AdvancedTrussSelf-stressLack of fitVirtual work

A member fabricated short: lack of fit

The workshop cut a diagonal 5 mm short, and the crew pulled it into place anyway. In a determinate truss that would cost nothing — the geometry would simply shift. In a redundant one the member fights back: over 100 kN of self-stress locks in, with no load on the structure and every support reaction exactly zero. Here we’ll pin down exactly how much.

Figure 1.The problem: a 3 m × 3 m X-braced panel — 4 nodes, 6 members (M1–M6). Member M5 (the diagonal 1–3) is fabricated 5 mm short. No applied load; the self-stress it locks in comes after we solve.
Given
Configuration
3 m × 3 m X-braced panel
Nodes / members
4 nodes · 6 membersindeterminate, degree 1
Area (all)
1500 mm²
200 GPa
300 000 kN
Diagonal 1–3 length
3√2 = 4.243 m
Misfit
5 mm shortentered −5
Applied load
none
1

Step 1 — Count the members: one too many

One more member than statics can resolve — and that is the whole story.

Before we touch the misfit, count. A plane truss offers two equilibrium equations at every joint — 2j = 8 here — against nine unknowns: six member forces and three reaction components. So m + r = 6 + 3 = 9 > 2j = 8, and the panel is indeterminate to the first degree — one member more than statics can resolve.

That one surplus member decides everything that follows. A determinate truss (m + r = 2j) has exactly one geometry compatible with whatever member lengths the workshop delivers: arrive with a diagonal 5 mm short and the truss simply assembles into a very slightly different shape, force-free. This panel has no such freedom. Its members must all agree on one shared geometry, so a diagonal that shows up short can only be connected by stretching itself and squeezing its neighbors. Force is coming — the next steps compute exactly how much.

2

Step 2 — Cut the misfit member, apply a unit pair

Release the redundant, load it with a unit tension.

The flexibility method turns that surplus member into the unknown. We cut the redundant — member M5 (the diagonal 1–3) — leaving a determinate truss statics can handle, and pull its two cut ends apart with a unit tension pair. The member forces this pair produces are the mode shape n. The classic X-braced-square answer, solvable at any joint:

(all four)

Read the shape: a unit pull across one diagonal squeezes the square’s four sides (M1–M4) and stretches the opposite diagonal M6 (2–4) — entirely internally. The support reactions under this self-equilibrated pair are already zero, which is the whole reason the finished self-stress state will read zero at every support too.

3

Step 3 — The flexibility of the cut, by virtual work

One sum condenses the whole panel into a single number.

How hard is it to close a gap at that cut? Virtual work answers with one sum. The flexibility of the cut — how far the cut ends separate per unit of the pair — is summed over every member: the four 3 m sides at , the two 4.243 m diagonals at :

Notice that n enters squared: every member adds flexibility whether its mode force is tension or compression — the squeezed sides help the cut open just as the stretched diagonal does. From here on, nothing else about the truss matters; the whole panel has been condensed into this one number.

4

Step 4 — Five millimeters becomes 103.55 kN

Five millimeters → over 100 kN, by virtual work.

Member M5 (the diagonal 1–3) must close a ΔL = 5 mm gap, so the redundant force is the misfit divided by that flexibility:

Every other member force follows from :

That is roughly 69 MPa of locked-in stress from five millimeters of mis-cut — before the structure carries a single newton of real load. Fabrication tolerance on a redundant frame is not pedantry.

Figure 2.The locked-in self-stress — members colored red (tension) / blue (compression), member IDs shown: both diagonals M5 and M6 at +103.55 kN tension, all four sides M1–M4 at −73.22 kN compression, and every support reaction zero. The panel strains against itself with nothing applied.
5

Step 5 — Why every support reaction is zero

Statics sees nothing; strain gauges see 69 MPa.

A self-stress state is internally balanced: the member set pushes and pulls against itself. At each joint the member forces close into equilibrium without help from a support, so ΣF at every support is exactly zero. Statics, looking only at the boundary, sees an unloaded structure — yet a strain gauge on any member would read a real, permanent force.

The proof

Hand calculation vs the solver.

Verified — hand calculation vs the solver, to round-off
QuantityBy handStructureCalcs
Short diagonal 1-3+103.55 kN (T) — X = ΔL/δ₁₁+103.553 kN
Other diagonal 2-4+103.55 kN (T) — n = +1+103.553 kN
All four sides−73.22 kN (C) — n = −1/√2−73.223 kN
Support reactionszero (self-stress is internal)0 / 0
Applied loadzerozero

Every value was worked by hand with the classical method, then checked against this site’s solver — the same engine the Try it button opens. This agreement is re-run automatically on every build.

Now make it yours

Open this exact model in the calculator — then change a load, drag a support, and watch every diagram update in real time. The best way to build intuition is to break it and see what happens.

Take it with you

Export this worked example as a PDF, or download it as a .screport and open it in the Report Builder — the model travels inside the file, so you can reconstruct it, re-solve, and build your own report from it.

Verifying your link…