The design of metal building roof and wall systems under uplift and suction wind loading is complicated because the laterally unbraced purlin and girt's free flange is compressed, and the cross-section rotates due to the shear flow. The objective of this thesis is to introduce a Direct Strength Method (DSM) prediction approach for simple span purlins and girts with one flange through-fastened under uplift or suction loading. This prediction method is also applicable for the case when rigid board insulation is placed between the metal panel and through-fastened flange. The prediction method is validated with a database of 62 simple span tests. To evaluate the prediction for the case when rigid board is used, 50 full-scale tests with rigid board insulation are conducted by the author of this thesis. In the experimental study panel failure, connection failure and member (purlin and girt) failure are observed, and they all limit the system's capacity. Another important contribution of this thesis is that it builds the foundation for future study of a general, mechanics-based limit state design approach for metal building roof and wall systems that can accommodate uplift and gravity loads, simple and continuous spans, and through-fastened and standing seam roofs. / Ph. D.

Z-sections are widely used as secondary members in metal building roof systems. Lateral restraints are required to maintain the stability of a Z-section roof system and provide resistance to the lateral forces generated by the slope of the roof and the effects due to the rotation of the principal axes of the Z-section relative to the plane of the roof sheathing. The behavior of Z-sections in roof systems is complex as they act in conjunction with the roof sheathing as a system and as a light gage cold formed member, is subject to local cross section deformations. The goal of this research program was to provide a means of predicting lateral restraint forces in Z-section supported roof systems. The research program began with laboratory tests to measure lateral restraint forces in single and multiple span sloped roof systems. A description of the test apparatus and procedure as well as the results of the 40 tests performed is provided in Appendix II. To better understand the need for lateral restraints and to provide a means of testing different variables of the roof system, two types of finite element models were developed and are discussed in detail in appended Paper I. The first finite element model is simplified model that uses frame stiffness elements to represent the purlin and sheathing. This model has been used extensively by previous researchers and modifications were made to improve correlation with test results. The second model is more rigorous and uses shell finite elements to represent the Z-section and sheathing. The shell finite element model was used to develop a calculation procedure referred to as the Component Stiffness Method for predicting the lateral restraint forces in Z-section roof systems. The method uses flexural and torsional mechanics to describe the behavior of the Z-section subject to uniform gravity loads. The forces generated by the system of Z-sections are resisted by the "components" of the system: the lateral restraints, the sheathing and Z-section-to-rafter connection. The mechanics of purlin behavior providing the basis for this method are discussed in appended Paper II. The development of the method and the application of the method to supports restraints and interior restraints are provided in appended papers III, IV and V. / Ph. D.