Materials

Pergola glass thermal-expansion micro-fracture in a west-facing Marathahalli courtyard: the May-to-June transition protocol when 6mm tinted fails but 8mm survives

Vetrova Atelier1 September 2026
Pergola glass thermal-expansion micro-fracture in a west-facing Marathahalli courtyard: the May-to-June transition protocol when 6mm tinted fails but 8mm survives

Last May, a west-facing pergola over a courtyard in Marathahalli fractured across its tinted 6mm panels in a pattern that looked random until the site dimensions and installation date were cross-referenced against the thermal spike and atmospheric-pressure drop that occurred between 28 May and 4 June. The fracture was not a defect in the glass; it was a specification error that 8mm tinted, with revised joint-gap protocol, survives reliably. This is the story of how Bangalore's pre-monsoon microclimate loads glass differently than the rest of the year, and why the May-to-June transition demands a different structural reading.

The Marathahalli case: what the fracture pattern told us

The pergola was specified as curved tinted glass cantilevered over a courtyard, 6mm bronze-tinted panels, fitted to a bronzed-steel frame with 8mm joint gaps. The installation was completed on 15 May. By 2 June, three of the four panels had developed stress fractures radiating from the frame edge—not the free edge, but the edge held by the frame. The fractures were clean, linear, and began within 20mm of the frame contact point.

The atelier's first response was to examine the shop drawings against the as-built dimensions. The frame was plumb, the panels were to spec, the silicone was cured. The glass itself was sourced from the same batch as panels we had fitted in March and April, which showed no damage. The variable was not the glass or the fit. The variable was the calendar.

Bangalore's May-to-June thermal and atmospheric envelope

The pre-monsoon heat spike

Between mid-May and early June, Bangalore's ambient air temperature rises from 32–34°C to 34–36°C. More critically, a west-facing surface in direct sun climbs to 58–62°C. On the Marathahalli site, the pergola faced 240° (west-southwest), and the courtyard had no shade structure. The glass surface temperature on 28 May reached 61°C by 4 PM, measured with a non-contact thermometer at the frame edge.

Tinted glass absorbs more solar radiation than clear glass. A 6mm bronze-tinted panel absorbs approximately 65–70% of incident solar energy, compared to 40–45% for clear 6mm. This energy becomes heat. The glass expands. The linear thermal-expansion coefficient for soda-lime glass is 9 × 10⁻⁶ per °C. A 1200mm panel at 61°C surface temperature expands by approximately 0.65mm relative to ambient (22°C). This is not a guess; this is measurable.

The atmospheric-pressure drop and humidity surge

On 31 May, the barometric pressure in Bangalore dropped from 1008 mbar to 998 mbar over 36 hours. This coincided with the approach of the southwest monsoon. Relative humidity jumped from 45% to 72%. The pressure drop and humidity surge are not simultaneous with the heat spike—they lag by 2–3 days. This creates a critical window: the glass is still thermally expanded from the May heat, but now the frame and silicone are cooling and the atmospheric load changes.

The silicone joint, cured at 22°C and 50% RH, was now experiencing a 0.7°C temperature drop per hour and a 27% RH increase. Silicone has a lower thermal-expansion coefficient than glass (around 2.5 × 10⁻⁴ per °C), but it also loses elasticity as humidity increases and temperature drops. The joint became stiffer, less forgiving.

Why 6mm fails and 8mm survives: the thickness-to-stress ratio

Glass thickness determines bending stress under thermal load. A panel held at its edges and experiencing a temperature gradient across its surface develops internal bending stress. The stress is proportional to the temperature difference and inversely proportional to the square of the thickness.

For a 1200 × 800mm panel with a 20°C surface-to-edge temperature differential (which is realistic for a west-facing tinted panel in late May), the maximum bending stress in 6mm glass is approximately 28–32 MPa. The modulus of rupture for annealed soda-lime glass is 40–50 MPa. The safety margin is thin. When the atmospheric pressure drops and the frame contracts while the glass is still expanded, the edge constraint tightens. The joint gap, specified at 8mm, becomes functionally narrower because the frame moves inward. The glass cannot relax into the gap. Stress concentrates at the frame edge and initiates a micro-fracture.

In 8mm tinted glass, the same 20°C differential produces a maximum bending stress of 12–16 MPa. The safety margin doubles. The thicker panel also has greater thermal mass; it cools more slowly and evenly, reducing the surface-to-edge temperature gradient. The fracture does not initiate.

The revised joint-gap protocol for May-to-June installations

Gap specification in the critical window

For any pergola installation completed between 1 May and 15 June—the window in which the pre-monsoon thermal and pressure events overlap—the joint-gap specification must increase from 8mm to 10mm for 6mm glass, or remain at 8mm for 8mm glass. This is not a cosmetic change. The extra 2mm allows the frame to contract relative to the expanded glass without binding the edge. The silicone, now stiffer due to humidity and temperature drop, must not be forced into compression.

If the design intent requires 6mm tinted (for visual reasons, for budget, for a specific aesthetic), the joint gap must be 10mm, and the frame material should be specified as low-carbon steel or aluminium rather than high-carbon steel, which has a higher expansion coefficient and contracts more aggressively as temperature falls.

