Tension–shear interaction domain for EVA-laminated countersunk point-fixing devices

  • Marco Cervio
  • Giovanni Muciaccia
  • Gianpaolo Rosati
Research Paper
  • 23 Downloads

Abstract

The present paper aims is to characterize the mechanical behavior of a new embedded point-fixing device for laminated glass for combined tension–shear load cases. This kind of support is based on the principle known as “interlayer junction” in which the anchor is bonded to the interlayer and embedded in the glass panes during the lamination process. The purposes of this innovative point-fixing device are nowadays mainly focused on interior architecture applications: e.g. balustrades, stairs, doors, elevator cars and wells, design furniture, etc., taking advantage of the beautiful aesthetic of an “all-glass” element. This is due to the fact that, contrary to traditional point-fixing bolted connections, the countersunk hole is not passing through the outer glass layer, so that the stainless steel anchor looks like embedded in the glass mass. A comprehensive experimental investigation on the load-bearing capacity under the combined action of tension and shear forces is performed by means of specifically designed test apparatus and procedure. The failure modes in the different direction of load applications are identified and discussed. Finally, in addition to the interaction domain given for combined load cases between tension and shear forces, a simplified analytical interaction expression is also proposed, intended to be used in the practical design applications.

Keywords

Point-fixing devices Laminated embedded fittings Interlayer junctions Interaction domains Glass connections design 

1 Introduction

Contemporary architecture is eager for transparency. As a matter of fact, curtain wall and structural glazing façades represent a standard solution for tall buildings all around the world, independently on the specific climatic conditions or on the cultural background of the site. The possibility to convey as much light as possible into a building is a constant objective of architecture since middle ages, as this was the main scope of gothic cathedrals. Unfortunately, the strength and ductility of construction materials were not sufficient to allow to satisfying at once the requirement of spanning big distances and catching light from the exterior. In most recent ages the two functions were uncoupled: the façade does not constitute anymore the main load-bearing structure; the introduction of steel for opposing tensile forces brought to light frame-resisting structures, while façades basically started to independently evolve from a technological point of view.

In the seventies architects introduced glass panels for continuous façades (Willis Faber and Dumas Headquarters, by Foster and Partners), while in the eighties point-fixing devices for glass panels were effectively engineered, developing the so called “rotules” system (Les Sevres at La Villette, by RFR) (Vyzantiadou and Avdelas 2004). Its key principle, from a mechanical point of view, consists in a passing through fastener which is clamped on both surfaces and transfers both tensile and shear forces to the underlaying structure. Technology successively developed with the main scope of improving air and water proofing of the system, while from a structural point of view the efforts were focused on the reduction of stress intensification at the edge of the hole in the panel which, being constituted by a brittle element at the panel scale, is crucial (Bernard and Daudeville 2009; Herrmann 2005). However, the main working principle remained basically unchanged.

Only in the last decade new systems were introduced, based either on the bonding capability of adhesive systems (Dispersyn et al. 2014; Hagl et al. 2012) or, exploiting innovative technologies for cutting the glass (Siebert and Pauli 2005), or on the use of non-passing through fasteners which develop an undercut mechanism in a single glass layer (Fischer FZP-G) (Unterweger et al. 2007). More recently, innovative solutions for fixing laminated glass panels have been proposed on the bases of the so called “interlayer junction” principle in which a metallic insert is bonded to the interlayer during the lamination process (Royer-Carfagni and Silvestri 2009; Puller and Sobek 2012; Santarsiero et al. 2016b). Finally, attempts of combining both the bond and the undercut mechanisms in a single fastening systems were developed and investigated (Cruz et al. 2014).

The present paper investigates the experimental behavior under tensile, shear or combined action of an innovative system of non-passing through point-fixing metal device embedded in a laminated glass panel during the autoclaving process. The paper is structured as it follows. Section 2 gives details of the point-fixing device investigated in the present research. Section 3 illustrates the test plan and describes the related experimental apparatus and test procedure. Section 4 reports the results of the experimental investigation and their statistical evaluation with special emphasis to design and verification purposes. Section 5 contains some conclusions and outlines ongoing developments that will place the present work in a larger context.

2 Materials

The aim of this section is to give an overview of the point-fixing device investigated in this work. The following paragraphs are to characterize its parts and the assembly.

