Study of edge strength of load bearing glasses

MATERIALS TECHNOLOGY  ANYAGTECHNOLÓGIA 62. évf. 1. szám  2010/1  építôanyagépítôanyag | 15 Study of edge strength of load bearing glasses KINGA PANKHARDT  Department of Civil Engineering, University of Debrecen  [email protected] GYÖRGY L. BALÁZS  Department of Construction Materials and Engineering Geology, BME  [email protected] Received: 21.10.2009.  Érkezett: 2009.10.21. A single glass layer can be considered as safety glass if tempered or reinforced with wire mesh. When tempered glass fractures it shatters into tiny pieces with blunt edges. Heator chemically strengthened glasses can not be considered as safety glass owing to the fracture pattern (fractures in shards), unless they are laminated. Bending tests were carried out (test arrangement according to standard EN 1288-3:2000) and the bending strength of single layer tempered and float glasses were determined with use of the well-known formulas. The calculated values for surface strength with use of ultimate strain were compared to the bending strength results in the case of float and tempered glasses. The question arise: the strains measured in the centre of the pane and near the edge are equally? Strain measurements of the surface of loaded specimens – at the middle of the pane and near the edge – were also investigated. The failure of glass originates from cracks with microscopic sharp tips. In spite of the careful manufacture and handling of glass panes, impacts by sharp particles or environmental impacts can cause defects on the surface. The glass manufacture e.g. edge finishing techniques influence the glass strength and can cause flaws which can propagate during the lifetime of the glass. Defects of edge surfaces caused by different edge work techniques were shown with scanning electron microscopy. Most of the glass strengthening methods are used to introduce residual compressive stresses into the outer layers by physical or chemical tempering. The compressed resulting layer helps to close cracks initiated on the surface, can stop crack propagation and can also increase the bending strength. Recent studies [6, 14] have shown that surface strengthening can lead to substantial improvements in degradation resistance. Therefore, in outdoor conditions, when the glass surface is exposed to humidity etc., tempered or heat strengthened glass should be used. Keywords: load bearing glass, edgework, strength, glass, tempered glass Kinga PANKHARDT MSc Civil Eng. at BME (scholarships Univ. Karlsruhe, Germany) in 1998. From 1998 to 2001 PhD student at Dept. of Building Materials, BME. In year 2000 at GKSS Research Centre, Geesthacht, Germany, at Dept. of Material Modelling (WMS), research fellow. From 2001 to 2003 Glasmetal Ltd., civil engineer, special glass constructions. From 2003 to 2006 SK Engineering office, civil engineer and lecturer at University of Debrecen. From 2007 associate professor at Department of Civil Engineering, University of Debrecen. Member of the Technical committee of Glass Working Group (MSZT/MB 112) of Hungarian Standardization Institute; Hungarian Group of fib; IABSE; IASS Working Group. Field of interests: load bearing glass constructions, recycling of building materials. Prof. György L. BALÁZS (1958), PhD, Dr.-habil, professor in structural engineering, head of Department of Construction Materials and Engineering Geology at the Budapest University of Technology and Economics. His main fields of activities are experimental and analytical investigations as well as modelling reinforced and prestressed concrete, fibre reinforced concrete (FRC), cracking in concrete, durability and resistance. He is convenor of fib Task Groups on “Serviceability Models” and “fib Seminar”. In addition he is member of several fib Task Groups or Commissions. He is President of the Hungarian Group of fib. Member of fib Presidium, deputy president of fib. 1. Introduction Float glass is widely used in the architecture because of the good optical quality of the surface. Th e fl oat process produces glass sheets with a uniform thickness and perfectly smooth surfaces that need no further polishing. Th e resulting glass will then be further treated in various ways. Soda lime silicate glass is mainly used for architectural purposes. Th e advanced treatment technologies are applied to fl oat glass products depending on the end-products and on the application. When fl oat glass is used as load bearing element [1, 2] safety glass is needed. Single glass layer can be considered as safety glass if tempered or reinforced with wire mesh. When tempered glass fractures it shatters into tiny pieces with blunt edges. Heator chemically strengthened glasses can not be considered as safety glasses owing to the fracture patterns (fractures in shards) unless they are laminated. Bending strength with the well known formulas was evaluated in case of non heat-treated fl oat and tempered glasses and was compared to the measured surface stresses. Th e question arise: if strains measured in the centre of the pane and near to the edges are equal. Strain measurements of glass surface of loaded specimens – at the mid of pane (Region 1) and near to the edge (Region 2) – were investigated (Fig. 