Standards & Safety

Glass-and-steel railing deflection proof: how to read the wind-tunnel report an architect needs for a Bellandur 18th-floor balcony

Vetrova Atelier8 July 2026
Glass-and-steel railing deflection proof: how to read the wind-tunnel report an architect needs for a Bellandur 18th-floor balcony

An 18th-floor balcony in Bellandur on a June afternoon: the monsoon wind pushes at the glass railing with a force most architects never calculate by hand. The frameless glass panel—say, 1200 mm tall, 10 mm tempered, held at top and bottom by a steel spigot—deflects. Not visibly to the eye. But it moves. The architect who specifies that railing needs to know exactly how much, under what load, and whether the movement stays within tolerance. The test report proves it. Most architects don't read it carefully. This one does.

Why deflection matters more than thickness

Glass thickness is not the answer to wind load. A 12 mm panel and a 10 mm panel will both deflect under wind pressure; the difference is one millimetre in the pane itself. What matters is the span—the distance between the top and bottom fixings—and the way the glass is held at those points. A railing on the 18th floor of a Bellandur tower, exposed on three sides, will see wind speeds that a ground-floor installation in Koramangala will never encounter. The wind-load test report quantifies this: it tells you, in millimetres, how far the glass will move under a specified pressure.

Deflection tolerance for frameless glass railings in India follows IS 6533 (Safety Code for Glass and Glazing). The rule is simple: the glass panel must not deflect more than span divided by 200. For a 1200 mm tall railing, that's 6 mm maximum. The test report proves whether your chosen glass thickness and fixing method stay within that limit. If the report shows 7.2 mm deflection at design wind pressure, the spec fails. You either thicken the glass, change the fixing detail, or reduce the span.

Reading the load cases: which ones apply to your Bangalore site

A proper wind-load test report will include three or four load cases. Not all of them apply to every balcony. You need to know which ones matter.

Perpendicular wind load (the primary case)

This is wind pushing directly at the face of the railing—the most common load case. The test report will specify the pressure in pascals (Pa). For a mid-rise balcony in Bangalore—say, 15 to 25 storeys—the design wind pressure is typically 1.2 to 1.5 kPa (1200 to 1500 Pa), depending on the building's exposure category and the site's wind history. The report will show a graph or table: at 1200 Pa, the glass deflects to X mm. At 1500 Pa, to Y mm. You compare Y against your tolerance. If Y exceeds span divided by 200, that spec doesn't work for your site.

Racking load (the secondary case)

Racking is lateral shear—the top of the railing pushed one way, the bottom held fixed. This load case matters less for a typical balcony railing, but it matters for tall, narrow panels and for railings that double as wind bracing. The test report will include racking deflection, usually expressed as a horizontal offset at mid-height under a shear force. For a 1200 mm tall frameless panel, racking deflection should stay under 5 mm. If your report doesn't include racking data, ask for it before you sign off the spec.

Impact load (the safety case)

This is not wind. This is a person falling against the railing. IS 6533 requires the railing to withstand a horizontal impact of 1.1 kN applied at mid-height without permanent deformation. The test report will state whether the glass panel passes this test. For frameless railings, the impact load is often the governing case—it determines whether you need 10 mm or 12 mm glass more than wind does. If the report shows the 10 mm panel passes wind but fails impact, you must specify 12 mm.

Understanding the test certificate and site-specific adjustments

A test report is only valid for the exact configuration tested. If the report covers a 1200 mm tall, 10 mm tempered, single-pane railing with spigot fixings at top and bottom, you cannot use that report to justify a 1500 mm tall railing or a laminated panel. The test is material and geometry specific. Before you rely on a report, verify three things: the glass thickness, the panel height, and the fixing method match your shop drawing exactly.

Bangalore's monsoon humidity (June to September) and the Cauvery water TDS (200–300 ppm) affect the durability of the steel fixings, not the deflection itself. But they do affect the long-term stiffness of the joint. A corroded spigot will lose preload over time, and a loose spigot will allow the glass to deflect more. Specify stainless-steel fixings (grade 316, not 304) for any railing in Bangalore. The test report should note the material grade of the fixings; if it doesn't, ask the manufacturer before you commit to the spec.

