Stress Distributions in the Adhesive within the Corner Zone

Because the stress levels shown in Figs. 15-21 are averaged across the bond ge­ometry for the purpose of the visualization of basic principles, these values do not directly represent the peak loadings within the adhesive, as the stress distri­bution of the adhesive, especially in the corner zone, is of a more complex 3D nature. Figure 22 gives an impression of the inhomogeneous characteristics of

FIG. 22—Normal comer stress distribution for laminated glass with degraded interlayer (left: top view; right: middle plane); surface load of 4.48 x 10~3 MPa (glass unit: 1.5 min length, 1.25 m in width, 2×6 mm in thickness; bonding: width — 20 mm, thickness — 9 mm).

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.

stress in the corner zone for the case of a bond geometry of 9 mm x 20 mm for the investigated laminated glass with a degraded interlayer (no shear stresses). The “hot spot" in the corner zone, which can be understood as a continuous rep­resentation of the concentrated corner force, is clearly evident for the normal stress distribution presented here.

Whereas the averaged corner stress level of the configuration discussed amounts to 0.26 MPa, the local stresses rise up to 0.35 MPa and higher, accord­ing to the FEA shown in Fig. 22. This finding of significantly higher local stresses is underlined by data in Table 2, presenting the differences with respect to peak stresses for selected configurations for a bonding geometry of 6.7 mm x 20 mm and glass panes 1.5 m in length and 1.25 m in width.

The ETAG 002 and the related ETA documents refer to a design stress of 0.14 MPa based on tensile test data of H-type specimens of the adhesive by applying a safety factor of six. However, it is well known that the H-type speci­mens favor stress concentrations along the edges and in the corners due to the interfacing of the adhesive with significantly stiffer materials such as steel, alu­minum, and glass, as has been discussed previously by one of the authors of this work [7]. Thus, the “real" performance of the adhesive material for other geometries is not measured by experiments on H-type specimens, as the failure load is triggered by local stress concentrations. However, testing with H-type specimens results in an “averaging" equivalent to using engineering stress and strain definitions, thus relating the local failure to global parameters.