1.). Th e bending strength of a single glass pane is infl uenced by the following factors [3, 5]: Fig. 1. Region 1 and Region 2 of glass surface strain measurements [3, 4] 1. ábra Üvegfelület alakváltozásainak mérése R1 és R2 jelű tartományokon [3, 4] a) heat treatment, b) surface condition (e.g. non-slip characteristics), c) rate and duration of loading, d) area of surface stressed in tension, e) relaxation, f) ambient medium, g) age, i.e. time elapsed from the last mechanical surface treatment, h) ambient temperature, i) edgework. Most of glass strengthening methods are used to introduce residual compressive stresses into the outer layers by physical or chemical tempering (Fig. 2). Surface stresses are related to the temperature gradient that results from cooling. Th e produced compressed layer helps to close cracks initiated on the surface (Fig. 3) and can stop crack propagation and can increase also the bending strength. ÉPA 2010_1.indd 15ÉPA 2010_1.indd 15 2010.04.05. 19:30:082010.04.05. 19:30:08 http://dx.doi.org/10.14382/epitoanyag-jsbcm.2010.4 ANYAGTECHNOLÓGIA  MATERIALS TECHNOLOGY 16 | építôanyagépítôanyag  2010/1  62. évf. 1. szám Fig. 2. Stress distributions over the thickness of tempered glass pane. Residual compressive stresses aft er tempering near the surface are: 120 to 150 MPa, tensile stress in the middle plane is about half of them [3] 2. ábra Jellemző feszültségeloszlás az edzett üveg vastagsága mentén mezőközépen és élek közelében (R2 és R1 tartományoknál). A felületi rétegek hőkezelés után maradó nyomófeszültsége általában: σ=120-150 MPa, az üveg belsejében lévő húzófeszültség: σ/2, [3]. a) b) Fig. 3. Vickers diamond pyramid indentation into a tempered glass surface a) section view, b) plane view. Indentation fi eld drives the crack, residual tempering fi eld opposes them [6] 3. ábra Vickers gyémánt piramisos behatás vizsgálat edzett üveg felületén a) függőleges metszet, b) felülnézet. A behatás által létrejövő repedéseket, az edzést követő maradó feszültségek összezárják [6] Th ere is a peak of tensile stresses at a small distance from the edge (typically 12–25 mm from the edge (Fig. 4.). Tensile stresses are indication of an undesirable edge cooling rate and a potential bending. Th erefore, surface strains in Region 1 (at centre of glass panes) and Region 2 (at edge region) were measured in present experiments. Fig. 4. Surface and mid-layer stress near to edges [7] 4. ábra Élekhez közeli tartományok felület közeli és belső rétegeinek feszültségei [7] Failure originates from cracks with atomically sharp tips. In spite of the careful manufacture and transportation of glass panes, impacts with sharp particles or environmental impacts can cause defects on the surface. Research results have shown [8, 9, 10, 11, 12] that aging of newly created fl aws should be benefi cial, and residual stresses will tend to relax in time (especially in reactive environments). Mould et al. (1959) [13] showed that the strength of specimens containing abrasion microcracks can increase with the aging time (i.e., time between abrasion and testing to failure). In case of glasses under such conditions those lifetime design procedures will be diffi cult which try to make predictions or estimations based on actual glass surface conditions. Further studies [6, 14] have shown that surface strengthening can lead to substantial improvements in degradation resistance, therefore, in outdoor conditions, when glass surface is exposed to humidity etc. tempered or heat strengthened glass should be used. 2. Materials and experimental procedure An experimental programme was carried out to analyse the load bearing capacity of single and laminated glass panes. Results of single glasses are discussed in present publication. Specimens were tested in four-point bending. In case of tempered glass, the stress must fi rst exceed the built-in compression stresses before tension develops. Th e infl uence of edge strength also infl uences the strength of glass pane. Th erefore, the deformations at mid of pane and at the edge regions were also studied. Th e results of tempered glass specimens were compared to those of non heat treated fl oat glass specimens. 2.1. Test parameters and test programme Test parameters of single glass specimens were the follo w ing: Constants: test arrangement, width and length of specimens, Variables: thickness, type of fl oat glass (non heat treated or tempered), rate of loading, temperature of specimens (not discussed here). Specimens tested in four-point bending were manufactured from soda-lime silicate fl oat glass with polished edges. Any intended changes to the condition of the test specimens like edge working, was completed at least 24 hours before testing [5]. Specimens were stored in laboratory conditions for min. 1 day before being tested. If the glass surface was modifi ed by abrasion, etching, edge working etc., it was necessary to allow the fresh damage to heal before the test is done. Th e continuous surface modifi cation by moisture aff ects the damage in a way that can reduce any weakening eff ect [8]. Single glass specimens with nominal thickness of 6 mm and 12 mm as well as 19 mm were investigated. Th e accuracy of thickness measurements were 0.01 mm. Th e measured values are the following: density of glass, ■ average ρglass: 2.50 g/cm3 length of specimen, ■ average L: 1099 mm, (1100 mm ± 5 mm) width of specimen, ■ average b: 358 mm, (360 mm ± 5 mm) thickness of specimen, ■ average hnom,6 mm: 5.87 mm, (nominal 6 mm) thickness of specimen, ■ average hnom,12 mm: 11.85 mm, (nominal 12 mm) thickness of specimen, ■ average hnom,19 mm: 18.99 mm, (nominal 19 mm) Th e required pieces of specimens to any combination of parameters were determined according to the standards (at least 3 specimens should be tested). Standard deviation of the tests results was at most 10% of the average of measured values. ÉPA 2010_1.indd 16ÉPA 2010_1.indd 16 2010.04.05. 