The research covered in the following report discusses the problems and limitations in the design of glass fins for structural glazing systems. The report covers and details the two models of analysis used in the design of the same system, to compare both the effectiveness and accuracy of each of the models. In doing so, the current procedure for analysis, as outlined in the relevant Australian Standard, can be appraised with respect to its conservativeness and precision for the design of the glass fins. The effectiveness of a second model of analysis, that being a “Finite Element Analysis”, is also covered and its suitability for use as a design tool is assessed.
For the design of the glass fins in question, a specific supporting situation was chosen, the results of which would be comparable between the two models of analysis. For this research, a “propped cantilever” configuration was selected, which was assumed fully fixed to the glass wall it was supporting along the full length of its windward edge. Any assumptions that were made in the analysis process were standardised between the two models, along with any calculations that needed to be performed. Discussion is also given as to the suitability of the finite element analysis model for this design situation, with respect to the assumptions that were made and the results that were obtained.
The analysis between the two models was able to yield data that was consistent, with any trend shown in one model reciprocated by the other. As the finite element analysis model was able to give an accurate prediction of the properties of the fin and the forces involved at the moment of buckling failure, the analysis of the difference between this model and the code of practice (Australian standard AS1288:1994), was able to provide an effective “factor of safety” between the two designs. In this respect, it was observed that the code of practice is indeed conservative in the buckling design for the particular fin and its connectivity analysed herein. Apropos the purpose of the research, an average factor of safety of 2.3 existed between the finite element analysis model and the un-factored AS1288:1994 provisions. This means that the results from the finite element model could be divided by 2.3 to produce safe bending moments in the glass fin that would avoid buckling and also, that the code of practice could be safely used with an un-factored critical elastic buckling moment. However, these results are only relevant for the particular “propped cantilever” fin covered by this research.
From the analysis, another feature was also apparent for the design of the glass fins. Whereas some of the fins analysed failed by buckling before the material strength became a concern, others suggested that a failure of the fins due to the material’s low strength in tension would pre-empt the buckling failure. This means that both the limit states of strength and stability need to be considered in glass fin design. Further analysis of the data showed that some of the provisions suggested by AS1288:1994 were pushing the design towards a failure in the strength limit state only, thus providing a much simpler way of designing the glass fins. For the finite element analysis model though, this means that the results would need to be analysed for buckling failure and then tensile stress failure to provide a safe design. This would require more work, but would be able to provide the designer with a very accurate model of analysis that could aid in producing a far more economical solution for the client.
Throughout this report, it has been proven that both models of analysis discussed are able to provide a suitable method for the design of glass fins. The investigation of the finite element analysis model in particular was able to prove that both accurate results are obtainable, coupled with the desired versatility required of an alternate design method. Even though the research showed that more than one limit state governs the design of glass fins, with both buckling failure as well as tensile strength failure requiring consideration, a satisfactory and safe design can be produced by either method. To be able to provide a unified design procedure though, further analysis is required of glass fins and structural glass systems that are both connected in different ways and contain tapering cross sections, to reflect the trends in current practice. However, this report is able to produce valid and valuable research into the use of a finite element analysis model as an alternative design method for glass fins, that is able to satisfy the major constraints of accuracy and versatility and thus economy.