangular truss with joints A B and C, load at C, pin support at A and roller support at B, each member force arrow labelled tension or compression
Triangular truss — method of joints on A, B, C A pin support B roller support C — 10 kN load F_AC = 6.01 kN C F_BC = 6.01 kN C F_AB = 3.33 kN T Every member is a two-force member — pure tension or compression, no bending
Derivation AWhere the determinacy equation m plus r equals 2j comes from

The goal is to show that m+r=2jm+r=2j is nothing more than a headcount of equations versus unknowns — the number of independent equilibrium equations a plane truss can supply must match the number of unknown forces if statics alone is to determine every member force and reaction, and seeing this clearly explains why the check is necessary but not sufficient for stability.

Step 1. Count the unknowns. Every member of a two-force member carries exactly one unknown, its axial force (tension or compression), so mm members contribute mm unknowns. The supports contribute rr unknown reaction components (a pin support contributes two, a roller one). Total unknowns: m+rm+r.

Step 2. Count the available equations. Each joint, treated as a particle in equilibrium, supplies exactly two independent scalar equations, ∑Fx=0\sum F_x=0 and ∑Fy=0\sum F_y=0 (a pin transmits no moment, so there is no third equation per joint). With jj joints, the whole truss supplies 2j2j independent equations.

Step 3. Match equations to unknowns. Statics alone can solve for all m+rm+r unknowns only if the number of equations equals the number of unknowns: 2j=m+r2j = m+r. This is exactly the determinacy condition. If m+rm+r is smaller, there are more equations than unknowns in a formal count, which in practice means the geometry has slack — a panel can move without any member force resisting it, so the structure is a mechanism. If m+rm+r is larger, there are more unknowns than equations, and the extra members are statically indeterminate; their forces depend on stiffness and deformation, not on statics alone.

Step 4. Note the gap between the count and true stability. The equation m+r=2jm+r=2j only checks that the numbers match — it says nothing about how the members are arranged. A truss can satisfy the count exactly and still be an unstable mechanism if the members happen to be arranged so that two of the 2j2j equations turn out not to be independent (for example, a quadrilateral panel with no diagonal: it has the right member count for a neighbouring triangulated panel to compensate, yet that one panel can still flex like a parallelogram). So the equation is a necessary screening test, not a proof of stability; proper triangulation of every panel is what actually guarantees it. Check against the worked triangular truss below: m=3m=3 members (ABAB, ACAC, BCBC), r=3r=3 reactions (pin at AA gives 2, roller at BB gives 1), j=3j=3 joints, so m+r=6=2j=6m+r=6=2j=6 — determinate and, since every panel is a single triangle, also stable.

Worked example

A triangular truss has joints A(0,0)A(0,0), B(4,0)B(4,0), C(2,3)C(2,3) in metres, pinned at AA and roller-supported (vertical reaction only) at BB, with a downward 10 kN load at CC. Find every member force by the method of joints, then use the method of sections on a second, larger truss to find three specific member forces directly.

Step 1 — external reactions. With no horizontal load, ∑Fx=0\sum F_x=0 gives Ax=0A_x=0. Taking moments about AA: ∑MA=0⇒By(4)−10(2)=0⇒By=5\sum M_A=0 \Rightarrow B_y(4) - 10(2) = 0 \Rightarrow B_y = 5 kN. Then ∑Fy=0⇒Ay+5−10=0⇒Ay=5\sum F_y=0 \Rightarrow A_y + 5 - 10 = 0 \Rightarrow A_y = 5 kN.

Step 2 — joint A. Member ACAC runs from (0,0)(0,0) to (2,3)(2,3), length 13=3.606\sqrt{13}=3.606 m, so cos⁡θ=0.5547\cos\theta=0.5547, sin⁡θ=0.8321\sin\theta=0.8321. Vertical equilibrium: FACsin⁡θ+Ay=0⇒FAC(0.8321)=−5⇒FAC=−6.01F_{AC}\sin\theta + A_y = 0 \Rightarrow F_{AC}(0.8321)=-5 \Rightarrow F_{AC}=-6.01 kN, i.e. 6.01 kN compression. Horizontal equilibrium then gives FAB+FACcos⁡θ=0⇒FAB=3.33F_{AB} + F_{AC}\cos\theta = 0 \Rightarrow F_{AB} = 3.33 kN tension.