What to ask the manufacturer before you sign the shop drawing

Three questions will protect your specification and your schedule.

  1. Does the test report cover the exact glass thickness, panel height, and fixing detail in my shop drawing? If not, can you provide a report that does, or can you certify that the variation is negligible?
  2. What is the design wind pressure for my site, and at what pressure does the glass deflect to the IS 6533 limit (span divided by 200)? I need a number, not a range.
  3. Has the railing been tested for impact load per IS 6533, and does it pass with the glass thickness I've specified? If it passes wind but fails impact, I need to know now.

A manufacturer who cannot answer these questions clearly should not be on your shortlist. The test report is not a marketing document; it is engineering proof. Treat it as such.

Specifying the joint tolerance and the shop drawing callout

The deflection test assumes the glass is held at the top and bottom spigot with zero play. In reality, there will be a small gap—a tolerance—between the glass and the spigot bore. This tolerance is critical. If it is too loose, the glass will rattle and deflect more than the test predicts. If it is too tight, the glass may crack during installation or thermal cycling.

For a 10 mm tempered railing in Bangalore, specify a joint tolerance of 0.5 to 1.0 mm between the glass edge and the spigot bore. The spigot bore diameter should be 10 mm nominal; the glass edge should be 9.0 to 9.5 mm. This tolerance allows the glass to sit snugly without binding. Call this out explicitly in the shop drawing. Include a detail section showing the spigot, the glass, and the gap. The installer needs to see it; the manufacturer needs to confirm it can be held to that tolerance. If the manufacturer says "we'll fit it tight," you have a problem. Tight is not a tolerance.

A worked example: HSR Layout, 16th-floor apartment, 1200 mm tall railing

An apartment on the 16th floor of a residential tower in HSR Layout, east-facing balcony, 1200 mm tall frameless railing, 10 mm tempered glass, stainless-steel spigot fixings at top and bottom. The architect specifies the railing and receives a test report from the manufacturer. The report shows:

  • Perpendicular wind load at 1200 Pa: 4.8 mm deflection
  • Perpendicular wind load at 1500 Pa: 6.2 mm deflection
  • Impact load (1.1 kN horizontal): passes, no permanent deformation
  • Racking load (shear): 3.1 mm deflection at mid-height

The IS 6533 limit for a 1200 mm span is 6 mm. At 1200 Pa, the railing is safe. At 1500 Pa, it exceeds the limit by 0.2 mm. The architect checks the wind-speed history for HSR Layout and the exposure category of the building. If the design wind pressure is certified at 1200 Pa or less, the 10 mm spec is valid. If the structural engineer has specified 1500 Pa, the architect must either move to 12 mm glass or reduce the panel height to 1100 mm (limit: 5.5 mm). The test report gives the data; the architect makes the call. This is how it should work.

Joint detail and the role of the spigot in controlling deflection

The spigot—the steel post that holds the glass top and bottom—is not just a fixing. It is part of the deflection-control system. A spigot with a loose bore will allow the glass to move laterally; a spigot with a tight, well-finished bore will keep the glass centred and reduce lateral deflection. The test report will specify the spigot bore diameter and the preload force (if any) applied to the glass. For a 10 mm tempered railing, a typical preload is 5 to 10 kN, applied through a set screw at the top of the spigot. This preload compresses the glass slightly and keeps it from rocking in the bore.

When you specify the railing, include a shop drawing detail that shows the spigot bore diameter, the preload method, and the tightening torque. For a stainless-steel set screw in a 10 mm bore, the torque is typically 8 to 12 N⋅m. Too little torque, and the glass will move. Too much, and you risk cracking the glass or stripping the screw. The manufacturer should provide a torque specification in the installation manual. If they don't, ask for it. This detail goes on site and is checked during handover.