19:30:082010.04.05. 19:30:08 MATERIALS TECHNOLOGY  ANYAGTECHNOLÓGIA 62. évf. 1. szám  2010/1  építôanyagépítôanyag | 17 2.2. Experimental procedure 2.2.1. Force measurement All glass specimens with constant span of 1000 mm and supported at width of 360 mm were tested in four-point bending. Th e load and defl ection at mid-span of the glass panes were measured in all tests. Th e test procedure was a semidynamic short-term test. Th e tests were carried out at specimen temperature of +23 °C. Th e temperature of the specimens and room temperature was continuously measured during the tests. Th e specimens were mounted as shown in Fig. 5. Fig. 5. Test method in four-point bending (EN 1288-3:2000) where, 1.: specimen: 1100×360 mm, 2.: bending roller, 3.: supporting roller, 4.: rubber strips (3 mm thick, according to ISO 48), 5.: custom-made transducer, symbols: Ls: 1000 mm, Lb: 200 mm, h: thickness of the specimen (6 mm, 12 mm, 19 mm) [3, 15]. 5. ábra Kísérleti elrendezés két vonal menti hajlítás esetén (EN 1288-3:2000), ahol a számok a következőket jelölik, 1.: próbatest: 1100×360 mm, 2.: terhelőhenger, 3.: alátámasztó henger, 4.: gumiszalag (3 mm vastag, ISO 48:1994 szerint), 5.: saját tervezésű erőmérő, Ls: 1000 mm, Lb: 200 mm, h: próbatest vastagsága (6 mm, 12 mm, 19 mm vagy 2×6, 3×6 mm) [3, 15]. Fig. 6. Fracture of single tempered glass 6. ábra Egyrétegű edzett üveg tönkremenetele Fig. 7. Fractured laminated glass consisting of three fl oat glass layers (cracks usually started from the edges) 7. ábra Vizsgált laminált három rétegű fl oat üveg (a repedések általában az élektől indulnak) Rubber strips of 3 mm thickness and hardness of 40 ± 10 IRHD (in accordance with ISO 48) was placed between the specimen and the bending and supporting rollers to avoid hard contacts [16]. Th e bending tests were carried out at (23 ± 5) °C room temperature with relative humidity between 55% and 65%. During the test the temperature was kept constant with ± 1 °C in order to avoid the development of thermal stresses. Load was measured with a self-made force transducer, developed by the authors [17] for Instron Type of 1197 testing instrument and calibrated with Hottinger Baldwin Messtechnik’s 200 kN force transducer (No. 76411). Th e self-made transducer was fi xed to the upper plate of the Instron type 1197 testing equipment. Th e authors developed a steel construction with hinge connections at defi ned places was constructed to transfer the loads from the test equipment Instron to the specimens with four-point bending. Displacements were measured with a Hottinger Baldwin Messtechnik type W50 displacement transducer. Signs of the instruments were transformed with soft ware Catman to the measured values in μm/m. Measured values during the tests were recorded by computer. Th e fracture process and crack pattern of glass specimens were recorded with digital optical methods (CMOS SONY Camera). Th e applied loading rates of glass specimens with thickness of 6 mm were: 50 mm/min, 20 mm/min and 5 mm/min as well as 1 mm/min (by Instron Type 1197 available). Th e applied loading rate of glass specimens with thickness of 12 and 19 mm was: 20 mm/min. Th e specimens were tested until their fracture (Figs. 6., 7.). 2.2.2. Strain measurement Strains at selected points on the surface (in R 1 and R 2 Region) of the glass panes with strain gauges Type HBM LY1110/120 were measured (Fig. 1). For temperature compensation another glass specimen with strain gauges on its surface was applied and stored at the same condition with the tested specimens. Th e change of resistance in mV/V of the gauges was measured and transferred to the digital channels of HBM Spider8 instrument. Th e soft ware Catman aft er calibration was able to transform the measured mV/V data in μm/m. Stresses at glass surface may be calculated with Hooke’s law for linear elastic materials. 2.2.3. Scanning electron microscopic analysis (SEM) To study morphologically the edge regions of single glasses four diff erent types of edges were prepared: a) manually arrised edge, b) machine ground edge, c) machine ground + acid etched edge, d) machine polished edge. Type of used scanning electron microscope was: JEOL JSM-5500LV. Th e edge samples were covered with Au-Pd vapour for electron microscopy. Th e parameters of electron microscopy were the following: high vacuum mode, using of secondary electron (SE) detector, acceleration voltage 25 kV. Digitally photos were taken with magnifi cation of ×50, ×100, ×300, ×1000. In the photos the scaling line is also printed. 