This kind of connection is based on the principle known as “interlayer junction” in which the anchor is bonded to the interlayer and embedded in the glass panes during the lamination process. In fact, contrary to traditional point-fixing bolted connections, the countersunk hole is not passing through the outer glass layer and the stainless steel insert looks like embedded in the glass mass. The above-mentioned product is suitable to join structures composed by laminated glass panels to other structures either made of glass or other materials. Although the present study focuses only on a single typology of specimens described more in detail hereinafter, generally the number, the thicknesses and the type of the glass layers may vary according to the required load-bearing capacity, impact performance and post-breakage behavior.

The specimens are made up of the following parts:
  • metal anchor

  • glass plies

  • polymeric interlayer film

Their geometry and dimensions are depicted in Fig. 1. The non symmetric geometry of the specimen represents the worst case scenario of an anchor located at the minimum manufacturing distance, which is 50 mm next to a corner with a loaded edge. Beside the main scope of this work this is also intended to obtain some preliminary low boundary results to be used in design. In fact, the PFDs are generally located at the corner and at the sides of the supported glass panels, well spaced apart from each other and the the panel edges.
Fig. 1

Specimen dimensions and detail of the stainless steel insert: 1. Anchor, 2. Upper glass layer, 3. Interlayer, 4. Lower glass layer

Fig. 2

Check of the surface compressive stress level by means of photoelastic method

2.1 Anchor

The anchor is made of AISI 316 stainless steel (material number 1.4401 in accordance with EN 10088-3:2005). The insert has the shape of a conical frustum obtained by turning. The frustum has a nominal height of 12 mm and a base and top diameter of 50 and 26 mm, respectively, so that the lateral surface is inclined to an angle of \(45^{\circ }\) with the base. On the top surface there is a M12 threaded hole, which accommodates a stud bolt used to join the point-fixing device and so the laminated glass panel to other structures. In the end, surfaces are deburred and polished.

2.2 Glass

The investigated specimens consist of two 6 mm tempered glass sheets. The lower one, which is usually facing the external side of the glazing, is continuous and it measures \(180 \,\hbox {mm}\times 180 \,\hbox {mm}\). The upper one, which is on the contrary usually facing the internal side of the glazing has a countersunk hole and it measures \(150\, \hbox {mm}\times 150\, \hbox {mm}\). The larger surface of the continuous layer is used to fix the specimen to the test rig. The hole is drilled before the heat treatment using diamond tools and it is shaped and sized to accommodate the insert leaving a small gap between the two. Despite the drilling technique, as it will be detailed in the interlayer paragraph, the key point is the small tolerances of this feature. In order to prevent side type of failures all the edges are polished. Additionally, the level of the surface compressive stress induced by tempering was measured on a sampling of 5 specimens by means of photoelastic method using a “Grazing Angle Surface Polarimeter” (GASP1), see Fig. 2. The measures detected almost constant level of tempering in the batch and the average surface compressive stress was 102 MPa. However, considering a single specimen the surface compressive stress are non homogeneous. In particular, it is worth to mention the loss of prestress in the neighborhood of the hole and along the edges of the glass layers due to the different geometry and so the different thermal regime in the cooling phase of the tempering process (see Watson et al. 2013).

2.3 Interlayer

The interlayer is made of a thermoset polymer film 1.52 mm thick made up of four single sheets each about 0.38 mm thick. The raw film before lamination consists of ethylene vinyl acetate copolymer (EVA) and other additives, among them the most important is the so called “Cross-linking agent”, usually an organic peroxide. The interlayer material not only represents an essential point improving strength, stiffness and safety of the glass panel but in this case it plays a fundamental role in the connection. In fact, when the laminate is heated inside a vacuum bag or an autoclave the material starts to melt and it flows into the gaps located around the insert. Successively, the laminate is brought to the cross-linking temperature, which is higher than the melting point, and cured for a proper amount of time. In this way, the chemical structure changes and the material develops new stable mechanical and physical properties suitable for practical building applications. At this point the anchor is bonded to the glass. This can be determined both by visual inspection and by test applying an average torque couple of 110 Nm without the anchor starts rotating inside the hole. Additionally, the thin layer made of elastomeric material prevents any steel-to-glass direct contact during the normal service life. Finally, it is worth to mention that EVA is not the most common material among different interlayer polymers for laminated glass in buildings. EVA films are mainly used as encapsulation material for photovoltaic panels lamination. The use of EVA for special purposes such as embedded joints has at least two main reasons. First, current experience (see Weller et al. 2005) shows that the presence of vinyl acetate provides greater adhesive resistance compared to other interlayer materials. Second, the manufacturing of laminated glass by means of EVA interlayers it is less sensitive to the interlayer storage conditions (e.g. effect of moisture).