Common mistakes architects make when reading deflection reports

The first mistake is assuming that a thicker glass always means less deflection. A 12 mm panel will deflect less than a 10 mm panel, but only by about 30 percent. If you need to reduce deflection by half, you cannot do it by thickness alone; you must also reduce the span or change the fixing method. The test report will show this clearly. If you see a report that claims 12 mm glass reduces deflection by 50 percent, the report is misleading or the test conditions have changed.

The second mistake is ignoring the joint tolerance. An architect who specifies 10 mm glass in a 10.2 mm bore is asking for trouble. The glass will rattle, the joint line will look sloppy, and the deflection under wind will exceed the test prediction. Tolerance is not a detail; it is a specification. Call it out in the shop drawing, and verify it during the first-fix inspection.

The third mistake is using a test report from a different climate or building type. A wind-load test conducted in a wind tunnel in a temperate country may not account for Bangalore's monsoon wind patterns or the thermal cycling caused by the Cauvery water and the summer heat. Ask the manufacturer whether the test was conducted with Indian design standards (IS 6533, IS 875) and whether the test pressure matches your site's design wind speed. If the report is based on European or American standards, convert it to IS equivalents before you sign off.

Questions we get asked

Can I use the same railing spec for a 10th-floor balcony and a 20th-floor balcony?

No. Wind pressure increases with height. A 10th-floor balcony in Whitefield might see 1.0 kPa design wind pressure; a 20th-floor balcony in the same building might see 1.4 kPa. The test report will show deflection at different pressures. You must confirm that the deflection at your site's design wind pressure stays within the IS 6533 limit. If it doesn't, you thicken the glass or reduce the span for the higher floor.

Does the test report account for the monsoon?

The test report accounts for wind speed, not season. The design wind pressure for Bangalore is derived from wind-speed data collected over many years, including monsoon months. IS 875 specifies the method for calculating design wind pressure based on the building's exposure category and the site's basic wind speed. The manufacturer's test report should cite the design wind pressure in pascals. If the report says "tested to Indian standards," it should be based on IS 875 and IS 6533. If you are unsure, ask the structural engineer for the design wind pressure for your specific site and floor level.

What happens if the glass deflects more than the test report predicts?

If the installed railing deflects visibly more than the test report predicts, one of three things has happened: the glass thickness is wrong, the span is longer than specified, or the spigot preload has loosened. Check the as-built dimensions against the shop drawing. Measure the glass thickness with a calliper. Check the spigot preload with a torque wrench. If the preload has dropped below the manufacturer's specification, tighten it. If the glass thickness or span is wrong, the railing may need to be replaced.

Do I need a separate test report for a laminated railing, or can I use the tempered report?

You need a separate report. Laminated glass has different stiffness and deflection characteristics than tempered glass. A 10 mm laminated panel (say, 5 mm plus 5 mm tempered with a 0.76 mm PVB interlayer) will deflect more than a 10 mm monolithic tempered panel. If your shop drawing specifies laminated glass, ask the manufacturer for a test report that covers laminated glass at that thickness and that span. Do not assume the tempered report applies.

Is the test report valid forever, or does it expire?

The test report is valid as long as the manufacturing process and the material specification remain unchanged. If the manufacturer changes the glass supplier, the tempering process, or the spigot design, the old test report is no longer valid. Before you specify a railing from a new supplier, ask for a current test report, not one from five years ago. Also, if you are commissioning a railing with a custom detail—say, a non-standard spigot or a different fixing method—ask whether the manufacturer has a test report for that detail. If not, they may need to commission a test.

Commissioning a tested and certified railing for your Bangalore project

The test report is the foundation of a good railing spec. It proves that the glass, the fixings, and the joint detail will perform within the limits set by IS 6533 under the wind speeds and impact loads your site will experience. Before you finalize the shop drawing, ensure that the test report covers your exact configuration, that the deflection stays within tolerance at your site's design wind pressure, and that the manufacturer has provided a clear installation and maintenance manual. The atelier can walk you through the test data and help you translate it into a site-ready detail. Talk to the atelier about your specific floor height, balcony exposure, and glass preference, and we will confirm the right spec for your project.