3. Results and discussion 3.1. Th e eff ectiveness of tempering In most of the references [18, 19] can be found that the load bearing capacity in case of tempered single glasses is 3 to 4 times higher than in case of fl oat glasses. Th e question arise: is the load bearing capacity of tempered glass always 3 to 4 ÉPA 2010_1.indd 17ÉPA 2010_1.indd 17 2010.04.05. 19:30:092010.04.05. 19:30:09 ANYAGTECHNOLÓGIA  MATERIALS TECHNOLOGY 18 | építôanyagépítôanyag  2010/1  62. évf. 1. szám times higher in case of diff erent glass thicknesses or in case of diff erent applied loading rates? Th e authors suggest to introduce the defi nition eff ectiveness of tempering (heat treatment). Th e eff ectiveness of tempering shows the proportion of load bearing properties (e.g. maximal force) of tempered glasses to non heat treated fl oat glasses with the same thickness. Th e authors have experimentally shown that the eff ectiveness of tempering depends on the glass thickness and the loading rate. Based on the laboratory four-point bending tests the authors have shown that the eff ectiveness of tempering decreases with the increase of glass thickness by loading rate of 20 mm/min. Th e relationship between eff ectiveness of tempering and glass thickness is linear (Fig. 8.). Fig. 8. Eff ectiveness of tempering vs. glass thickness in case of loading rate of 20 mm/min 8. ábra Edzés hatékonysága az üvegvastagság függvényében 20 mm/min terhelési sebesség esetén Th e authors have experimentally shown in case of nominal glass thickness of 6 mm that the eff ectiveness of tempering decreases with decrease of loading rate from 20 mm/min to 1 mm/min (Fig. 9.) and no signifi cant changes with increase of loading rate from 20 mm/min to 50 mm/min. Explanation for that is: the development of cracks starting from surface scratches needs time which is rather available in case of loading rates of 5 or 1 mm/min than in case of loading rates of 50 or 20 mm/min. Fig. 9. Eff ectiveness of tempering vs. rate of loading by 6 mm glass thickness 9. ábra Edzés hatékonysága és a terhelési sebesség összefüggése 6 mm vastagságú üvegeknél With the eff ectiveness of tempering (heat treatment) it is possible to choose the appropriate and economic glass thicknesses in the fi eld of glasses which were heat treated (tempered or heat strengthened). 3.2. Deformations of glass surface at mid of pane (Region 1) and edge region (Region 2) Strains were measured at the bottom (tensioned) surface of specimens. In case of tempered specimens the stress must fi rst exceed the built-in compression stresses before tension develops, therefore, the so called prestressed layers of tempered specimens help to reduce the strains caused by defl ection at same force level. Results of strain measurements indicated that the measured strains are higher in Region 2 than in Region 1, both in case of fl oat non heat treated glasses and tempered glasses with thicknesses of 6, 12, 19 mm (Fig. 10.). Fig. 10. Maximal force vs. strain at bottom surface in Region 1 (mid) and Region 2 (edge) of single tempered and fl oat glass specimens thicknesses of 6, 12 or 19 mm, with loading rate of 20 mm/min 10. ábra Törőerő és törési alakváltozások az R1 és R2 tartományoknál 6, 12 és 19 mm vastagságú fl oat és edzett üvegek esetében, 20 mm/perc terhelési sebességnél Th e ratio of maximal strain and ultimate force is illustrated as a function of glass thickness in Fig. 11. Fig. 11. Ratio of maximal strain and ultimate force at bottom surface in Region 1 (mid) and Region 2 (edge) of single tempered and fl oat glass specimens vs. thicknesses of 6, 12 or 19 mm 11. ábra Törési alakváltozás és törőerő aránya R1 és R2 tartományoknál 6, 12 és 19 mm vastagságú fl oat és edzett üvegek esetében, 20 mm/perc terhelési sebességnél Fig. 11. indicates that the ratio of maximal strain and ultimate force decreases with increase of thickness for both fl oat and tempered specimens calculated at Region 1 and Region 2, respectively. Th is ratio also indicates the eff ectiveness of tempering in case of single glass specimens. To determine bending strength in four-point bending the following formula can be applied: (1) ÉPA 2010_1.indd 18ÉPA 2010_1.indd 18 2010.04.05. 19:30:102010.04.05. 19:30:10 MATERIALS TECHNOLOGY  ANYAGTECHNOLÓGIA 62. évf. 1. szám  2010/1  építôanyagépítôanyag | 19 where, b-width of specimen; h-thickness of specimen; Ls-distance between the centre lines of the supporting rollers; Lb-distance between the centre lines of the bending rollers; y-central defl ection of the specimen; k=ke-dimensionless factor as function of y/h to determine the stress at the mid of span k=1; σbB-bending strength; σbG-bending stress imposed by the self-weight of the specimen. Th e bending stresses should be calculated by applying a factor k to take into account non-uniformity of the stress fi led, (see factor k in Eq. (1)) and the calculated bending stress is called eff ective bending stress. Factor k is used when it is required to determine the bending strength of glass where the eff ects of the edge conditions are important. For calculating the overall bending strength or equivalent bending strength of the surface area, including the edges the value k=1 shall be used. For calculating