3 Experimental method

The main scope of the present work is to characterize the mechanical behavior of this new embedded point-fixing device, hereinafter PFD, for combined tension–shear load cases. Numerical methods have been already extensively calibrated and validated for well-known PFDs and simple load cases (see Herrmann 2005; Maniatis 2006). German national standard (DIN 18008-3 2013) also provides some modeling criteria. In addition, there are few commercial finite element codes implementing specific glass fixing features taking advantage of the mentioned current state of the art. However, finite element analyses lack of reliability when they are applied to such innovative products without a proper input parameters identification and discussion of results. Hence, the necessity to resort to experimental investigation and “Design assisted by testing” procedures (EN 1990 2005).

3.1 Test programme

A total of seven different angles of the applied action were accounted. The set of experimented angles is as follows \(0^{\circ }\), \(15^{\circ }\), \(30^{\circ }\), \(45^{\circ }\), \(60^{\circ }\), \(75^{\circ }\) and \(90^{\circ }\). Positive angles have a shear component pointing toward the loaded edge of the glass plates. For each direction five tests were carried out, excluding only \(30^{\circ }\) and \(60^{\circ }\) angles, for which three tests only were carried out. All the tests were load controlled and the load rate selected in order to reach the failure in less than 120 s. The tests have been performed at normal room temperature of \(23\pm 2{}^{\circ }\hbox {C}\) and 55% relative humidity. All the specimen were left exposed to the room environment for at least 24 h before testing. The mechanical behavior of laminated glass is significantly affected by the service condition, that is the temperature and the duration of load. This is because of the polymeric interlayer material. Therefore, the same is expected to apply to embedded PFDs based on the interlayer junction principle (see Santarsiero et al. 2016a, c, 2017). Consequently, the design and the adoption of such type of joints in situations that deviate from the above-mentioned testing conditions necessarily requires further investigations. Nonetheless, also durability aspects (e.g. effect of humidity and solar radiation) shall be taken into account. Additionally, as previously mentioned in Sect. 2 the level of surface compressive stresses induced by the tempering was checked on a sampling of 5 specimens by means of photoelastic method. On the same batch, fragmentation tests in accordance with EN 12150-1 (2015) method have also been carried out as detailed in Sect. 4. Knowing the actual level of the surface prestress is of great interest for the analytical and/or numerical prediction of the connection’s resistance with respect to the failure mechanism involving the upper glass layer with the countersunk hole (see Watson et al. 2013). In view of such theoretical development, the characterization of the upper glass panel by means of coaxial double ring tests may represent a way for validating the calculations.

3.2 Test apparatus

Existing test protocols for PFDs only partially fit the scope of the present investigation (Maniatis 2006). In fact, they usually take into account tension or shear load cases separately. However, it is very common in the current practice to have combined in-plane and out-of-plane load, as for the case of point-supported vertical glass panes subjected to self-weight and wind actions, respectively. Recent development in the study of innovative glass fixtures has started investigating the combined behavior under different actions (eg. combined tensile–shear, tensile–torsion) (Unterweger et al. 2007; Dispersyn et al. 2015), while for traditional point-fixing devices (DIN 18008-3 2013) may be applied. In this context, a specific testing apparatus consisting in a steel reaction frame which allows an hydraulic jack to be inclined to different angles with respect to the PFD axis was developed and it is shown in Fig. 3. The apparatus is designed such that the direction of the applied action passes through the centroid of the anchor’s bottom surface by means of a specific load transfer device screwed to the stud bolt of the same PFD. The specimen is clamped to a support plate only on the lower glass layer, which is larger than the upper one and presents an appropriate clamping surface. This allows a smooth assignment of the boundary conditions and it limits stress concentrations in the upper layer, which is weaker because of the presence of the hole. This vertical restraints act only on the two sides perpendicular to the loaded edge. Additionally, for load cases with shear component there is also a lateral restraint acting on the loaded edge. Because of the specimen’s shape, only the bottom glass layer is directly loaded in shear. The direct contact between the specimen and the steel restraints is prevented by interposing lead stripes. This is to avoid undesired possible indentation failures. A load cell of full scale capacity 25 kN was interposed between the reaction frame and the load transfer device. The relative displacement of the anchor with respect to the reaction frame was monitored by two LVDTs with a maximum stroke of 20 mm placed both in the plane of the glass sample (horizontal, \({\mathrm {H}}_{1}\)) and transversally (vertical, \({\mathrm {V}}\)). In addition, in order to detect possible shear slip, the side of the lower glass layer opposite to the loaded edge (\({\mathrm {H}}_{2}\)) was also monitored by an LVDT with a maximum stroke of 10 mm. A detail of displacement transducers arrangement is shown in Fig. 4. All data were acquired with a HBM Spider 8 data acquisition system.
Fig. 3

