[gtranslate]
Structural Silicone Bite Depth and Cure Time: How to Set a Safe Handling Window
A structural glazing bead has two numbers that decide whether the panel stays on the building: how much sealant touches the glass, and how long that sealant has cured before someone lifts the panel. Most site failures we are asked to diagnose are not chemistry failures. They are scheduling failures — a panel moved on day two, a bite measured on the chamfer instead of the contact face, a second side glazed before the first side had cured.
This guide explains how bite depth, cure rate and handling time connect, and how to build a handling window you can defend in a submittal.
What bite depth actually measures
Bite depth (also called bite, adhesion width or contact width) is the width of sealant contact on the glass, measured perpendicular to the joint length. It is not the joint gap you see from outside, and it is not the bead thickness.
- Bite — contact width on the glass surface, hidden behind the panel edge.
- Bead thickness — the depth of sealant in the gap between glass and frame; usually 6 mm or more.
- Joint width — the visible gap, which for a weather seal is a movement dimension, not a strength dimension.
For a structural joint the bite is the load path. Wind load enters the glass, crosses the bite, and leaves through the frame. If the bite is short, the sealant is asked to carry a stress it was never designed for, and it will read as an adhesion failure later — even though the real cause was geometry.
How bite depth is calculated
The usual design approach follows ASTM C1401 practice: divide the design wind load by the allowable design stress of the structural silicone, then check the result against the minimum bite rule.
Two constraints apply at the same time, and the larger one governs:
- Minimum bite: 6 mm on each contact face, regardless of load. Below this, application tolerance alone can consume the design.
- Calculated bite: design pressure × the short span of the panel, divided by the allowable design stress of the sealant. A common allowable figure for structural silicone is 0.14 MPa (about 20 psi) after the safety factors in the standard are applied.
The calculation gives the number. The shop drawing gives the tolerance. What matters on the line is that the tape spacer, the setting block and the glazing robot all deliver the calculated contact width — measured, not assumed.
Cure rate: the number the cartridge does not print
A one-part neutral structural silicone cures from the exposed surface inward at roughly 2 mm per 24 hours at 23 °C and 50 % relative humidity. That is a laboratory figure. On a winter facade it can fall to 1 mm or less, and in a hot dry shop it will not speed up much because the reaction needs moisture from the air, not heat.
Two-part structural silicone is a different schedule: mixing starts the reaction, and full cure of the bite happens in hours rather than days. That is why two-part systems are used where panels must ship quickly — but the handling window still has to be written down, because “fast” is not a number.
The cure rate is measured on a slab, not on your joint. A bead with two contact faces cures from two directions, but the deepest point — the far edge of the bite — is still the last to cure, so the bite depth is the governing distance.
Turning bite depth into a handling window
The working formula we recommend to fabricators is deliberately conservative:
Handling time = (bite depth ÷ measured cure rate) × 2
The factor of two covers the difference between a 23 °C lab slab and a real shop, plus the tolerance on the bite itself.
| Bite depth | Theoretical cure at 2 mm/24 h | Recommended handling window | Where it is used |
|---|---|---|---|
| 6 mm | 72 h | 6 days | Low-rise, small panels, low wind load |
| 8 mm | 96 h | 8 days | Typical mid-rise curtain wall |
| 10 mm | 120 h | 10 days | High wind load, large modules |
| 12 mm | 144 h | 12 days | Four-sided units, tall buildings |
| 16 mm | 192 h | 16 days | Oversized panels, severe exposure |
Read the table as a planning baseline, not a specification. The number that governs your project is the cure rate on your sealant’s technical data sheet, measured at your shop temperature, applied to the bite on your shop drawing.
Where the handling window is lost on site
1. The panel is lifted the next morning
The bead looks and feels set because the surface has skinned. Skinning is a surface event; cure is a depth event. A panel lifted while the inner bite is still soft can shift a fraction of a millimetre — enough to leave a thin, stressed adhesion line that fails two winters later.
2. The bite is measured on the chamfer
Glass edges are chamfered or seamed. If the contact width is measured from the outer edge instead of the true flat contact face, the effective bite can be 1–2 mm less than drawn. On a 6 mm design that is a 20–30 % loss of load path.
3. The tape spacer is compressed
Oversized spacer tape, or a gun pressure that pushes the glass outward, reduces the bead thickness and can push the sealant into a wedge. A wedge-shaped bead has a thin edge that cures fast and a thick edge that cures slowly, so the panel reads “cured” when it is not.
4. Primer on a damp or dusty edge
Primer is a bridge, not a cleaner. Applied over dust, moisture or cutting fluid, it becomes a weak layer between glass and sealant, and the joint fails cohesively at the primer line no matter how long it cured.
5. The second side is glazed too early
In four-sided structural glazing, the first two sides must carry the panel while the second two are applied. If the handling window is not respected between passes, the first bite is loaded before it is cured, and the unit is delivered with a pre-stressed joint.
6. Night temperature is ignored
Cure rate follows substrate temperature and humidity, not the day’s peak. A shop at 28 °C in the afternoon and 8 °C at night cures much more slowly than the average suggests. Log the shop temperature through the cure, not at the start of it.
How to verify cure before you move a panel
Production panels cannot be destroyed, so the verification is done on control coupons:
- Glaze a coupon with every batch. Same glass, same primer, same sealant lot, same bite, same day. Keep it next to the panels.
- Pull the coupon at the planned handling time. A hand-pull or a peel test should show cohesive failure inside the rubber, not clean glass.
- Snap-test the squeeze-out. A thin ribbon of squeezed-out sealant should snap with elastic recovery; if it smears, the depth is still uncured.
- Record it. Coupon photo, lot number, shop temperature log and the date the panel was released. This is the file you hand to the façade consultant when a joint is questioned.
What to ask your sealant supplier for
- Cure rate at two temperatures — 23 °C and 5 °C. One number is not enough to plan a winter programme.
- A bite-depth design table with the allowable design stress behind it, referenced to ASTM C1401 or the equivalent local standard.
- Primer recommendation by substrate pair — coated glass, anodised aluminium, mill-finish aluminium, stainless steel, and the non-porous plastics that need a different primer or none.
- Movement capability and the test behind it — a printed class is worth nothing without the cycle count and the post-test inspection.
- Retained samples and lot traceability — so that when a question appears in year four, the actual batch can be tested.
- Structural certification to the standard your project is specified against — GB 16776, ASTM C1184, or the applicable European technical assessment.
Quick checklist for the shop floor
| Check | Target |
|---|---|
| Bite on the shop drawing | ≥ 6 mm, and ≥ calculated value |
| Measured bite on first unit of the shift | Within ±1 mm of drawing |
| Bead thickness | ≥ 6 mm, no wedge |
| Primer | Correct grade for substrate, dry, flashed off |
| Shop temperature during cure | Logged, ≥ 5 °C, ideally 15–30 °C |
| Control coupon | Glazed per batch, pulled before release |
| Handling time | (bite ÷ cure rate) × 2, written on the traveller |
The short version
Bite depth is a design number that the shop either delivers or does not. Cure time is a function of that bite and the shop’s real conditions. Neither is on the cartridge label, and both decide whether the joint survives its first decade. Write the handling window on the production traveller, verify it with a coupon, and keep the record — that is the difference between a facade that ages and a facade that is repaired.
If you are specifying structural glazing and want the cure-rate data, the bite-depth table and the primer matrix for your substrate pair, our engineering team can send them with the technical data sheet.