the bending strength or equivalent bending strength of the free edges of the glass pane k=ke shall be used. Th e appropriate value of ke for use in Eq. (1) shall be obtained from Fig. 12., which gives the value of k=ke as a function of the value of y/h. Fig. 12. Dimensionless factor ke as a function of y/h of 6, 12, 19 mm thickness of glass, continuous lines as given in EN 1288-3:2000, dashed lines are measured values with loading rate of 20 mm/min, rectangles indicate measured values for 6 mm glass with loading rate of 50 mm/min 12. ábra Szorzótényező, ke 6, 12, 19 mm vastagságú egyrétegű üvegekhez a lehajlás/ vastagság (y/h) függvényében 50 és 20 mm/min terhelési sebességeknél. A folytonos vonalak az EN 1288-3:2000 szabványban megadott görbéket jelzik. A szaggatott vonalak a 20 mm/perc terhelési sebességhez javasolt értékeket jelzik Based on the laboratory results the authors have shown that the value of factor ke is infl uenced by the loading rate. Th e authors determined the values to factor ke with applied loading rate of 20 mm/min in case of single glasses thickness of 6 mm, 12 mm and 19 mm. Fig. 12. indicated that in case of reduced testing speed from 50 mm/min to 20 mm/min ke increases. Defects caused by edgework are initiations location of cracks and the nucleation of cracks needs. With increase of loading time more cracks can develop, therefore the stresses in Region 2 increase. Further investigations on the eff ect of glass strength in Region 2 with diff erent type of edge works and testing speed should be done. Stresses at selected points on a glass surface can be determined from strain measurements. Surface stresses have been calculat ed based on Hooke’s law using the theoretical Young’s modulus of glass E=70 000 N/mm2. Th e average of calculated surface strength, σ of individual single glass specimens and the bending strength, σbB calculated with Eq. (1) are indicated in Table 1. Mid pane strength, σbB (in Region 1) and edge strength, σbB,edge (in Region 2) have been calculated with Eq. (1) on the bottom surface of single non heat treated fl oat (F) and tempered (E) glass specimens. Table 1. gives the surface stresses at maximal force which are diff erent from the calculated bending strengths of single glass specimen with thicknesses of 6, 12 and 19 mm, respectively. Th e calculated surface stresses at maximal force are lower in case of single glass specimens with thickness of 6 mm than the bending strength. Th e maximal surface stress should be considered in case of glass panes thinner than 6 mm for both tempered or fl oat glasses [4]. In case of specimens with large deformation (no appropriate shear resistance) the calculated strengths (both surface and bending) are overestimated for thin (6 mm) specimens with use of Eq. (1), see also [4, 20]. In case of single glass specimens thicker than 12 mm or with appropriate shear resistance Eq. (1) can be applied. Th e surface stresses are more infl uenced by the surface condition of glass element than the bending strength. Impacts on glass surface by hard, sharp particles e.g. scratching are the initial locations of cracks, which can develop and quickly propagate. When the surface stresses will reach the surface strength of the pane, fracture occurs. Th e probability of glass failure starting from the edge region is higher than of mid region. Edges of panes or bore holes contain more defects. Specimens h Measured values (avg.) Theoretical surface strength, σCalculated bending strength Fmax, m ymax,m εmax,m Mid (Region 1) Edge (Region 2) σbG σbB (Region 1) y/h ke σbB edge (Region 2) Mid (Region 1) Edge (Region 2) mm kN mm μm/m N/mm2N/mm2N/mm2N/mm2- - N/mm2 E 6 1.58 61.63 1846.14 2063.92 129.20 144.50 3.80 157.50 10.50 1.12 176.40 12 7.94 46.15 2563.46 2921.36 179.40 204.50 1.90 191.40 3.81 1.14 218.20 19 15.52 24.73 2075.88 2274.99 145.30 159.20 1.20 145.40 1.29 1.10 159.90 F 6 0.43 16.20 508.80 590.55 35.62 41.34 3.80 45.60 2.76 1.16 52.90 12 2.66 17.41 1053.20 1157.63 73.70 81.00 1.90 65.40 1.42 1.10 71.90 19 7.21 11.87 1125.55 1188.82 78.80 83.20 1.20 68.20 0.62 1.06 72.30 Table 1. Bending strength and surface strength of tested single glass specimens, where symbols are: E - tempered glass, F - non heat treated glass (average values are avarages of four measurements) 1. táblázat Egyrétegű üvegek számított hajlító szilárdsága és alakváltozásmérések eredményeiből számított felületi szilárdságok (rövidítések: E - edzett üveget, F - fl oat üveget jelölnek ) ÉPA 2010_1.indd 19ÉPA 2010_1.indd 19 2010.04.05. 19:30:102010.04.05. 