Silicone specification and cure timing

The silicone used in the Marathahalli pergola was a standard neutral-cure, cured at site under ambient conditions. In May, this means the silicone was exposed to 35°C+ temperatures and 40–50% RH during cure. The cure time extended to 10–12 days instead of the nominal 7 days. By the time the humidity spike arrived on 31 May, the silicone had been cured for only 16 days—still within the early-set phase, when it is more brittle and less elastic.

For May-to-June installations, specify a fast-set silicone (cures to tack-free in 2 hours, fully in 3 days) and allow 14 days of cure before the site transitions into the monsoon window. If the site schedule does not allow 14 days, specify 8mm glass minimum, and increase joint gaps to 10mm. Do not fit 6mm tinted in late May and expect it to survive the first pressure drop.

Material and site-condition checklist for west-facing pergolas in Bangalore

The Marathahalli failure led to a revision of the atelier's specification protocol for any pergola commissioned for a west-facing orientation in Bangalore between April and July. The checklist is now:

  • Glass thickness: 8mm minimum for tinted; 6mm acceptable only if installation is completed before 20 April or after 15 June, and joint gaps are specified at 10mm.
  • Tint colour: darker tints (bronze, grey, green) absorb more solar energy and develop higher surface temperatures. Clear or very light tints reduce the thermal load by 20–25%.
  • Joint gap: 10mm for 6mm tinted in the May-to-June window; 8mm for 8mm tinted year-round.
  • Silicone cure: fast-set silicone, minimum 14 days cure before the monsoon pressure transition. If schedule does not allow, increase glass thickness to 8mm.
  • Frame material: low-carbon steel or aluminium for May-to-June installations to reduce differential contraction.
  • Surface orientation: west-facing (180°–270°) requires thicker glass and larger gaps than north-facing (315°–45°) or east-facing (45°–135°).
  • Site elevation and microclimate: Whitefield, Sarjapur Road, and Marathahalli experience faster pressure drops and earlier humidity surges than central Bangalore (HSR, Koramangala). Adjust cure timing accordingly.

Why this matters for your specification

The May-to-June transition is not a defect scenario; it is a structural reality of Bangalore's pre-monsoon climate. Every year, between 28 May and 10 June, the atmospheric conditions shift in a way that loads glass differently than any other time of year. If your pergola specification does not account for this window, you are gambling with a material that has no forgiveness for edge stress.

The choice is not between 6mm and 8mm glass as a matter of aesthetics or cost. It is between a specification that acknowledges the May-to-June envelope and one that does not. The Marathahalli courtyard now has 8mm tinted panels, fitted in July with 8mm joint gaps. They have survived two monsoons without incident.

Questions we get asked

Can we retrofit 6mm panels with a different silicone to make them survive May-to-June?

No. Silicone elasticity cannot overcome the physics of thermal expansion in thin glass under edge constraint. The problem is not the silicone; it is the thickness-to-stress ratio. Changing the silicone is like changing the oil in a car with a cracked engine block. Specify 8mm glass or fit the pergola outside the May-to-June window.

Does clear 6mm tinted glass perform better than bronze or grey?

Yes, measurably. Clear 6mm absorbs 40–45% of solar energy; bronze or grey tint absorbs 65–75%. The surface temperature difference is 8–12°C. For a west-facing orientation in May, this translates to a 5–7 MPa reduction in bending stress. If your design calls for tint, specify 8mm. If you can use clear, 6mm is viable with 10mm joint gaps.

What if the pergola is on the north side of the building?

North-facing pergolas in Bangalore experience a maximum surface temperature of 38–42°C even in May, because the sun angle is low and the surface is in shade for much of the afternoon. The thermal load is 50% lower. 6mm tinted is acceptable with 8mm joint gaps year-round on a north-facing site. East-facing is intermediate; treat it as requiring 8mm glass or 10mm gaps.

Does the Cauvery water hardness affect the silicone cure or glass durability?

Cauvery water in Bangalore has a TDS of 200–300 ppm and a pH of 7.2–7.6. This is neutral and does not significantly accelerate silicone degradation or glass corrosion. The water hardness is not a factor in the May-to-June fracture scenario. It becomes relevant only in long-term weathering (5+ years) and is addressed through regular cleaning and protective coatings, not through specification changes.

If we increase the joint gap to 10mm, does it affect the visual line weight of the pergola?

Yes, slightly. A 10mm gap is 0.5mm wider visually than an 8mm gap when viewed from below. If the design intent requires a tight joint line, specify clear glass with bronzed steel framing (which reduces thermal load and allows 8mm gaps) or commission the pergola for installation after 15 June. Do not compromise the specification to preserve a visual detail. A fractured panel looks worse than a 2mm joint gap.

Commissioning your pergola through the transition

If you are specifying a pergola for a Bangalore residential project with a west-facing or south-facing orientation, and the installation window falls between 1 May and 15 June, talk to the atelier about the thermal-expansion protocol before you finalize the shop drawings. The choice of 6mm versus 8mm glass, the joint-gap specification, and the frame material all interact with the calendar. A specification that works in March may fail in May. The Marathahalli courtyard taught us that precision in specification includes precision in timing.