Simplified Treatment of Bond Loading: Case Study ETAG 002
SSG typically involves complex mechanical characteristics of its components and the interplay among them. One of the key issues for performance assessment consists in the adequate treatment of the mechanical behavior of the adhesive material, which typically belongs to the class of elastomers. Elastomers
differ significantly from metal alloys and glass in terms of their behavior at room temperature; special topics to be addressed are incompressibility and large strain, the Mullins effect, visco-elasticity, and creep. It is obvious that design rules are needed to simplify the material properties of the adhesive for practical applications. Thus, ETAG 002 is based on an assumed load distribution that does not require detailed knowledge of the adhesive properties. However, the price to be paid for this simplification is significant restriction of the applicability of ETAG 002, for instance, in terms of bond geometries.
Figure 9 and Fig. 10 show the underlying geometric and load assumptions of ETAG 002 with the bond cross-section being of a rectangular shape. Please note that in this assumption the corner loads are vanishing, which is contradictory to thin plate theory as presented in the preceding section. The bond load assumption of Fig. 10 directly leads to a stress equation that can be exploited for sizing. The maximum stress value along the edge, rcenter, is related to the
FIG. 9—Four-sided bonded geometry as assumed for ETAG 002.
Copyright by ASTM Int’l (all rights reserved); Tue May 6 12:07:08 EDT 2014 Downloaded/printed by
length of the smaller side a, the constant surface load p, and the bond bite hc by Eq 5.
Complementary to the design rules, design stress limits are defined for approved silicone adhesives in related documents, called European Technical Approval (ETA) documents. Table 1 shows two representative and widely used two – component silicone adhesives approved by the European Organisation for Technical Approvals for structural sealant glazing in the frame of ETAG 002.
Using the design stresses in Table 1, the minimum width hc of the bonding geometry can be calculated by the re-arranged formula shown in Eq 6.
Given the limited numbers of input parameters for this quite simple design rule, it is obvious that this formula has to neglect a variety of phenomena. The most important approximations are related to the following issues (see also a previous publication by one of the authors of this work ):
• The non-linear distribution of loading along the edges as supposed by plate theory (including sign changes in the corner area) is neglected (see the preceding section).
• Stress variations in the bond width direction that are evoked by edge bending moments caused by the deformation of the attached glass panes are not taken into account.
TABLE 1—European Technical Approvals for representative two-component silicone adhesives.
• The impact of facade design element stiffness/flexibilities, such as of the supporting frame and spacers, on the bonding interface with the frame or in the insulating glass seal is neglected.
• The geometry and bending properties of the glass panes leading to deformations of the bonding interface with the glass are not taken into account.
• The effective mechanical properties of the adhesive in addition to the limit stress values for a more accurate determination of load paths in view of hyperstatic analysis are not exploited for sizing.
A sketch illustrating the complexity of SSG behavior is shown in Fig. 11.