The end of the economic service life (useful life of multi-pane insulating glass around 30 - 40 years) often manifests itself as "blindness", i.e. the progressive formation of condensation in the space between the panes.
Not so in this case: immediate bursting brought this insulating glass to an abrupt end. And it is not a simple crack running through the pane or a spider's web that indicates a mechanical "impact" on the glass surface. No. In this case, the fracture pattern looks absolutely clear: it is a pressure jump triggered by a strong change in pressure in the insulating glass, a so-called burst fracture.
Ekkehard Wagner, glass expert
Typical for a pressure crack is its starting point in the centre of the pane and the splitting of the fractures in their running direction towards the corners of the pane. This is exactly how the fracture pattern appears in the present case.
See also: Key points about the connection of frame and glass in security windows
Ekkehard Wagner, glass expert
Whether it is an overpressure or underpressure crack can only be determined on closer examination. For this purpose, the centre of the pane with the beginning of the fracture was generously masked to be able to hold the fragments together and then this masked part was cut out of the pane.
The 547 x 1755 mm insulating glass, designated by the manufacturer as thermal insulation glass with the marking "G" for gas-filled, which was required on the spacer at that time according to ÖNORM, consisted of:
· 8 mm float glass,
· 16 mm SZR with inert gas filling,
· 4 mm float glass with LowE coating on position 3.
Ekkehard Wagner, glass expert
The pane was manufactured at 289 m above sea level and installed at 360 m above sea level.
Typically, glass breakage occurs on the thinner inner pane, which can withstand significantly less load than the 8 mm outer pane, which is twice as thick.
Examination and analysis
After receiving the fragments, they were carefully freed from the adhesive foil and the fracture surfaces were examined. Since the beginning of the fracture was exactly in the middle, it was easy to determine the direction of travel on the basis of the fracture surfaces.
Ekkehard Wagner, glass expert
This is where one of the 3 laws of fracture comes into play: "Fractures always divide only in their direction of travel". Since clear "Wallner lines" were recognisable, these fracture surface markers made it possible to clearly determine where compressive and where tensile stress occurred.
Due to strong concavity, tensile stress was visible on the surfaces facing the SDR and compressive stress on the surfaces facing outwards. A further indication of this was the slight run-out of some fractures slightly parallel to the glass surface, instead of the often present shell-shaped spalling.
Ekkehard Wagner, glass expert
Very strong and apparently prolonged bulging (concave) led here to total failure of the thinner glass pane.
What makes this break special?
Interesting was the fact that on the fracture surface at the beginning of the fracture there was a clearly visible outward opening fracture mirror. According to Orr's equation and the assumption that the fracture mirror constant for float glass (soda-lime-silicate glass) can be assumed to be approx. 2.0, this resulted in a fracture stress of approx. 44 N/mm² with a fracture mirror radius of 2.05 mm.
Ekkehard Wagner, glass expert
Due to the differing specifications for the fracture mirror constant in the literature, these specifications are relatively accurate at +- 20%.
This is not a very high breaking stress, but since such a compressive stress is usually a longer-lasting stress, the glass breakage can be explained.
If you take a closer look at the fracture mirror, you can quickly recognise the exact fracture exit by a very small irregularity.
What cannot be explained here, however, is "Why does this pressure jump occur only after 27 years?"
The fact that it is a slow, so-called "subcritical fracture growth" can be clearly denied on the basis of the typical fracture appearance with many splits.
The fact that the filling gas has diffused out over the years can be rejected, as this gas permeation would also diffuse moisture to the outside, as it were, and the pane would have visibly gone blind.
Also interesting: "Window manufacturers also need to think about anisotropies"
The only explanation the authors of the article have is based on the assumption that the desiccant filled in has absorbed a small proportion of the argon over decades and thus the pressure in the SZR has slowly but steadily decreased more and more - a slowly rising, persistent negative pressure forms in the SZR. This could plausibly explain the pressure jump due to concave deformation.
The examination of the desiccant present in the spacer also showed no abnormalities. The granularity, dryness and colour of the molecular sieve material corresponded to the appearance of an intact desiccant that had just been conveyed from the big bag. A more detailed examination with regard to preloading was dispensed with due to lack of relevance.
The expert's conclusion
Research into the causes of glass breakage does not always lead to a clear, plausible result. However, whether it was actually the case, as assumed here, could only be determined with disproportionately high testing effort. A plausible explanation, however, is argon sorption by presumably 4°A or similar desiccant.
The authors
Ekkehard Wagner and Manfred Beham are glass experts and have been active in the glass industry for decades.
