Glass · Safety critical
Glass balustrade calculator
A free-standing glass balustrade is a cantilever, so the interlayer decides everything. Stress at the base is σ = 6·F·h/t² and deflection at the top is δ = 4·F·h³/(E·t³) — and t is the effective thickness, which for the same 12+12 laminate can be 15 mm or 24 mm depending on how much shear the interlayer transfers. That is a factor of four in deflection. This page gives you both bounds and never pretends to know which one you have.
How this is calculated
A free-standing balustrade is a vertical cantilever, clamped at the bottom, pushed horizontally near the top. Everything is worked out per metre of width, which is how the line load is given.
The effective thicknesses are the two bounds of a laminate, exactly as in the thickness calculator:
Design strength follows EN 16612, with kmod = 0.663 · t−1/16 for a load lasting t hours — which gives 1.00 at five seconds, 0.89 at thirty and 0.74 at ten minutes.
| Glass type | fbk | Design strength, 5 s |
|---|---|---|
| Annealed | 45 N/mm² | 25.0 |
| Heat-strengthened | 70 N/mm² | 45.8 |
| Fully tempered | 120 N/mm² | 87.5 |
The broken-ply check, which almost nobody does
This is the part that matters and the part most online calculators skip entirely. A balustrade is not judged only on whether it holds when it is intact. It is judged on what happens when one ply breaks — because glass does break, and a barrier that lets someone through when it does has failed at its only real job.
The page checks the remaining ply on its own, carrying the full line load, and reports the stress. Read it as a warning light, not as a certificate:
- Tempered glass all but disappears when it breaks. It dices into small fragments, and the residual capacity comes from the interlayer holding those fragments, not from the glass. A calculation of "the remaining ply" flatters it badly.
- Heat-strengthened breaks into large pieces that stay bonded and keep interlocking, which is why barrier standards lean towards it despite the lower strength.
- Real post-breakage performance is demonstrated by test, to BS 6180 and EN 12600, not by arithmetic. Codes generally allow a reduced load and a longer duration for the residual case; this page does not attempt to pick those for you.
What this page does not do
- Free-standing cantilever only. A continuous top handrail changes the model completely — the rail spans between panels and sheds load, and the glass sees far less.
- The clamp is assumed rigid. In real life the base channel rotates, and that rotation often adds more to the top deflection than the glass itself. Ask the channel supplier for their rotation figure and add it.
- Line load only. No uniformly distributed infill load, no point load, no wind, no crowd dynamics, no impact or soft-body test.
- No fixings, no bolted or clamped local stresses, and no check on the glass around holes.
- The interlayer bounds are bounds. Where you sit between them depends on the product, the temperature and the load duration — a stiff ionoplast interlayer on a cold day is near the top, soft PVB on a hot summer afternoon is near the bottom. If the design only works at the upper bound, it does not work.
Worked example you can check by hand
1100 mm free height, 0.74 kN/m, 12 + 12 laminated, heat-strengthened, 5 s.
So it passes on both counts at the conservative bound — but the deflection is at 96 % of its limit, which is exactly the kind of margin that disappears once the base channel rotates.