19:30:10 ANYAGTECHNOLÓGIA  MATERIALS TECHNOLOGY 20 | építôanyagépítôanyag  2010/1  62. évf. 1. szám In case of fl oat glass specimens the edge strength is more infl uenced by thickness and edge condition than in case of tempered specimens. Reaching the ultimate strain in edge region, fracture occurred. Th erefore, the eff ect of the edge quality is important on load bearing capacity and durability of glass. Load bearing capacity of a glass pane with same thickness decreases with decrease of edge strength. Although the tested specimens were manufactured with machine polished edges, further glasses with diff erent edgeworks were investigated for scanning electron-microscopic observation. Fig. 13. indicates that manually arrised edges contain defects in macroscopic scale. Fig. 13. Macroscopic defects as initiators of cracks at “manually arrised” edge of a glass pane 13. ábra Makroszkopikus zámolt élmegmunkálási hibák, mint a repedések kiindulási helyei Fig. 13. indicates that the edge region of glass was the most damaged by manually arrised edgework which can be initiator of cracks. Th e roughness of edge surface decreases with use of fi ner abrasives or with acid etching. Edgework reduces the initiation of cracks caused by cutting of glass pane, see as cut edge in Fig. 14. Fig. 14. Typical edge in “as cut” samples a) surface defects ×50, b) ×300 as potential initiator of cracks (photos taken by B. Koczka, Department of Inorganic and Analytical Chemistry, BME) 14. ábra Jellemző vágott él a) felületi hibák ×50, b) ×300 repedések potenciális kiindulási helyei Th e traditional edge work process requires a steady stream of water and an abrasive compound. Abrasives are available in many diff erent sizes (called grits), ranging from around mesh size 60 (= 250 micron, which is a very rough grit used for initial grinding) to around mesh size 600 (= 30 micron, an extremely fi ne grit). Generally, achieving a highly polished fi nish involves using a series of fi ner and fi ner abrasives (diamond discs or pad, SiC silica carbide slurry, Cerium dioxide etc.). A fi nished ground surface will appear whitish and dull (Figs. 15. b, c), but polished surface will shine with no visible scratches (Fig. 15. d). a) manually arrised edge, left : ×100; right: ×1000 very coarse abrasive size around 420 to 250 micron, (diamond wheel speed in rpm’s 1600-2400) a) zámolt él, bal: ×100; jobb: ×1000 b) machine ground edge, left : ×100; right: ×1000 to remove large rough areas of glass the process begins with abrasive about 105 micron and require further processing down to medium size around 53 to 48 micron, (speed of feed from 0.5 up to 4 m/min) b) gépi csiszolt él, bal: ×100; jobb: ×1000 c) machine ground + acid etched edge, left : ×100; right: ×1000, medium abrasive size around 53 to 48 micron, acid type of hydrogen-fl uoride c) gépi csiszolt + savmaratott él, bal: ×100; jobb: ×1000 d) machine polished edge, left : ×100; right: ×1000 very fi ne Cerium dioxide (CeO2) abrasive, size around 37 to 29 micron d) polírozott él, bal: ×100; jobb: ×1000 Fig. 15. Typical edge fi nishing of a) manually arrised, b) machine ground, c) machine ground+ acid etched d) machine polished samples (photos taken by B. Koczka, Department of Inorganic and Analytical Chemistry, BME) 15. ábra Jellegzetes élmegmunkálások a) zámolt él, b) gépi csizsolt él, c) gépi csizsolt + savmaratott él, d) polírozott él (a fotókat a BME Szervetlen kémia és analítikai tanszékén Koczka B. készítette) Fig. 15. indicates that the manually arrised edge has the roughest surfaces and will contain the most defects. By cutting ÉPA 2010_1.indd 20ÉPA 2010_1.indd 20 2010.04.05. 19:30:102010.04.05. 19:30:10 MATERIALS TECHNOLOGY  ANYAGTECHNOLÓGIA 62. évf. 1. szám  2010/1  építôanyagépítôanyag | 21 process initiated sharp crack tips can be eliminated or the size of initiated cracks can be reduced with use of edge fi nishing techniques. Transportation of glass can also cause fl aws on the surface (Fig. 16.), which can propagate during the lifetime of glass. Th e results [14] have shown that the nature of fl aws is an important factor in fatigue characterization. A fl aw of given size may respond in diff erent ways, depending on whether there are residual stresses present or cracks have been formed. Th is is also important in the context of lifetime design. Surface scratch ×50, caused by steel razor, F=0,4 kN Fig. 16. Typical surface of fl oat glass surface scratch with ”shell-like” fracture ×50, (photo taken by B. Koczka, BME Department of Inorganic and Analytical Chemistry) 16. ábra Üveg felületi karcolás jellegzetes „kagylós törési” képe 50×szeres nagyításban 4. Conclusions 1. Th e conclusions for the eff ectiveness of tempering can be summarized as follows: It is suggested herein to introduce the defi nition eff ectiveness of tempering (heat treatment). Th e eff ectiveness of tempering shows the proportion of load bearing properties (e.g. maximal force) of tempered glasses to non heat treated fl oat glasses with the same thickness. Th e higher the thickness the lower is the eff ectiveness of tempering. Th e tempering is more eff ective in case of single layer glass specimens with thicknesses lower than 12 mm. Th e results indicated that the eff ectiveness of tempering decreases with decrease of loading rate from 20 mm/min to 5 mm/min and 1 mm/min. 2. Th e conclusions for the maximal strain at various regions and the infl uence of edgework on the edge strength can be summarized as follows: Strength of a glass pane should be investigated at least at 2 diff erent regions. Region 1 at mid of pane and Region 2 at edge region. Th e eff ect of edge work should be also studied. Th e maximal strain of edge region (Region 2) is higher than of mid of pane (Region 1). In case of fl oat glass specimens the edge strength is more infl uenced by thickness and edge condition than in case of tempered specimens. Reaching the ultimate strain in edge region fracture occurred, therefore, the eff ect of the edge quality is important on load bearing capacity and durability of glass. Th e edge region of glass contains more defects caused by edgework. When glass fractures, it fails practically at the edges fi rst and the crack propagates in the direction of mid of pane. In case of reduce the testing speed from 50 mm/min to 20 mm/min ke increases. Defects caused by edgework are initiations location of cracks and the nucleation of cracks needs time. With increase of loading time more cracks can develop, therefore the stresses in Region 2 increase. 