Step 3 — joint B and the check at C. By the identical geometry on the other side, FBC=6.01F_{BC} = 6.01 kN compression. Checking joint CC: the vertical components of the two 6.01 kN compressive forces pushing into the joint sum to 2×6.01×0.8321=10.02 \times 6.01 \times 0.8321 = 10.0 kN, exactly balancing the 10 kN applied load, and the horizontal components cancel by symmetry — confirming FAB=3.33F_{AB}=3.33 kN (T), FAC=6.01F_{AC}=6.01 kN (C), FBC=6.01F_{BC}=6.01 kN (C).

Competitive stretch — method of sections

A section cut through a larger truss passes through three members: top chord DEDE (horizontal, height 3 m), diagonal DFDF, and bottom chord CFCF (horizontal, y=0y=0), with D(4,3)D(4,3) and F(6,0)F(6,0), so tan⁡θ=3/2\tan\theta=3/2, θ=56.3°\theta=56.3°. The isolated left portion carries only a 15 kN upward reaction at x=0x=0. Find FDEF_{DE}, FDFF_{DF}, FCFF_{CF}, assuming tension.

Step 1. Take moments about DD, eliminating FDEF_{DE} (passes through DD) and FDFF_{DF}'s line does not pass through DD but FCFF_{CF}'s moment arm is the height 3 m: ∑MD=0:15(4)−FCF(3)=0⇒FCF=20\sum M_D=0: 15(4) - F_{CF}(3) = 0 \Rightarrow F_{CF}=20 kN tension.

Step 2. Take moments about FF, where both FDFF_{DF} and FCFF_{CF} act along lines through FF and so contribute zero moment: ∑MF=0:15(6)−FDE(3)=0⇒FDE=30\sum M_F=0: 15(6) - F_{DE}(3) = 0 \Rightarrow F_{DE}=30 kN by the assumed-tension sign convention — meaning the top chord is actually in 30 kN compression, typical for a top chord under sagging.

Step 3. Vertical equilibrium isolates FDFF_{DF}: ∑Fy=0:15−FDFsin⁡56.3°=0⇒FDF=15/0.832=18.0\sum F_y=0: 15 - F_{DF}\sin56.3° = 0 \Rightarrow F_{DF} = 15/0.832 = 18.0 kN tension.

Each of the three member forces came from a single equation — one clean cut and three moment/force choices, versus tracing joint by joint from the support all the way to DD and FF.

Common traps

  • Forgetting a pin joint gives only two equilibrium equations. No moment equation applies at a frictionless pin — only ∑Fx=0\sum F_x=0 and ∑Fy=0\sum F_y=0.
  • Sign confusion between tension and compression. Always assume tension as positive; a negative result means compression. Do not flip the assumed direction mid-calculation.
  • Cutting through more than three unknown members. The method of sections needs at most three unknowns per cut, or the three equilibrium equations become simultaneous instead of one-shot.
  • Treating m+r=2jm+r=2j as sufficient for stability. The count can match exactly while a panel that isn't triangulated still flexes as a mechanism — always check the geometry, not just the arithmetic.

Ask the tutor: If you're unsure which joint to start with or want a walkthrough on spotting the fastest section cut before grinding through algebra, ask the AI tutor to work through a truss diagram with you step by step.

Why does m plus r equal 2j guarantee I can solve for every member force, but not guarantee the truss is actually stable?

The equation only counts equations against unknowns: 2j2j equilibrium equations from the joints against m+rm+r unknown member forces and reactions. When the counts match, statics has, in principle, enough equations to solve everything — but that assumes the equations are all independent. If a panel isn't triangulated (say a plain quadrilateral with no diagonal), that panel can flex like a parallelogram even though the overall member count elsewhere happens to make up the difference, so two of the equations turn out not to be independent after all. That's why you always check triangulation by eye in addition to the arithmetic.

When should I use the method of sections instead of the method of joints

Reach for the method of sections whenever you need only a handful of member forces buried deep in a large truss and don't want to solve every joint between the support and your target member. Pick a cut through at most three unknown members, choose a moment centre where two of them intersect to isolate the third, and you often get each force from a single equation — as in the section-cut example above, where all three forces came from one cut. The method of joints is better when you need every member force, since it naturally sweeps through all of them.

Why is the answer negative for F_AC in the worked example, and what does that mean physically

Solving joint AA with tension assumed positive gave FAC=−6.01F_{AC}=-6.01 kN. The negative sign is not an error — it is the equilibrium equations telling you the member is actually being pushed, not pulled. A member in compression resists being shortened by pushing outward on each pin, which is the opposite sense to the tension direction you assumed when setting up ∑Fy=0\sum F_y=0. Report the magnitude, 6.01 kN, and state "compression" rather than carrying the minus sign forward into later calculations, where it is easy to mishandle.

Before this

Where this leads