Overview of the testing machine

Fig. 4

Detail of the displacement transducers arrangement

Table 1

Test results

\(\theta \)

\(R_{1}\)

\(R_{2}\)

\(R_{3}\)

\(R_{4}\)

\(R_{5}\)

\(R_{{\mathrm {m}},\theta }\)

(\(^{\circ }\))

(kN)

(kN)

(kN)

(kN)

(kN)

(kN)

0

8.03

7.18

7.90

6.70

7.70

7.50

15

8.02

9.04

7.80

8.45

9.91

8.44

30

9.03

8.42

8.19

8.55

45

9.41

9.31

8.92

9.66

9.57

9.37

60

11.24

11.68

10.00

10.97

75

10.38

9.58

11.79

10.17

9.76

10.34

90

8.32

9.84

7.48

8.34

9.41

8.68

4 Test results and discussion

Test results are synthetically reported in Table 1, where \(\theta \) is the angle of the applied action with respect to the anchor’s top surface outward normal and \(R_{i}\) are the recorded failure loads, while \(R_{{\mathrm {m}},\theta }\) are the mean values associated to each test series. Positive angles have a shear component pointing toward the loaded edge of the glass plates. It is worth to remember that the present experimental investigation has the aim to directly establish the ultimate load-bearing capacity at different angles of the applied action for the considered PFD. This quantitative data should be considered valid only for the product specifications described in Sect. 2 and for the investigated type of load and environmental condition. However, the resulting interaction domain is of more general interest.

The specimens have been visually inspected both before and after the tests, while during the tests the upper glass layer was monitored by means of a digital camera. Before the tests, the specimens were found free of macroscopic flaws and/or alterations. During the tests, a few interlayer “bubbles” have been detected in the anchor’s neighborhood at a load level approaching the failure. This phenomenon is commonly regarded as an evidence of glass layers delamination. The failure mode observed in all the tested specimens is the breakage of the upper glass layer into small fragments. The crack patterns show both the peculiarity of indentation and splitting failures (that is the maximum principal stress arranged in the circumferential direction). Figure 5 shows a typical specimen after failure, while Fig. 6 depicts a detail of the anchor’s bottom surface showing start of the loss of adhesion. The fragmentation pattern is in accordance with thermally toughened safety glass specifications (see EN 12150-1 2015) as it was expected on the basis of compressive surface stress measurements. Although several deviation from the standard test method are reported, the compliancy was checked performing fragmentation tests on some broken specimens. An example of marked fragmentation pattern is also depicted in Fig. 5. However, fragments hold together and the anchor shows a minimum post-breakage restrain capacity, this is also because the countersunk shape of the hole.

Figure 7 shows a comparison among five typical load versus displacement curves for different angles \(\theta \) of the applied action, while Fig. 8 depicts the load versus displacement curves of a complete test series at an angle \(\theta \) of \(45^{\circ }\). A perfectly elastic response is detected for angles close to \(0^{\circ }\) (pure tension), which are typically related to sudden failure of the upper glass layer. The introduction of a shear component increases the stiffness and allows for a small non-linear phase prior to failure, which probably indicates that debonding at the steel-to-interlayer interface is being activated. The study of this non-linear behavior and of possible inelastic displacements need further investigation. In this context a test procedure that accounts for several loading and unloading cycles at different load levels may provide the necessary information (e.g. DIN 18008-3 2013).
Fig. 5

Typical type of failure with marked fragmentation pattern (number of fragments \(=\) 264)

Fig. 6

Detail of the anchor’s bottom surface for a typical type of failure

Fig. 7

Typical load versus displacement curves for different angles of the applied action

Fig. 8

Load versus displacement curves for \(\theta = 45^{\circ }\)

