Structural Sealang Glazing Parameter Studies with a Focus on the Structural Sealant Bond

Nevertheless, the thickness of the glass units is typically related to stress levels experienced by the glass panes and thus is independent of the corner loading of the adhesive. Consequently, the glass thickness has to be considered as a bound­ary condition for bond sizing. Design parameters that are directly linked to the adhesive are the bite (or width) of the bond geometry and the height (thickness). Starting with the variation of the height of the adhesive, three different bond configurations were analyzed featuring heights of 6.7 mm, 9 mm, and 20 mm. Thus the investigated configurations with respect to the bond geometry limits of ETAG 002 ask for height-to-width ratios between 1:3 and 1:1 for the selected width of 20 mm. Typically, the bond height is determined by thermal loads in order to limit related thermal shear strains to acceptable values.

Figure 20 shows stress distributions for the investigated configurations based on a monolithic glass pane of 6 mm thickness. According to the results presented in this figure, the greater the bond height, the lower the corner peak stress values. This behavior can be explained by the relationship of bond flexibil­ity and bond height (glue thickness), as the joint has a lower modulus when the aspect ratio is closer to 1:1 than to a 1:3 ratio of bond height to bite. An absolute displacement due to glass deformation results in less stress in the joint with greater bond height (glue thickness). If the bond is more flexible, the glass bend­ing due to the twisting moments in the corner region has a lower impact, as the bond deformations are generally higher, and thus the uneven deformation pat­tern of the glass panes in the corners is of less importance. Please note that the effective tensile or compressive stiffness of the bond geometry is not exactly lin­ear with respect to the reciprocal value of the bond height as expected by a sim­ple 1D stiffness formula. The suppression of lateral contraction of the almost

Copyright by ASTM Int’l (all rights reserved); Tue May 6 12:07:08 EDT 2014

Downloaded/printed by

Rochester Institute Of Technology pursuant to License Agreement. No further reproductions authorized.

FIG. 20—Stress distribution for varying adhesive height; surface load of 4.48 x 10~3 MPa (glass unit: 1.5 m in length, 1.25 m in width, 10 mm in thickness; bonding: width — 20 mm, varying thickness).

incompressible adhesive leads to inhomogeneous deformation patterns inside the bond, thus resulting in non-reciprocal characteristics for increasing height.

The second major parameter for the bond geometry is the bond width or bite. Please note that this parameter is determined based on the wind (and other surface) loads. Thus, in order to compare different bites, the load case has to be adapted in accordance with the ETAG 002 sizing rule. Two different configura­tions were selected: in addition to the baseline bond geometry of 9 mm x 20 mm, the bite has been reduced from 20 mm to 10 mm, which leads to a similar reduc­tion of the surface loadp according to the sizing rule (see Eq 6). The reduction of the bite leads to a more flexible bond. Figure 21 presents a reduction of the peak corner loading for this configuration. This result is in line with the statements made before for the case of varying height. Similarly, the relationship between bond width and tensile/compressive bond stiffness is non-linear in bite due to in­homogeneous loading of the adhesive.

FIG. 21—Stress distribution for varying adhesive width; surface load of 2.24 x 10~3 MPa for a 10 mm bond width and 4.48 x 10~3 MPa for a 20 mm bond width (glass unit: 1.5 m in length, 1.25 m in width, 10 mm in thickness; bonding: varying width, 9 mm in thickness).

Copyright by ASTM Int’l (all rights reserved); Tue May 6 12:07:08 EDT 2014 Downloaded/printed by

Rochester Institute Of Technology pursuant to License Agreement. No further reproductions authorized.