A 1970s single-family house, two identical covered canopies, and cracks running through the rendered masonry right above the timber edge beams. Before proceeding with the planned re-roofing, the owner wanted a documented answer: where do these cracks come from, are they active, and what do they mean for the structure?
Instrumented findings
Crack widths were measured with a half-millimetre graduated rule; timber moisture was checked with an electrode moisture meter.
| Crack | Canopy | Width | Pattern |
|---|---|---|---|
| S1 | South (terrace) | 2–3 mm at the base | Full height, near-vertical, offset between lips |
| S2 | South (terrace) | 0.5–1 mm | ≥ 35 cm, zigzag |
| S3 | South (terrace) | Not measured | Full height, near-vertical |
| N1 | North (entrance) | 1.5–2 mm | Full height, near-vertical |
A 5 to 10 mm gap was also found between the top of the south beam and the masonry band above it: light passes through the joint — the band no longer bears continuously on the beam. Timber moisture was about 15 %, normal for sheltered exterior wood.
Four possible origins, one verdict
Each hypothesis was tested against its expected signature. Timber movement (deflection and creep) matched: cracks starting at the beam, wider at the base, plus the gap. Shrinkage of the render alone and support movement were ruled out; differential material movement acts as an aggravating factor only.
The mechanism, quantified
The edge beam alone carries the masonry band and the roof edge — a permanent load of roughly 3.5–5.2 kN/m. Under sustained load, wood keeps deflecting for decades: this is creep. Rendered masonry cracks at just 0.8–1.6 mm of deflection. Eurocode 5 calculations across the full plausible range of parameters (sections 16 × 20 to 20 × 25 cm, spans 3.5–4.5 m, softwood or oak) give a final deflection of 12–43 mm against an allowable 7–9 mm (L/500) — exceeded in 99.997 % of 100,000 simulated combinations. Independently, the measured crack widths and the gap imply at least 8–17 mm of deflection. Both approaches agree.
The visible steel bar is not a working lintel: a 3 mm crack would require about 1,500 MPa in a bonded bar — three times the yield strength of ordinary steel.
Re-roofing: four priced scenarios
Without constituting a work recommendation, four sequences were assessed: simply filling the cracks (likely reopening); stripping and re-laying the roof without propping (probable worsening, ratchet effect, design strength exceeded once in two under site loads); the same with a heavier covering (worse); and propping the beams first, repairing, then treating cracks — the only sequence without an identified aggravating factor.
What this case reveals about our method
- Systematic elimination over first impressions — each origin is tested against its expected signature.
- Numbers over adjectives — Eurocode 5, 100,000 simulated combinations, 99.997 % exceedance probability, cross-checked against site measurements.
- Stated uncertainties and how to lift them — laser survey, resistograph, graduated tell-tales.
- Total independence — no works, no materials sold, never commissioned by insurers; no work recommendations.
- Fast turnaround — written report delivered within 48 hours of the visit.
This case study is published with the owner’s consent and fully anonymised: no city, address, visit date or photograph of the property appears in it. This case study is published for educational purposes; it does not replace an inspection of your own property, which depends on its configuration and history.
📅 Last updated 02/11/2026 - Cassini Expertise, independent building expert – departments 06, 83 and 13.