The evaluation of the test results follows the “Design assisted by testing” approach. In this context, the assessment of the 5th percentile characteristic values \(R_{{{\mathrm {k}}},\theta }\) has been carried out accounting for a Gaussian Normal Distribution, even though typical statistical method for the assessment of the strength of glass elements is generally based on the Weibull approach. This is because the present work deals with the behavior of the connection, which is a complex system made of different materials and with several competitive resistant mechanisms. It is worth to mention that also other probability density functions may be used (e.g. Log-normal distribution), however considering the relatively low scattering in the results the shape of the tension–shear interaction domain would not be significantly affected. Table 2 presents the statistical analysis of the data, where \(\sigma \), \(\delta \) and \({\mathrm {k}}_{n}\) are the standard deviation, the coefficient of variation and the characteristic fractile factor, respectively. The derivation of the anchor’s load-bearing characteristic values \(R_{{\mathrm {k}},\theta }\) from the experimental mean values \(R_{{\mathrm {m}},\theta }\) is performed by means of the quantity \((1-{\mathrm {k}}_n\delta )\) computed for each different test series. This quantity takes into account the scatter of the test data and the statistical uncertainty associated with the number of tests. Regarding these two aspects it can be seen that the scattering of the results is lower than the 11%, while the relatively limited number of investigated samples has a significant impact. In fact, the characteristic fractile factor \({\mathrm {k}}_{n}\) is equal to 2.46 for test series with 5 samples and 3.15 for test series with 3 samples only (ISO 12491 1997). This is assuming a 75% confidence level and separately evaluating each test series so that the coefficient of variation \(\delta \) for the considered angle is supposed to be known on the basis of “prior knowledge” coming from other tests in comparable situations.
Table 2

Statistical analysis

\(\theta \)

\(R_{{\mathrm {m}},\theta }\)

\(\sigma \)

\(\delta \)

\(1-{{\mathrm {k}}}_{n}\delta \)

\(R_{{{\mathrm {k}}},\theta }\)

\(\overline{R}_{{\mathrm {k}},\theta }\)

(\(^{\circ }\))

(kN)

(kN)

(%)

(−)

(kN)

(kN)

0

7.50

0.55

7.39

0.82

6.14

5.50

15

8.44

0.54

6.40

0.84

7.11

6.19

30

8.55

0.44

5.11

0.84

7.17

6.27

45

9.37

0.29

3.06

0.92

8.67

6.87

60

10.97

0.87

7.96

0.75

8.22

8.04

75

10.34

0.87

8.45

0.79

8.19

7.58

90

8.68

0.94

10.85

0.73

6.36

6.36

Finally, the results are plotted on a polar chart (see Fig. 9), which has the virtue of providing a direct representation of the tension–shear interaction domain as it was experimentally investigated. In fact, for a given point belonging to this plane, the distance from the pole is the measured resisting force while the angle corresponds to the inclination of the applied action according to the previously mentioned reference system. The present approach keeps constant the probability of failure at each level of interaction, on the contrary it has the main drawback that the shape of the domain changes when considering either mean or characteristic values. For design convenience, an homotetic interaction domain is proposed to overcome this issue, where the characteristic resistance \(\overline{R}_{{\mathrm {k}},\theta }\) is calculated accounting for the lowest of all the \((1-{\mathrm {k}}_n\delta )\) values, which is 0.73. However, the resistance verification of the connection is usually put in terms of tension, \(N_{{\mathrm {E}}}\) (axial) and shear, \(V_{{\mathrm {E}}}\) (radial) forces acting on the anchor, thus the results can conveniently be expressed in the Cartesian plane \(N_{{\mathrm {R}}} - V_{{\mathrm {R}}}\) by a simple change of the coordinate reference system (see Fig. 10).
Fig. 9

Polar interaction domain

Additionally, although the present graphical descriptions provide a prompt overview of the interaction domain they are not very effective for practical use. Hence, the following analytical design expression by means of “Partial factor method” is proposed:
$$\begin{aligned} \biggl (\frac{N_{\mathrm {Ed}}}{N_{\mathrm {Rd}}}\biggr )^{2} + {\mathrm {k}}_s \biggl (\frac{V_{\mathrm {Ed}}}{V_{\mathrm {Rd}}}\biggr )^{2} \le 1 \end{aligned}$$
(1)
where, \({\mathrm {k}}_{s} = 0.75\) is experimentally derived and the design resistances \(N_{{\mathrm {Rd}}}\), \(V_{{\mathrm {Rd}}}\) are computed assuming a material partial safety factor \(\gamma _{{\mathrm {M}}} = 1.25\) as recommended in Haldimann et al. (2008) for adhesive joints. The adopted partial safety factor applies to the specific testing conditions and the considered end-use. In particular, the manufacturer has an appropriate factory production control and the anchor is considered subjected to short-term static loads under normal temperature. The region of the allowable design values analytically determined has an elliptical shape fitting the experimental domain with the following constrains: it returns exact values for pure tension or shear and it always represents a lower bound with respect to the experimentally determined values for all the combined tension–shear load cases.
Fig. 10