3. Th e conclusions for the relationship of bending strength and surface strength can be summarized as follows: Th e surface strength (calculated by Hooke’s law) is more infl uenced by the surface condition of glass element than the bending strength (calculated from the maximal moment). Surface strength results earlier failure than bending strength, especially in the case of thin fl oat (h < 10 mm) glass panes. In case of fl oat non heat treated and relatively thick (h > 10 mm) specimens the strength is considerably infl uenced by the size eff ect. 4. Th e conclusions for diff erent types of edgework can be summarized as follows: By cutting process initiated sharp crack tips can be eliminated or the size of initiated cracks can be reduced with use of edge fi nishing techniques. Th e size (roughness) of the edge defects decreases with use of fi ner abrasives in the edge fi nishing process. Th e roughness of the edge surface decreases with use of acid etching. 5. Acknowledgement Th e authors would like to express their thanks to RÁKOSY GLASS Ltd. especially to Eszter Rákosy for providing the specimens. Th e authors would like to thank also for the support by SEM measurement to Dr. Béla Koczka, Faculty of Chemical Technology and Biotechnology, Department of Inorganic and Analytical Chemistry, BME and Dr. Katalin Kopecskó Dept. of Construction Materials and Engineering Geology, BME. Th e authors would like to thank to Dr. Salem G. Nehme for his intellectual support. Th e authors would also like to thank to Dr. Sándor Fehérvári, Mr. András Eipl, Mr. Mátyás Varga, Mr. Dávid Diriczi, Mr. Gábor Kovács for their technical support. References [1] Pankhardt, K.: Development of load bearing glass structures (Az üveg tartórendszerek fejlődése). Alaprajz. 7 évf. 5. szám (2000), pp. 14–18. [2] Pankhardt, K.: Special load bearing glass structures in Hungarian building industry (Különleges üveg tartószerkezetek a magyar építőiparban). Építőanyag. 55. évf. 3. szám (2003), pp. 106-111. [3] Pankhardt, K.: Investigation on load bearing capacity of glass panes. Pe - riodica Polytechnica - Civil Engineering, Vol.52, No.2 (2008), pp. 73-82., doi:10.3311/pp.ci.2008-2.03 [4] Pankhardt, K.: Investigation on load bearing safety glass. Eligehausen R., Gehlen C. (Ed.), Proceedings of 7th International PhD Symposium in Civil Engineering, Stuttgart (2008), pp. 53-62. ÉPA 2010_1.indd 21ÉPA 2010_1.indd 21 2010.04.05. 19:30:112010.04.05. 19:30:11 ANYAGTECHNOLÓGIA  MATERIALS TECHNOLOGY 22 | építôanyagépítôanyag  2010/1  62. évf. 1. szám [5] EN 1288-1:2000: Glass in Building – Determination of the bending strength of glass – Part 1: Fundamentals of testing glass. CEN Brussels (2000), pp. 7-8. [6] Marshall, D. B. – Lawn B.R.: An indentation technique for measuring stresses in tempered glass surfaces. Journal of the American Ceramic Society, Vol. 60. No. 1-2. (1977), pp. 86-87. [7] Redner, A. S.: Automated measurement of edge stress in automotive glass, Strainoptic Technologies Inc., Glass Proceeding days (2000), Poster 24. [8] Wiederhorn, S. M.: Infl uence of water vapor on crack propagation in soda-lime glass. Journal of the American Ceramic Society, Vol.50 (1967), pp. 407-414. [9] Michalske, T.A.: Th e stress corrosion limit: Its measurement and implications. Fracture Mechanics of Ceramics, Vol. 5, Surface Flaws, Statistics and Microcracking, (Ed.) Bradt, R.C., Evans, A.G., Hasselman, D.P.H., and Lange, F.F., Plenum Press, New York (1977), pp. 277–289. [10] Ito, S. — Tomozawa, M.: Crack blunting of high silica glass. Journal of the American Ceramic Society, Vol.65, No.8 (1982), pp. 368-371. [11] Marshall, D.B. — Lawn, B.R.: Surface fl aws in glass. Kurkjian, C.R. (Ed.), Strength of inorganic glass, Plenum Publishing Corporation, New York (1985), pp. 171-180. [12] Wiederhorn, S.M. — Dretzke, A. — Rödel, J.: Crack growth in soda–lime– silicate glass near the static fatigue limit, Journal of the American Ceramic Society (2002), Vol.85, No.9, pp. 2287–2292. [13] Mould R. E. — Southwick, R. D.: Strength and static fatigue of abraded glass under controlled ambient conditions: II . J. Amer. Ceram. Soc., Vol. 42., No.12. (1959), pp. 582-592. [14] Lawn, B.R. — Marshall, D.B. — Dabbs, T.P.: Fatigue strength of glass: a controlled fl aw study. Kurkjian, C.R. (Ed.), Strength of inorganic glass, Plenum Publishing Corporation, New York (1985), pp. 249-259. [15] EN 1288-3:2000: Glass in Building – Determination of the bending strength of glass – Part 3: Test with specimen supported at two points (four-point bending), CEN Brussels (2000), pp. 8-11. [16] ISO 48:1994: Rubber, vulcanized or thermoplastic — Determination of