Cartesian interaction domain

5 Conclusions

Extreme transparency drives innovation in building components. Innovative PFDs as non passing-through type are embedded and hidden in the glass panel with the aim of preserving the beautiful “all-glass” aesthetic and improving the glazing performances. The new PFD presented in this work takes advantage of most recent advances in interlayer material knowledge combined with a fine tuned manufacturing process. In this context, EVA rubber-like behavior avoids the necessity of placing an additional bushing material to prevent the steel-to-glass direct contact. On the contrary, the small tolerances required for the countersunk hole may be pointed out as a drawback with respect to traditional systems. In spite of the above-mentioned enhancements there are several aspects that have still not been investigated thoroughly. Among these, it is worth to mention the effect of service conditions in terms of temperature, load duration and durability, as well as the cyclic behavior. Although, interaction laws are usually less sensitive to such deviations, the extrapolation of the result presented in this paper to situations different from the reported testing conditions requires further investigations.

The main scope of the present paper is thus to characterize the mechanical behavior of this new embedded PFD for combined tension–shear load cases. Resistant mechanism for such fastener proved to be very complex due to its mixed nature: partly shared with adhesive joints and partly common to mechanical undercut anchors as the load level approaches the failure. The adhesive character of the connection plays an important role smoothing the stress peaks that usually occur in the neighborhood of point-fixing joints.

Although finite element analysis are nowadays extensively applied to the design of glass structures and their joints, designer must be aware that it is not easy to predict the load-bearing capacity without a proper calibration and validation of the models. Therefore, in view of practical design applications in buildings a typical “Design assisted by testing” methodology of investigation was adopted and illustrated in detail in the paper. In doing so it is worth to remember that also non structural parts and products that have independent structural function must be designed and installed in compliancy to the safety levels and performances prescribed by the adopted building code. Hence, the necessity of providing adequate “tools” to the designers.

A comprehensive experimental investigation on the load-bearing capacity under the combined action of tension and shear forces has thus been performed. The analysis of the results clarified the previously mentioned hypothesis about the complex resistant mechanism. Moreover, although quantitative data should be considered product dependent the resulting interaction domain is of more general interest and increases the general knowledge on this topic. Finally, the proposed design criteria bridged a gap between technology progress and engineering practice.

In the end, a further step to allow a proper use of such point-fixing devices in structural application will consist in assessing the interaction between the local stress regime induced by the same PFD in the glass panel and the stresses function of the flexural behavior of the panel, which is currently under investigation.

Footnotes

  1. 1.

    GASP is a registered trademark of Strainoptics Inc.

Notes

Acknowledgements

The authors would like to thank Vetreria F.lli Paci, Seregno (Italy) for the manufacturing of the samples. The cooperation in the tests execution of M. Cucchi and M. Antico from Testing Laboratory for Materials, Buildings and Civil Structures of Politecnico di Milano is also gratefully acknowledged.

Compliance with ethical standards

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

References

  1. Bernard, F., Daudeville, L.: Point fixings in annealed and tempered glass structures: modeling and optimization of bolted connections. Eng. Struct. 31, 946–955 (2009)CrossRefGoogle Scholar
  2. Cruz, P.J.S., Carvalho, P.L., Silva, E., Rocha, P.: Embedded point fittings in laminated and double insulating units. In: Louter, C., Bos, F., Belis, J., Lebet, J.-P. (eds.) Proceedings of Challenging Glass 4 and COST Action TU0905 Final Conference. CRC Press, London (2014)Google Scholar
  3. DIN 18008-3: Glass in Building—Design and Construction Rules—Part 3: Point fixed glazing. Beuth Verlag GmbH, Berlin (2013)Google Scholar
  4. Dispersyn, J., Martens, K., Van Lancker, B., Belis, J.: Experimental investigation of multi-axial loading on adhesive point-fixings. In: Proceedings of Glass Performance Days 2015, pp. 270–274. Tampere (2015)Google Scholar
  5. Dispersyn, J., Santarsiero, M., Belis, J., Louter, C.: A preliminary study of the nonlinearity of adhesive point-fixings in structural glass facades. J. Facade Des. Eng. 2(1–2), 85–107 (2014)Google Scholar
  6. EN 12150-1: Glass in Buildings—Thermally Toughened Soda Lime Silicate Safety Glass—Definition and Description. CEN, Brussels (2015)Google Scholar
  7. EN 1990: Eurocode—Basis of Structural Design. CEN, Brussels (2005)Google Scholar
  8. Hagl, A., Dieterich, O., Wolf, A.T., Sitte, S.: Tensile loading of silicone point supports—revisited. In: Bos, F., Louter, C., Nijsse, R., Veer, F. (eds.) Proceedings of Challenging Glass 3. IOS Press, Amsterdam (2012)Google Scholar
  9. Haldimann, M., Luible, A., Overend, M.: Structural use of glass. IABSE (2008)Google Scholar
  10. Herrmann, T.: Design rules for glass panes fixed by embedded point fittings. In: Proceedings of Glass Processing Days 2005. Tampere (2005)Google Scholar
  11. ISO 12491: Statistical Methods for Quality Control of Building Materials and Components. Geneva (1997)Google Scholar
  12. Maniatis, I.: Numerical and experimental investigation on the stress distribution of bolted connection under in-plane load. Doctoral Dissertation, Technische Universität München (2006)Google Scholar
  13. Puller, K., Sobek, W.: Load-carrying behaviour of metal inserts embedded in laminated glass. In: Bos, F., Louter, C., Nijsse, R., Veer, F. (eds.) Proceedings of Challenging Glass 3. IOS Press, Amsterdam (2012)Google Scholar
  14. Royer-Carfagni, G., Silvestri, M.: Fail-safe point fixing of structural glass. New advances. Eng. Struct. 31, 1661–1676 (2009)CrossRefGoogle Scholar
  15. Santarsiero, M., Louter, C., Nussbaumer, A.: Laminated connections for structural glass applications under shear loading at different temperatures and strain rates. Constr. Build. Mater. 128, 214–237 (2016a)CrossRefGoogle Scholar
  16. Santarsiero, M., Louter, C., Nussbaumer, A.: Laminated connections for structural glass components: a full-scale experimental study. Glas. Struct. Eng. 2, 79–101 (2016b)CrossRefGoogle Scholar
  17. Santarsiero, M., Louter, C., Nussbaumer, A.: The mechanical behaviour of SentryGlas ionomer and TSSA silicon bulk materials at different temperatures and strain rates under uniaxial tensile stress state. Glas. Struct. Eng. 1, 395–415 (2016c)CrossRefGoogle Scholar
  18. Santarsiero, M., Louter, C., Nussbaumer, A.: Laminated connections for structural glass applications under tensile loading at different temperatures and strain rates. Int. J. Adhes. Adhes. (2017).  https://doi.org/10.1016/j.ijadhadh.2017.09.002
  19. Siebert, G., Pauli, W.R.: New generation of point fixings—architecture and safety aspects. In: Proceedings of Glass Processing Days 2005. Tampere (2005)Google Scholar
  20. Unterweger, R., Beyer, J., Wörner, D., Hilber, R., Bergmeister, K.: A test and design concept for the Fischer FZP-G undercut point fitting for glass panels. In: Proceedings of the 2nd International Symposium on Connections between Steel and Concrete, University of Stuttgart, vol. 2, pp. 827–836. Stuttgart (2007)Google Scholar
  21. Vyzantiadou, M., Avdelas, A.: Point fixed glazing systems: technological and morphological aspects. J. Constr. Steel Res. 60, 1227–1240 (2004)CrossRefGoogle Scholar
  22. Watson, J., Nielsen, J., Overend, M.: A critical flaw size approach for predicting the strength of bolted glass connections. Eng. Struct. 57, 87–99 (2013)CrossRefGoogle Scholar
  23. Weller, B., Wünsch, J., Härth, K.: Experimental study on different interlayer materials for laminated glass. In: Proceedings of Glass Processing Days 2005. Tampere (2005)Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of Civil and Environmental EngineeringPolitecnico di MilanoMilanItaly

Personalised recommendations