hardness (hardness between 10 IRHD and 100 IRHD)., TC45/SC2 [17] Pankhardt, K.: Load-bearing glass structures. Periodica Polytechnica, Civil Engineering. Vol. 48. No.1-2. (2004), pp. 157-172. [18] „Glass terminology” B&G glass, http://www.bgglass.com/glass%20terms. htm Retrieved in 2009. [19] „Th ermal stress, Part two”. Achitect’s guide to glass & metal (2009), http:// www.glassguides.com/index.php/archives/610# [20] Pankhardt, K. — Balázs, L. Gy.: New opportunities of structural glazing, loadbearing glass structures. Proceedings (Report) Responding to tomorrow’s challenges in structural engineering. IABSE Symposium Budapest 2006; Vol. 92, No. 34-35 (2006), the whole paper on CD with 11 pages. Teherhordó üvegek éleinek szilárdsága Az egyrétegű üvegeket akkor hívhatjuk biztonsági üvegeknek, ha edzettek (hőkezelési eljárással) vagy dróthálóval erősítettek. Ha az edzett üveg eltörik, apró tompa szilánkokra esik szét. A hőkezelt vagy kémiailag edzett üvegek törésképeik miatt nem tartoznak a biztonsági üvegek közé, csak akkor, ha laminált formában alkalmazzák őket. Hajlító vizsgálatokat végeztünk (EN 1288-3:2000 szabvány szerinti elrendezéssel), és meghatároztuk az egyrétegű edzett és float üvegek hajlító szilárdságát az ismert összefüggésekkel. Továbbá a számított értékeket összehasonlítottuk az üveg felületén mért törési alakváltozásokból számított szilárdsági értékekkel nem hőkezelt float valamint edzett egyrétegű üvegeknél. Felmerült a kérdés: vajon az alakváltozások egyenlők lesznek-e az üvegtábla síkjának középpontjában vagy a tábla élekhez közeli helyen mérve? Alakváltozási méréseket végeztünk az üvegtábla közepén és a tábla élhez közeli tartományokban. Az üveg tönkremenetele a mikroszkopikusan kicsiny repedéscsúcsokból indul ki. Az üvegtáblák felületén az elővigyázatos gyártásuk és szállításuk ellenére az éles tárgyakkal való érintkezésükkor és a környezeti hatások következtében karcolások keletkezhetnek. Az üvegfeldolgozás (pl. vágás, élmegmunkálás) által számos további olyan behatás éri az üveget, melyek repedéseket eredményezhetnek. Pásztázó elektronmikroszkópos vizsgálatokkal kimutattuk különböző élmegmunkálási eljárásokkal kialakított élek felületének hibáit. A legtöbb üveg erősítési módszer célja, hogy maradó nyomófeszültségeket vezessen az üvegtábla felületéhez közeli rétegeibe fizikai vagy kémiai úton. A keletkezett, nyomott réteg segíti a felületi repedések (karcolások) összezárását, ezáltal gátolja a repedések továbbterjedését, és növeli a hajlítószilárdságot. Tanulmányok [6, 14] kimutatták, hogy az üvegfelület erősítése jelentősen javítja a degradációval szembeni ellenállását, így különösen, teherhordó üvegek kültéri alkalmazásánál, edzett vagy hőkezelt üvegek alkalmazása javasolt. Kulcsszavak: teherhordó üveg, élmegmunkálás, szilárdság, üveg, edzett üveg A belés külföldi vállalatok érdeklődése a BAU kiállítási területei iránt eddig soha nem látott méreteket öltött. A 2011. január 17. és 22. között az új Müncheni Vásárváros területén megrendezésre kerülő BAU 2011 területe gyakorlatilag teljes egészében elkelt. „Az építészet, anyagok és rendszerek világkiállítása“ alkalmából, újból több mint 40 országot képviselő mintegy 1900 kiállító mutatkozik majd be. A vásárvezetés 210 ezernél is több látogatót vár, mintegy 150 országból. Vezető témák  Fenntartható építés: a fenntartható épületeknek gazdaságilag hatékonynak, környezetbarátnak és erőforrás-takarékosnak kell lenniük, használóiknak kényelmet, kellemes közérzetet és egészséget kell biztosítaniuk, ezenkívül pedig illeszkedniük kell szocio-kulturális környezetükbe is.  Akadálymentes építés: épületek, lakások és környezetük oly módon megvalósuló építése, kialakítása és berendezése értendő alatta, hogy a valamilyen fogyatékossággal vagy anélkül élők, nagyok és kicsik, fiatalok és idősek lehetőleg önállóan, biztonságosan és kényelmesen élhessenek abban.  Épületek felújítása, renoválása és modernizálása az építőipar sok vállalkozója számára időközben a forgalom nagy részét teszi ki. Csak Németországban már a teljes lakásépítési beruházások 60%-át ez teszi ki. Vásári fórumok – BAU plus! A kiállítók bemutatói mellett a BAU 2011 ismét exkluzív szakmai kísérő programot kínál. A vásári események kellős közepén elhelyezett fórumok szabadon látogathatók.  „Az építés jövője“ fórumon a világ minden tájáról érkező neves tervezők, építészek és építőmérnökök fogják projektpéldák segítségével bemutatni, hogyan fog kinézni a jövőben az építés és kialakítás.  A „MakroÉpítészet“ az építészek jelenlegi és jövőbeni szerepét világítja meg. Délelőttönként építészek és vállalati képviselők közös projekteket mutatnak majd be. Délutánonként pedig egy „Trendpanel“ prezentálja az interdiszciplináris alapkutatás eredményeit. Ennek során arról lesz szó, hogy az olyan mega-trendek, mint a fenntarthatóság vagy a digitalizáció milyen hatással lesznek a jövőben az építészetre, a termékekre és a folyamatokra.  Időközben számos belés külföldi vállalkozás esetében az üzlet nagy részét az épületek modernizálása és karbantartása teszi ki. Éppen ezért ebben a témában is önálló fórumot szerveznek. További információ: www.messe-muenchen.de. TELE CSARNOKOK – CSÚCSON A BAU ÉPA 2010_1.indd 22ÉPA 2010_1.indd 22 2010.04.05. 19:30:112010.04.05. 19:30:11